c s~s: S&-2 : C <* /// REPORT OF THE CLIMATE DATA MANAGEMENT WORKSHOP SPONSORED BY: THE NATIONAL CLIMATE PROGRAM OFFICE AND THE ENVIRONMENTAL DATA AND INFORMATION SERVICE NATIONAL OCEANIC AND ATMOSPHERIC ADMINISTRATION ?<°>+ C \ % 8 S >*T £ S & + s CLIFFSIDE MOTOR INN HARPERS FERRY, WEST VIRGINIA MAY 8-11, 1979 ^MOSgg- r toEm of CLIMATE DATA MANAGEMENT WORKSHOP May 8-11, 1979 CI iff side Motor Inn Harpers Ferry, West Virginia United States Department of Commerce Juanita M. Kreps , Secretary National Oceanic and Atmospheric Administration Richard A. Frank, Administrator National Climate Program Office Edward S. Epstein, Director Environmental Data and Information Service Thomas D. Potter, Acting Director b: Digitized by the Internet Archive in 2012 with funding from LYRASIS Members and Sloan Foundation http://www.archive.org/details/reportofclimatedOOclim EXECUTIVE SUMMARY The National Climate Program Act (PL 95-367), as the enabling legislation, sets out the basic policy for this Program. Among its findings pursuant to the Act, Congress declared that: "Information regarding climate is not being fully disseminated or used, and Federal efforts having given insufficient attention to assessing and applying this information." The Congress further found that: "The United States lacks a well-defined and coordi- nated program in climate-related research, monitoring, assessment of effects, and information utilization". Thus, the Act mandates that, among other elements, the Climate Program contain: • "Global data collection and monitoring and analysis activities to provide reliable, useful and readily available information on a continuing basis. • "Systems for the management and active dissemination of climatological data, information and assessments, including mechanisms for consultation with current and potential users. • "Measures for international cooperation in climate research, monitoring, analysis, and data dissemination." In the spirit of the Act and in the interest of developing an effective, comprehensive program to serve the Nation's needs for climate information, a Workshop was held to bring together a broad range of climate data specialists — both data managers and data users. The goals of the Workshop were: 1. To further the development of an integrated inventory of climate data. 2. To make preliminary assessments of the status of the various types of climate data in relation to requirements. 3. To outline steps/efforts to enhance accessibility and utility of climate data. 4. To recommend steps to be taken after the Workshop to further progress toward a national inventory. in The primary effort of the Workshop took place in Working Groups organized as follows: 1. Basic Atmospheric Data Co-Chairmen: Robert J. Fox Gordon A. McKay 2. Hydrology, Precipitation, Snow and Ice Chairman: Eugene Rasmusson 3 . Ocean Data Chairman: Kent Hughes 4. Radiation, Physics, and Chemistry Co-Chairmen: Albert Arking Otto Thiele 5. Proxy and Non-Instrumental Data Resources Co-Chairmen: Alan Hecht George Smith 6. Geographical, Land Use, and Assessments Data Chairman: James R. Anderson 7. Climate Inventory Chairman: Kenneth Hadeen The Reports of the Working Groups form the core of this Workshop Report . The following summarizes certain of the recommendations which were common to several Working Groups. Analyses, justification, and further amplification as well as additional specific recom- mendations are in the Working Group Reports. This summary pre- sents the recommendations in relation to the goals of the Workshop SUMMARY RECOMMENDATIONS I . Regarding an Inventory of Climate Data Existing data must be identified, described, and its presence and utility made known to the user community. — Data set identification must be continued and expanded to include State and private sources; — A permanent mechanism must be established for coordination, evaluation, publication, and update of the inventory. Specifically, it is recommended that NAWDEX be expanded to an international inventory and exchange system for streamflow and groundwater information; and that WATSTORE be expanded into an archive for such information. iv This recommendation, along with the earlier plan (Ref. 1) to incorporate all appropriate data files into NOAA ' s Environmental Data Index (ENDEX) sets a clear framework for progress toward a complementary inventory system. The Report of Working Group 7 lays out further guidelines toward an inventory. II. Regarding Assessments of the Status of Data in Relation to Requirements These assessments are analytical in nature and peculiar to each data regime. Please refer to the Working Group Reports for the assessments. III. A. Regarding Steps and Efforts to Enhance Accessibility of Climate Data 1. Noting the continued growth in user requests and the need for timely response, agencies and repositories should prepare readily available sets of basic data which are in high demand based on experience with users' request. Data set preparation should include improvements in quality, ease of access, and completeness of the data files for both existing data and those to be archived in the future. 2. In view of the scarcity of Data Management resources (manpower and budget) and the potentially large demand upon them, an objective, program-wide, coordinated system must be developed to prioritize the compilation of data sets. Net social benefit, feasibility, costs, and utility of the data set should be factors that determine priorities between data management development and services. For instance, passive microwave data from the NIMBUS satellite series has great promise of yielding a global picture of precipitation — especially over the oceans. However, systematic efforts to compile, reduce, and publish the full set of data should await further results on the accuracy, interpre- tation, and reliability of the data. In view of the great potential value of the data set, verification tests and analyses should proceed immediately. 3. Operational programs (e.g., weather and river flow forecast) are sources of vitally important data for Climate Program purposes. It is recommended that: a. Operational data products (both input data and the results of analyses) be reviewed as soon as possible to ensure that all data valuable for climate purposes are saved. v b. Agencies take steps to document changes in operational data analysis procedures as such changes have great impacts on the analyses of the data record for climate purposes. c. Enough input data be saved to permit new analyses, using uniform methods, in order to resolve problems that are important enough to justify the expense of re-analysis. These requirements should be made known to operational data acquisition and analysis elements in a systematic and well coordi- nated way. For example, the Climate Analysis Center in NOAA could be the appropriate interface for communicating Climate Program interests to operational elements such as the National Meteoro- logical Center and the Office of Hydrology. Additional inter- faces need to be identified for other major data analysis activi- ties such as in DOD, USDA, USGS, and so forth. B. Regarding Steps to Increase the Usefulness and Use of Climate Data 1. Data must be acquired and records maintained in the most complete and comprehensive manner practicable. Improvements in this area will require United States efforts at the local, national, and international levels. a. The maintenance of a valid climatic record must include accurate station histories particularly regarding changes in and calibration of instru- mentation and changes in the physical surroundings of the measurement site. b. As data are collected by a variety of agencies and institutions, there is a clear need to develop and observe uniform standards for station inspection and calibration. c. The United States must actively cooperate and promote quality, conformity and completeness in the acquisition of climatic information on a global basis. This will include active support of programs leading to receipt of quality conven- tional, climatological and upper air observations not presently being obtained on a scheduled basis by the United States. Vigorous support must also be given to WMO efforts to enhance data reported and its timely receipt such as by new codes and message content standards. VI 2. Because of the high cost of retrofit and the possible loss to the user community of valuable information, the Data Management objectives and requirements of the National Climate Program must be incorporated into the planning and budgeting for new observation systems and the modification of existing systems. a. Because new observation systems such as the National Oceanic Satellite System (NOSS) and System 85 will be important sources of climate data, effec- tive mechanisms must be developed to acquire the data. b. Economically important user programs may be greatly disadvantaged by inadvertent changes in the data acquisition and data management system. Due consideration must be given to such impacts which may be counter to National Climate Program objectives . Specifically, there should be a thorough study to define objective requirements for climatic measurement networks as a firm basis for main- taining and supporting them. Such a study should include data requirements for assessments, appli- cations and research; measurements should include both basic atmospheric and hydrologic data needs. 3. The information and service requirements of current and potential users must be clearly defined as a basis for further development of responsive data management and dissemi- nation programs. a. The user community (Federal, State, academic, private) must be defined in terms of their needs for information and the most appropriate means of access. A first step should be the systematic consolidation and evaluation of the results of the several previous conferences to identify those needs and means of access. b. There should also be joint efforts, with users, to exploit historical data resident in the archives but not presently reduced to climatic parameters or accessible through the system. VII 4. As part of their information services program, agencies and other data repositories should increase their efforts to inform user communities of the availability, means of access to, and ways of applying climate data and information. This requires: a. Instruction and information on the optimal use of the system in the form of handbooks, handouts, and active user assistance programs, and b. Promotion, such as by articles in trade journals, user workshops, and speakers at technical meetings of potential user communities. IV. Regarding Steps to be Taken After the Workshop to Further Progress Toward an Inventory and Other Goals of the Workshop The preceding recommendations have implications for all suppliers and users of data management services considered with- in the National Climate Program. 1 . The National Climate Program Office The Act establishes the National Climate Program Office (NCPO) as the lead entity for administering the Program. This was done in the context of formulating a well-defined, coordinated National Program from the several Federal and State activities which have been underway. Thus, the NCPO can and should serve as a catalyst in bringing together appropriate parties and stimulating action to move the Program along. The Office is the logical central agent to act as an advocate for the recom- mendations presented here and see that they are respected in reports, plans, and budgets relative to the Climate Act and Program. Specifically, the NCPO should: a. Formalize the Workshop's climate data inventory project. A first step should be the final prepara- tion and publication of the data inventory gathered in preparation for the Workshop. b. Advocate, promote, and where possible support activities -- including studies -- directed toward improved data acquisition, archival, and dissemina- tion and user service programs. Internationally, this may entail working through the State Department and lead agencies which normally interface with international bodies (e.g., NOAA with WMO) . vm c. Take the initiative to establish coordination of interagency, intergovernmental, and public/private activities as required to implement these recommenda- tions. 2. Other Federal Agencies/Departments Implementation resides mainly with operating agencies. The NCP Act requires that each Federal officer, employee, depart- ment, and agency involved in the Program shall cooperate with the Secretary of Commerce in carrying out its provisions. The cooperating agencies have, therefore, the responsibility to develop appropriate data management systems, having an account- ability extending from acquisition to delivery for data within their jurisdiction. All the preceding recommendations have rele- vance to these agencies. Furthermore, interagency action is essential in the development of the most appropriate user oriented data base and information delivery systems. Specifically, the cooperating agencies and departments should: a. Identify, describe, and make known their holdings of climatic and climate related data/information; b. Clearly define the needs of their current and potential user community; c. Develop methods for prioritizing services on the basis of net benefit to the Nation, feasiblity, and resources required. d. Improve the quality and content of their data holdings and enhance their utility by incorporating station history and geographic data; e. Publicize their holdings and their utility and provide active user assistance; f . Cooperate with WMO and other international agencies for the acquisition of complete, timely, and quality data on a global basis; g. Incorporate data management considerations in major new data systems or alterations thereto. IX 3 . Intergovernmental Since State and inter-State climate programs are an integral component of the Act, they too must be developed consistent with the NCP data management requirements. Contracts, grants, and other support must be negotiated with due recognition for data management objectives. State data management systems should be operated as components of a hierarchal national data management system. 4 . Private Sector, Universities, Insitutions Most of the beneficiaries of the NCP Act are within this group. Tailoring of the appropriate data management system requires extensive user involvement in the definition of needs, system planning, and redesign. Furthermore, members of this group operate special data systems which should be included within the National Program. Mechanisms must be developed for their close involvement, possibly through professional and technical organi- zations that speak on their behalf. x PREFACE In some ways, this Workshop was "an idea whose time had come." First, the need for improved access to climate data had been noted by the Climate Research Board of the National Academy of Sciences. Their review (NAS*1979) of agency planning for the National Climate Program cited the need for "a unified though not necessarily centralized data archive." This need was noted in the context of growing user demand for comprehensive and complementary sets of data, portions of which may be housed among several insti- tutional structures. Second, leading climatologists, such as Professor Helmut Landsberg, had been calling for a comprehensive inventory of climate data holdings — especially in the context of upgrading the quality and coverage of the data. Third, the U.S. Climate Program Plan (NOAA 1977) called for "a significant initiative to plan a major expansion of activities to locate, inventory, and index climate data." The plan specifically called for incorporation of all appropriate U.S. data files into NOAA's ENDEX system and the interfacing of ENDEX with international data indexes such as UNEP/IRS and IODE/MEDI. Subsequently, the NOAA Climate Program (NOAA 1977) called for implementation of these aspects of the National Plan. Then finally, the National Climate Program Act (Public Law 95-367, 1978) established a program element for "global data collection, and monitoring and analysis activities to provide reliable, useful, and readily available information on a continuing basis." Thus, in September of 1978, just after the signing of the Climate Program Act, the National Climate Program Office (NCPO) and NOAA's Environmental Data and Information Service jointly called together representatives of agencies participating in the National Climate Program, to plan a Data Management Workshop. The goals established for the Workshop were: 1. To further development of an integrated inventory of climate data. 2. To make preliminary assessments of the status of the various types of climate data in relation to require- ments . 3. To outline steps/efforts to enhance accessibility and utility of climate data. 4. To recommend steps to be taken after the Workshop to further progress toward a national inventory. In considering the conduct of the Workshop, two points bear special notice: First, the question of "requirements" arose time and time again. The bases of the Workshop were that (a) many statements of requirements were already in existence and that (b) a comprehensive reformulation of requirements would call for a Workshop or forum of its own. This Workshop XI was therefore to focus on the solution of problems associated with the data, per se, in the context of present requirements. Second, some of the recommendations relate to measurement activities and to information services. While these areas are strictly distinct from data management, the data managers are aware of gaps in the data record which limit the use and degrade the reliability of the record. The data managers also know how to improve the interface with users (the Services Program) , for they are often the final recipients of user requests for data and information. Because of the breadth of climatic data among geophysical regimes, the Workshop adopted a Working Group structure basically along the lines of the regimes. Further, the Working Groups themselves were composed of discipline area specialists (both data managers and users) within the regimes. The main efforts in achieving the goals of the Workshop were to be the special studies done by Working Group members in advance of the Workshop and the preparation of Working Group Reports at the conference itself. A final word on the data set information sheets. In pursuit of goals relating to the inventory, we collected over 900 data sheets describing existing climatic data sets and certain other data useful in climatic assessments. The original plan was to review, edit, and publish this collection as an appendix to the Proceedings. However, the review and editing task was too extensive to be completed in time for these Proceedings. The data sheets do form a solid first core of inventory informa- tion which can serve as the basis for the inventory project the Workshop recommended. Dudley G. McConnell, Associate Director Data Information Services National Climate Program Office National Oceanic and Atmospheric Administration XII REFERENCES Interdepartmental Committee for Atmospheric Sciences (ICAS) , 1977: A United States Climate Program Plan. Federal Council for Science, Engineering, and Technology, July. ICAS, 1977: A United States Climate Program Plan by a drafting group of the Interdepartmental Committee for Atmospheric Sciences (ICAS) , Federal Council for Science, Engineering, and Technology, July 1977. Available from the National Climate Program Office, NOAA Headquarters, Rockville, MD 20852. NAS, 197 9: Toward a U.S. Climate Program. Report of the Workshop to Review the U.S. Climate Program Plans. 1979. Available from the Climate Research Board, National Academy of Sciences, 2101 Constitution Avenue, Washington, D.C. 20418. NOAA, 1977: NOAA Climate Program. NOAA. December, 1977. Available from the NOAA Climate Program Office, NOAA Headquarters, Rockville, MD 20852 Public Law 95-367, 1978: The National Climate Program Act, Sept. 17, 1978. 15 USC 2901 through 2908. Xlll ACKNOWLEDGEMENTS As Chairman of the Workshop, I wish to express my deep and sincere gratitude for the devoted effort of the Workshop participants. Many traveled great distances and withstood troublesome inconveniences to be a part of the effort. In that same spirit of cooperation and constructive contribution, the Working Group Chairmen deserve special appreciation. They all, individually, have important and demanding professional responsibilities. Yet, since the time they were named (January 1979), they have given hour upon hour of intense, productive attention to the tasks of the Workshop. Truly, the output of the Working Groups is the substance of the Workshop and the body of this Report. I want especially to thank Ernest Neil, Director of the Meteorological Program Office at NASA's Goddard Space Flight Center, for his expert guidance and counsel in administering the Workshop. And finally, I wish to thank Susan Orchin, Suzanne Karam, and Mildred Blackwell, of the Systematics General Corporation. They served as Conference Manager, Conference Coordinator, and Clerical Staff Supervisor, respectively. Their efforts were vital to a productive, smoothly run, effective Workshop. Dudley G. McConnell Chairman xiv CONTENTS Page Executive Summary iii Preface xi Acknowledgements xiv I. WORKING GROUP REPORTS Basic Atmospheric Data Working Group 1 1 Hydrology, Precipitation, Snow and Ice Working Group 2 11 Ocean Data Working Group 3 57 Radiation, Physics, and Chemistry Working Group 4 77 Proxy and Non- Instrumental Data Resources Working Group 5 115 Geographical, Land Use, and Assessment Data Working Group 6 127 Climate Inventory Working Group 7 159 II. SPECIAL TOPICS Open Letter to the U.S. Department of Agriculture - Kenneth D. Hadeen 17 3 Open Letter to the Department of Defense - Roy L. Jenne 175 Disposition of Data Inventory Information - Dudley G. McConnell 177 Research Strategy for the U.S. Climate Program - John E. Kutzbach 179 xv CONTENTS (Continued) III. WORKSHOP PROCEEDINGS List of Participants and Observers 187 Agenda 19 3 Welcoming Address - Edward S. Epstein 195 Opening Address - Thomas S. Austin 199 National Climatic Center Current Status and Future Plans - Daniel B. Mitchell 205 Data Requirements for Research and Ways to Promote The Use of Climate Data - Roy L. Jenne 225 AGENCY STATEMENTS The United States Department of Agriculture - "Requirements for Weather (and Climate) Data, Service and Information" - Galen F. Hart 2 39 The United States Department of Commerce - "Uses and Sources of Climate Data and Information in the Department of Commerce" - Joshua Z. Holland. . 2 57 The United States Department of Defense - USAF - Environmental Technical Applications Center Elbert W. Friday 269 The United States Department of Energy - Harry Moses. 275 The United States Deparment of Interior - George I. Smith 283 "Some Remarks Concerning Climate Service Programs in the States and the Data Needs for These Programs" - Wayne L. Decker 287 "Presentation on the Survey of NASA Data on Potential Applicability to the Climate Program" - Karen W. Posey 29 3 xvi Oa O 4H X rr! CO ■U M rd c u ^ ■H $ tn-U £ n CD •H o^ TJ a -P -H > rd tn c M to •H ft £ H 3 CO £, i 6 CO (D ^ +J CO CD CO ■H U >\A (D CO H 13 •H -'-J CO e . c CD 4-J e C O) (1) n £ cu 3 -H ,Q 10 !h rd QJ CD Qa-P e X cd CO M-l CD h U 0) 'H &i-C 3 c 4-> ■H rd CO M 0) CD £ rci T3 4-> G iH rd GJ rd T3 h £ M-i CO H - CD +J >1 CO rd G > s ■H U CD o P" A 53 T) ■p w K D H XVI 1 I. WORKING GROUP REPORTS BASIC ATMOSPHERIC DATA Working Group 1 Co-Chairmen: Robert J. Fox Gordon A. McKay Members: Harold Crutcher Amos Eddy Richard E. Felch James D. Laver August L. Shumbera Louis Westphal INTRODUCTION Central to the issues that led to the promulgation of the National Climate Program Act is the need for climatological data for planning and decision making in a wide spectrum of social, economic, and environmental activities. It follows that the manage- ment of basic climatological data — its acquisition, archiving, transformation into needed information, and dissemination -- is a fundamental undertaking within the Act. The U. S. Climate Program Plan identifies two types of users of climatic information. The major group comprises almost all sectors of the economy, including farmers, heating and ventilating engineers, water resource managers, designers, etc. This group needs climatic information in a variety of forms ranging from simple summaries of particular data to complex expressions of climatic expectations and the results of impact studies. The second group is interested in the past climate and data pertaining to physical processes relating to climatic fluctuations, i.e., that information needed for scientific research, in development of climate forecasting, and similar problems. The requirements for climatic data by users are not always well understood. The user community is not generally well informed on what information exists and how it may be obtained and used because of deficiencies within the supply system. The result is that available data are not being adequately exploited and data base management planning falls short of national objectives Yet major benefits are realizable now from the use of these existing data and they may be greatly increased at a very low cost. Exploi- tation of this relatively untapped source is a major realizable benefit within the concept of the National Climate Program. Relevance, credibility, utility, and accessibility are four aspects of climatological data that require particular attention. Raw data are often by-products, i.e., acquired for other purposes, and they may not be the best set or in the best form to solve the problems of a specific user. Until the user requirements are adequately known and evaluated, the data manager cannot objec- tively set management priorities, or be expected to respond in an effective and efficient manner. Credibility relates to representativeness, quality, and comparability. Standards must be rigorously imposed for these purposes so that the user fully understands the quality of data sets and the context in which measurements were made. Stringent quality control measures must be implemented to ensure that the data are not degraded in the observational, transmission, and processing phases. The utility of data and user confidence therein can be quickly undermined by inadequate control. Quality must be clearly identifiable, and a feedback system should exist so that the effectiveness of these measures can be evaluated. All data have limitations; the user must be made aware of these and of ways in which he can use data most effectively. Inadequate knowledge of the reliability and representativeness of data has led to its improper use or neglect. If the objectives of the Climate Program are to be realized, guidance on the use of climatological data/information is imperative -- logically, this must be readily accessible as part of the delivery system. Utility relates to relevance and credibility -- but more is implied. Traditionally, data have been provided as measured values., means, or similar statistics. This is often inadequate: For many decisions the data must be transformed into information, i.e., converted into values whose utility is easily recognizable and which can be used directly in decision making. This conversion is an essential function of the data management system if the objectives of the National Climate Program are to be reached. The data/information must also be in formats and on media that make them readily understandable, convenient, and accessible to the user. A variety of techniques is required because of the diversity of needs of the user community. Although there is a shift toward computer access, microfiche and hard copy are still preferred forms for many users. Use implies availability of the right kind and amount of data. In fact, only limited resources are available for any endeavor and the processes of climatic data acquisition, storage, retrieval, and use are no exception. Value trade-offs must be routinely made and will continue to be made relative to the quantity and type of climatic data to be acquired, and services to be provided. Both the value system and the trade-off decision-making can and should be objective. The users must feel their needs are being farily considered in this determination, i.e., prioritization procedures should be quantitative and perceived as impartial. Decision models or other objective techniques are needed to rationalize and optimize the value accruing to the Nation as a result of these required trade-offs. Gross national product, national objectives, quality of the environment, and health are among the many factors upon which such decisions could be made. Whatever the value assessments and decisions, it is clear that for the foreseeable future the hard core of basic atmospheric data for climatology will be conventional weather observations — surface, upper air, and satellite-derived thermodynamic, moisture, and wind data -- routinely collected for operational purposes and available in digital form on communication circuits. Additionally, the cooperative or otherwise operated climatological data acqui- sition networks which supplement the core system also need strengthening, both on a global basis and within the United States. The Federal, State, and sub-State networks constitute continuing and growing data sources of major economic value which must be recognized as an integral part of the total system. All such data should be regularly collected on a world-wide basis by the climatology system. Furthermore, this Nation must take strong positive action to improve the collection, efficiency, and quality of the basic world-wide data set for its use by United States' agencies . Finally, the user must participate even after receiving his data; he must "close the loop" for an effective data manage- ment system. Feedback to the data-providing agency on data relevancy and adequacy provides the basis for future enhancements of the overall system. RESPONDING TO SOCIO-ECONOMIC NEEDS FOR DATA Basic atmospheric data are widely used directly or in con- junction with other geophysical and social information. Present inventories show that the available data are extensive. The value of making full use of these data can be readily demonstrated. Some major sectors of use are agriculture, energy, water resources, transportation, land-use, health, fisheries, and recreation. These sectors are interested in time scales ranging from short, for operations, to long for strategic planning. Their require- ments often differ from those of the weather services system in that there is a great need for site-specific information, risk statistics, packaged data, and transfer functions that relate data to economical, physical, environmental, health, and other parameters. Another major use within the Climate Program is the support of monitoring, prediction, and research. Typical uses are in model development, diagnostic studies, and impact assess- ment. The need for data exists on regional to global scales, and for elements or parameters that identify with physical processes, e.g., cloud cover, temperature, and wind. The data management issues identified in this report are considered in a generalized manner because of the diversity of users and applications. Three specific aspects considered here are: (1) data needs, (2) improving the data management system, and (3) availability of and access to data. For each of these aspects problems are identified, principles are enunciated and, finally, remedial actions are proposed. The report ends with a list of recommendations that follow from the deficiency and action stations. DATA NEEDS Problems Requirements for climatological information are not always understood by data users or managers. Furthermore, not all segments of the user community are well informed on what informa- tion exists and how to obtain it. The needs of most users are for site-specific information, i.e., detail that is often not contained in national archives. Urbanization, environmental assessment, and more effective use of natural resources is increasing the demand for such information. Estimation techniques are a logical partial solution to the expanding demand for new information, but enhanced observing networks are required for their development as well. Definition of this need and response are often effected best on the regional to local level. Mechanisms for this are presently obscure and not commensurate with needs. But site-specific information is not the only requirement. Climatological analyses and impact assessment on the regional to global scale are now of extreme importance to governments, industry, and in research. Research needs can only be adequately defined by the research community, an activity which is fairly advanced as a result of national and international experiments and climate programs. Principles The full spectrum of user needs must be defined, and mechanisms for providing information must be developed. This should be done in full consultation between the user community and agencies which serve their interests , and should be implemented in a way that is most cost effective, on a priority basis. The resulting system must be responsive to the need for relevant, timely informa- tion, the limitations of which are clearly understood. Major social and economic issues within a region or nation will have highest priority in the development of the climatic information system. Remedial Actions Remedial actions include the development within service agencies of needs-oriented, product-generating systems tailored to user requirements. Initally, workshops, interviews, etc., must be held to find out what the needs are. Having established the needs, priorities must be set. Objective methods must be developed wherever possible for setting priorities and evaluating alterna- tives. The user community, Federal, State, university and private, and all the major economic sectors, should participate in all phases of decision making. An intergovernmental advisory group should be established to see that the users of regional to local scale information are adequately served by the integrated data management system. IMPROVING THE DATA MANAGEMENT SYSTEM Problems The National Climate Program Act states that information regarding climate is not fully disseminated or used and that Federal efforts have given insufficient attention to assessing and applying this information. A wide range of problems exist within the data system that limits the application of climate data and information to meet user needs. Examples are incompleteness of records (both manuscript and digital), non-standard records, unrepresented observing sites, inefficient file organization^ and difficult access. As a result, the data are frequently of questionable quality in the eyes of the user, or are ignored. Over the years hundreds of basic atmospheric data sets have been established in response to and in a mode dictated by then existing observation systems and agency needs. These data exist in both digital and hard copy form. The large data files are often cumbersome, expensive to use, and contain numerous duplicates and specialized products. The data vary in quality due to changes over time in observing codes and practices, technology, and data requirements. Many of these changes are undocumented. There is also a serious lack of inventory and guidance material on the use of available data. Gaps and inconsistencies in and between digital sets require specialized editing to make them usable. These and similar problems make it difficult and expensive (in many cases prohibitive) to provide users with data and information that are essential for assessment, applications, and research. Something must be done to improve the data system! Principles Data management systems must be responsive to the great diversity of needs imposed by the user community if the objectives of the National Climate Program Act are to be reached. This requires an integrated approach with regional acquisition or data management systems complementing the national system. The climate data management system should facilitate access to climate data/information in a cost-effective manner, and respect the needs of the user to be able to exploit its products with confidence. A dynamic integrated system is needed. Credibility and use demand appropriate levels of quality control, documentation, inventory, and easy access. An efficient, integrated system will minimize the cost of supplying the data and thereby enhance its use. Remedial Actions Editing, quality control, and organization of basic atmospheric data to be compatible with established standards and consistent with user requirements are basic remedial needs. Enlightened data management must consider the spectrum of users, recognize the merits of global data sets and the desirability of regional data manage- ment for specialized applications. The existing data base must be upgraded by: (1) editing, (2) determining and recording validity of data, (3) completing digital files by filling in gaps within periods of record, (4) identifying a basic set of selected stations and upgrading their quality for ultra-long periods of record, and (5) completing the inventory of climatic data sets. To further enhance the usefulness and availability of data, station files must be expanded to include history (e.g., changes in instrumenta- tion types, placement, and geographic information) for the sites. Concurrent with the upgrading of the historical data sets the data acquisition system must be upgraded. Networks must match prime needs. Instruments should be calibrated and compared with national (ANSI) meteorological standards (as developed) at regular time intervals and the results systematically reported to the archiving data center. Measurements should be made comparable by using instruments and practices that meet WMO and ANSI specifi- cations. The opportunity for efficiency and economy in management and systems offered by new and existing technology must be fully exploited in the interest of fulfilling national and regional needs, within resource limitations. An example is the proposed use of touch-tone telephone pads and INWATS to upgrade the acquisition quality of c limatological data. Some data currently being acquired for operational applications are only partially archived, e.g., meteorological satellite data. These should be evaluated to see if they can be further exploited to enhance the utility of the current archive. AVAILABILITY OF AND ACCESS TO DATA Problems As noted the availability of information and systems through which climatological data may be accessed are generally inadequately understood within the user community. On the other hand climate data management has not always been well tuned to user require- ments. In many instances, data formats do not meet the user needs, e.g., card formats are still used extensively, whereas the scientific community finds a different format to be incomplete or to include parameters that are not appropriate to the user's problem. There are lags in interfacing technology, and inadequate direction to best sources. Dozens of independently operating, self-contained data systems are functioning on national or regional scales providing access to basic meteorological data for various communities of users. Some of these systems operate relatively unaware of each other's existence and lack the benefits of pooled knowledge and facilities of an integrated system. Principles An effective climate data management system must serve the diversity of users. It must be integrated for efficiency and effectiveness. The system must be compatible with other infor- mation systems so that the needs for impact or interagency studies, which involve several disciplines, can be readily met. The system should also do more than simply provide a data retrieval service; it should also supply probability statistics, extreme values, comparative information as required for climatic evaluations, and climatic information tailored to decision processes. The nature and capabilities of the system (s) should be well known and for that purpose it is necessary to undertake appropriate promotions or publicity. Furthermore, "linkage" functions and general edu- cation on use are essential if the system is to play its proper role in the community. Remedial Actions To satisfy climate data needs it is necessary to Droperly archive, retrieve, and transform the basic atmospheric data. To achieve this goal it is necessary to develop a hierarchal system of integrated data centers and referral systems. There should be linkages between those centers serving global purposes and those serving national, sub-national, and/or regional functions. Path- ways to different types of data centers, through an efficient referral and cross-referencing system, must be easily accessible to the user community. However, mere availability of a data set does not make it a useful climatological data set. Also necessary are the data set "characteristics", e.g., regression equations, analysis scheme changes, or station location changes, to which only the data set generator or climatological community may be privy. Lack of knowledge of these data set "characteristics" could lead to misinterpretation or misuse of the data set and deny the user the additional benefit of information/ as opposed to uninterpreted data. Availability of data sets should be handled by making the user community aware of systems such as ENDEX, OASIS, and DIALOG (see Working Group 7 Report) . Questions of user access to the data should be addressed first by studying successful systems already in existence. Some examples are: the Indiana/Purdue Agro- climate computer terminal network which reaches into some 50 counties at present with second order terminal linkages within a county to individual users, the HISAR system for hydrology, and the Georgia based Forest Weather Information Service. State and independent systems must be recognized and supported as component parts of regional to national data management systems. Worth examining are the user services supplied by the California Extension Service and extrapolating such services to meet the widespread need for linkages between users and the data management system. Finally, if only a poor quality data set is available, its use should not be denied to users until it can be upgraded, especially if it would require several years to do so. Rather, an intermediate version, properly flagged as to limitations, should be made available for use at the user's risk. RECOMMENDATIONS 1 . Take immediate action to assure more complete collection and quality assurance of world-wide conventional surface and upper air data . These have been and will continue to be the preeminent data for climate services, applications, and assessments. 2 . Espouse a data system dedicated to national needs. Insist upon a strong commitment, including resources and promotion by all U.S. climate-oriented agencie's . Inadequate knowledge of data, data sets, data limitations, how to access data sets and their proper use are major obstacles to the success of the National Climate Program. Enhance and widely publicize the system so that all potential users know how to obtain and use climate "information." 3 . Develop quantitative bases for rational decisio ns in the acquisition, collection, archiving, and use of specific data for climate purposes . At present many major decisions in data management are based on gut-feelings, lobbies, and similar non-objective methods. The development of a quantitative approach is imperative because of the cost of data system operation and the need for optimum service to the Nation. 4. Undertake a program to determine and record station histories and local geography for U.S. data observing sites . The instrumentation and site characteristics (latitude and longitude, elevation, exposure, land-use characteristics, types of instruments, ...) of world-wide observing loca- tions, and the history of changes in same, are vital infor- mation for the climate community. Such a program for U.S. stations must be promoted and fully implemented as a com- ponent of any centralized archive facility. Further, the National Climate Program Office (NCPO) should strongly advocate within the World Meteorological Organization (WMO) the develop- ment of a world-wide program with similar goals. 5 . Define the full spectrum of actual and potential users of climate data/information and their requirements . 6 . Continue the strong emphasis on the search for and classi - fication of new data sets for inclusion in the climate data inventor y^ This search should continue until it is no longer cost- effective. The current status of the inventory has already revealed many potentially valuable data sets which have had little previous exposure within the user community. 7 . Improve the standardization of global climatological data . World-wide practices in the collection, encoding, and transmission of conventional meteorological and climatological data vary extensively from WMO standards, creating severe decoding and processing problems. These in turn lead to the discard of valuable data, more costly quality control or recovery procedures, and reduced utility of the overall data set. Current and proposed actions of the WMO such as the revision of the synoptic code (1981) and proposed changes in the CLIMAT code have the potential of further enhancing the value of the global data set. Such WMO procedures and intentions deserve the strongest possible support, and the NCPO should be a major advocate thereof. 8 . Determine the economic and social value of climatic informa - tion . Studies must be funded or otherwise promoted and enabled through the NCPO. The results of the studies will assist top administrators in setting priorities for the allocation of resources, and assist the NCPO and cooperating agencies in the evaluation of plans and the development of strategies. 9 . Establish a Federal/State advisory group to make specific recommendations on climatic data management at the region / State level . Climatic data sets exist at the Federal level, at the State level, and at sub-State levels, each to serve user groups operating on different though not mutually exclu- sive space-time-economic scales of activity. The advisory group recommendations are to be referred to agencies con- cerned for agreed-upon action. 10 . Assess the cost effectiveness of unattended data acquisition systems specifically designed for and devoted to obtaining climate data . Current technology has made remote and/or unattended acquisition of climate data feasible, and perhaps cost effective. This and related technology also offer the potential for meso- or microscale networks to become cost effective because they can free personnel and other resources for application to other high priority requirements. Land-use and similar concerns, particularly in urbanizing areas, create a need for understanding climatology on this small scale. Such systems could be used to offset the degradation in clima- tological networks that now appears to be resulting from demographic and life-style changes. 11 . Evaluate from an archival point of view all new federally funded programs and projects that may generate environmental data . This should be done during the concept and design phases. If archiving is deemed to be in the national/regional interest, the cost of preparing and archiving the data should be funded as a basic system acquisition cost. Consideration for archiving climate data has traditionally occurred well after the data were being (had been) collected and inadequate funding has led to general deterioration in data services. So that future data collections are more applicable to a wider range of user needs, the NCP should advocate legisla- tion requiring this evaluation. 10 HYDROLOGY, PRECIPITATION, SNOW AND ICE Working Group 2 Chairman: Eugene Rasmusson Members: Roger Barry David Curtis Richard Guthrie Charles Laughlin James Ormsby Ward Seguin Charles R. Showen John L. Vogel Donald Wiesnet INTRODUCTION In order to deal effectively with the broad range of topics assigned to this working group, the following reviews were prepared by Working Group members prior to the Workshop: "Streamflow Data" by C. R. Showen, USGS "Satellite Measurements of Precipitation" by C. Laughlin, NASA "Contribution of Operational Satellite Data to the Problems of Hydrology - Current and Near Future Recommendations" by D. R. Wiesnet and M. Matson, NOAA/NESS "Outlook for Remote Sensing of Hydrologic Parameters in the 8 's" by J. Ormsby and J. Theon , NASA "Atmospheric Hydrologic Data" by E. Rasmusson, NOAA/NWS "Current and Projected Status of the Surface Marine Data Base in Terms of its Use for Estimating Oceanic Evaporation" by W. Seguin, NOAA/EDIS "Availability of Precipitation Data Over Land - U.S. and Global" by W. Seguin "Snow and Ice Data" by R. Barry, INSTARR, U. of Colorado "Small Scale Variability of Land Surface Hydrologic Parameters", by J. Vogel, Illinois Water Survey 11 "Land Use and Land Cover Data Requirements for Modeling of Surface Hydrologic Processes" by D. Curtis, NOAA/NWS Most of these reviews were circulated to working group members prior to the Workshop. During the Workshop, the papers were discussed, revised, and in some cases rearranged or consolidated around more appropriate topics. The final result was seven summary reports on the following topics: • Streamflow and Groundwater Data • Conventional Precipitation Data Over Land • Satellite Measurement of Precipitation • Snow and Ice Data • Requirements for Modeling of Surface Hydrologic Processes • Ocean Evaporation • Atmospheric Hydrologic Data These summary papers are included as part of this report. After reviewing and discussing the content of these papers, and coordinating overlapping areas of interest with other working groups, the members of Working Group 2 wish to make the following recommendations: RECOMMENDAT IONS STREAMFLOW AND GROUNDWATER DATA 1. Expand the National Water Data Exchange (NAWDEX) of the USGS to an international inventory and exchange system for streamflow and groundwater information . The structure for expanding NAWDEX to an international system is in place, and some Canadian and Mexican infor- mation is already incorporated into the system. U.S. members of NAWDEX are required to sign a memorandum of understanding. In the case of foreign data, an informal exchange program is established and the foreign organi- zations are identified as affiliates. It is not effective to develop an international system of data exchange through a series of individual bilateral arrangements. Rather, as in the case of meteorological data, an overall structure for data exchange should be developed through an appropriate international organization such as the World Meteorological Organization (WMO) or the United Nations Educational, Scientific and Cultural Organization (UNESCO) . It is recommended that the National Climate Program Office (NCPO) , working closely with NOAA and USGS, assume the lead role in the development of this structure in the United States. 12 2 . Expand the Water Data Storage and Retrieval System (WATSTORE) of the USGS into an archive for international streamflow and groundwater data . Although WATSTORE should serve as the central archive for basic streamflow and groundwater data, it will undoubtedly be desirable to prepare subsets and special summary products from these data for specific climate appli- cations. These derived products should be developed as the need arises. 3. Develop a consolidated station history file for stream gaging stations . Information on stream gaging stations is at present scattered throughout a variety of publications and organi- zations. Auxiliary information such as gaging station history and changes in channel characteristics/ diversions, and storage are required for proper interpretation of streamflow data within the context of the climate program. A comprehensive assembly of these data, for both United States and foreign gaging stations, should proceed along with the indexing and exchange of these data. CONVENTIONAL PRECIPITATION DATA OVER LAND The following recommendations are made to improve the accessibility and utility of the conventional precipitation data sets. They are made within the framework of data: (1) acquisi- tion; (2) editing, validation and enhancement; (3) archiving; (4) distribution; and (5) evaluation. The recommendations for the acquisition of data are further subdivided into the inter- national, national, and regional scale. The recommendations apply to all levels of precipitation data. 1 . Take vigorous action to bring about increased compliance with synoptic coding and transmission standards for precipitation data . Precipitation data is required on a global basis for applications such as crop assessment. Although subsynoptic scale coverage is needed, the reports from the World Weather Watch (WWW) Synoptic Network now constitute the basic source of real time global precipitation data. Failure to transmit these data in accordance with established international standards results in data far short of its potential. The NCPO should take vigorous action through the WMO to bring about increased compliance with synoptic coding and transmissing standards, and assure that the effective transmission of precipitation data be a prime consideration when changes in the synoptic code are considered. 13 2 . Conduct a systematic evaluation of the requirements for the U.S. substation precipitation network . It is recommended that (1) the NMC and cooperative sub- station networks be maintained in their present state until a clear demonstration can be made of the precipita- tion requirements of the National Climate Program (NCP) , and (2) a systematic evaluation of the precipitation network requirements be made using economic and climate- dependent models. This evaluation process would be similar to network studies performed in the late 1960s to determine the optimum distribution of upper-air reporting stations required for numerical modeling efforts. 3 . Establish an inventory of data from special precipitation networks . On the local or regional scale the NCP should establish a dynamic inventory system of all special precipitation networks, whether they be short-term or continuous. Careful preparation must be made to ensure that such an inventory gives a complete description of each network, the data type, and its availability. Archiving of special network data should be encouraged, although not necessarily at a single data center. Particular emphasis should be placed on data derived from very high density gage networks, or high resolution mappings of the precipitation field derived through the combination of quantitative radar data and data from conventional precipitation gages. In addition, the NCP should encourage the establishment of local and regional precipitation networks to establish the natural variability of precipitation and its relation to hydro- climatic parameters such as soil moisture, runoff, and soil erosion. 4 . Include proper documentation of network characteristics and processing procedures with precipitation data sets . Proper documentation for the observational network and data processing procedures is essential for the effective use of precipitation data. It should be an accepted principle of data management that such supporting information constitutes an integral a part of the data base as the data values themselves. Such information includes documentation of site locations, instrument types, length of record, data processing operations, and known deficiencies in the data. 14 SATELLITE MEASUREMENT OF PRECIPITATION The possibility of estimating rainfall from satellite data on a global basis exists through the use of passive micro- wave radiometry, active radar, and visible and infrared imagery. No other reasonable possibility now exists for obtaining comprehensive coverage over the world oceans. Even over land, the conventional point precipitation networks are generally inadequate to resolve small-scale variations and provide reliable areally averaged values. Microwave, as employed by the Nimbus-5 Electronically Scanned Microwave Radiometer (ESMR) , may be suitable for global ocean studies but unsuitable for local land-based hydrologic applications. Visible and IR cloud-top temperature techniques used together with other ancillary data and information may be most applicable to con- vective rainfall situations, and most effective for local and regional studies. 1 . Develop sites for evaluating satellite precipitation estimation techniques . It is essential that the accuracy and sphere of usefulness of existing techniques for satellite estimation of precipitation be evaluated and objectively documented, since these methods are still in the development state. Test sites for carrying out this verification over land areas should be designated for several climatic zones (tropical, temperate, arctic, etc.). The temporal/ spatial frequency of surface precipitation observations should be comparable to at least that of the GARP Atlantic Tropical Experiment (GATE) B array. In deference to fiscal constraints, these test sites should be selected from currently well-instrumented test areas wherever possible. Examples over the United States are (1) the South Dakota State University weather radar and rain gage network; (2) the precipitation gage network at Chickashaw, Oklahoma; and (3) the flash flood network being developed by NWS in the Appalachian area. As an initial step in testing passive microwave radiometry methods over ocean areas, further studies should be made using the GATE ESMR and radar data to determine both the capabilities and limitations of ESMR in providing rainfall rates over oceans, and from this determine the desirability of processing the remaining ESMR data. Such studies would also suggest future directions for research in passive radiometry, and provide a basis for judging the potential of the improved measurement accuracy offered by radar techniques. 15 2 . Develop a precipitation data base for use in satellite systems design . A need exists for basic data on the statistical properties, both in time and space, of precipitating systems for use in satellite/sensor systems design. As a result, a reasonable basis for determination of system specifications such as the microwave antenna size required (and corresponding beam- width and ground resolution) is lacking. It is, therefore, further recommended that the most interested agencies, NOAA and NASA, take the lead in developing the requirements for such a data set, and take steps to acquire and assemble the required data. SNOW AND ICE DATA 1 . Digital data bases of variables characterizing global snow cov e r and sea ice must be developed in grid form at. Initially, (1) Operational snow and ice chart products should be evalu- ated for possible digitization. (2) The new series of 7-day Composite Minimum Brightness (CMB) Charts from NOAA/NESS should be recorded in digital form at NESS prior to archiving at EDIS. 2 . High priority should be given to continuing the development of inventories for the major variables characterizing global snow cover and sea ice . 3 . Recognizing the valuable property of microwave sensors to penetrate clouds, the usefulness of ESMR microwave observa - tions of polar area snow and sea ice should be evaluated . 4 . Support must be given to providing the auxiliary data needed for studies involving interaction with snow cover . These data include geographic data on terrain characteristics, vegetation cover, land use, as well as temperature, radia- tion, and wind speed data. 5 . To facilitate international exchange of ice and snow data , national agencies are encouraged to send data, information , and user requests to the WDC-A for glaciology (Snow and Ice ) . REQUIREMENTS FOR MODELING OF SURFACE HYDROLOGIC PROCESSES (Also see recommendations of Working Group 6) 1. Develop coordinated soil data base management systems . Several Federal and State agencies are independently undertaking the tasks of digitizing soil survey maps and 16 construction of computer oriented soil data base systems. The purpose, scope, and execution of each agency's effort is varied, as are the levels of financial support. Therefore, to optimize the utility of soils data to the National Climate Program, the Working Group recommends a coordinated national emphasis on the development of soil data base management systems. No new data collection programs are proposed by this recommendation. Rather, existing information found in tabular and map form should be digitized and formatted for interactive availability. Additionally, the soil from which data have been acquired should be properly classified before the data are archived. Soils data from local, regional, and international sources should be included. 2. Develop a central archive of time series of soil moisture measurements . About 200 soil-vegetation records have been uncovered in a recent survey of available soil moisture data. They include several studies mainly at Science Education Administration-Agricultural Research (SEA-AR) research stations scattered throughout the United States. A 5-year soil moisture pilot study is proposed by the Soil Conservation Service (SCS) in cooperation with SEA-AR and State Agricultural Experiment Stations. It is recommended, that time series of directly measured soil moisture, which • are becoming available in increasing numbers, be centrally archived, and that consideration be given to making these data available for general use, perhaps through interactive computer systems. It is recommended that USDA assume the lead in this effort. 3. Encourage research programs for the development of technologies to remotely sense soil moisture . Recognizing the difficulty of obtaining useful coarse grid estimates of soil moisture, it is recommended that the NCPO serve as the focal point for coordinating and encouraging research programs for the continued develop- ment of technologies to remotely sense "grid-scale" soil moisture on a regional and global scale. OCEAN EVAPORATION See recommendation of Working Group 3 concerning the development of the surface marine data base. 17 ATMOSPHERIC HYDROLOGIC DATA The operational rawinsonde network continues to provide the basic moisture and wind data for evaluating the atmospheric hydrologic cycle. Because of special data problems and require- ments, much of the archived rawinsonde data is inadequate or only marginally adequate for these computations. Recommendation The N.C.C. should make the archived rawinsonde data more useful for atmospheric water balance computations . (1) the corresponding surface observation be archived with the upper level data; (2) data be regularly computed, transmitted and archived at 50-mb intervals, at least for the layer below 850 mb; and (3) documentation of instrument type, including humidity element, known biases, and established methods for correcting the data be included as an integral part of the rawinsonde data archive. SUMMARY PAPERS STREAMFLOW AND GROUNDWATER DATA Source of Data The U.S. Geological Survey (USGS) collects streamflow information at about 9,000 sites, which represent about 80 percent of the streamflow data that is collected in the United States. The Office of Water Data Coordination (OWDC) of the USGS impaneled a Federal Interagency Work Group for Designation of Standards for Water Data Acquisition. The report of the work group on surface water, including storage, stage, streamflow, and gage data, will be published in late 1979. The establish- ment of standardized methods for acquiring surface water data will enable comparison of data collected by different individ- uals or agencies. Without standards, the validity of much irreplaceable data that was costly and time consuming to acquire would be questionable. Streamflow data at international gaging stations on the borders of Canada and Mexico are available from the Water Survey of Canada and the International Boundary and Water Commission. The availability of streamflow data on a global basis varies country-by-country. For example, the Hashemite Kingdom of Jordan collects streamflow information, but due to lack of experienced personnel the data are not computed and published on a regular basis. In other countries, such as Australia and New Zealand, the data are no longer published but finalized data are available from the agency that collected the data. Data Formats Since about 1968, most of the streamflow data collected in the United States have been collected in digital form. The data are automatically punched on tape at 15-, 30-, or 60- minute intervals. This detailed hydrograph information is available in machine-readable form through the local USGS office, which is generally located in the State Capitol. On a global basis, most streamflow data are available in manuscript form, such as strip charts or daily readings by local observers. Data accuracy will vary considerably depending on the physical conditions at the site and the techniques employed by the country that collects the data. Future Plans The USGS is evaluating satellite technology that is expected to provide a cost-effective technique for the automatic collection of data from hydrologic stations. These data include water stage, water quality, precipitation, and snow depth. This technology, which is referred to as satellite Data Collection Systems (DCS) , provides an opportunity to collect data from inexpensive battery-operated radios located at literally tens of thousands of hydrologic stations distributed over national or continental areas. The availability of real time-processed hydrologic data will aid in flood forecasting and provide for more effective water-resources management. Data Management The user of U.S. streamflow data faces a bewildering problem in trying to determine if the specific information he needs has been collected, and where it is available. Because of the lack of a requirement for large-scale international data exchange, arrangements such as those developed through necessity for meteorological data do not now exist for streamflow and ground- water data. To help solve these problems on the national level, the National Water Data Exchange (NAWDEX) was established within the USGS. NAWDEX is a national confederation of water-oriented organizations working together to improve access to water data. Its primary objective is to assist users of water data in the identification, location, and acquisition of needed data. It consists of two data bases. The first is a Master Water Data Index which identifies specific sites for which water data are available, and provides information on types of data, periods for which data are available, parameters, frequency of observations, and media. The second data base consists of the Water Data Sources Directory, which identifies organizations that are sources of water data. The NAWDEX Program Office 19 encompasses four major areas of operation: (1) maintaining an internal data center, including access to automated data processing facilities for maintenance and use of its information files, (2) indexing water data held by participating organizations, (3) providing facilities and personnel for responding to requests for water data, and (4) formulating recommended water data handling and exchange standards. The function of NAWDEX is not to become a repository of water data. Instead, the Program Office indexes the data held by NAWDEX members and participants to provide a central source of water-data information available from a large number of organizations. These data may be in computerized and non- computerized form. The National Water Data Storage and Retrieval System (WATSTORE) was established in November 1971 to modernize the USGS existing water-data processing procedures and techniques, and to provide for more effective and efficient management of its data releasing activities. The WATSTORE System consists of several files in which data are grouped and stored by common characteristics and data collection frequencies. Currently, files are maintained for the storage of (1) surf acewater , quality-of -water, and groundwater data measured on a daily or continuous basis; (2) annual peak values for streamflow stations; (3) chemical analyses for surface- and groundwater sites; and (4) geologic and inventory data for groundwater sites. The daily values file includes data for streamflow values, river states, reservoir contents, water temperatures, sediment discharges, and groundwater levels. Recomme ndations 1. Expand NAWDEX to an international inventory and exchange system for streamflow and groundwater information . The structure for expanding NAWDEX to an international system is in place, and some Canadian and Mexican infor- mation is now incorporated in the system. U.S. members of NAWDEX are required to sign a memorandum of under- standing. In the case of foreign data, an informal exchange program is established and the foreign organiza- tions are identified as affiliates. It is not effective to develop an international system of data exchange through a series of individual bilateral arrangements. Rather, as in the case of meteorological data, an overall structure for data exchange should be developed through an appropriate international organization 20 such as the WMO or UNESCO. It is recommended that the NCPO assume the U.S. lead role In the development of this structure. 2 . Expand WATSTORE into an archive for international stream- flow and groundwater data . Although WATSTORE should serve as the central archive for basic streamflow and groundwater data, it will undoubtedly be desirable to prepare subsets and special summary products from these data for specific climate applications. These derived products should be developed as the need arises. 3 . Develop a consolidated station history file for stream gaging stations . Information on stream gaging stations is at present scattered throughout a variety of publications and organizations. Auxiliary information such as gaging station history, and changes in channel characteristics, diversions, and storage are required for proper interpretation of stream- flow data within the context of the climate program. The USCPO should assure that comprehensive assembly of this information for both U.S. and foreign gaging stations proceed along with the indexing and exchange of these data. CONVENTIONAL PRECIPITATION DATA OVER LAND U.S. and Global Introduction Conventional precipitation data are very important to the climate initiatives of the 1980s and beyond. Precipitation data are used for agriculture, energy and water resource purposes, for predicting the overall health of the world's economies, and to specify the hydrologic cycle which is fundamental to climate forecast models. The data are used to evaluate soil moisture, evaporation rates, and erosion for land resource studies, and provide a baseline with which we can study and evaluate potential climate change whether natural or man-made. Precipitation data are used by many diverse groups including government units, architects, attorneys, engineers, and people in health, insurance, manufacturing, recreation, transportation, and utilities. These groups derive their data from international, national, and regional sources, whose requirements vary with the users. 21 Global precipitation data are required to aid in the forecast of global crop production and to determine the ability of nations to produce energy using hydroelectric energy production. Precipitation data are needed on a national scale by government and private groups for use in long-range planning, policy, design, management, education, and documentation of factual records. On a regional scale the data are used for a multitude of specialized requirements ranging from the determination of precipitation effects on soil moisture and other hydrologic effects and the possible influence by man upon urban environments or local climate through planned weather modification. Therefore, the National Climate Plan should include a comprehensive plan to ensure the systematic collection, processing, and dissemination of precipitation data and information. This includes providing quality controlled data in both raw and processed form. Additional emphasis should be placed on the quality, density, and continuity of reporting stations, the accessibility of data and information, and the timeliness of the data . Present Data Status The National Climatic Center (NCC) receives precipitation data for thousands of stations in the United States and the world. These data are received on different media including logs or forms, strip charts, punched paper tape, publications, and magnetic tape. The time resolution varies from continuous (e.g., strip charts) to monthly summaries (e.g., CLIMAT reports). Some data are edited carefully, while others are little more than scanned, and some simply are inventoried and archived. NCC archives the manuscript data, computes and prepares digital tape files, and publishes a number of monthly and annual summaries. Some tape files serve specific users such as NOAA's Office of Hydrology. The publications, history, and the information contained in various precipitation data bases are summarized in an EDIS report: Selective Guide to Climatic Data Sourc es. Canada publishes a similar report entitled Handbook on Climatological Data Sources of the Atmospheric Environment Service . International data are relatively scarce. A monthly summary of precipitation data for hundreds of stations throughout the world, which are derived from CLIMAT reports, are published each month in Monthly Climatic Data for the World . The publication originated in May 1948 under the title Monthly Climatic Data for the World by Continents . It is jointly sponsored by the World Meteorological Organization (WMO) and NOAA. 22 The corresponding digital tape file is the Monthly Global L and Surfac e data for about 1,500 global stations. Parts of i he data extend as far back as the 1700s and contain total precipitation, among other variables, for each month. Recently the NCC began to receive decoded daily and shorter term global precipitation data from the Center for Environmental Assessment Services (CEAS) of EDIS, who decode global tele- communications as part of its crop assessment service. However, these data are incomplete because of data loss in transmission and poor international coding practices. Documentation of data problems, instrumentation and observing practices is either nonexistent or at best lacking. Precipitation data have been processed and documented for a number of short duration experiments such as the Barbados Oceanographic and Atmospheric Experiment (BOMEX) and the GARP Atlantic Tropical Experiment (GATE) . These are available from NCC. Precipitation data for national networks are available in several forms from the NCC including raw manuscript forms, publications, and computer tape files. The four monthly publications summarizing the U.S. precipitation data by State are listed here: 1. Local Climatological Data (monthly) , and the Local Climatological Data, Annual Summary with Comparative Data. The monthly issue presents basic climatological data together with a table of hourly precipitation by month for each of the first order weather stations (approximately 300) . Predecessor issues were first published as the Monthly Meteorological Summary in 1897. The first A nnual Meteorological Summary was published in 1909; today it contains sequential and annual tables for vain and snowfall, among other variables . 2. Climatological Data (monthly and annual) is published for each State or combination of States (Maryland- Delaware, New England; the Pacific area; and Puerto Rico and. the Virgin Islands) . It was first published by the Weather Bureau as the Climatological Service Bulletin beginning in February 1906. Monthly issues contain a daily precipitation table for some 9,000 stations, and annual issues contain monthly and annual total precipita- tion and departures from normals. 3. Hourly Precipitation Data is published for each State or combination of States (Maryland-Delaware, New England) except Alaska. The predecessor publication known as the Hydrologic Bulletin was issued by drainage districts, 23 and began publication in January 1940. The monthly issue presents daily and hourly precipitation data from about 3,000 stations equipped with automatic recording gages. The annual issue contains monthly and annual totals of precipitation. 4. Climatological Data, National Summary is issued monthly and annually, and contains selected climatological data on a national basis. It began with the 1950 issue, but prior to that much of the data appeared in the Monthly Weather Review , the U.S. Meteorological Yearbook , and the Report of the Chief of the Weather Bureau . It contains, among other variables, the greatest and least monthly totals of precipitation for each State, together with the names of the locations at which they occurred; and maps depicting total precipitation and percentage of normal precipitation. The annual issue gives maps of total annual precipitation and percentage of normal annual precipitation. The principal computer precipitation tape files in the NCC library are listed below: 1. Summary of the day data files for the cooperative observer network contain daily precipitation and extend back to the late 1800s for selected stations. Currently, precipitation data for approximately 9,000 stations are being added to the files. 2. Summary of the day data files for the first order National Weather Service stations contain daily precipitation beginning in 1948. 3. Hourly precipitation data files for the 3,000 universal and Fischer-Porter gages begins in 1948. 4 . Monthly United States Land Surface data contain monthly totals of precipitation data for approximately 10,000 current cooperative stations with data for some stations extending back to 1900. The National Weather Service has a number of different observing networks which fall into one of two general classes: the first order stations, which include the principal NWS forecast and observing stations as well as certain FAA stations, and the substation networks. The substation networks are made up of the A-network, the basic climatic network; the B-network, the basic hydrologic network; and the C-network, a local service network. A few stations have been placed in an X-network category which consists of long-term stations transferred from the C-network. These are to be kept open through 1980, at which time their usefulness will be reexamined. 2 4 Station histories are generally available for all of the national stations, although many need to be updated. Likewise, document a ion on instrumentation used is available, but documen- tation on data quality or problems of individual stations generally is not available. Regional and local precipitation network data are available from a variety of sources ranging from private organizations and local governments to State and Federal agencies. Some networks are designed to supplement the cooperative substation network on a State or regional basis and are operated on a continuing basis. Examples are networks operated by the State of California and the TVA. Other networks are operated to describe local or subregional precipitation processes such as urban precipitation networks (Chicago and San Francisco) to monitor and identify urban hydrology problems. Other networks are designed to determine various relations between precipi- tation and hydrologic parameters such as runoff, soil moisture, and soil erosion. The Atmospheric Environment Service (AES) of Canada is developing a national network of C-band quantitative weather radars with a digital recording and transmission capability. Digital data from this network are used to create a national archive of good quality, easily accessible, quantitative radar data. Single time, software development permits the initiation of a radar climatology program for the entire network and special analysis programs for one area of the country are readily usable at the other sites in the network. Additional special networks are operated on a short-term basis for specific purposes. Some have detailed rainfall information, such as METROMEX, which documented the rainfall distribution around a large urban-industrial complex for a 5-year period to evaluate inadvertent precipitation effects and the Thunderstorm Project networks of the mid 1940s, which were designed to help understand thunderstorm dynamics. Other net- works, operated by local or regional groups, measure rainfall amounts to determine the optimum harvesting period of crops, to evaluate weather modification experiments, and for other special purposes. The quality and accessibility of those records range from wholly inadequate to ready retrieval of the data. Recommendations The following recommendations are made to improve the accessibility and utility of the conventional precipitation data sets. They are made within the framework of data: (1) acquisition; (2) editing, validation, and enhancement; (3) archiving; (4) distribution; and (5) evaluation. The recommendations for the acquisition of data are further sub- divided into the international, national, and regional 25 scale. The remaining recommendations apply to all levels of precipitation data. The necessary flow of precipitation data to evaluate global needs requires that (1) the international community be approached to ensure the continued flow of summarized and raw data; (2) efforts should be made to upgrade the quality and quantity of international precipitation data; and (3) standards be established and followed for the international transmission of global precipitation data. The last recommendation is especially important in light of the proposed changes in the synoptic code scheduled to occur in the early 1980s. Nationally, it is recommended that (1) the NMC and cooperative substation networks be continued in their present state until a clear demonstration can be made of the precipita- tion requirements of the National Climate Program; and (2) a systematic evaluation of the precipitation network require- ments be made using economic and climate dependent models. This evaluation process would be similar to studies performed during the 1960s to determine the optimum distribution of upper-air reporting stations for numerical modeling efforts. On the local or regional scale the National Climate Program should establish a dynamic inventory system of all special precipitation and quantitative weather radar networks, whether they be short-term or continuous. Careful preparation must be made to ensure that such an inventory gives a complete description of each network, the data, and its availability. In addition, the National Climate Program should encourage the establishment of local and regional precipitation networks to measure the natural variability of precipitation and its relation to hydro-climatic parameters such as soil moisture, runoff, and soil erosion. Additional local or regional networks should be established and operated for a sufficient length of time to define hydro-climatic effects of new technologies with the potential of inadvertently modifying the local or regional hydrologic cycle. For proper editing, validation, and enhancement of precipitation data, it is encouraged that there be (1) careful documentation of all data processing and irregularities; (2) documentation of instrument types, accuracy, and site locations; (3) documentation of the data processing and history of each site automatically made available to users; (4) re- processing of historical data sets to eliminate or correct errors; and (5) an effort to maintain all data sets with the highest possible temporal and spatial resolution. 26 For the proper archiving of precipitation data it is recommended that (1) all data be duplicated with the originals and duplicates stored separately; and (2) all data sets should be made complete and extended back into time if feasible. The efficient distribution of precipitation data requires that (1) on-line data inventories be provided and that they should be continually updated; and (2) data should be made available as fast as is practical. An evaluation mechanism should be established so the data user can indicate data problems and inadequacies, and this mechanism must provide procedure to ensure that remedial actions will be taken promptly. SATELLITE MEASUREMENTS OF PRECIPITATION Visible and Infrared Imagery Several methods for estimating rainfall using satellite imagery have been developed. These methods depend on empirical relationships between the cloud brightness and rainfall amount which, at best, apply to limited convective rainfall situations, These methods generally define the area of cloudiness very well However, due to the rapid growth of convective clouds, rainfall estimates based on images every 30 minutes or longer are questionable . Visible and infrared radiation from polar orbiting and geostationary satellites have been used to give areal and time averages of precipitation mainly by monitoring the growth rate of the convective clouds, and cloud top brightness temperature and significant cloud features. Geostationary Operational Environmental Satellite (GOES) visible images have been used to estimate areal 24-hour total rainfall. On a finer scale, 30-minute rainfall has been estimated using NOAA visible and IR data. The error of the estimates was on the order of a factor of two for both of these studies. Future polar orbiters that carry visible and IR scanners along with microwave sensors could calibrate rainfall estimates from GOES 30-minute VISSR data and thereby provide better temporal rainfall coverage. A simple but effective technique was developed which uses daily satellite visible images of clouds to infer rainfall over land areas during daylight hours. Cloud type, extent, and persistence over a given region are determined from the (low orbiting) satellite images. These factors are used empirically, together with a surface gage network (sparse though it may be) , to "calibrate" the precipitation associated 27 with the observed clouds. Integration of the semidiurnal values and patterns then provides total rainfall. The region to be calibrated is carefully selected to account for orographic and maritime influences. This technique has been used in underdeveloped countries where the number of surface rain gage stations are very limited. The technique is claimed to be capable of providing detailed patterns and accuracies to within 75 percent. Active Microwave (Radar) Microwave radar offers the most promise for observations of rain over land. Using narrow pencil beams and range gating, effects of ground reflectivity can be estimated. Concept studies have found that a high resolution multibeam radar can accurately profile precipitation over land and ocean for rain rates in excess of 40 mm/hr. Multibeam radars may be either of the fixed beam (pushbroom) Dr scanning beam type. The fixed-beam phased array antenna Is a more complex electromagnetic design. However, being stationary, it poses less problem to the spacecraft than a nechanically-scanned antenna. The mechanically-scanned multi- beam radar is the simplest antenna system design, and single ind dual reflector configurations are feasible, but mechanically- scanned antennas generally present considerable momentum- compensation problems. Radars offer unique advantages over radiometers: (a) the precipitation profile in range may be determined rather than just' a height-averaged value; (b) the range of rain rates observable extends from 1 to as much as 200 mm/hr; and (c) system designs have been analyzed that offer reasonably high confidence in rain-over-land applications. These advantages are, of course, acquired at the expense of system complexity, cost, size, and weight. Passive Microwave Radiometry The microwave brightness temperature as observed from a satellite depends on the emission from the Earth's surface as well as that of the intervening atmosphere. The emissivity of the ocean surface is more or less uniform at the 1.55 cm wavelength and has a low value of 0.4. This low emissivity provides the needed contrast to distinguish between the emission from the ocean surface and the emission from atmospheric constituents. In an atmosphere with rain, three major con- stituents contribute to the emission, i.e., molecular oxygen, water vapor, and liquid water droplets. When water droplets are present, they are the main source of scattering of microwave radiation in the atmosphere. The microwave brightness temperature 28 for different rainfall rates can be calculated by numerically solving the radiative transfer equation. Comparisons of simultaneous measurements of rain rate by radar and microwave brightness temperature over the ocean area have been made to confirm the theoretical model. Microwave radiometric observation of rain over the ocean is made possible by the low background brightness temperature of the highly reflective (low emissivity) ocean. Studies have shown that rain areas can often be mapped over land when the ground is wet and temperatures are above 15°C. Algorithms and models presently under development for retrieval of rain from microwave radiometers suggest that improved observation of rain can be achieved with microwave radiometers at both 37 GHz (SMMR and Nimbus-6 ESMR) and 94 GHz. Sampling Errors The present estimate of the passive microwave accuracy is based partly on subjective considerations and partly on observations. Between 1 and 2 mm/hr, a 1-km error in the assumed freezing level can result in an error in rainfall rate of about 20 percent. Beyond this measurement error, a time series limited to a finite number of observations, such as that produced by the ESMR, can result in a sampling error that might even overwhelm the instrumental error for single observations . However, the standard error for satellite-derived monthly mean estimates is reduced by virtue of the number of independent samples obtained over a month's duration. Current estimates are that the monthly mean fluctuation errors over ocean areas are on the order of 8 percent. Another major source of error results from the nonlinear nature of the brightness versus rainfall relationship. Present limitations on antenna dimensions result in a footprint area that exceeds the typical size of rain cells. As a result, the precipitation often does not fill the beam pattern and a bias is produced. Current estimates are that the monthly mean rainfall may be underestimated by an amount of about 60 percent of the true mean. Status of Satellite Precipitation Data Weekly, monthly, seasonally, and annually averaged rainfall maps for ocean areas derived from the Nimbus-5 Electrically Scanned Microwave Radiometer (ESMR) instrument have been published in the form of an atlas. These are in general agree- ment with available conventional rainfall data. They depict known characteristics of global patterns and they suggest new and interesting climatological features, including a potential for predicting the onset and development of the South- east Asia monsoon. 2 9 Because of limitations in measurement accuracy, and others due to methods used in generating the rain atlas, the authors make no claim for accuracy to better than a factor of two, although in a relative sense, they believe the results are much more dependable when changes are considered. In spite of the fact that estimates of the validity of the ESMR-derived rain maps cannot yet be quantitatively stated, and it cannot be said with certainty that their accuracy is either limited by instrumental errors, sampling errors, or indeed a combination of the two, they do appear to be significant and valuable. Several interesting aspects of global climatology are evident from these maps. The monthly maps reveal the displace- ment of rain belts and changes in intensity of the ITCZ rain pattern (the forking in the Pacific with the southern limb merging as it runs southeast with the polar front) throws further light on the observations of a dry zone near the Ellice Islands in the mid-Pacific. The rainfall rate over the oceans appears to be less in the Southern Hemisphere than in the Northern Hemisphere, particularly in the Atlantic. There appears to be a regular evolution of the maxima with the progression of the seasons. In the Pacific, the ITCZ shows a general displacement to the north of the Equator and is more sharply defined, and its seasonal movement more pronounced, in the Atlantic. The polar front near 40ON appears more active from September through November, whereas its weaker complement in the Southern Hemis- phere is most active from March through May. Two main maxima appear in the Indian Ocean, the one at higher latitude growing at the expense of the other at lower latitudes as the monsoon advances r and vice versa as it retreats. The weekly maps portray the onset and retreat of the South- west monsoon. According to the official Indian weather summary, in 1973 the date of onset of the monsoon to penisular India was June 10, and the date of withdrawal from the main part of that region was October 16. This closely corresponds to the ESMR data and indicates the feasibility of satellite monitoring to predict its onset and later development. Precipitation estimates are currently made entirely on a research basis at NOAA, but are evolving into an integral part of a newly developing flash-flood warning system. NESS in cooperation with NWS is using the Oliver-Scof ield technique based on GOES/VISSR data at half-hourly intervals. The Griffith- Woodley technique is used in Florida. None of the existing techniques have been adequately evaluated, although comparisons with individual storms have been made. GOES IR data on which these techniques are based are currently archived by NOAA/EDIS. 30 Data Management Strategy While all current techniques for the estimation of precipitation from satellites are still in the realm of research, enough promise has been demonstrated to permit consideration of a strategy for management of the data. At the same time, it is senseless to expend large amounts of money on processing and storing data whose reliability is questionable and whose utility is yet unknown. A reasonable estimate for careful analysis of the ESMR-5 data is in the neighborhood of one million dollars. While this is a small amount compared to the cost for its original collection, it is still considerable and would take resources that could otherwise be devoted to other data. As a result, a detailed plan needs to be developed based on an overall strategy leading from initial assessment and validation to final processing and archiving only when the value has been established at each step. Case studies and physical process studies may be individually justifiable and in general do not require treatment of the entire data set. On the other hand, climatic data sets involving temporal and spatial averages must take into account the finite sampling limitations of the satellite data. In addition to the temporal sampling, satellite measurements are spatial averages over the area in view and validations through ground measurements are difficult. As the resolution capabilities of polar-orbiting remote sensors are improved to provide the kind of information needed to describe the processes accurately on a local, regional, or global level, the volume of data will grow considerably. With global coverage repeated one or two times per day (visible vs IR and microwave) , the quantities of data available can soon become overwhelming. (See chart on following page.) All too often in the past, the data processing and management systems of satellite missions have been neglected. Initially, this neglect probably resulted from the failure to recognize the magnitude of the problem. Once the data were collected no provisions were made to fully process, archive, and distribute them to users. Too often, overruns in the costs of hardware systems drained the funds budgeted for data processing. Thus, we find ourselves with thousands of magnetic tapes containing satellite observations which are not fully processed and cannot be retrieved readily. It is true that some of these data are not suitable either because of limitations in the instrument, the orbit, or the spacecraft. In other cases, a large and expensive effort must be mounted to process data which meet the accuracy and coverage requirements . In any case, recognition and acceptance of the true costs of data processing and management must be a part of the planning of such 31 missions. It will only be possible to reap the true potential of remote sensing capabilities when the large body of data is made available for applications to the problems of climate. DATA RATES Data Swath Satellite Sensors Channels Rate (b/s) Width Resolution Current Nimbus-7 SMMR 5 2K 8 00 km 30-150 km Seasat-A SAR 1 15-24 M 100 km 2 5 km TIROS-N AVHRR 4,5 600 K horiz to horiz 1 km GOES VISSR 2 28 M full disc 1 km, VIS 8 km IR Under Study Imaging radar 21 cm 50 M 100-200 25-100 M 1980s L-band km Multichannel K to low 100 km 20-50 km Passive P band Microwave Recommendations The possibility of estimating rainfall from satellite data on a global basis exists through the use of visible and infrared imagery, and active and passive microwave radiometry. No other reasonable possibility now exists for obtaining comprehensive coverage over the world oceans. Even over land, the conventional point precipitation networks are generally inadequate to resolve small scale variations to the level required for reliable areally averaged values. Microwave (e.g., ESMR) techniques for precipitation estimation may be suitable for global studies but unsuitable for certain local hydrologic forecasts. IR or cloud-top temperature techniques, such as the Oliver-Schof ield or Griff ith- Woodley methods are most applicable to convective rainfall situations, and may be most effective for local and regional studies 32 1 . Develop sites for testing satellite techniques fo r estimating precipitation . Since these methods are still in the development state, it is essential that the accuracy and sphere of usefulness of existing techniques for satellite estimation of pre- cipitation be evaluated and objectively documented. Test sites for carrying out this verification over land areas should be designated for several climatic zones (tropical, temperate, arctic, etc.). The time/space frequency of surface precipitation observations should be comparable to that of the GATE A/B array. In deference to fiscal constraints, these test sites should be selected from currently well-instrumented test areas wherever possible. Examples over the United States are (1) the South Dakota State University weather radar and rain gage network; (2) the precipitation gage network at Chickashaw, Oklahoma; and (3) the flash flood network being developed by NWS in the Appalachian area. As an initial step in testing these methods over ocean areas, further studies should be made using the GATE ESMR and radar data to first determine the capability of ESMR to provide rainfall rates over oceans, and from this determine the desirability of processing the remaining ESMR data. 2 . Develop a precipitation data base for use in satellite systems design . A need exists for basic data on the statistical structure, in both time and space, of precipitating systems for use in satellite/sensor systems design. As a result of this shortage of climatological data, a reasonable basis for the selection of a microwave antenna size (and corresponding beamwidth and ground resolution) is lacking. It is therefore recommended that the most interested agencies, NOAA and NASA, take the lead in developing the requirements for such a data set, and take steps to assemble or acquire the required data. SNOW AND ICE DATA Introduction Snow and ice cover in polar regions is recognized to be a sensitive index of global climate. Study of the processes involved in atmosphere/surface exchanges involving the 33 cryosphere are identified as a significant problem area for Climate Program research. Major seasonal changes in the extent, thickness, and surface characteristics of snow cover and sea ice influence the energy fluxes at the interface which, in turn, contribute to the forcing of both atmospheric circulation and local mesoscale systems. Long-term changes in sea and land ice cover, including the glacial maxima and so-called Little Ice Age(s), reflect global climatic states rather different from those of the present day. Investigation of the processes involved in these inter- actions, and of their spatial and temporal scales, will require adequate digital, gridded records of the major cryospheric parameters. For energy exchange studies at the snow/ice surface that involve the atmospheric and the oceanic boundary layers, synchronous data sets will be needed, such as synoptic atmospheric and oceanic soundings, and measurements of horizontal transport. Data Requirements In the Climate Program, cryospheric data may be used in: Climate research Boundary conditions for input to models (GCMs and SDMs) (snow/ice, sea ice extent, concentration, thickness, age, surface characteristics needed for albedo and roughness; land ice elevations.) Short-term monitoring and prediction of climate and surface hydrology Daily to seasonal variations in extent, depth, water equivalent, and free water content of snow cover; extent, concentration, and thickness of sea ice; pack ice and iceberg drift Long-term monitoring of climate Climatic fluctuations due to natural and anthropogenic causes, extent and thickness of snow cover, glaciers, sea ice, and permafrost. Design/planning assessments Baseline data on cryospheric variables; cryospheric hazards related to weather and climate (avalanches, icebergs, surging glaciers, jokulhlaups, freshwater or sea ice damage to structures.) 34 The specific variables and required resolutions, accuracy and frequency of observations are summarized in Table 1 below (modified from WMO, 1976). For some purposes, information with less than the stated accuracy is also useful. Table 1. Data Requirements for Cryospheric Parameters Variable Horizontal Resolution* (km) Accuracy Frequency A B C D X Snow cover extent 0.3 1.0 3.0 50 1-5 + 5% 1-3 day Snow cover water equiv. 0.1 0.3 1.0 50 1-5 + 10% (3cm) 1-3 day Snow cover free water 0.1 0.3 1.0 50 + 10% (3cm) 1-3 day Sea ice extent/con- centration 50 10-50 + 2% 1-3 day Sea ice thickness 50 + 10% (20cm) 1-3 day Glaciers/ice sheets extent 0.1 1.0 3 50 1-5 + 5% Yearly Ice sheet thickness, surface elevation 0.1 1.0 3 50 1-5 + 5% dm) Yearly Ice sheet horizontal velocity Point + 10% (5 0m/yr) Yearly Permafrost extent °- 1 0.3 v'° 50 1-5 + 5% Yearly *Plrai nano ha c i r> c: < -Ton l-m'Z (&} • i nr -i nnn Vm^ WRl->innn i mZ (D • Global climate studies (D) ; Impact Assessment (X) 35 Data Status The major sources of cryospheric data are: snow cover - ground observations (synoptic and climate stations, snow courses, special programs) - aircraft remote sensing (gamma ray) - satellite remote sensing (visible and IR, microwave; composite minimum brightness) - coastal or ship observations - aircraft reconnaissance and remote sensing (active and passive microwave) - satellite remote sensing (visible and IR, active and passive microwave) glaciers and ice sheets - ground observations (glacier fluctuations and mass balance, geophysical profiling) - aircraft remote sensing (radio-echo sounding) - satellite remote sensing (visible and IR, microwave, and laser and radar altimetry) Satellite data on global snow and ice cover exist since 1966/67, supplemented by ground observations of snow cover or sea ice for tens of years, upwards to a century or more at a few stations, mainly in Europe. These conventional records are not yet centrally available, and most of the satellite data are only in map form. Table 2 summarizes the major national and international data sets on regional to global scales. Ancillary information on surface properties and vegetation types needed for albedo analyses and as modeling inputs are relatively poorly known and are not routinely mapped. Digiti- zation of the basic data sets is just beginning. G. Kukla (Lamont-Doherty Geological Observatory) is preparing snow cover and sea ice and point digital data from NESS and U.S. Navy data. J. Walsh (U. of Illinois) has arctic sea ice concentra- tions for 1953-77. Other gridded ice data are being prepared by M. Kelly (U. of East Anglia) and K. Hasselmann (Max Planck Institute, Hamburg) from 1901. 36 Table 2. Summary of Major Snow Cover and Sea Ice Data Sets Location Parameter Period Source Global Snow depth maps 1975 USAF GWC N. Hemis- Snow/ice boundaries 1966 NESS sphere and 3 brightness classes Global Composite minimum 1968 NESS brightness maps (format variable) U.S. Depth of snow maps 1935 NOAA (Dec-March) (Weather and Crop Bulletin) U.S. Snow pack water 1930s SCS, USDA 1978 Snotel System Canada Monthly snow depth Environ- ment , Canada Global Sea ice limits 1970 U.S. Navy and concentrations (Weekly) Global Sea ice boundary and concentrations (Microwave bright- ness temperature) Dec 1972 NASA Canadian Sea ice to 70°N 1958 Environ- Arctic Sea ice to 80°N Late May-Oct ment, 1960 Canada N. Hemis- Sea ice limits and 1960 British sphere concentrations 1957 Meteor- ology Met. Inst. 37 • Archiving There is no national center for snow and ice data. Most sets of global snow and ice cover data are archived by the (operational) agencies responsible for producing them. These include NESS, NWS, NOAA-ERL, NASA, U.S. Navy, Air Force, and USDA. NCC (SDSB) archives pertinent satellite imagery, and climate records at NCC in Asheville contain much information on snow cover in the United States. Canadian snow cover and sea ice data are being organized into data sets and atlas information by the Atmospheric Environment Service, Canada. World Data Center-A for Glaciology (Snow and Ice) in Boulder, Colorado is currently undertaking an inventory study of snow cover and sea ice data sets. Glacier and ice sheet data are the focus of the following activities : (a) World Glacier Inventory being conducted by a Temporary Technical Secretariat of IHP (UNESCO AND UNEP) , located in Zurich. (b) Permanent Service on the Fluctuations of Glaciers (including mass balance data) - data published in three volumes (1959-75) , also located in Zurich. (c) Satellite image atlas of glaciers of the world, being prepared by USGS. (d) Ice Core Data - currently being inventoried by World Data Center-A for Glaciology. Data Management Strategy Cryospheric data exist currently in scattered locations in diverse forms; literature references, hard copy lists (often unpublished) of point data, imagery, and chart series of sea ice and snow cover. In a few selected areas digitization is beginning. However, the immediate need is to develop inventories as a basis for developing data bases. From the standpoint of the Climate Program, the priorities should be as follows for candidate data sets: • Studies of present climate and physical processes : Elements Variables Priorities Priorities Snow Cover Extent, thickness, water equivalent, water content, surface temperature 38 Sea Ice Extent, concentration, thickness, age structure, surface features Land Ice Extent, surface elevation, thickness Permafrost Extent, active layer depth, thickness Studies of climatic change : Elements Variables Priorities Priorities Ice Cones Isotopes, particles, stratigraphy, chemistry Land Ice Extent, mass balance, thickness Sea Ice Extent Snow Cover/ Extent/occurrence Snowfall Permafrost Temperature profiles, extent, thickness The Climate Program will require gridded data with time resolution as stated in Table 1, especially for snow cover and sea ice. A free format is desirable so that overlay of cryospheric/atmospheric/oceanic/terrestrial parameters in various geographical displays is possible. For climatic change studies, time series of quality-controlled historical records, mainly at point sites, are especially important. The annual records from ice cores are a high priority item. The dating techniques used to establish the chronology are critical. The use of polar-orbiting satellites to gather data on snow, ice, and sea ice has resulted in an unprecedented accumu- lation of data. Users requesting ground-based data collections ought to be alerted to the availability of concurrent satellite data . Recommendations 1. Operational snow and ice chart products should be evaluated and subsequently digitized. 2 . The ESMR microwave observations of polar area snow and sea ice should be evaluated, recognizing the valuable property of microwave sensors to penetrate cloud cover. 39 3. High priority should first be given to continuing the development of inventories for the major variables characterizing global snow cover and sea ice, and, second, to developing digital data bases in grid format. WDC-A for glaciology is already involved in the first state. 4. The new series of 7-day Composite Minimum Brightness (CMB) charts from NOAA/NESS ought to be recorded in digital form at NESS prior to archiving at EDIS. 5. Support must be given to make available geographic data on terrain characteristics, vegetation cover, land use, etc., and data on temperature, radiation, wind speed, etc., that are necessary for studies involving interaction with snow cover. 6. To facilitate international exchange of data, national agencies are encouraged to send data, information, and user requests to the WDC-A for glaciology (snow and ice) . Other specific recommendations are contained in the report "Workshop on Mapping and Archiving of Data on Snow Cover and Sea Ice" (See Glaciological Data, Rpt. GD-5, 1979, in press.) REQUIREMENTS FOR MODELING SURFACE HYDROLOGIC PROCESSES As indicated in the National Climate Program - 5-Year Plan , there are five physical ingredients which are fundamental to the hydrologic behavior of a land surface: 1. Atmospheric Evaporative Capacity - the potential rate of evaporation as a function of the gradient in water vapor mixing ratio, of wind speed, and of surface roughness. 2. Evapotranspiration Efficiency - the ratio of actual to potential rates of evapotranspiration as a function of the soil moisture concentration, the hydraulic properties of the soil, and the density and species of vegetation. 3. Surface Temperature - the result of the energy balance in the surface layer. 4. Soil Moisture Concentration - the result of a water mass balance in the vegetal root zone. 5. Runoff - the sum of surface runoff and gravitational percolation as a function of precipitation, soil properties, and soil moisture concentration. 4 When snow-covered surfaces exist, a sixth ingredient can be added: 6. Density of Ripe Snow - the density of snow (containing the maximum amount of liquid water) as a function of the thermal quality of the snow. To adequately develop, test, and verify climate models that include a dynamically coupled land-atmosphere system which incorporates the preceding physical constituents, the following kinds of data are needed: Global Data Sets - (grid-scale averages) • vegetal canopy density (seasonal variation) • species or broader biological grouping to differentiate water demand characteristics • hydraulic and thermal properties of soil — • effective porosity — saturated hydraulic conductivity -- residual soil moisture — thermal conductivity — specific head (2) • albedo (seasonal variation) • surface temperature (seasonal variation) (3) • snow cover (seasonal variation) (3) snowpack water equivalent (seasonal variation) ice cover (seasonal var urbanization density^ ' (3} ice cover (seasonal variation) v ' cultivated agriculture density storm precipitation statistics ^ ' — storm duration — time between storms — storm intensity — storm depth Addressed by Working Group 6: Geographical, Land Use, and Assessment Data. (1) (2) Addressed by Working Group 4: Radiation, Physics, and Chemistry ' Addressed by other elements of Working Group 2. 41 To specifically verify parameterizations, coordinated data sets at grid scale within major river basins are needed. The data necessary are: • all types listed under GLOBAL DATA SETS • soil moisture in the surface boundary layer • streamflow Data in support of many of these requirements exist in various forms throughout the world. The key issue then is to identify these data sources and in some manner coordinate their availability and retrieval for general use. Also, in view of the large spatial scales involved when modeling on a global basis, the promise of remotely sensed data must continue to be explored. Satellite observations will provide some of the spatial integrations that are necessary for grid-scale parameterizations . Status of Soil Moisture Data About 200 soil-vegetation records have been uncovered in a recent survey of available soil moisture data. They include several studies mainly at SEA-AR research stations scattered throughout the United States. A 5-year soil moisture pilot study is proposed by the Soil Conservation Service (SCS) in cooperation with SEA-AR and State Agricultural Experiment Stations . Soil moisture and climatic data are used in the SPAW soil moisture model to predict soil water consumption by various crops during the growing season. The major data requirements of this model include the following: • daily rainfall • open pan evaporation • daily net radiation • daytime wind travel • weekly soil moisture (for development and verification) • estimated rooting depth Hydraulic and Thermal Propertie s Data that provides measurements of effective porosity, saturated hydraulic conductivity, and residual soil moisture 42 are available in many technical publications, but generally have not been made a part of a national data base. Much of the research has been done independently by university and Federal researchers. Data generated at the National Soil Survey Laboratory in recent years have been stored in a soil data base system. Major needs for these kinds of data can best be identified only when the available data are made a part of a data management system. Data on thermal conductivity of soil-water systems and surface temperature (seasonal variations) are very limited. Coordinated data sets in major river basins are needed. Recommendations Several Federal and State agencies are independently undertaking the tasks of digitizing soil survey maps, and construction of computer-oriented soil data base systems. The purpose, scope, and execution of each agency's efforts are varied, as are the levels of financial support. Therefore, to optimize the utility of soils data to the National Climate Program, the working group recommends a coordinated national emphasis on the development of soil data base management systems. No new data collection programs are proposed by this recommendation Rather, existing information found in tabular and map form should be digitized and formatted for interactive availability. Additionally, the soil from which data have been acquired should be properly classified before the data are archived. Soils data from local, regional, and international sources should be included as soon as they are validated. It is recommended that the USDA take the lead role in this effort. Soil moisture is a critical component in establishing a statement of water balance in the land surface boundary layer. Therefore, it is recommended that time series of soil moisture data be centrally archived and made available for general use through interactive computer systems. It is recommended that the USDA take the lead role in this effort. Recognizing the difficulty in obtaining useful soil moisture information on a global basis, it is further recommended that the National Climate Program Office be a focal point for coordinating and sponsoring research programs for the continued development of technologies to remotely sense soil moisture data on an international scale. 43 OCEAN EVAPORATION Data Requiremen ts There is little likelihood of direct measurement of evaporation over the world oceans in the foreseeable future. Rather, parameterization methods, using ship deck level vapor pressure and wind speed, and sea-surface temperature will continue to be the prime method for obtaining routine estimates of oceanic evaporation. The time/space resolution of these data should be adequate to resolve synoptic scale atmospheric variations. The most stringent requirement for sea-surface temperature observations is the resolution of the strong grad- ients in the neighborhood of the Gulf Stream and Kuroshio. General Status of the NCC Marine Data Base Currently, the data for estimating oceanic evaporation is derived primarily from surface ship observations, i.e., the Marine Data Base. The Marine Data Base at NCC is one of the most complicated and least organized of all NCC files. Problems associated with marine files stem from a number of factors. Enthusiasm and support for marine data has been sporadic. In 1973, after a major funding effort to produce an updated marine atlas, the Navy terminated funding and as a result data on forms began to pile up in boxes. In 1976, the Navy and NOAA jointly renewed support of marine data processing but on a much smaller scale. In order to have as complete a file as possible, NCC obtains marine data from numerous sources including NMC , the U.S. Air Force Global Weather Central, the Navy, and foreign countries. This means that NCC must eliminate duplicate observations from all of these sources. It also means the NCC is receiving observation months and even years after they are taken. There are over 1,500 tapes in the marine file and nearly 300 in what NCC terms its main library. Because the main library data are sorted by area, all 300 tapes must be mounted when new data are received for all areas. The best and most comprehensive marine files are the so-called Marine Atlas Files (generated for the Navy) and the Historical Sea Surface Temperature (HSST) File. The Atlas files cover the following periods for each ocean: 44 North Atlantic 1850-1972 Indian 1850-1973 North Pacific 1850-1974 South Atlantic 1850-1976 South Pacific 1850-1978 The HSST File contains bucket SST observations for 1860-1960. HSST processing for the Atlantic Ocean has not yet been completed. Availability of Data to Compute Evaporation To compute evaporation from the oceans, the bulk aero- dynamic equations can be used provided air temperature, wet bulb temperature, and wind speed are available for ship deck levels and the sea-surface temperatures are also available. No comprehensive inventories of the observation variables exist. However, temperature, wind speed, and sea-surface temperature observations are generally available in all observations going back to the late 1800s. When NCC produced a marine atlas in 1954 based on observations from before 1900 to 1952, only 14 percent of the observations contained wet bulb temperatures, and a cursory check has shown that most of these occurred after 1940. When NCC produced another marine atlas in 1973, the percent of observations containing wet bulb temperatures increased to approximately 45. These statistics provide no information on the quality of the observations. Data Management Strategy The NCC has set up a task team to design a new data processing scheme using the systems analysis approach. The processing scheme will be used to reprocess historical marine files as well as incoming data. The NCC hopes to eliminate the numerous marine files now in existence and to create two cleaned up files, one in a synoptic sort and the other in an area sort. At the same time detailed inventories will be generated to the element level about quantity and quality. Because this is viewed as a 3- to 5-year task at a time when the data are needed, an intermediate objective is being defined. The marine atlas files represent a relatively good data base for the period from the 18 50s to approximately 1973. From 1973 to the present, the data are in several files which cannot readily be used to service customers. For this reason, the NCC plans to process and merge the data in this latter period first in an attempt to provide marine data which compliments the pre-197 3 data. 45 ATMOSPHERIC HYDROLOGIC DATA INTRODUCTION The vital common element of hydrology and meteorology is water. Both sciences are concerned with synoptic observations of water in its various forms, and with the movement and phase changes of water through natural processes. The transfer of water at the air-surface interface is of prime concern to both hydrologists and meteorologists. A similar common interest between oceanographers and meteorologists exists at the air-sea interface. Evapotranspiration serves as an all-important "input" to the atmosphere in the determination of weather and climate, while representing "output" or water loss from the hydrological or oceanographic point of view. Conversely, precipitation is atmospheric "output" while representing the replenishment of the water resource to hydrologists. It therefore seems clear that consideration of the continents, oceans, and atmosphere as interacting components of a single system is fundamental to a basic understanding of the role of the global hydrological cycle in climate and climate change. The equation for the water balance of the earth, which is derived from the more general statement of the continuity of water substance, has served as a central concept for hydro- logists. A similar equation can be derived for the water balance of the atmosphere. During the past few decades an improving network of aerological stations has produced progressively more detailed and accurate measurements of the various terms of this balance equation. It has been clearly demonstrated by a number of studies that the atmospheric water balance can be accurately evaluated, but the smallest areas and shortest time periods for which useful computations can be made are determined by the density of observation stations, the accuracy of observations, the frequency of sampling, and the local terrain. DATA STATUS Atmospheric Moisture Flux Computations from Rawinsonde Data The potential use of satellite data for the evaluation of the atmospheric water balance will be discussed in the next section. However, it appears that rawinsonde measure- ments of wind and humidity will continue to serve as the primary data source for atmospheric moisture and moisture flux 46 computations over the land areas of the earth for some time to come. The operational rawinsonde network is primarily designed to resolve synoptic scale (1,000 km) meteorological features. This spatial resolution is compatible to that of existing general circulation models of the atmosphere. The operational hydrologist is primarily interested in basins smaller than 105 km^ ; scales which cannot be adequately resolved by the rawinsonde network. Nevertheless, the larger scale studies can provide information useful in the calibration of empirical relationships which can in turn be applied to the smaller basins. The network of operational aerological stations over North America, Eurasia, and Australia is adequate for a number of important studies involving the atmospheric branch of the hydrologic cycle. The upper air network over the remaining areas of Earth severely limits the range of studies which can be undertaken. Some improvement in this situation can be expected for the augmented network of global observations during the current Global Weather Experiment. Recent field experiments, which began in 1969 with the Atlantic Tradewind Expedition (ATEX) and the Barbados Oceano- graphic and Meteorological Experiment (BOMEX) , provide data for a new generation of atmospheric water budget studies. Unlike data from the routine operational network, the field experiments were carefully designed to provide data for meaningful budget computations over relatively small areas, and short time periods. Specifically designed or calibrated instrumentation was used, and observational arrays and schedules were tailored to provide answers to the specific problems under investigation. Rawinsonde observations are sometimes augmented by aircraft observations and data from special boundary layer observational systems. Although of limited extent in both time and space, these special data sets provide useful information on many aspects of the atmospheric hydrologic cycle. A number of important problems must be carefully considered when developing a rawinsonde data base specifically tailored to the rigid requirements of atmospheric water balance computations : 1. The accurate evaluation of the atmospheric water balance requires relatively high vertical resolution of the specific humidity and wind profiles in the lowest few hundred millibars of the atmosphere. Data only at the mandatory reporting levels (1,000 mb , 850 mb , 700 mb) are often inadequate for these computations. Fifty or even 25 mb resolution is sometimes required from the surface to 900 or 800 mb to define the profiles. The 5-year MIT data set, which has been the data source for many 47 of the global scale water balance studies, consists of data at 50-mb intervals for Canadian and United States controlled stations, but only mandatory level data, without surface observations, are available for most of the remaining stations. Efforts have been made by some investigators to supplement these data with low altitude climatological data. 2. The monthly average values of aerological humidity parameters included in the MIT data base were computed as averages of the available observations during the month. Thus a station average for any particular month may be based on anything from a few observations to a full set of observations. Since hydrological events have a highly non-uniform distribution throughout the month, e.g., the major vapor flux convergence may be associated with two or three major precipitation events, the lack of a full set of observations may lead to significant errors in individual monthly values. In order to obtain accurate monthly values, it is necessary to begin with relatively complete station data sets, and apply methods of interpolation for observations which are missing. 3. Prior to the 1960s, operational rawinsonde humidity measurements exhibited known deficiencies. The change of U.S. rawinsonde humidity sensors from a lithium chloride to a carbon element in the early 1960s was expected to significantly improve the quality of the observations, but deficiencies in the design of the rawinsonde package, not discovered until the late 1960s, led to significant errors in humidity measurements during ascents when the instrument was exposed to direct sunlight. Thus, the daytime observations from U.S. or U . S . -controlled stations are of uncertain quality during the period when this instrument package was in use (early 1960s - 1970) . Documentation of and adjustment for data biases such as this is critical if atmospheric vapor flux data are to be effectively used in climate studies. Application of Satellite Products Two types of meteorological satellites currently provide operational products and research data for a wide variety of users. Geostationary satellites are positioned over the equator at a fixed point above the Earth's surface. Polar orbiting satellites are in sun synchronous orbits, with periods of slightly less than 2 hours. Operational meteorological satellites currently provide remote sensing of the atmosphere in the visible and infrared, and Nimbus experimental satellites provide remote sensing of the atmospheric in the infrared and microwave portions of the electromagnetic spectrum. 48 Wind Estimation Imagery from geostationary satellites have provided meteorologists with a method for estimating winds from satellite data. Several pictures taken during the lifetimes of trackable cloud patches are used to obtain cloud motion vectors To the extent that these clouds are advected by the wind, the method provides an estimate of the wind velocity at cloud level . Since the geostationary satellites are positioned over the equator, computations of this type are generally limited to a 35-degree latitude band on either side of the equator. Within this latitude band, computations have generally been limited to two levels: (1) a low level at which the cumulus patches can be used as tracers and (2) a high level, at which cirrus can be used as the cloud tracer. The accuracy of these wind estimates has been difficult to assess due to the lack of adequate "ground truth" with which they can be compared. However, a recent study by Hubert and Thomassel indicates that NESS winds computed from low level cloud vectors contain errors of 4.7 m s~l. Use of more sophisticated methods can, however, reduce the errors. Never- theless, this method has a number of features which limits its value in providing wind data for vapor flux computations. These factors are enumerated below: 1. Vertical resolution As previously discussed, a minimum vertical resolution of 50-100 mb is required for vapor flux computations. The cloud motion vectors provide data for vapor flux computations primarily at a single low level, since the cirrus level winds are too high to be of much practical value in vapor flux computations. In the past, when only picture data were available, it was difficult to determine even the height of the clouds, i.e., the level to which to attribute the motion vector. This situation has improved considerably with the inclusion of infrared sensors on the SMS satellite, which can be used to determine cloud top temperature, and thus, indirectly, cloud height. Using these data, it may also be possible to derive some additional cloud motion vectors at different levels near disturbances. 2. Biased statistics Computations are obviously made over areas where tracer clouds can be observed. For instance, good low level cloud tracers frequently cannot be discerned within 49 disturbances, where middle and upper clouds obscure the lower layers. As a result low level cloud vectors are derived only in undisturbed areas, and on the periphery of disturbances. Such observations, interpolated across the active portion of disturbances, may severely under- estimate the intensity of the system, and lead to large errors in the computed divergence field. Thus, increasing the density of cloud vector computations does not necessarily increase the meteorological information in disturbed areas. 3. Random sampling Coupled with the biased sampling discussed above is the general problem which random sampling introduces in the computation of statistics for a fixed drainage basin. Random observations are difficult to summarize without accumulating grid point statistics from a map analysis for each observational time. Our conclusion is that data derived from this method of wind estimation, in its current state of development, is of secondary value for vapor flux computations. Even when used as supplementary data at a single low level, the additional wind estimates may do more harm than good if they result in a biased sampling of synoptic conditions. This statement in no way reflects on the value of these wind estimates for defining the low level synoptic patterns for more general meteorological analyses over data-sparse areas. Humidity Data Estimates of specific humidity can be obtained from infrared data, and estimates of both liquid water and water vapor can be obtained from microwave data currently being collected from polar obiting satellites. These measurements are made by vertical sounders, which are basically instrument packages designed to measure radiances in a number of spectral regions corresponding to various layers of the atmosphere. The measurements can be used to derive atmospheric temperature and moisture profiles of the radiating columns. Instrument packages currently in orbit scan only a finite distance across the orbital track, resulting in gaps at low latitudes between successive orbital passes. The satellite radiance measurement is closely related to the total integrated water vapor content of a volume of atmosphere several kilometers in both the horizontal and vertical dimension. On the other hand, the radiosonde measures the water vapor content of small parcels of the atmosphere. Although the radiosonde is capable of giving SO much higher vertical resolution than the remote satellite sounder (meters compared to kilometers) , the satellite sounding system has the potential for achieving much higher spatial resolution and more meaningful horizontal gradients than that achievable with the current operational rawinsonde network. The fact that the same instrument observes at all geographical locations contributes to the relative accuracy of the satellite. The observational systems and computational techniques currently in use appear to be sufficient for computing only the vertically integrated water vapor content to a reasonable degree of accuracy. Even then, problems arise in the use of infrared data for this purpose in the presence of clouds. On the other hand, microwaves are unique in their ability to yield measurements of tropospheric water vapor in the presence of clouds. Unfortunately, because of the high emissivity of the land surface, the use of passive microwave radiometry is limited over land. GENERAL DATA MANAGEMENT STRATEGY Data Types To build a data base specifically designed for evalua- ting the atmospheric hydrologic cycle, the following types of atmospheric data should be routinely archived: • Rawinsonde data — wind, temperature, and humidity; surface to at least 300 mb; 50 mb vertical resolution highly desirable, but decreased resolution acceptable above 700 mb. • Surface synoptic meteorological data — at least 6 -hour ly resolution. Wind, temperature, humidity, surface pressure, cloud cover, weather, and precipitation amount. • Satellite data — derived humidity and wind data as available. Quality Control Past experience has demonstrated the necessity of quality control over the above, exercised on the operational data 51 when these data are to be used for computing sensitive, derived quantities such as vapor flux divergence. It is sense- less to expend large amounts of money on processing and storing atmospheric hydrologic data whose reliability is questionable and whose utility is correspondingly compromised. A well- conceived and implemented program of data validation may spell the difference between success and failure for many applications. This validation should include the following subtasks: 1. Maintenance of a historical documentation file which includes for each observing station: (a) A station description, including a description of the type of instrumentaton used and the period during which it was operational. The description should include the data characteristics and data collection rate, and the associated methods of processing. (b) Results from intercomparison programs. (c) Description of normalization adjustments applied to data from each sensor or system type. The maintenance of a file of this type and the proper normaliz- ation of current data requires a continuous monitoring of the observational network for changes in instrumentation, changes in procedures, or changes in exposure or station location The file should serve as a part of the basic data documentation and be available to the individual investigator. 2. Gross quality control analysis of the basic data in order to identify and flag or correct obvious errors. The basic data archive should include the normalization adjustments. These should be included in a manner which gives the user the option of applying them, or modifying them in the light of later studies and information. CONCLUDING REMARKS The most general application of vapor flux data is in the routine computation of regional water balances, and the sub- sequent development of monthly, seasonal, and long-term regional water balance climatologies. In their most complete form these computations consist of the simultaneous evaluation of the individual terms of the terrestrial and atmospheric water balance equations in order to obtain estimates of area-averaged evaporation and total surface and subsurface storage change. 52 The accumulation of these basic data and derived data products results in a data base of considerable value. A multi-year data base of this type can be used for such diverse projects as (1) examination of the variability of the hydrometeorological parameters and its relationship to such problems as drought, precipitation augmentation, and other water supply problems; (2) development of hydrologic and heat balance models, including the development, testing, and calibration of empirical methods for estimating areally averaged evapotranspiration; and (3) base- line data for environmental impact and climatic change studies. The use of atmospheric flux data for projects with more limited objectives may often be desirable. Examples of such studies are: 1. Evaluation of the magnitude of the suspected negative bias in the values of "measured" precipitation. Cases would be selected for which the error in estimation of the areally averaged evaporation would be a small percentage of the total precipitation. Examples would be composite balances computed from data covering a number of short periods spanning major rainstorms, or periods of heavy snowfall. 2. Estimates of regional snowpack. Computations of this type may be of value for major mountain areas or remote regions for which there is an adequate distribution of rawinsonde stations on the perimeter. Reasonably accurate independent estimates of evapotranspiration during the winter season are required for these computations. 3. Evaluation of the atmospheric water balance over oceanic regions or large inland water bodies which exert a direct effect on the hydrology of the adjacent region, e.g., Sea of Japan, Great Lakes. Within the framework of current World Weather Watch (WWW) observational programs, the rawinsonde remains the primary observational tool for obtaining the humidity and wind data required for the computation of the vapor flux and vapor flux divergence. Rawinsonde data are typically archived only for the mandatory levels, i.e., 1,000, 850, 700, 500 mb... . The accurate evaluation of atmospheric water balance quantities often requires a surface observation, and at least a 50-mb resolution of the wind and moisture profiles in the lowest few hundred millibars of the atmosphere. The use of these data is further complicated by the existence of system biases in the moisture measurements, which creates difficulties when using data acquired by a number of different national measure- ment systems. A well-known example is the negative bias in the daytime humidity measurements obtained from the U.S. rawinsonde in use during most of the 1960s. 53 In the current "state of the art," meteorological satellite products appear to be of value for obtaining estimates of total water vapor content, but appear to be of marginal value for vapor flux computations. Their major shortcomings at this time are inadequate derived wind data, and inadequate vertical resolution of the humidity profile. Progress continues in this field, and continuing research in the application of satellite data to atmospheric water balance computations should be monitored and encouraged. RECOMMENDATION The operational rawinsonde network continues to provide the basic moisture and wind data for evaluating the atmospheric hydrologic cycle. Because of special data problems and require- ments, the currently archived rawinsonde data is often inadequate or only marginally adequate for these computations. In order to overcome the major shortcomings, it is recommended that (1) the corresponding surface observation be archived with the upper level data; (2) data be regularly computed, transmitted, and archived at 50-mb intervals, at least for the layer below 850 mb; and (3) documentation of instrument type, including humidity element, known biases, and established methods for correcting the data be included as an integral part of the rawinsonde data archive. 54 5-1 00 cu r- >1 X\ •P X an -H CU fd rH -P 5^ u •H 5-i tn cu .Q ,3 rd > C •H rd -P 5-1 4-1 0) u fd o o > tr cu c H £ -H rd -P T3 u 'O ■H tn a; w C A >1 ■H CO .c £ a 0) .C rC 10 -P -p >1Ti 01 M CH CU fO £> tn rd rd g P e fd rH •H (D rd n > -p . -P CO CO CO ■H ■H CU 0) 4H •H rH C U a, 3 cu e o •H £ -P CO o cu fd -h £ g 4^ . -p 5-1 rt rH . 4H 4-1 rd < C -H • ■H CO cu U CO £ cu -P u U g fd p 3 g CO CO fd o U cu CO cu • & CO -P *0 < c fl Z -H O rd rd CM w PI D O H h 55 CN _ C I _< < 2 5S *C k 9 O < cr 1 co $ h- ^ <0 CO < UJ 0- co LJL TER OF ENTS: 5 QC o -J < CO -J CO O ■ > LU ■ ■ ICAL DIAME FACE CURRl \% K Co c •H u u o 4-1 CO •H CO a c a c CN M 56 OCEAN DATA Working Group 3 Chairman: Members : Kent Hughes James Churgin Philip Cohen Wayne Darnell Steacy Hicks Merton Ingham Sydney Levitus Douglas McLain Stephen Pazan Robert Quayle Vince Zegowitz SUMMARY The Ocean Data Working Group was comprised of ocean data users, data managers, and data generators representing several U.S. government agencies and a single academic institution. The group considered the influence of oceanic conditions, both surface and subsurface, on atmospheric climatology and the effects of varying atmospheric climate on ocean processes. It was felt by a minority of the group that the climate of the ocean itself is important and should be a part of the NCP. Certain data types comprising general categories of oceanographic parameters (noted in Table 1) were suggested for consideration by the group from the standpoint of relevance to the objectives of the U.S. Climate Program. Important climatological aspects of each data type are described in the following material. The group prioritized each data type with respect to the objectives of the U.S. Climate Program. The data needs under each category were considered, keeping in mind their relevance to assessment and application functions. Data types of broadest interest were: Surface Marine, Subsurface, and Ocean Current data; priority rankings are indicated in Table 1. 57 TABLE 1 RELATIONSHIP OF MARINE DATA MANAGEMENT PRIORITIES TO NCP PROGRAM OBJECTIVES* IMPACT DATA ASSESSMENT NO'S RESEARCH INFORMATION, SERVICES PROSPECT Surface Marine Id lh li la lc 2a 3b 3c 1.2b 2a 3b 3e High Subsurface Ocean Data Id lh la lc 2a 3b 3c 1.2b 2a 3b 3e High Currents Id lh la lc 2a 3b 3c 1.2b 2a 3e Med-High Time Series Ocean Data Id la 1.2b 2a 3b 3e Medium International/ Foreign Exchange lh 3b 3c 1.2b 2a 3e Medium Remotely Sensed Ocean Data Id lh 3c 1.2a L.2b 1 .4a 2a 3e Medium Fisheries Data Id li 1.2b 2a 3e Medium Sea Level Id li lc 1.2b 2a 3b 3e Medium Bathymetry Id lh 3b 2a Low *Task numbers as identified in Chapter 3 of the Draft Preliminary 5-Year Plan, National Climate Program, April 20, 1979. Key: Climate Impact Assessment Task Id - Biological response models related to marine fishery production lh - Climate impacts related to military systems and strategies li - Studies related to climate impacts of societal concern Research Task la - Improve historical data sets lc - Interactions between climates in geographical location 2a - Ocean heat storage, transport and atmospheric coupling 3b - Develop ocean models 3e - Develop coupled ocean-atmosphere climate models Climate Data, Information and Services Task 1.2a - Satellite system for global coverage of sea-surface temperature, sea-level (topography) , and wind stress 1.2b - Expand in-situ ocean monitoring 1.4a - Satellite measurements of polar ice 2a - Prepare data sets that will have broad use and enhance availability of implied information 3b - Diagnose recent and current climate fluctuations 3e - Continue and expand statistical descriptions 58 RECOMMENDATIONS Recommendations made by Working Group 3 are felt necessary to achieve National Climate Program goals. Surface Marine Data • Primary — Digital data from Surface Marine Platforms (primarily ships) for the period 1973 to the present require consolidation, validation, duplicate elimination, and compaction into forms that are readily accessible and updatable by geography time and in platform sequence. The data should be inventoried by space, time, platform, and element. NCC will perform this work. — Work similar to the above recommendation should be completed for the period 1850-1971. In addition, foreign digital data that are not presently in the archives should be surveyed, selectively acquired, and included. — U.S. data prior to WW II, which are mostly undigitized, should be inventoried, selectively digitized, and processed through the steps included in the first recommendation above. • Secondary Recognizing the importance of Sea-Surface Temperature (SST) and certain other data elements to remote sensing calibration as well as direct NCP needs, it is recommended that : -Efforts should be increased to selectively obtain ship data from data-sparse areas; -Acquisition and archival of data from fixed buoys, drifting buoys, and satellite remote sensing systems should be stimulated. -Further development and use of automated data acquisition systems (such as SEAS) and automated satellite data collection systems should be encouraged to improve the quality of ship-of -opportunity data. 59 Subsurface Ocean Data Historical data from WDC-A for oceanography should be processed by the National Oceanographic Data Center on a selective basis to fill time and geographical gaps in the record; accelerated accession of data for these gaps should also be encouraged. Newly developed quality control procedures must be applied to data in the NODC to assure high quality data sets. Compressed data formats must be developed and adopted at the NODC for improved accessibility of data. With respect to C/STD data, uniform formats must be adopted and definite processing procedures must be followed at the NODC. Currents • For the derived surface current file (SCUDS) , an on-line inventory should be developed to improve accessiblity to those data. • Contemporary data should be added to the SCUDS file. Those data added must have good geographic registration. Time Series Ocean Data • In light of the paucity of time series ocean data, a feasibility study should be undertaken to examine estab- lishing synthetized nearshore subsurface oceanographic time series of measurement sequences for a period of record of greater than 20 years. International/Foreign Exchange • Relevant international groups, i.e., IOC, WMO, UNEP , and others, should support efforts to exchange ocean data in uniform accepted formats. • A comprehensive inventory system should be supported — like the Marine Environmental Data Inventory Data System (MEDI) . Such a system is being undertaken for the FGGE year by the RNODC - FOY. 6 Remotely Sensed Ocean Data • In light of the great cost associated with processing this type of data and the great amounts of data generated, dedicated support should be provided for insuring proper processing, archiving, and retrieval of these data sets. Fisheries • An inventory should be conducted of all fisheries data held in archives and elsewhere. • A special effort should be made to identify and acquire long time-series of year class strength data; and informa- tion on bio-environmental linkages for major species. The identity of biologists cognizant of the strengths and weaknesses of the data should be recorded and climate- fisheries analyses must not be undertaken without expert fisheries biological assistance. Sea Level • Attempts should be made to develop better sources of international sea level data. Concomitantly provisions should be made for domestic archival and servicing of this data type. Bathymetry • An inventory of U.S. holdings of bathymetric data will aid in the selective accessioning and processing of such data in data-sparse areas and in areas of critical need. • Request from researchers and modelers an expression of the minimum density of observations per unit area of ocean bottom necessary for climatic modeling purposes. CONSIDERATION OF DATA TYPES Introduction There are a number of needs for oceanic data. The oceans as a component of the Earth's climate system appear to play a major, if not totally determined, role in the annual and inter-annual variability of our climate. Specifically, the ocean plays a major role in the Earth's heat balance, acting as a medium for both storage and transport of heat. Climate researchers are presently using observed ocean data in the following ways: 1. For diagnostic studies of the Earth's heat balance in conjunction with atmospheric and radiation data. 61 2. Verification of numerical and analytical models simulating the Earth's climate. Once the annual cycle of the Earth's present climate can be reasonably simulated then sen- sitivity studies such as the effect of increased CO2 on climate can be conducted with greater accuracy. 3. Use in diagnostic numerical ocean models. Because of computer time required for integrating numerical ocean models to an equilibrium state, analyses of observed temperature and salinity data as well as climatological wind stress data are used as driving forces to speed up equilibrium. Model predicted velocity fields are then used for studies such as simulating observed tracer dis- tributions such as tritium. Once reasonable simulations are achieved the models can be used for simulating oceanic transport of other climate related variables such as C0 2 - 4. Use in understanding and parameterizing relevant climate related processes such as mixed layer dynamics. Climatological ocean data provides environmental baseline information for engineering design criteria, long-range event planning, feasibility analysis, and impact assessment. Examples are outer-continental shelf environmental assessment programs; design of ships based on ocean wave climatologies; and design of marine structures based on storm, ocean current, wave distribution, and thermal conditions. Other uses include the study of climatic scale variability of living marine resources and pollution studies. Considering these needs and the objectives of the U.S. Climate Program for ocean data, several data types were addressed: • Surface Marine Data • Subsurface Ocean Data • Currents • Time-Series Ocean Data • International/Foreign Exchange • Remotely Sensed Ocean Data • Fisheries Data • Sea Level • Bathymetry 62 Focusing on each of these data types, the following questions were posed. 1. What data are needed? 2. What data can/should be obtained from other working groups within the working group? 3. What is the present status of this data type? 4. What is the accessibility of this data type to prospective users (both data and products); including ease of access, cost, inventory of data and products, etc.? 5. How well does this data type fit known user needs, both present and projected? Surface Marine Data Surface marine data considered essential to the climate program are as follows: (a) Remotely sensed - winds, ocean currents, ice, SST, clouds, waves, precipitation radiant fluxes (b) Surface platforms: Primary - SST, winds, air temperature, humidity Secondary - Clouds, waves, precipitation, sea level pressure Data needs from Working Groups 1 and 2 should be considered when discussing ocean data. User groups are primarily scientists interested in time series or long-term analyses for climatology and engineers whose main concerns are usually long period studies of small geographic areas. For the remotely acquired surface data, accessibility is adequate for sea- surface temperature and clouds. However, some improvement is needed in sea-surface temperature accuracy. Definite improvements are required for elements as this relates to accessi- bility. With respect to data acquired from surface platforms, the need is recognized for accessible, validated surface data, primarily SST, wind, air temperature, and humidity — secondarily, cloud, wave, precipitation, sea level, pressures and other mis- cellaneous elements that are collected for operational meteorology. The highest priority should be to organize available digital files. 1. Organize the U.S. International Digital Data Set for the period 1973-present . 63 2. Organize the U.S. International Digital Set for the period 1850-1973. 3. Punch and organize U.S. Data for the period prior to WWII, most of which is presently unprocessed. Recognizing the large gaps in data coverage away from the major shipping lanes, data from fixed buoys, drifting buoys, and satellite remote sensing systems are recognized as essential. With respect to the accessibility of data, cost for processing unprocessed or partially processed data is an essential limiting factor. Availability of the data as reflected in the above para- graph is also a problem, as are gaps in time and space. A thorough search of available surface data from worldwide sources should be conducted. With respect to user needs, derived products should be established and maintained, preferably using objective analysis techniques, to assure space and time continuity. Temperature, SST, wind, clouds, surface fluxes, wind stress, and pressure should be included. Selected time series should be produced as well as products for large scale, long-term climatology. These data are of vital importance to the National Climate Program as they represent the major avenues by which energy and momentum are exchanged between the oceans and atmosphere. Knowledge of these processes is necessary to support efforts to forecast climatic variations. Subsurface Ocean Data Various types of subsurface ocean data are required for climate purposes. Some of these data types are: water temperature, salinity, dissolved oxygen, and nutrients. These data are for a variety of purposes, including (1) mapping of ocean conditions; (2) computation of ocean currents, heat content, and advection; and (3) study of biological productivity. Such data have been collected for over 100 years and constitute a major source of our knowledge of the ocean, its currents, and its variations. Particularly important to the climate program are variations of major ocean currents and the variable amounts of heat they transfer from tropical to temperate region. Traditionally, subsurface ocean data were taken from Nansen bottle casts, but in the last 30 years sophisticated profiling instruments such as mechanical BTs and STD instruments have become available. In the very recent past, expendable profilers such as XSTDs have been available. Also in the past these missions were commonly made from research vessels. Recently, to contain escalating research vessel costs, many measurements are now made from platforms of opportunity such as merchant ships or moored buoys. 64 Ideally, users need such data on various scales of spatial resolutions. For global climate modeling one would like adequate data in every part of the world ocean. As noted previously, avail- able data is sparse in mid-ocean areas, particularly in the southern hemisphere. Thus a need exists for more data in tropical and southern hemisphere areas. Intensive sampling of XBTs from mer- chant ships is now available in the North Pacific and off the east coast of the United States. In certain selected areas, we want more frequent, more closely spaced observations than the above global dispersion of observations More concentrated observations are desired in waters near the United States. An inexpensive means of monitoring major ocean currents is possible by taking repeated XBT sections from ships of oppor- tunity in lines normal to such currents. Such repeated XBT sections can be made monthly or more frequently. Location New york to Burmuda Ocean Current Gulf Stream New Orleans to Loop Current St. of Florida Period of Record 1970-1975 1970 to Present Organization NAVOCEANO NMFS San Francisco California Current 1966 to Present NMFS (NSF) to Honolulu Honolulu to Tahiti Equatorial Current 1972 -to Present NMFS-Univ. of Hawaii We suggest other nations could do similarly in waters of interest to them. For example, the French are starting such sections between Japan and New Caledonia across the Kuroshio. Relative to Question 4, data obtained from other working groups are not applicable. The prime area repository of this type of data is the National Oceanographic Data Center; its holdings are generally good. There are areas of sparse or non-existent coverage. But efforts should be taken, however, to fill in gaps in areas and/or time periods of prime interest to be decided by project personnel. Additional data could come from foreign sources, but problems exist in format diversity, medium of exchange, and the expense of entering data into our system. 65 For existing domestic files, the access is good with response times usually on the order of days to weeks. For foreign data, however, the cost of obtaining and/or reformatting such data is often significant. Access is good to station data via NODC, but access to C/STD data is very poor. Better means are needed to support efforts to get C/STD data from institutions and in through the National Oceanographic Data Center. It was noted that Fleet Numerical Weather Central is now merging NBT, XBT Hydro- graphic Cast Data, STD and Message Data. It was suggested that NODC and FMWC use similar formats to inventory and store data. Subsurface ocean data sets are directly related to the need for the following National Climate Plan tasks as described in the Draft System Plan. Climate Impact Assessment Tasks Id, In; Research Tasks la, lc, 2a, 3b, 3c; Data Information and Services Tasks 1, 2b, 3b and e. Numerous problems exist with subsurface data: 1. Data at NODC are accessible by existing programs, but user cost and convenience will be considerably aided by compression and quality control of the data. 2. Better coverage is required in southern tropical oceans. 3. Potential problems exist with calibration of XBTs below 400 meters. 4. More salinity data is needed to better understand density structure and hence ocean currents. 5. Deep currents below 1,000 meters need to be studied for understanding of reservoir of carbon dioxide in oceans . 6. Cost effectiveness of digitizing pre-1954 mechanical BTs should be investigated. Priorities must be set for selective accession of data in areas of sparse data coverage and/or converting available but unautomated data into accessible form. General oceanographic files need general housekeeping for quality control purposes. Chemistry components of station data files must be examined for errors. Currents Surface and subsurface currents needed for climate purposes include derived surface currents from ship set and drift observa- tions (SCUDS data sets) , drifting buoy observations measured subsurface current meter observations and in the future, tomography 66 Processes of major importance in climate studies are the exchange of heat between the ocean and the atmosphere and the transport of heat between the tropics and temperate regions by ocean currents. Observations of ocean currents by direct means are possible using drifting buoys, current meters, and by the drift of ships from a fixed course. Normally, the direct observations are too sporadic in time or space and indirect measures of ocean currents (such as geostrophic transport computations from tempera- ture and salinity data or Ekman transport computations from wind and pressure data) must be used. The derived surface current data set from ship set and drift observations is reasonably well organized. Data and summaries are programmed and available but without an on-line inventory — only hard copy is available. Data is sparse outside of shipping lanes and data is of low quality for individual observations. Subsurface current data sets suffer from widely differing formats, differing observational standards, and highly varying degrees of quality. The National Oceanographic Data Center (NODC) has data sets from national and international programs and maintains a detailed inventory of U.S. data. There is a need for screening individual data sets and for incorporating those of apparent good quality into a uniformly formatted, easily retrievable quality control data set. Accessibility to the SCUDS file is good except that a detailed on-line inventory is not available. Both data and standard products, however, are available. The subsurface current data, for the most part, is costly and difficult to use due to the lack of uniformity in data formats. Ocean current data are needed for the following tasks as stated in the National Climate Program Draft System Plan: Climate Impact Assessment Tasks Id, lh; Climate Research Tasks la, lc, 2a, 3b, 3c; Data, Information and Services Task 1.2d, 2a, and 3e. Tomography and drifting buoy data may present problems similar to that of other remotely sensed data. The usefulness of drifting buoy data as a measure of surface current may be questionable because some buoys have been undrogued. More informaton about data acquisition for these new types of data acquisition systems is needed. For SCUDS - more work is needed to look at the availability for interannual variations and to add new data from contemporary sources from platforms with good navigational control. There is now a unique opportunity to collect far more accurate ship drift observations from ships which have satellite navigation facilities. Traditionally, ship drift measurements were made from ships using celestial navigation 67 (accurate to 50 nm?) and over 12- or 24-hour periods of drift. With modern satellite or OMEGA navigation techniques, positions can be obtained to an accuracy an order of magnitude better (hundreds or thousands of feet) . Thus ship drift observations can now be made over ship drifts of only 2 or 4 hours and, with automation of the calculations, require only minimal effort by ship personnel. Therefore, ship drift obser- vations can now be made much more accurately than previously and far more abundantly at less cost and effort than in the past. Such abundant data will allow monitoring of ocean current fluctuations on a frequent basis. Consideration should be given to analyzed surface currents. Geostrophic current fields have been calculated from Transpac XBT data using T-S analyses from 20 degrees to 40 degrees north. These surface currents are easily accessible in the form of monthly 5-degree grids. Perhaps an even more useful data set would be Ekman current fields analyzed from geostrophic wind forcing. Such analyses have been made at the University of British Columbia. Synoptic geostrophic wind field analyses recently made for 1946-76 from the Northern Hemisphere must be used to create Ekman flow analyses which can be averaged over space and time to make a long term monthly climatology. International/Foreign Exchange Data required coincides with data listed under most of the other headings: surface, subsurface, sea level, fisheries, etc. Input from other countries is vitally needed for the construction and maintenance of global ocean data sets. Perhaps most important is the need for a good inventory of foreign data sets such as that being undertaken for the GWE/FGGE YEAR by RNODC-FOY. We are not sure whether other groups feel that international exchange is critical to their needs. Data are exchanged through mechanisms established by the IOC and WMO. Non-telecommunicated, subsurface data are exchanged using the World Data Center System. Although there has been a marked increase in the amount of data entering this system, there are often long delays, format incompatibility, and differences in the exchange medium. For marine data, an internationationl marine data set inventory called MEDI has been adopted by the IOC, WMO, ICES, FAO, IHO, and IAEA. Input has been slow from the centers associated with these organizations primarily because funding for operation of the system is minimal, and because the international exchange mechanism is rusty and not being utilized in its proper capacity. 68 Data are accessible through the WDC-A for oceanography, NCC (Surface-Marine) , and NWS (Telecommunicated Data) . Although data Input through the WDC system is good, data forms, formats, and locumentation are extremely variable. Climate program partici- pants should support current IOC efforts to standardize exchange formats (GF-3 Format) and improve services (NODC's responsibility) md joint IOC/WMO efforts toward compatibility of data set develop- ment for the World Climate Program. User Needs — We cannot afford to be insular in our outlook for a National Climate Program, recognizing that oceanic and atmos- pheric processes do not cease to operate beyond our national boundaries In order to adequately describe ocean-wide or global processes, lata from foreign sources must be used. These seem to be par- ;icularly needed for the NCP, 5-year Plan Objectives as follows: Impact Assessment Tasks lh, Research Tasks 3b, and 3c, and )ata, Information and Services Tasks 1.2b, 2a, and 3e. lemotely Sensed Ocean Data Ocean data taken by remote sensors provide important information :or climatic research. Remote sensors on satellites can produce lata which provide global distributions of ocean-wave characteris- :ics, sea-surface temperature, ocean currents, sea-level elevation, surface winds, and fresh water inflows. Aircraft and in situ data :ollection platforms can also be used to provide complementary •emote data in areas where the satellites do not provide sufficient :overage. : Oceanic and atmospheric data must be combined to define the lir/sea interface which is vital to the climatic research model, "herefore, secondary ocean data required (along with the Atmos- )heric Working Group) are: atmospheric temperature soundings, :loud distributions, surface air density, precipitation, and fluxes >f salt and water across the sea-air interface. Currently, remotely sensed infrared data exist on a near- [lobal, synoptic basis, beginning with 1973 from a series of satellites . These data should be retained, consolidated, and irchived as a common data set for time-series use. Remotely sensed oceanographic data were taken during the >eriod of July to October 1978 by Seasat. Due to a lack of funds >nly small portions of these data were processed for experimenters ise. Now processing of these data into geophysical units for >road use is underway by JPL, but on a limited basis. Completed lata sets are being forwarded to the Environmental Data and Information Service. Processing will not be completed until liscal year 1980. 69 Future remotely sensed oceanographic data are anticipated from the National Oceanic Satellite System (NOSS) in 1984-1989. To assure the proper processing and availability of these data, special funding considerations are recommended. These data will be very important to oceanographers and should be archived in both geographic and time-series data sets. Synoptic ocean-air interface measurements are urgently needed for climatic studies. Historical surface data are sparse in both space and time and do not yield enough information on inter-annual variability. Although the capability for remote sensing in the area of interface measurements is only developing, the pursuit of these data should be stressed and data sets from satellites should be anticipated. Fisheries Data Certain data on marine fisheries are considered important in order to ascertain the effects of climate on these fisheries. These fisheries data include harvest statistics (catch or landing data) , survey estimates of abundance (usually research findings) , production estimates (estimates of biomass) , and process-oriented linkage data (from ECO systems studies) . Harvest statistics are available from the late 1800s for some species off the United States. Data from surveys of abundance are available from about 1950. Production, or biomass estimates are available for only a few species for a very few years. Process-oriented linkage data (or ECO system studies) are available for even fewer species and for fewer years. An inventory of data sets in the above mentioned four categories is vitally needed. Harvest statistics, however, are published by the National Marine Fisheries Service routinely. In the other three categories, data are available from the various research centers of the National Marine Fisheries Service. Users of the data are cautioned that many inhomogeneities exist in these data, and thus users are advised to work closely with biologists who understand how the data were collected. Users are not often able to obtain fisheries data in computer compatible form. Harvest data is only marginally useful without catch-per-unit effort statistics and age breakdown. Long-time series of year class strength (abundance data) are required for major species, as well as more information on bio- environmental linkages for early life stages of major species. It is important to emphasize that climatic/fishery analyses should be conducted in conjunction with biologists familiar with bio- environmental phenomena and data, in order to avoid incorrect inferences that may be drawn otherwise. An inventory of available fisheries data, both U.S. and foreign, should be developed. 70 Time Series Ocean Data This item refers specifically to coastal and nearshore, non-estuarine, physical oceanographic, surface and subsurface data as they affect fisheries assessment and atmospheric processes. Needed are a series of U.S. coastal stations with long-term (greater than 20 years) continuous sampling of temperature, salinity, currents, and water temperature. U.S. Tidal Stations can fill some of these requirements, but no other networks exist as such. As an example, data at 30 tide stations with a period of 40 years have been collected, but for the bulk of these data only summaries are available. These data, as well as others, require digitization. A pilot program to seek out and evaluate data sets currently described in NODC ' s environmental data base directory will be needed, as well as procedures and programs to perform quality control of these data. Ocean time series data may be available but are not necessarily in a form that is directly applicable to user needs. If sufficient data exist to construct a good time series data set it would be useful to National Climate Program objectives cited in the draft preliminary 5-year plan under Climate Impact Assessment Task Id; Research Tasks la; and Data, Information and Services Tasks 1.2a, 1.2b, 1.4a, 2a and 3c. A cooperative program of NOS, NMFS, and EDIS has started to develop a tape of monthly mean sea surface temperature, surface density, and sea level (tide height) at NOS coastal tide stations in the U.S. The resulting data set should be updated and merged with foreign data. This data set will be useful in climate studies and is analogous to data tapes developed for land station data by INCAR. Sea Level Tides in the ocean have been monitored at many stations worldwide since before 1900. These data constitute an important climatological record as the long-term signals present in long period sea level records are affected by such climatic factors as wind, ocean currents, and global freezing and melting of ice. Ocean current fluctuations are important in understanding heat advection from the tropics to temperate regions. Since historical records of ocean current fluctuations are almost non-existent, information on past currents could be inferred from such indirect historical records as sea level. 71 Very long term climatic fluctuations such as glacial-eustatic variations due to variations in the Antarctic and Greenland ice caps and other factors can also be obtained from sea level records . Standard statistical and analytical techniques exist for removing tides and other relatively high frequency factors, and noise may be effectively reduced by low-pass fitters . Vertical tectonic movements are eliminated by adjacent station comparisons with independent geodetic verification. This leaves series (corrected for gravity) that can be meaned to provide a composite record for glacial-eustatic variations and apparent secular trends. Also collected at many tide gages are daily sea- surface temperature, density, and air temperature observations. Long records of these data are also useful for studies of coastal climate but are subject to numerous problems due to sampling errors and non-representative location of gages in shallow water. Tide stations in the United States are operated by the National Ocean Survey which has uninterrupted data series extending back to 1939 or earlier at 29 locations along the coasts of the United States. The oldest uninterrupted series, at San Francisco, dates from 1860. At each tide station data, data are recorded continuously on a marigram or are recorded digitally on a punched paper tape. All processing, reductions, analyses, and archiving are performed by the Tides and Water Levels Division, Office of Oceanography, NOS . The sea level values are available in hard copy tabular form. These yearly mean sea level values (as well as monthly values) may be obtained from the Tidal Datum and Information Branch NODC should try to assemble foreign data in cooperation with Permanent Service for Mean Sea Level, Birkenhead, U.K. Bathymetry The best obtainable bathymetric data are required on a global basis and in digital form for computer modeling. Water mass movement can be influenced by sea floor configuration (e.g., submarine canyon extent and alignment, through narrow straits and ridge abruptness). Recognizing constraints imposed by time and funding, plus the fact that given geographic regions may not have coverage to the accuracies and densities required, priority processing of bathymetric data include continental shelf areas, and those exhibiting rapidly changing configurations as 72 well as in and about major ocean currents. Bottom roughness information may also be critical in certain situations, although specific qualitative significance to modeling goals is unknown at this time. Basic information required from other working groups includes specification of the degree of density of coverage for geographic areas of Climate Program interest. Generally, data are available in a 600-mile band off the eastern United States from eastern Georges Bank south to Bermuda, thence to a point east of the Florida Keys; the Gulf of Mexico; the West Coast of the United States to about 200 miles; in an irregular band of 1,000 miles from Hawaii to the Aleutian Islands; the Red Sea and the Persian Gulf; in and about Japan and southeast of Japan; and the Eastern Atlantic from Norway to Casablanca to about 200-600 miles offshore. Data of lesser reliability and/or density are available off the east and west coasts of Mexico; about the Philippines; off China and Korea; in the mid- and- south Atlantic; off Africa from Morocco to Liberia; and in the Black Sea and Mediterranean Sea. Sparse data are found in vast expanses of the south and southeast Pacific; large portions of the Arctic and Antarctic Oceans; and in many smaller areas within the regions mentioned above On a case by case basis, isolated areas, within those cited as having less reliable or sparse data, may contain useful data. Bottom roughness information has been analyzed in portions of the Pacific and Atlantic. Data on bottom roughness will be required in many key areas. The extent to which this parameter plays an important role in computer modeling, vis-a-vis climate, requires further elucidation by computer modeling experts. The only large areas where bathymetric data exist in digital form are off the east and west coasts of the United States (data from about 1930) . Some data are digitized off Oregon and Washing- ton, and in selected areas of the North Pacific and North Atlantic. Considerable amounts of useful data are found in contoured form. In many cases these data sets are amenable to digitization. There is, however, no program to accomplish this. Worldwide bathymetric data holdings are maintained by NOAA's National Geophysical and Solar-Terrestrial Data Center in Boulder, Colorado and the DOD Bathymetric Data Library in Washington, D.C. However, DOD data are partially classified and thus unavailable. 73 Known user needs should be ascertained with respect to bottom roughness requirements. Basic modeling criteria will be required before any significant bathymetric data accession program is undertaken. The basic density criteria as discussed by the group seems to be on the order of one observation per 20 by 20 nautical miles as a minimum. Greater densities, however, will be required in certain critical areas. A more definitive inventory of bathymetric data holdings should be obtained by EDIS. 74 c o ■H +J rd H •H o u H CU o e id £i +> H id w d) +J C id c o •H 4-1 fd ■H rd n c p to CD x: fd m w D H 75 o r',' 5g UJ oi 2£ *t D jj O CC ISI E 8 a 5~ o CO - ■* cc O 2 < CO O CO a ffi 3 IN. CD *" < ECTED LAR ui «e O + o < cc CO CO o O u ■ ./■■■■_:mam ■ m 'IS* v v ; • :.' . s 1 £* d a cq 5 CO UJ 0) 0) UJ O O CC CL UJ > 5 < cc -J CC < o O CL UJ < >8E I < O co — UJ (D CO O CO 2 uj O O t£ O a cc > o- z i o 2 ui QC Ui Z 0. CO UJ o < u. cc D CO i H co -P 1 CO CO 0) -H p c g c CD X! CD & a M O "—* o U £ CO CO J3 CD p XI p u 1 to . c rH cd U ip -p (d co tr> CL3 cd u M-) •H £J CD -p H -H a> c O CN •H *. -p CO CD •H a> nd -P CO ffi g >1 ^ rH CD -H •P > c Sh u CD -p cd rH rH U o rd CD CD u •> U £ •H rH ■H fd rd cd o a sh 5-1 U •H "* •H U c *-* tn ^— - rH H TS a b 3 a) H rH a rH 1 •H CO CO +J 4H G o rH ^ -P ■P rH CD u CD rH 3 c -H to P •H cd ■p cu u rd CD XI CO £ •H fd g CO ai - -P a 3 ^ -C a CD •» a CQ CO 3 •H CD u CO XI CD o rH }-l fi to CO ro CO ~,c -P X3 u -P £ ■H -p fd -P T3 CO to rd O T5 — * fd P CD rH cn +J x: o CO •H rH -p O -P C c 4J c H ft o fd CO O DvH o 3 a. fd *. fd cd fd rH CD >1 1 CO -P g t3 (ti ■H -p U tr CO u CD CN U CD -H tn co rH c o 3 M a> a as •rH O u -p o o -P 1 CD rd P CD CD U-i ,C g rH u -p CO fd U CO CO > >H CO CO tn iH CD u CD u < &H o CO in • • CO • * * * •^r Oh S CO u CD o „ — . X to 3 H tjl rH rH c CD m H •H > -P r* CD +J o }H a c cd •H CD rH 3: rH a 10 -H XJ l *- CO (d CO 3 -P *-* m -p CD •H •p -H u H-> PJ u rH c -P •h fd fl -P (U CD rH -P CD cd rH rH CD rH c c 4=1 -P fd W C a -H fd -p cd Qa •H a. w fd r-l a, fd — c rH to -p g -H •p CO rH g v CVH « H c rd g -P CO 0) c 3 -p rx TJ cd rH u ■p fd c nrj -P u £ rd u H d rH T3 fd •H M-t fd U CD £ rd CD C *H CD £ to •H h M » £. \ 1 s-.S -3^ sir- .gi.^-f* ^ CO Z o < J— CO < t— < Q Z o < D < ac at < O CO Q IXI LU a O z o CO « l:|'i-"=S 5«^3 -5 5, * •! ! i /-/"? 7^. £ z° ° 2 z o ° o jv> i. c ix « i \ L si f i * 1 J 2 ^5 XTi, 1 z o rS-'rf'c 1 i * sy sis i ;--j s-s ' x > — f s -^l" "' z ' .:- a >~- -f - - o j I g s „ 2 -, Z o 1— < =) IXI I ►- o KOROR ISIAND, PACIFIC KWAJAIEIN ISIAND. PACIfIC SAN JUAN. PUERTO RICO WAKE ISIANC, PACIFIC z ° ■« i,2o«> <• - ■ « 2« « ; ! 1 2 * - "' *.< a J 1 k« 8' ° i * i. .5^^ o * ••=■»< o " - 2 i -» ^b i 2/5 «< o =-„ ' • o ',o 1 o 1 s ■« ° i S /*2 *z - / i i-^,-« z * .-' z z 2 "> 1 :2 # l«0< ; 0- J 5 ' » Z- »^' ;.f ! * j 1- I ir-r •- 'IJ g v-1 1 1 :... ! '■ ?. , = i Li|...s = = -s O ' ~- z I .' • * 1 1 i o •» / _ z • »o •- ! o ■» *o * i «» • - 52 z, °i, ^ « i ' o ? H /•« 3 i / .5 2 I o 3 Si 0«£/o S v z r 3 L 'I 1 / ! '" fS * 5 52. ! ?! 5 , L 5 „ si —*f ---..?• s - /•: ,.| #'r--^.., s - 1/ * ' -— -v / Pi- S " li i fii * * *•/ ' o • £ 2- *o ; if, C - - J -^ * ? / l • )s § 2 ' '§!. o - -s -is § ! / < / < z S 3 < ^ /-""" „ ^> o 1 x • • / - 2» - j 5 5 *• / «• * / 1 3 x *) ~ < v," o v^ z 1 X ^ ^l REV 8/78 109 SET #3 Title: Geophysical Monitoring for Climatic Change (GMCC) Solar Radiation Data Description: 1. Type of Data = Global and normal incidence pyrheliometer observations through quartz domes or windows and colored filters GG22, 0G1 , RG2, RCB and UV. 2. When = Data were taken from January 19 7 6 through June 197 8. Global radiation data are given in two-minute integrals. Normal incidence data are discontinuous and are taken during clear sky conditions. 3. Space Coverage and Resolution = Data are taken at Mauna Loa Observatory, Hawaii; Barrow Observatory, Alaska; Amundsen-Scott Station, South Pole; and Samoa Observatory, Pago Pago, American Samoa. 4. Data Characteristics, Status, Quality = Instrument calibratons have shown that the 0G1 and RGB hemispheric domes direct solar beam transmission may change with time of day and also with different solar elevation from season to season. Inhomogeneity in the domes is the cause of the changes. The magnitude of the changes are on the order of + or 1 percent with a maximum of + or 5 percent. Data analysis should take this into account. The absolute values for the data from the UV pyranometer are provisional. Obstructions from towers prevent continuous unobstructed measurements of irradiances at the South Pole over a 24-hour period. An intermittent grounding problem caused noise in the global radiation data during 18 days in October and November 1976 at Mauna Loa. Mauna Loa's GG2 2 pyranometer was periodically shaded by a tracking disk to obtain diffuse radiation measurements during 1976. Filter factors (about 1.10) were applied to normal incidence data from the four locations. 5. Relationships Between Data Sets = The original data are assembled at the National Oceanic and Atmospheric Administration's Environmental Research Laboratories in Boulder, Colorado. 6. Data Organization = Global and normal incidence data are on separate tapes. Data are arranged in chronological order. Each logical record contains measurements from up to five sensors/filters. A tape may contain more than one file. 110 7. Media = Data are on 9-track magnetic tape. 8. Volume = The current data set (as of 4/19/79) resides on ten magnetic tapes. South Pole 1976 and 1977 global data are on one tape. The 1977 normal incidence data for all four stations are on one tape. Otherwise, tapes contain one station-year of data. Global data are available for 1976, 1977, and January-June 1978 at Barrow and Samoa and for 1976 and 1977 at Mauna Loa and the South Pole. Normal incidence data are available for 1977 only. For information on tape copies, write to the Director, National Climatic Center, Federal Building, Asheville, N.C. 28801 or call (704) 258-2850, ext. 203. For information about the data collection or processing, write to Bernard Mendonca, N07AA-ERL Air Resources Labs. (R329) , Boulder, Colorado 80303 or call (303) 499-1000, ext. 6811. Ill SET #4 Title: Solar Radiation Measurement (Old Data Set) Description: 1. Type of Data = Measurements are on file for hemispheric (global) solar radiation data as hourly and/or daily, normal incidence, net exchange and ultraviolet, The hemispheric sensors in use were Eppley pyranometers (CSGI 1958) . The normal incidence (NIP) instruments were Eppley pyrheliometers . The instruments responded to a wavelength of 0.3 to 3 microns. Net exchange (net) and ultraviolet (UV) networks were small and consisted of private euqipment. 2* Where = Solar radiation records began in the U.S. in March 1903. The bulk of the data began in July 1952 and containued through December 1976, when the old type data set ended. 3. Space Coverage and Resolution = From a one-station opera- tion in 1903 the network expanded as follows: 1920s, 5 stations; 1930s, and 40s, 24 stations; 1950s, 83 stations; and 1960s and 70s a maximum of 90 stations. Stations were well grouped over the contiguous states, with a few overseas locations. 4. Data Chracteristics , Status, Quality = Only gross quality control was done for continuity on the hemispheric data. These data were also checked against computed extraterrestrial radiation values. No quality control was done on NIP, UV, or net exchange data. Data accuracy was probably not greater than 14 percent. Hemispheric pyrometers were subject to color change and moisture, and were uncompensated for temperature change. Changes in other equipment are unknown . 5. Relationships Between Data Sets = The data came from National Weather Service (NWS) stations and private cooperators. The data were received as paper manuscript or punch cards as well as original strip or circular (NWS only) charts. Hemispheric data were key punched, edited and placed on magnetic tape. For the area involved the size of the data set is minimal. 6. & Data Organization and Media = Data are available as paper 7. copy of original charts or observer tabulations, as computer printouts, or as copies of magnetic tapes. Manuscript records are filed by subject and name. The magnetic tapes are in a five group number identifier. Daily hemispheric, normal incidence, ultraviolet, and 112 net exchange data were published in the Monthly Weather Review 1914-49 and Climate-logical Data National Summary 1950-1972 and 1975-1976. Monthly and period of record averages of hemispheric data are printed in the annual Local Climatological Data by station for the years 1963-71 Volume = The digital data are in two families: TD-9726 (Daily) consisting of 5 reels and TD-9725 (Hourly) consisting of 10 reels of magnetic tape. Tapes are 9 track. For more information address the Director, National Climatic Center, Federal Building, Asheville, North Carolina 28801, or call (704) 258-2850, ext. 683. Caution : The United States National Weather Service considers the accuracy of these data questionable. Data described in "Rehabilitated Solar Radiation Data" are recommended over those contained in the old data set. 113 REFERENCES Air Resources Laboratories, 19XX: Geophysical Monitoring for climatic change summary report. NOAA Air Resources Laboratories, U.S. Department of Commerce, Boulder, Colorado. V. Cinquemani, J. R. Owenby, & R. G. Baldwin, 1978: Input Data for Solar Systems. Report of the National Climatic Center, U.S. Department of Commerce, Asheville, North Carolina, 192 pp. National Climatic Center, 1978: Hourly solar radiation-surface meteorological observations. SOLMET Vol. 1 - User's Manual TD-9724, National Climatic Center, U.S. Department of Commerce, Asheville, North Carolina, 46 pp. National Climatic Center, 1979: Hourly solar radiation-surface meteorological observations. SOLMET Vol. 2 - Final Report TD-9724, National Climatic Center, U.S. Department of Commerce, Asheville, North Carolina, 184 pp. National Oceanic and Atmospheric Administration, Report & Recommendations of the Solar Energy Data Workshop held November 29-30, 1973: National Oceanic and Atmospheric Administration, Environmental Research Laboratories, Silver Spring, Maryland. B. B. Williams, J. R. Cristina, and E. A. Carter, 1979: Solar Radiation Observation Stations, updated to 1979. Kenneth E. Johnson Environmental and Energy Center, University of Alabama, Huntsville, Alabama. 114 PROXY AND NON- INSTRUMENTAL DATA RESOURCES Working Group 5 Co-Chairmen: Alan Hecht George Smith Members: L. Dennis Delorme Henry Diaz Marion L. Parker Marilyn Shartran George Smith Thompson Webb, III Peter Woodbury INTRODUCTION Understanding the climate system and testing reliable climate models requires improved documentation of climatic change on a wide range of time scales (NAS 1978; Hecht 1979) . Historic and paleoclimatic studies must therefore be an important research component of the National Climate Program. Such studies place the relatively short meteorological record in the perspective Of the historic and geologic past, and provide clues regarding the causes and mechanisms of both short-and long-term climatic variations and changes. In particular, the paleo- climatic record makes it possible to establish the patterns in time and space of changes in the climate that affect the oceans, ice sheets, and continents, and to relate the variations recorded in the sluggish parts of the system that respond to climate to the more rapid variations in the atmosphere and on the land surface. Historic and paleoclimatic data are derived from both direct and indirect sources. Examples are primary historic documents, geologic and biologic indicators such as variations of the widths of rings in trees, changes in the ratios and abundance of diagnostic flora and fauna in marine and terrestrial environments, and the size of lakes in closed basins. Over the past 10 years, two major developments have greatly advanced analysis of paleoclimatic data. First, the development of more-precise methods that date paleoclimatic records have provided a firmer chronology for dating past climatic changes. Second, the development of quantitative statistical methods make it possible to relate changes in the many kinds of paleoclimatic indicators to changes in climatic variables. Using such statistics, for example, the 115 changes of abundance of marine organisms preserved in deep- sea cores can be read as a record of past ocean temperatures. Analysis of historic climatic data has also been signifi- cantly advanced in recent years; but generally, this field of study is handicapped by too few researchers. Historians have a major opportunity to contribute to climate studies by interpreting the older written records that convey climate information from historic sources. Researchers in paleoclimatology require conventional climatic data to calibrate their paleoclimatic models. This need places certain demands on conventional data centers and requires interaction between these centers and paleoclimatologists In the long run, the interests of the National Climate Program are best served by an understanding and appreciation of available historic and paleoclimatic data resources, and by an overall national strategy to augment and preserve this valuable resource. RECOMMENDATIONS 1. Historic and paleoclimatic climatic data are an essential national resource. An appropriate fraction of national funding for climate research must be designated for the maintenance and expansion of paleoclimatic data centers. 2. Many raw materials from which paleoclimatic information is extracted (e.g., ocean-bottom cores, polar-ice cores, tree cores, and lake-bottom cores) are collected at considerable expense. New measuring techniques, or new ideas on what to measure, frequently call for reexamination of these materials. Funds to ensure their preservation should be provided as an economically sound alternative to re-collection. 3. Data centers currently exist for several types of paleoclimatic data. These centers include the Laboratory of Tree Ring Research (University of Arizona) , the Deep Sea Drilling Program (Scripps Institute of Oceanography) , and the Glacial Photographic Collection (U.S. Geological Survey, Tacoma) . Such data centers should be nationally recognized and maintained with accepted standards for security, documentation, storage, and exchange. In cases where such centers are located at universities, appropriate national funding must be provided to maintain the activities of these centers. Fiscal and coordination arrangements should be made between the paleoclimatic data centers and National data centers to handle the distribution of bulk data requests. 116 Paleoclimatic researchers are users of conventional climatic data. The data centers handling these data are often not able to provide it in a form that is suitable for paleoclimatic research. Conventional data, for instance, are not always adequately documented as to sources, formats, length of records, gaps, levels and types of error, and cautions against misinterpretation. At the same time, paleoclimatologists are not fully aware of all available services provided by conventional data centers. In view of these circumstances, it is recommended that improved methods be developed for making conventional data, and information about the existence and availability of these data, more accessible. Specific examples of conventional data required by paleoclimatic researchers are given in Appendix B. An advisory group on historic and paleoclimatic data representing all areas of paleoclimatic research should be established and charged with identifying data management problems, facilitating interchange between data-bank- management centers, and acting as a technical resource for investigators in need of data management assistance. An inventory of historic and paleoclimatic data should be assembled. This inventory will be prepared in cooperation with International Council of Scientific Unions (ICSU)/ International Quaternary Association (INQUA) and it will be international in extent. A survey questionnaire for this inventory is now being prepared. A relatively unexplored facet of paleoclimatic research is the availability and archiving of historic data. There exists therefore, a major opportunity to examine primary historic documents for climatic information. The Global Atmospheric Research Program (GARP) terminology should be extended for describing historic and paleo- climatic data levels as shown below: Level Description Examples of Special Documentation Requirements The physical sample: collected sediment, tree core, or ice core. Sample identification, sample location. Primary data (a) "raw" observations or measurements: thickness of rings for a particular tree, counts of micro-fossils within written record of lake level. Descriptions of procedure used for measurement, description of taxonomic classifications used. 117 Level Description 1 (b) "final" working data set: data that have been standard- ized, filtered, or otherwise adjusted for quality - data set used for generating peleo- climatic estimates. Examples of Special Documentation Requirements Description of criteria used to average or adjust data and to deter- mine quality of data. Pal^oclimatic estimates de- rived from a particular site. Two or three dimensional fields of estimates, derived from one or more sources of paleoclimatic data. Documentation of tech- niques or procedure used to transform primary data, including accuracy. Description of grid scheme and resolution criteria used to integrate different classes. 9. All data sets require sufficient documentation and assurance of their accuracies. Specific recommendations for historic and proxy data are given in Appendix A. 118 REFERENCES National Academy of Science, 1978, Elements of the Research Strategy for the U.S. Climate Program: Report of the Climate Dynamics Panel to the U.S. Committee for GARP , 46 pp. Hecht, Alan D. (editor), Paleoclimatic Research: Status and Opportunities, Quaternary Research, July 1978-in press. 119 APPENDIX A DOCUMENTATION A. Data file documentation is of direct concern to: (1) Data producers - because they are the primary source of information concerning descriptions and limitations of the data. (2) Data (banks) centers - because their repositories and disseminators can provide a useful function in developing practical guidelines for docu- mentation. (3) Funding agencies - so that they can require documen- tation of the specific purpose of their funding. Documentation can be subdivided into two major types: (1) General - not specific to any one data set, and intended primarily as background guidance for outside users. (2) Specific - unique to an individual data set. Each of these types can be further subdivided into two parts: (a) Descriptive - what, where, when, how. (b) Informative - quality, resolution, cautions, limitations . B. The requirements for basic data documentation of an individual data set includes as a minimum a title and abstract and some information about the data, tape layout, and data format details. 1. Title A short name for the data set or sets. 120 2. Abstract The length should be not more than a single-spaced page. This abstract can also be used in preparing inventory information about the data. A different abstract will often be needed for the reconstructed climatic grid data, or the map data, as compared to the other data levels. Suggested contents for each abstract are: • Title, giving data type. • Whether the data consist of raw material, basic measurements, grid analyses, or maps. • Number of sites, periods of record, areal extent of sites, spacing of grid points. If different sites have very different periods of record, comment on this. • Brief comments on data status, sources of errors, limitations on the data, and statements of caution as necessary. • Approximate data volume . C. Supplemental information More detailed information above minimum requirements might include the following: 1. Extended abstract and general information • Type of data. • Space-time coverage, e.g., "Thirty lake- bottom core sites, mostly to 7,000 years BP, some 10,000 to 18,000 years, for the eastern U.S.A. " • Quantity of data. • Map showing locations of sites and lengths of record. 2. Time resolution of data. "Samples taken each 100 years for 3,000 years, then every 500 years." 121 Note that the minimum sample spacing in time should be appropriate to the type of proxy material involved. If natural processes stir the mud, or if material diffuses through the layers, frequent sampling with depth (time) might not be meaningful. 3. Indication as to whether the data are based on some- thing measured, counted, or weighed. 4. Information about the data gathering process. 5. Comprehensiveness of this data set compared to other available data of this type. Are these data included in more comprehensive sets? 6. Statement about the quality of the data, to include comments on: • Quality of control procedures used. • Flags, if any, to discriminate general quality of one site vs. another. • Approximate level of dating uncertainties . Indication of how subsequent refinement of dates is handled in the data set. Give references as necessary. • Cautionary statements about using the data, as appropriate. If users might reasonably think they are getting certain data quality and quantities which they aren't, caution them on data quality and data gaps. 7. Procedures: • Procedures used to adjust the data, or to make climatic estimates. Give references for details. • What data (coverage) were used as input to the reconstruction? • Comments on the spatial representativeness of the data. Is the "site" just one sample of a consolida- tion of several? What is the geographic variability of samples from neighboring sites? What is the amount of spatial variability of samples from neighboring site? 122 D. Tape layout and formats • Information on tape density, number of tracks, and character code. Note if the archive tape may be different from the tape sent to the user. • File structure of the tape including number of files header format, length of physical records, and blocking factors. • Number of records, if known. • Structure and logical layout of a record, including location, size and content of each data field. E. Data flags Information on the quality of data would allow sample sites or paleoclimate information to be used with different weights (or not used) in a given reconstruction. F. Data separation Since formats are quite different for Level 3 data compared to Levels 1 or 2, these types of data should be kept in separate files or on separate tapes. G. Documentation within a laboratory For each data set, the basic documentation should be in a small folder which includes: 1. Tape numbers and backups that have the data. Usually a tape creation date also is desirable. 2. The data set writeup. 3. Pointers to more detailed inventory listings. 4 . Dates when changes were made to the archive and what the changes accomplished. 5. Pointers to simple computer programs that can be used to access the data. 6. The computer program used in each processing step. 7. Pointer to the program listings that accomplished the processing step. 8. Dates when data were sent to users and which version of the data was sent in each case. 123 9. Pointer to a more detailed set of publications. 10. Pointer to data listings on print or microfilm. If there are many data sets in the laboratory, a listing of data set names, perhaps including abstracts, will be useful. 124 APPENDIX B Paleoclimatologists need to know what conventional data regarding the present climate are available, what form the data are in, how they can be obtained, and what are the costs to the user wishing to obtain them. At present, more complete information is needed in all such respects, as well as information on the spatial coverage and resolution of readily available data sets. For example, the following data sets are of use to paleoclimatic researchers, but are not readily available: 1. Uniformly accepted global record of ocean-surface long-period average temperatures. 2. Standardized record of long-period mean surface albedo values. 3. Divisional averages of climatic data for years prior to 1931. 4. Measurements on the isotopic content of precipitation 5. Monthly soil temperatures. 6. Long continuous records of temperature and precipi- tation in monthly averages from 1700 to present for use in snowline and glaciological studies. 7 . Mean ablation-season temperatures and mean accumu- lation-season precipitation in alpine regions. 8. Altitudes of summer freezing temperature levels. 9. Mean lapse rates and their inter-annual and inter- decadal variance for various types of climate areas. 10. Thirty-year averages of monthly data for the coopera- tive network of U. S. stations, including mean maximum temperatures, mean minimum temperatures, number of days with 0.01 inches of rain, number of days with 0.5 inches of rain. 125 While in some cases work is in progress, to make the above data sets available, there are data problems which also need to be corrected. Specifically: • Individual pieces of data missing from tape or other storage media files should be filled in. The missing values are often recoverable from a hard copy document. • The data currently available on tape need to be better organized and documented. • The data need to be supplemented, where possible, with an updated edition of station or sub-station histories. 126 GEOGRAPHICAL, LAND USE, AND ASSESSMENT DATA Working Group 6 Chairman: James R. Anderson Members: Robert Bailey Raymond B. Daniels John E. Estes Fred B. Warren Arthur Woll INTRODUCTION Working Group 6 of the Climate Data Management Workshop held at Harpers Ferry, May 8-11 was assigned the following topics: Land Use and Land Cover — James R. Anderson Vegetation Types — Robert Bailey Soils — Raymond B. Daniels Land Elevation and Surface Roughness — Arthur Woll Crop Acreages, Yields, and Prices — Fred B. Warren Population (Number, Distribution, Trends, etc.) -- James R. Anderson Use of Remote Sensing Technology for Acquisition of Data -- John E. Estes Biomass Amounts and Changes Energy Production and Use The selection of members of Working Group 6 was made with the intention of providing appropriate expertise on each of the topics assigned to the Working Group. However, William Osburn from the Department of Energy was unable to participate in the Workshop. Therefore, no presentation on biomass amounts and changes, and energy production and use has been made in this report as origi- nally planned. Tasks assigned to Working Group 6 were to: 1. Develop an inventory of geographical and related data sources. 2. Make a preliminary assessment of the existence and quality of available geographical and related data sources to determine the relationships of these data sources to the requirements for such data for use in climate research, modeling, and impact assessment. 3. Submit recommendations to enable appropriate and effective use of existing and any additional data needed in climatic studies. 127 A review was made of the requirements, uses, and needs for geographical, land use, and assessment data in climatic studies from the following sources of information: Jenne, Roy L. , The Global Data Base for Climatic Research: A Report to the WMO-ICSU Joint Organizing Committee for the Global Atmospheric Research Program (GARP) , Preliminary draft dated 2 3 November 1978. Public Law 95-367 — 95th Congress — Sept. 17, 1978 (National Climate Program Act) . Draft Preliminary 5-Year Plan — National Climate Program, March 28, 1979. The Global Atmospheric Research Program (GARP) Climate Dynamics Sub-Programme: Report of the Meeting of the Joint Organizing Committee (JOC) Working Group on Land Surface Processes, Dublin, 8-12 May, 1978. Published in Geneva, December 1978. Climate Data Requirements, Uses, and Needs of Federal Agencies and State Climatologists based on papers submitted and presentations made at the Climate Data Management Workshop, Harpers Ferry, May 8-11, 1979. Generally, geographical and related data will be used in: (1) conducting climatic research (i.e., how changes in biomass affect the CO2 content of the atmosphere) ; (2) atmos- pheric and hydrologic modeling; (3) making climate impact assessments (i.e., the impact of a 3-year drought on crop production in the Great Plains) ; (4) climate related programs of the Department of Commerce, Department of Agriculture, Department of Energy, Department of Defense, Department of the Interior, Department of Housing and Urban Development, Department of State, Council on Environmental Quality, Environmental Protection Agency, National Science Foundation, National Aeronautics and Space Administration, State Climate Program Offices of the 50 States, and the private sector. Data Requirements Some specific requirements or uses of geographical and related data are: •1 Statistics on the areal distribution of soils and their physical and mineralogical properties relating rainfall, evapotranspiration data, etc. for climatic research, hydrologic modeling, and impact assessments. 2. Destruction of biomass in relation to the CO2 problem. 3. Distribution of the use of fuels of different types in relation to air pollution conditions. 128 4. Energy release rates in relation to land use, land cover, socio-economic, demographic, and biophysical patterns . 5. Vegetation type, land cover, surface roughness, elevation data for use in the modeling of land surface processes as these influence climate. 6. Data on terrain features, land use and land cover, soil surface properties and soil moisture, vegetation types, etc., is needed for studies on the albedo, roughness height, and climatic impact assessments. 7. Average terrain data is needed for synoptic state atmospheric modeling. 8. Vegetation type and density are important data sets for the indirect estimation of global hydrologic data. 9. Constant parameters that describe the properties of the land surface are required as input to general circulation models. 10. Analysis of climatic states or variations in physi- cally explaining climatic variations, in predicting, and in developing and testing models for predicting variations or in assessing impacts of climatic variations on economic, social, or ecological conditions . 11. Historical crop areas and yields can be used, together with soil and topographic information, to model the possible effects of climatic fluctuations on future production of food and fiber. Data Quality and Constraints Although it is difficult to be too specific about quality and constraints of geographical, land use, and assessment data, a number of generalizations may be made: 1. The vast majority of these data are compiled with no thought of their use for climate related purposes. 2. These data are highly variable in terms of: spatial scales (levels of generalization) , terminology (concepts of classification), temporal scales, methods of data display and the accuracy not only of the data but in the representation as well (map and/or tabular or statistical form) . 3. Few of these data are global let alone continental or even national. 129 4. Gaps in coverages and inconsistencies in classi- fication schemes, grids and projections employed along with the factors mentioned in 1 to 3 above makes comparison of such data difficult even for data Which purport to display the same or a similar theme (e.g., soils, vegetation, land use). Data Formats Geographical, land use, and assessment data are generally presented in maps and statistical tabulations. Maps of soils, vegetation types, land use and land cover are widely produced and used for many different purposes. Some soil characteristics may also be presented in statistical formats. Tabulations showing areas occupied by different categories of land use and land cover have generally been available for counties and States in the United States. Crop areas, yields, and prices are generally presented in tabulations. Remotely sensed data are now becoming widely available in both digital and analog formats. Similarly, data on topography, soils, vegetation types, land use and land cover, and crop areas, yields, and prices are becoming more widely available in digital formats. Presently, land use and land cover maps and soil maps are being digitized in polygon formats by the U.S. Geological Survey and the U.S. Soil Conservation Service. Formerly and still in some State and Federal agencies such maps are being inputted into geographic or georef erenced information systems in grids of various sizes. Initially, this approach was used by the U.S. Geological Survey. However, when users began to ask for data by grid squares ranging from one acre to one square mile or more, it was soon realized that such an input procedure lacked needed flexibility in using the data. Therefore, a change to digitizing the actual boundaries of different categories of land use and land cover was made. Then software programs were developed to grid the maps that were in polygon formats to produce cells or gridded areas of various sizes. Thus the data can be delivered to users at varying scales of generalization, which has proved to be much appreciated. USE OF REMOTE SENSING The inventory, analysis, and assessment of surface cover types and the level of human use of land encompasses the application of information from many disciplines, including agriculture, hydrology, forestry, rangeland management , oceano- graphy, demography, sociology, economics, climatology, geology, and 130 geography among others. Remote sensor systems designed for environmental monitoring attempt to detect , identify, monitor, map and measure parameters that affect the complex relationships of peoples' interactions with their surroundings. This may involve the direct interpretation of the areal extent of snowpack of the mountains of the western United States, or use of remote sensor data as surrogate or proxy parameters for getting at information not directly observable from space (e.g., inferring amounts of irrigation waters applied from an identi- fication of crop types in irrigated agricultural areas in arid regions of the southwestern United States) . Today manual and semi-automated cartographic techniques are employed in planning, implementing, and monitoring nearly every parameter and/or activity involving human interaction with the physical environment (NAS/NRC, 1976). These data exist for many areas, or a wide variety of themes, at diverse scales and levels of accuracy. Inventorying, formatting, and processing these data for use in climate investigations is a difficult task. In recent years, however, we have developed the remote sensing capability to map land use and land cover categories at regional scales for nearly the entire world. LANDSAT data are potentially available for all areas from essentially 85°N latitude to 85°S latitude. We certainly can provide categories of land use and land cover data at scales and levels of specificity compatible with the need for climatic modeling at global, national, state, and regional levels within reasonable time frames. This program has been acquiring data for the United States and many areas of the world since July 1972. Many areas have been covered many times, and some areas of the globe have yet to be covered as no requirement to do so exists. Remote sensor data from LANDSAT* employed together with data from collateral sources offer the potential of providing data on current and changing land use and land cover patterns in either digital and/or analog form. These data can be struc- tured and placed into georeferenced information systems so that the functional and spatial dynamics of land resources and their relationships to fluctuations in climate may be effectively and efficiently studied. In evaluating the existing LANDSAT data on a global, continental, and local basis it is important to remember the inherent nature of the data itself. LANDSAT data in digital or analog form contains information on a variety of parameters. Until these data are processed, analyzed, interpreted, and the significance and accuracies of the interpretations verified, their use will be limited. * Radiometer data on the AEM01 satellite is covered in another Working Group Report. 131 Some gaps do exist in LANDSAT coverage at the global scale. Good data coverage exists for the United States, Canada, and Mexico. Very good data coverage exists throughout the United States , except for parts of Alaska and Hawaii. LANDSAT data are or can be made available in common analog and digital formats for use at a variety of scales. Although the data are typically formatted into images 185 km x 185 km, both analog and digital mosaicing capabilities exist which can provide accurate data sets of varying spatial dimensions. These data can be registered to a variety of coordinate systems. It should be emphasized, however, that this capability, although proved for areas as large as a State (e.g., California, Arizona), have not been tested on a national level as yet. In addition the costs of compiling, processing, registering, and storing such digital data sets at the present time would be considerable. Thus, while analog LANDSAT data may be somewhat more difficult to register, interpret, digitize, and reformat for input into climatic models, it may be a cost-effective alternative for producing land use and land cover data for large areas for use in climatic studies. Indeed, analog LANDSAT mosaics currently exist for several States in the United States (e.g., California, Nevada, Utah, New Jersey, Florida, and Georgia) and several countries (e.g., the United States, Mexico, Italy, and Algeria). Companies such as Earth Satellite Corporation, General Electric, and IBM all have produced such mosaics in the past and continue to do this kind of work. Accessibility - Information on how to obtain LANDSAT data for anywhere in the world may be obtained from: EROS Data Center, United States Geological Survey, Sioux Falls, SD 57198. Quality of Data - Although the quality of LANDSAT data is uniformly high, depending upon whether the user requires raw, uncorrected, or radiometrically and geometrically corrected data, a variety of problems may exist when using a particular data set. The nature and extent of these problems can be ascertained through contacts with personnel of the User Services Division of the EROS Data Center, Sioux Falls, SD 57198. For all areas within range of ground receiving stations (and there are now a number of them in operation around the world and more are planned) , data are potentially available every 18 days at present, weather permitting. For areas outside the range of tracking stations data availability is limited at present by the use of wide band video tape recorders. These re- corders have been a handicap inhibiting progress towards total coverage for the area of the Earth beneath the orbits of the LANDSAT satellites; however, future implementation of a Tracking Data Relay Satellite System (TDRSS) will free coverage from tape recorder dependence and open the way to more effective global acquisition of LANDSAT data. 132 Processed but unanalyzed LANDSAT data are currently readily available and accessible in raw form on a global, continental, and local basis through the EROS Data Center in Sioux Falls. By "unanalyzed" it is meant that no thematic classification of these data has yet been done. As such, the potential user of information from this data set will either of necessity become aware of the problems inherent in thematic classification of these data in a manual or semi-automated mode or contact or contract with someone who is knowledgeable. If the users themselves are to work with these data, they should be aware that manual image interpretation procedures can be obtained by referring to Chapter 14 of the Manual of Remote Sensing , of the American Society of Photogrammetry . (Estes and Simenett, author-editors 1975) . In the area of semi-automated analysis the user should be aware that many factors affect the ability of a machine to accurately provide a thematic classification of LANDSAT data. Among these factors are: the classification scheme itself, feature development and selection, preprocessing functions, generation of class training statistics, classification algorithms, and performance evaluation. For more detail in each of the above the reader is referred to Appendix Once LANDSAT data have been classified they should be rectified, registered, and input into some type of georeferenced data base for most efficient use of land use and land cover data with other data sets used in climatic modeling. SOILS Soils are the medium of plant growth and are a physical, chemical, and biological interface between the atmosphere and the underlying geologic materials. The soil climate, material from which the soils are derived, vegetation, and site history all have an influence upon the kind of soil found at any one point. When these factors are reasonably uniform over an area, the same soil, or closely related soils are found. A soil map is an attempt to show those parts of the landscape that have had a reasonably uniform soil environment. Because no area of the Earth's surface is uniform, soil maps show the distribution of mapping units that, while not uniform, are dominated by one or two kinds of soils. The inclusions of other soils depends upon the complexity of the hydrologic and geologic factors, the scale of mapping (soil maps at a scale of 1:20,000 do not show areas smaller than 5 acres) experience of the mapper, and purpose of the survey. Mapping units are named for the dominant or co-dominant taxonomic unit(s). Interpretations for a map unit are based upon the taxonomic units present. Soil survey pedon data are classified by taxonomic unit. 133 Soil survey data in the form of maps, descriptions, and laboratory data are available for most of the world through FAO- UNESCO. There are large variations in quality and density, especially in areas where little detailed mapping has been done as contrasted to areas where soil surveys of varying intensity have been made for 80 years or more. Soil survey data in the United States and Canada are abundant and of excellent quality. Although different classi- fication systems are used in the two countries, it is relatively easy to correlate soils between the systems. Abundant published and unpublished physical, chemical, and mineralogical data are available in both Canada and the United States. Interpreta- tions of soil data and map units are well advanced in both countries. If the FAO-UNESCO soil map of the world were digitized, much useful soil data would be available for climatic studies. In the United States, soil map data should be digitized and expanded on natural regional soil zones such as major land resource areas or physiographic areas. A similar approach should be used for the Canadian soil survey data. Plant available water has been measured for several soils and can be predicted from other physical properties of soils. Extrapolation of data for studies that need soil moisture estimates should be made based upon major land resource areas, not upon political boundaries. VEGETATION TYPES Vegetation may be defined as the mosaic of plant communities in the landscape (Kuchler 1964) . It consists of a given combination of life forms (trees, shrubs) and a given combination of taxa (genera, species) with relatively uniform ecological requirements. Numerous classifications of vegetation are in use. Each of the many natural resource-related disciplines has numerous classifications, each land management agency has several systems, and each geographic area varies in the applicability and use of different classifications. The same types are sometimes described by different terms, and some terms are applied with numerous meanings. Many countries have been making maps of vegetation types for many years. The three-volume International Bibliography of Vegetation Maps (Kuchler 1965) contains hundreds of entries. They vary in scale, projection, and principle of classification so that transforming one system into another one is extremely difficult or impossible. There are two basic approaches to the problem of classifying vegetation types. First, there is potential natural vegetation. This approach views vegetation in terms of which type would exist today if man were removed from the scene and if plant 134 succession were telescoped into a single moment. Its value derives from the indication it provides of the biological potential of all sites. The second way of classifying vegetation is by the actual, or real, vegetation that occurs at the time of observation. It may be natural (not appreciably affected by man), seminatural, or cultural vegetation depending on the degree of human inter- ference. Such schemes have wide application because they are of value not only to users of the vegetation itself such as foresters, but also to those who view vegetation mainly as an obstacle, a source of cover, or an amenity. The problems involved in developing a standardized approach to vegetation classification in all its complexities are not likely ever to be solved. However, it is not necessary to solve these problems completely and practical solutions have already been made. A good example of this is the UNESCO system for classifying and mapping of vegetation (UNESCO 1973) . It is designed to map the world's vegetation at the scale of 1:1,000,000 or smaller. While the classification is intended primarily for the preparation of new maps, it may also be used for transforming existing vegetation maps into this system. Maps for the continents of Africa and South America at a scale of 1:5,000,000 are in preparation for world-wide comparison. The classification is based primarily on physiognomy and structure of vegetation. At the continental level, potential vegetation maps have been developed using standardized definitions for the United States (Kuchler 1964) and Canada (Rowe 1972) . Forest cover-type classifications have been developed on the basis of the composition of tree species in a stand. A national classification was published by the Society of American Foresters (1954) . The Forest Survey staff of the U.S. Forest Service has developed a similar classification for analyzing timber inventory data. Forest community classifications (based on tree and undergrowth composition) have been developed for many geographic areas, by numerous criteria, and for many purposes. A relatively standard approach in the western United States is the "habitat type" classification developed originally by Daubermire (1952) , and extended now to at least half of the forested lands in the West by the U.S. Forest Service (Pfister 1976) . None of these data are available in digital format. The Forest Service is by no means the only organization to publish vegetation maps. Both Federal and State agencies have become aware of their usefulness and, as a result, the number of vegetation maps has grown enormously. The Bureau of Land Management, the Soil Conservation Service, the Fish and Wildlife Service, the National Park Service, river basin authorities, planning agencies, state geological surveys, and others have undertaken their preparation. 135 Small-scale mapping is essential in order to tie various parts of continents together and establish the bases for detailed climatic research. Therefore, efforts should be made to expedite the UNESCO Mapping project, particularily for North America. Along these lines, a pilot project to determine the usefulness of LANDSAT data in vegetation mapping should be initiated. A study should also be undertaken to determine if the UNESCO classification of vegetation lends itself not only to the scales for which it was designed but also to much larger ones. Large-scale mapping is needed just as urgently as mapping at small scales or even more so. As a framework for coordinating this work, a study should be conducted to determine the feasibility or transforming existing maps into the UNESCO system for use in climatic research, modeling, and impact assessment. LAND USE AND LAND COVER MAPS AND DATA As used in this report, land use refers to "man's activities on land which are directly related to the land" (Clawson and Stewart, 1965). Land cover, on the other hand, describes "the vegetational and artificial constructions covering the land surface" (Burley, 1961) . Also included in the concept of land cover are areas of barren land and perennial snow or ice. Many countries have been making land use and land cover maps and collecting statistical data for many years. Yet today there is really not a good worldwide set of land use and land cover maps. Therefore, there is no common global set of land use and land cover data that can be keyed to relatively small areas of the Earth's surface. Likewise no such maps or data are available for the North American continent. Thus data in digital format suitable for climate research, modeling, and impact assessment do not exist on a global or continental basis at the present time. The capability to make maps and obtain data pertaining to land use and land cover has improved rapidly in the past 2 decades. Developments in remote sensing equipment, interpretation techniques, and data processing now make it possible to produce the maps and data that are needed for climatic studies. Also present is a growing recognition that some degree of standardi- zation in classification of land use and land cover is important if a really comprehensive and useful data set is to be obtained over extensive areas of the world (Anderson et al. 1976). Although many individual countries such as those of western Europe and Japan have very good land use and land cover maps and data, there are really no such maps in the scale range of 1:250,000 to 1:1,000,000 for the world as a whole from which statistical data can be extracted. Data on land use and land cover collected by using enumeration techniques such as for the United States Census of Agriculture are available 136 only at the county level, which means that only a very gross generalization of land use and land cover patterns is possible. Similarly the Conservation Needs Inventories carried out under the lead of the U.S. Soil Conservation Service in 1958 and 1967 by taking an average 2 percent sample of the total area of the United States, may not yield data suitable for gridding into 10 feet x 10 feet squares such as those being used by the Office of Climatology of the U.S. Navy (Conservation Needs Inventory, 1967) . The only maps of land use and land cover for the world as a whole were published in the World Atlas of Agriculture (World Atlas of Agriculture , 1969) . Map scales are either 1:2,500,000 or 1:5,000,000 depending upon availability of information for the preparation of the maps. Sixteen main categories of land use and land cover are shown. These include arable land, woods and forest, rough grazing land, and non- agricultural land such as urban areas. Although these maps of land use and land cover in this Atlas were carefully compiled, there simply was not enough existing information to make more detailed maps at the time the Atlas was prepared. Turning to North America, land use and land cover data are becoming available at scales appropriate for coverage of the United States and Canada. However, at the present time neither country is completely mapped. In the early 1960 's land use maps for the more densely settled part of Canada or about 800,000 square miles were prepared at a scale of 1:50,000. These were later generalized to a scale of 1:250,000 and then digitized. Ten major categories of land use were identified from aerial photographs (McClellan and others, "A Guide to the Classification of Land use for the Canada Land Inventory, Department of Regional Economic Expansion", unpublished report, Department of Energy, Mines & Resources, January 1968). The digital data which is stored on magnetic tape at either 1600 or 6250 bpi consists of 750 million bytes of data. No land use and land cover mapping comparable to that done for the more densely settled part of Canada is available for the remaining 3 million square miles of Canada. For the United States, the U.S. Geological Survey is preparing and publishing land use and land cover maps for the entire Nation at a scale of 1:250,000, except for a relatively small part of the country, which will have maps published at 1:100,000. These maps are being compiled from remotely sensed data using a classification system that has a framework of 9 general Level I categories that are further subdivided into 37 more specifically defined Level II categories (Anderson et al. 1976). The land use and land cover mapping is to be completed by 1982 with an update of the more dynamic parts of the country to begin in 1980. Approximately 1.7 million square miles of the 3.6 million square miles in the United States have already been mapped. The maps are being digitized in a polygon format. Polygons can be converted to grid cells of varying sizes when desired. A computerized geographic 13 information system has been developed to store and retrieve data for analytical and other uses (Mitchell et al. 1977). Digital tapes are now available for about 600,000 square miles of the United States. The land use and land cover maps and data which are available for parts of Canada and the United States are of high quality for use in climatic studies. The maps and digital tapes are readily accessible. However, the use of land use and land cover maps and digital data in climatic research, modeling, and impact assessment will probably require adaptations in data formats. Considerable flexibility exists once such data have been incorporated into a geographic information system, which should make it possible to accommodate reformatting needs more easily. At the local level many land use and land cover type maps have been made by city, county, and multicounty regional planning agencies. Such maps vary greatly in quality and many would not be very useful for micro-level climatic studies. In rural areas very little has generally been done to map land use at larger scales such as at 1:24,000. Some digital data are available for land use at local or microscales. Such data would generally be found in city, county, or multicounty regional planning offices. The U.S. Geological Survey has been conducting studies which may provide guidance for using land use and land cover data in micro-level climatic studies (Reed and Lewis, 1978) . It is also necessary to recognize the need for adding new categories or combining categories in existing data bases in order to accommodate specific needs. For example, Working Group 2 in addressing the need for land use and land cover data to develop, test, and verify climate models that include a dynamically coupled land-atmosphere interface identified the need for the following categories of land cover: vegetal canopy density (including seasonal variations, species or broader biological groupings, density of urbanization, density of agricultural activity (intertilled crops, close sown annual crops, hay crops, and pasture. Sometimes useful data surrogates can be derived from existing categorizations. In other instances new categories may need to be mapped. TOPOGRAPHY (Land Elevation and Surface Roughness) Several data sets are available that give hypsographic data by geographical location. These data sets may be in a digital or map format giving contours and spot elevations. 138 Digital terrain tapes have been developed from 1:250,000 scale base maps with available elevation spot locations every 200 feet. There is complete coverage of the United States and a major part of Europe and Korea has also been completed. The record is on a multifile 9-track magnetic tape with odd parity at 1600 bpi. The Defense Mapping Agency has several small-scale maps available to the Executive Branch of the U.S. Government for global use. The relief is usually shown by contours or gradient tint with heights in meters. Map scales range from 1:1,000,000 to 1:5,000,000 and stated contour intervals of 200 and 1000 meters. Status maps are available showing 44 maps by geographic area and by scale. None of the elevation data is in a digital format. The U.S. Geological Survey has a basic topographic map series at a scale of 1:24,000 with gradual future conversion to 1:25,000. Such maps should have applications in some micro- climatic studies. These maps present the following elements by geographic location: hypsography (including contours and spot elevations) , vegetation (woodland) - culture, hydrography, and boundaries. Tne usual contour interval is 10 and 20 feet, where not more than 10 percent of the elevations tested shall be in error more than one-half the contour interval. A concerted effort is underway to' digitize elevation data from 1: 24, 000-scale photoimage bases. The U.S. Geological Survey has a 1 : 250 , 000-scale topo- graphic quadrangle series. The digital terrain tapes referred to above were to be made from this series. It is the largest scale series for which maps are available for the entire United States. Because of the emphasis in climatic studies on digital formats, this map series is an important one. The new USGS 1 : 100 , 000-scale topographic maps have very limited coverage at this time. However, rapid progress is being made in producing such maps for the United States except for Alaska, which does not yet have the 1 : 24 , 000-scale base maps from which the 1 : 100 , 000-scale series is being derived. CROP AREAS AND YIELDS AND PRICES RECEIVED BY PRODUCERS Principal sources of data on crop areas and yields, both in the United States and in other countries, and prices received by producers are as shown on the following page. 139 Data Set Name Contents Contact USDA Official USA Estimates Crop area and yield estimates to the country level within the U.S. Prices received at the State level, from 1866. Data Services Branch Survey Division Economics, Statistics and Cooperatives Service U.S. Dept. of Agriculture Washington, D.C. 20250 202-447-7017 USDA Foreign Crop Crop areas and yields Estimates for all foreign countries, from 1960. Data Systems Division Foreign Agricultural Service U.S. Dept. of Agriculture Washington, D.C. 20250 202-447-5255 CANSIM (Canadian Socio-economic Information Management) Canada yield, acreage, energy supply/demand, fish catch population and price indices. Mr. Michael Williams User Adivsory Services Division Statistics Canada 25 St. Clair Avenue, East Tenth Floor Toronto M4T 1M4 CANADA Phone: 966-6574 Information Management United States: Official U.S. Department of Agriculture estimates for the following crops grown in the United States are available in the following formats for all years beginning with the year indicated below. Item Magnetic Tape Major crops, including potatoes 1954 Non-citrus fruits 1963 Tree nuts 1963 Citrus 1974 Vegetables 1969 All other crops 1975 Prices received by farmers 1978 County estimates 1972 Tabular 1866 1889 1909 1909 1919 varies 1866 mid-1930s 140 Information on the areas and yields of crops in foreign countries is contained in an online indexed sequential file by the U. S. Department of Agriculture. This file may be accessed directly through remote terminals or copied onto magnetic tape. Data for some earlier years are also available in tabular form. Insofar as possible, the data in the file are official estimates released by the individual countries. The quality of these estimates varies from very good to otherwise, depending upon the country. If official estimates are not available, then the "best available" data, e.g., attache reports, trade publica- tions, or inhouse estimates are used. This file is updated monthly to reflect changes in current prospects and to revise historical data as better estimates become available. This file has only country level data. Canada maintains crop area and yield data, energy requirements, and other data for the different provinces in an online file. The Center for Environmental Assessment Services (CEAS) of the Environmental Data and Information Service, National Oceanic and Atmospheric Administration at Columbia, Missouri, has sub- national yield and weather data for major crops in the following countries: Argentina, Australia, Brazil, Peoples Republic of China, and the Soviet Union. The data are in computer processible form. Until 1964, crop area and yield estimates in the United States were based upon a combination of non-probability mail survey indications plus, in some States, State Farm Censuses, and were benchmarked against the national 5-year Censuses of Agriculture. Since 1964, the State and national estimates of crop areas, and the yields of major crops, have been based largely upon probability surveys. Relative sampling errors of the major crops (corn and wheat) have been less than 2 percent at the national level and are in the 4 to 6 percent range for most major states. Country level estimates of land area in crops are mostly derived from non-probability mail survey indications and are less reliable than the State or national estimates. (Statistical Reporting Service, 1975) Quality of the individual yield estimates is highly correlated with the proportion of the crop which passes through recordable marketing channels. Therefore, the national estimates of yield for food grains and for cotton would be the most accurate, those for feed grains would be less precise, and yield estimates for the various types of forages would be least accurate. As with the yield data, the quality of the estimated prices received by farmers would be correlated with the proportion of the crop which passes through recordable market channels and would be more accurate at the national than the State level. The ESCS data base is also updated monthly to reflect current changes in potential production and to revise historical data as indicated. 141 CEAS also has yield and weather data at the crop reporting district (sub-state) level for major crops (wheat, corn, soybeans, and barley) in major producing States. Major gaps in these data bases would include: • Much historical data are not in computer processible form. • In most foreign countries, there is a dearth of infor- mation about prices received by farmers and about areas and yields of crops at sub-national levels. • In the United States, there is a potential need for site specific data which would include such data as soil characteristics, crop yields, weather, and cultural practices, over years. There is now a multi-agency effort within the U.S. Department of Agriculture to establish minimum requirements for, and to establish a point specific data base for crop modeling research work. The leader of this effort is Dr. Donald E. McCormack, Director, Soil Survey Interpretations Division, Soil Conservation Service, U.S. Department of Agriculture, Washington, D.C., Telephone 202/447-9218. In order to make effective use of the historical crop data, foreign data for years before 1960 should be converted to computer processible form. There should also be a continuing project to secure crop data, at least currently, at the sub-national level in foreign countries. Also, presently undigitized historical crop data for the United States should be added to the computer data banks as it is needed by climatic researchers. POPULATION DATA Population data are available for all of the world's nations However, there are great variations in quality, particularly on such characteristics as age, per capita and family income, ethnic origin, race, and level of education. Even the actual number of people occupying given areas in some countries is not very reliable. Generally the most reliable data on actual number of people are available for the large metropolitan areas. Excellent population data are obtainable for Western European countries, Japan, Canada, the United States, South Africa, and some other countries in various parts of the world. For example, Brazil and Costa Rica have much better population data than some other Lation American countries. The Population Reference Bureau of Washington, D.C., annually provides a very useful and reliable "World Population Data Sheet." Of course, the data are obtained 142 from the most reliable available sources, which as previously indicated vary greatly from country to country. Data include: Rate of natural increase Population projection to the year 2,000 Infant mortality rate Population under 15 years Population over 64 years Life expectancy at birth Urban population Per capita gross national product For the more advanced countries population data are available in digital format. Canada and the United States have excellent population data compared with the world as a whole. The actual counts of number of people are reasonably accurate, although problems do exist in obtaining a complete enumeration in inner city areas and in some rural areas such as on Indian reservations. More difficulty exists in obtaining accurate population data on race, ethnic origin, per capita, and family income, and level of education. In 285 standard metropolitan statistical areas (SMSA) data are available for census tracts which are small areal units that sometimes are less than a square mile (640 acres) in size. Such data are published in Urban Atlases which are available for 65 metropolitan areas. Nearly all population data in the United States and Canada are available in digital format. It is suggested that some carefully designed pilot studies be initiated to test the usefulness of population data in digital formats, particularly metropolitan areas for climatic studies. RECOMMENDAT IONS 1. Explore the feasibility of preparing generalized land use and land cover maps and digital data for the world from LANDSAT data for use in climatic modeling and impact assessments. 2. Initiate a project to examine the practicality of digitizing data from existing global and national thematic maps such as the FAO global soil survey and the UNESCO vegeta- tion type maps. 3. Undertake a pilot project to obtain crop area and yield data for selected homogeneous areas in other countries to facilitate more effective use of such data in climatic 143 studies. Also improve the use of the present U.S. system of county estimates of crop areas and yields in climatic impact assessments. 4. Initiate a pilot project to merge, generalize, and grid existing digital data in polygon formats such as land use and land cover maps and terrain data for use in climatic modeling. 5. Conduct experiments of limited scope with existing remotely sensed data and available digital land use and land cover data to determine if specific categories of land use and land cover such as seasonal variations in vegetal canopy density can be extracted on a cost-effective basis. 6. Expedite the UNESCO mapping of vegetation types through the use of different kinds of remotely sensed data including LANDSAT and high resolution color infrared photography. 144 SELECTED REFERENCES Use of Remote Sensing Anderson, J. R. , Hardy, E. E. , Roach, J. T. , and Witmer, R. E., 1976, A land use and land cover classification system for use with remote sensor data: U. S. Geological Survey Professional Paper 964, 28 p. Estes, J. E. and Simonett, D. S., 1975, Fundamentals of image interpretation, in Reeves, R. G. , ed., Manual of remote sensing, v. 2- — interpretation and applications: Falls Church, Va., American Society of Photogrammetry , p. 869-1076. National Academy of Sciences/National Research Council, 1976, Research and environmental surveys from space with the thematic mapper in the 19 80 's, Land Use Management: National Academy of Sciences, p. 43-51. U.S. Urban Renewal Administration, Housing and Home Finance Agency, and Bureau of Public Roads, 1965, Standard land use coding manual, a standard system for identifying and coding land use activities: Washington, D. C, Department of Commerce, 111 p. Soils Canada Department of Agriculture, 1976, Glossary of terms in soil science: Canada Department of Agriculture publication no. 1459, revised, 44 p. Canada Department of Agriculture, Research Branch, 1978, The Canadian system of soil classification: Canada Department of Agriculture publication no. 1646, 164 p. Soil Survey Staff Bureau of Plant Industry, Soils, and Agricultural Engineering, 1951, 1962, Soil survey manual: Washington, D. C. , U. S. Department of Agriculture Handbook no. 18, 503 p. Soil Survey Staff, Soil Conservation Service, 1962, Soil survey manual: Washington, D. C, U. S. Department of Agriculture, Supplement to Handbook no. 18, replacing pages 173-18 8. 1975, Soil taxonomy, A basic system of soil classi- fication for making and interpreting soil surveys: Washington, D. C. , U. S. Department of Agriculture Handbook no. 436, 754 p. Working Group on Soil Survey Data, 1975, The Canadian soil information system (Can SIS) manual for describing soils in the field: Ottawa, Ontario, Canada Department of Agriculture, Soil Research Institute, 170 p. 145 Vegetation Types Daubenmire, R. , 1952, Forest vegetation of northern Idaho and adjacent Washington, and its bearing on concepts of vegetation classification: Ecological Monograph, v. 22, p. 301-330. Kuchler, A. W. , 1964, Potential natural vegetation of the conterminous United States (map and manual) : American Geographical Society, special publication 36, p. 116. ed. , 1965-68, International bibliography of vegetation maps: Lawrence, Kansas, University of Kansas Library Series, no. 21, 26, and 29. Pfister, R. D. , 1976, Land capability assessment by habitat types, in America's Renewable Resource Potential - 1975, the turning point: Society of American Foresters, National Convention, Washington, D. C, 1975, Proceedings, p. 312-325. Rowe, J. S., 1972, Forest regions of Canada: Ottawa, Canadian Forest Service Publication 1300, 172 p. Society of American Foresters, 1954, Forest cover types of North America (exclusive of Mexico): Washington, D. C, Society of American Foresters, p. 67. United Nations Education, Scientific and Cultural Organization, 1973, International classification and mapping of vegetation: Paris, United Nations Education, Scientific and Cultural Organization, p. 37. Land Use and Land Cover Anderson, J. R. , Hardy, E. E., Roach, J. T. , and Witmer, R. E., 1976, A land use and land cover classification system for use with remote sensor data: U. S. Geological Survey Professional Paper 964, 28 p. Burley, T. M. , 1961, Land use or land utilization?: Professional Geographer, v. 13, no. 6, p. 18-20. Clawson, Marion, and Stewart, C. L. , 1965, Land use information, A critical survey of U.S. statistics including possibilities for greater uniformity: Baltimore, Md. , The Johns Hopkins Press for Resources for the Future, Inc., 402 p. Mitchell, W. B. , Guptill, S. C, Anderson, K. E., Fegeas, R. G., and Hallam, C. A., 1977, GIRAS: A geographic information retrieval and analysis system for handling land use and land cover data: U.S. Geological Survey Professional Paper 1059, 16 p. 146 Reed, W. E. and Lewis, J. E., 1978, Land use and land cover information and air-quality planning: U.S. Geological Survey Professional Paper 1099-B, 43 p. Conservation Needs Inventory Committee of the U.S. Department of Agriculture, and Bureau of Indian Affairs, 1971, Basic Statistics — national inventory of soil and water conserva- tion needs, 1967: U.S. Department of Agriculture Statis- tical Bulletion No. 461, 211 p. The Committee for the World Atlas of Agriculture, eds., World atlas of agriculture, 4 volumes, Novara, Istituto Geogra- fico De Agostine, 527 p. Topography Clarke, P. F., Hodgson, H. E., and North G. W. , 1978, A Guide to obtaining information from the USGS 1978: U.S. Geological Survey Circular 777, 36 p. National Cartographic Information Center, 1978 , Digital terrain tapes, user guide (2d ed.): U.S. Geological Survey, 12 p. Crop Acreages, Yields, and Prices Statistical Reporting Service, 1975, Scope and methods of the Statistical Reporting Service: U.S. Department of Agriculture, Miscellaneous Publication no. 1308. Population U.S. Bureau of Census, 1978, Mapping for censuses and surveys: U.S. Department of Commerce Statistical Training Document ISP-TR-3, 353 p. 147 APPENDIX A Some factors affecting machine processing of LANDSAT data for thematic classification of land use/land cover types: • The Classification Scheme Itself : To date most land use/land cover applications have made use of previously defined schemes with, in limited instances some adaptation to remote sensing capabilities (U.S. Government, 1965). A land use/ land cover classification scheme specifically designed with remote sensing in mind has been proposed by Anderson (Anderson et al., 1976). • Feature Development and Selection : Spectral and spatial features are the primary inputs of remote sensor systems into the classification process. Attributes of significance include spectral, spatial, temporal, and quantitization resolutions. To date most work has made exclusive use of spectral features, with limited but growing interest in spatial features such as texture and spectral transformations. With more (and better selected) spectral bands and spectrally derived features it will be necessary to reduce the overall dimensionality of data sets through various transformations. In many instances the temporal dimension is required for accurate discriminations; the addition of multiple dates to the analysis brings added emphasius upon the need for transfor- mations which reduce dimensionality. • Preprocessing Functions : Several radiometric and geometric characteristics of remotely sensed data are usually corrected or standardized prior to classification analyses. Sun angle and haze correction are two preprocessing functions that can significantly aid classification performance. Image geometric fidelity, necessary for multidate/ source data overlay, is of critical importance to assure that features are always assigned to the same spatial location. There have been a limited but revealing set of studies concerning the impact of image registration accuracy and its impact on classification performance. Since geometric trans- formations involve resampling there has been a great deal of debate, but with little study, concerning the ways in which various resampling algorithms (such as nearest neighbor, bilinear and cubic convolution) affect classification performance. In addition, several investigations have opted for rectified but non-rotated imagery to minimize both radiometric resampling errors and computational costs associated with rotation. 148 G eneration of Class Training Statistics : Nearly all classification techniques involve the development of class training statistics which act as prototypes for pattern recognition purposes. The training stage of classification is considered by many to be the single most important component affecting overall classified performance. Several techniques have been developed for training site selection. Perhaps the most conventional training procedure in use at present involves interpreter selection of training areas representative of each class. The experience of many researchers, however, has demonstrated that such directed training can significantly bias results because the analyst typically selects only those areas where confident interpretations are possible. In order to redress this bias, both systematic and random sample techniques have been tested for training purposes with much improved results. Clustering is an unsupervised approach that attempts to identify the natural spectral classes present in a scene. A problem with many early cluster programs was the limited number of pixels that could be processed and the large computational requirements in general. Several algorithms have. been implemented for clustering and both procedural steps, such as manual "seeding" and computational efficiencies, such as "weighted clustering" (similar in effect as a table lookup within a classifier) , have resulted in a greater efficiency and larger scene capacities. Image stratification using either collateral data or an image itself have also been employed with some success for training site selection since these data can serve to localize training class signatures to those areas for which they apply. To date, most automated stratification procedures have not proved successful, but manually defined strata have demonstrated that performance can be substantially improved through stratification. It has long been recognized that training class statistics should be as representative as possible of the entire area being processed. One measure of their representativeness is often noted in the difference between training and test site classifi- cation performances. Since test sites are labeled but not used in the training process, their classifi- cation performance is generally lower than training sites, but usually more representative of overall classification performance. The higher performance 149 of training sites suggests that if total image training (i.e., total ground truth) were possible, classification accuracies could be optimized. While seemingly redundant (why classify if you have a map?) , this approach can be effectively adapted somewhat to capitalize upon previous land use maps to develop training class statistics. Since time periods can be selected during which only a small proportion of changes occur it becomes possible to generate statistics effectively using a nearly complete set of ground truth (changed pixels may be avoided while training) . Classification : Several algorithms, ranging from simple parallel - to computationally complex quadratic equations have been implemented to process remote sensing data. Classification accuracies using these various algorithms have been compared in many instances. When considering the goal of high classification accuracy within land resources the computational costs associated with more complex algorithms appears warranted due to their generally better performance accuracy. Although the assumptions of many of these classifiers appear worthy of further study it seems that greater attention is needed in the development of sequential classification procedures (e.g., tree structures) and algorithms that can effectively integrate multisensor data and incorporate collateral information. Since some collateral data are continuous (such as terrain elevation) , it can be treated as a "pseudo- spectral" channel in the classifier. Other collateral data, such as soils, however, cannot be incorporated in this manner. One method currently being investigated is to develop individual class "a priori" values for each state of the discrete variable (in this case soil types) and modify the decision rule in a Bayesian optimal manner according to the "a priori 's". Performance Evaluation : The criteria for performance evaluations are often minimal and haphazardly applied to classifications developed from remote sensing data. Most analysts have been satisfied with statements of training and/or test site classification accuracies. Classification contingency tables are also commonly employed, but additional criteria, perhaps related to classification confidence statistics, need to be developed and employed. 150 Techniques for map classification accuracy assessment have been documented by USGS . In addition to standard per pixel techniques for assessing map accuracy is the potential use of variable grid sizes that contain several pixels within each grid cell (e.g., the 10 or 40 acre grid cell sizes in common use) . When per pixel land use/land cover data have been aggregated to larger grid sizes it will be necessary to apply accuracy assessment techniques that account for this transformation since apparent per pixel accuracy would otherwise appear signi- ficantly lower than is the real case. Again, it should be remembered that once LANDSAT data have been classified they should be rectified, registered, and input into some type of georeferenced data base for most efficient use in detecting long-term changes in land use/land cover categories and interaction with other data sets used in climatic modeling. 151 . V > CJ ■H , — ^ >irsi ■H m •H Q o QJ a-. c > ^H +J r-i > o OJ P o P n3 Cr- r« U rg co OJ o CD 3 C 03 D w» a rN ^ w CN N T3 ■H ai 'a ■H QJ T3 C CO •H c rO 1-1 -p 03 u P (C Jh ro P c • (13 a, O (0 C 03 03 C co QJ ^ •H OJ G CJ > P M (T3 p M U fO U X! 03 C 'C a ■ H M-l >1 >H CJ r~ w C u r^ a, u >H •H ro O G U [fl rH C P U o o rH P w T3 fO ■H CT< a > a-. < 03 u -P O CPQ >1 O •P o C >-i cr M u cc » En iH >1-p s aj c < 0\ ■* 03 QJ c < < CP •H • QJ CN C 2 p p H u QJ rH -P CO > Eh U OJ ^H H > H CO O (13 •H p 03 3 (0 X > p ro CJ C c U ca c • qj C (fl u 5 "* CO QJ co Sh 3 co UH •H IW > ■T3 TJ •H 03 c J c C T3 ■H 03 QJ QJ • p E r^ • X •H M C c > P QJ a 3 C c > P E •H CO co >i •H c cn M C qj 03 03 C P rH a QJ (0 c 4-1 03 Xi • Qi 03 nj Cm 3 > a PI H O u (0 CM Q CO a\ >i 0> >! >i >1 CTi >. Ol > H rH rH rH rH J W o o o o O o < S e CM E CM CM m E CN E CN > H 1 1 1 i 1 1 Eh iH LD P in in o iH O p O Eh fa -h Pm rH rH CM Pm rH fa rH ,^ ^^ w c C CO Eh « rH 03 rH 03 -p rH 03 M ro 1 P D-iTi X 0-T3 P >1 rH Eh J? > rrJ (0 c •H rH •H 03 c QJ (0 C ■H rH o . P 4-1 a, w < W O c o * C o 03 o o 03 U ■H E < K u CO in 03 o o 03 in g TJ m in +J co P C CO D * *. in r\i ^ ■H r\i CN (0 rH •H (0 ■a P-, CN in •• • • O JH a rH CO U c O Q • • rH rH o O rH •-i o (0 P (0 2 H rH rH U XI CO P ■^ < CC J rH MH 5 w >i"0 C o rH to o rH 03 rH QJ (0 03 • • 03 g 03 TD 0) r-t T3 TD X < < U *— X! 03 CO P P 03 03 4J — * • • 1 QJ T3 w O C c -P p C C M CO CO C TTJ C > rH 03 QJ QJ 03 03 03 o • • OJ c o u U T3 CO O, U U c UD D D C (j-j H u •H CD P -P 1 U QJ +J U 03 M 1 u M-t C 1 co T3 C (T3 CO OJ CJ 2 rjn en c 03 03 M QJ T3 C < D --^ J3 TS -H c co P > U 03 QJ u 03 a •H MH T3 0) c c CO QJ 03 P C > M >H 03 D CO 0J S 03 03 QJ u « o CO > w tl u ■n > T3 a,vo D M U w 03 — ' c >4H C 03 OJ QJ >1 J rH ■a rt3 u O 03 S UH ■a > CO P Eh P QJ c PI hJ c o C H < U ro QJ LO c (13 u p Eh ►h" 03 U u 03 J3 J L 1 c T3 -p -a 03 •H T3 -H P T3 03 V rH H na 03 HH P 03 P >H CO C p r^ > u 3 c C M CO C 03 o 03 ro ^c C u (0 (13 S •H 03 E 2 CO J Q o> H s u CO U D ^ 152 > ro •H ro Q (N O Q) CM P 3 tn • P 3 U rH to • 3 C P H U QJ U •H CJ - < u c H CPMH O 2 < -P O tn U -PC m-i ro -h o 4-1 4J U -CM to QJ -P •H CO CU • +J P 0J U >i C rd CQ U • 0) OJ P a) +■> a u e en > to (li P •H P v 4J£C CO (0 QJ to (0 H-> O rd O. P o P , rd in M CO 03 rH rH ^^ rH rH >i to in -h >, to < w •H p P E H rd p P > s rd 0) rC O 1 > 0) rd H > > QJ P O rd > QJ E-i < i Ph -H — QJ >i < • QJ • >1 m P U CO QJ o> O c rH CO H • > o ^ 03 si o c > D In CN >i OJ u o 4-1 H U 3 cn P o CO rH CM o QJ

u QJ • >1 c r-\ •H u P P 10 U p QJ P to rd c 0) QJ 3 QJ . •H 6 to (0 u •rH > to O OS a to QJ QJ •H -H CT> ■rH (/) rH CP >H ■P tji P C p QJ to •. QJ C OJ O < O •H D to OS c > C bn rH > QJ •rH ■H rd rH O - (0 P P c p C p S C fd qj - P (0 O to on D rd 03 c o c QJ to tn QJ E c ^ •P -rH o rH c P -H ». to tO t3 -p . +J xi H! • O U XX QJ rd QJ 03 -P to 0) o « to fn w to S CO (H C m • QJ 3 • ro O O rd 2 D I tf M G D to s PS &H U "O" c IP O P r\j <£> -H O ■H fC r^ as P QJ <^ H T3 CD rd >i rH QJ E > OS 4H tO •rH P in HH r» O -H o •P QJ rH O r^ to 1 rH CO 0) <^ -P XI o to < u rH < rH < H M ro rd rH rH 3 3 X) XI to rd rd QJ En Eh P QJ •H 10 U -H >h 03 x: C to X) P QJ -P QJ 03 QJ P tnX) -O E E HH 3 03 OJ 3 o O P E QJ -P C <4H to p rH O rH -P U QJ O M UH to 3 P (d 03 QJ 3 O iw p -p 03 rH OP C E rd to •H QJ CO >i P rH O P +J 3 P 73 "O -H rd O O •H rd 03 0) Cn C QJ C c Q U rH H-> MH O CJ < 03 n-t o P P 03 rd Cu a rH a •-{ m rd 3 rd 3 s g XI rd Eh £ XI rd Eh o o O o o o O o o o o o ^ •. - to *■ CO o o to o \£> O o QJ < o < r-i in o rH \ o \ •. » ^ 2 » 2 m r~ rH P O in r-\ rH t-^ rH W C >. QJ o • -H XI -H g < CO P •T3g>,C w CO C P rH 3 > • rd rd C o D rH MH O u *-* T3 c rd rH 1 MH MH QJ CO p 3 3 CO CO -P QJ C rH CO QJ 3 D U •H P «sr "W P r- r- rj^ •r-if^ IP •H 03 U rH re > 03 o T3 P 3 U S - P 3 id to 3 « c rH to c c +J *- rH T3 CO rd c rd O «0 03 C -H > U-H c c C 03 -P P w •H 01 o o m CQ C7^ >i OJ rH -H •P C -P QJ QJ QJ XI PS 03 -P ■POO) P U OS •H (0 03 -h en P rd tO 4J -P -P C -P QJ &H 3 -P T3 to to C QJ P 03 > O rd to 03 rd qj QJ tT QJ CJ < CO ■P QJ C E P •P QJ 4J-H4H a qj rd P rd o o > C *P CO OS a O U •H fa Cm H D fe CQ 153 0) (N +) 1 • Q Eh C O 03 m P E t0 4-> W rH > 00 rH U 05 P c r- CD CD >i U -P O C0 CD C < ■H H 03 1 BHO£ 1 in id iH -H -P ro w 3 E C > -P r~ •H 03 CN 3 -P H • 03 05 C -P (T> • o -mo • • hUO» S h3 PJ W O o> OS CO Ph S CN ps a c H G a •H o P rg rc > ■ P u 0) 10 o • to c en Q H CO a> u «N ■> ■H c C ^H X rd ■H -P a ■H 03 tn w > C • P • -H m • OJ o x: , w to to • 03 tc b Ph s 2 < a a- H 10 > § to to P P 03 03 0J CD >i >i to iH 03 CD >1 o in to to x: P >H iH p 03 03 ro c Q) CD QJ >i >i >i E rH T3 rH Eh 03 C 03 <" pi 03 -P S -H •H 2 tj> (X tP o -H 03 -H fe a £ Q CO J o M o O o o CO w ». o o i-3 o o o < o *o U o o * CO ■* in o in cm in -O >-i H C 03 03 03 H P 3 •rH axi Cn 03 03 -H 2 Eh a O P o o f o CO o * * o- in t rH CM p 03 rH a 3 03 i3 s 03 Eh o o o to o p o C o •H ». m Cm •• W o 03 03 Si 'O v x: ~ 03 05 -P Z w C C 3 > 05 03 O o U U to >X> u w C 03 TD T5 TS r- -p TS rH rH 03 VD P r-i P P C 03 U 03 1 4-1 c £ 3 U to s i 03 U 03 CO U U 0) u 3 p QJ a 03 x: 03 P -p to g CD a ia 03 a UH 03 ■P •-> E o s CD 03 •H QJ U Q Eh a, >1 3 CO ro 0) P Q) a ^ s > 03 to P T> .* •H P pi •H -H 0) O 1-t 3 03 rH N in CO •H CO Q S H •H T3 CD P X >i 03 CO rH rH T3 rH 4J •H rH P P •H C •H 03 Cno •H ro O 03 CO •H in ro D < s 10 CO Q Q fe 154 5) 1-1 . CO CD X) CO (0 C oo ro CD J o c o rH O P U - >ltN CD >1p -H O -P •rl P > o T3 CD (0 VD P f\| ■H x: > m P CUP 3 tN Q) P Vi rC > • L0 (0 C CO EC p CD ^ i-i u c l-i 0) rO CO w co i«H P C Oi Q m^Tj P c O -H u <1) -P U X) 00 c u rO m P •P CP U rsi *. (T3 p C r> (0 -H Cn P H N CO rP 2 c C E rO ro ro U P -H 2 rH •H D -H X D QJ U CM (0 H > O rH - O •H O p Q (0 | ^ e •u X (0 CO C >H cq en CD ■* m rO P CD - c rH CO c C K rO CT1 c o co a c • o • o jC t •H rO s o> O 4-1 O CO -P CO u -p • O £ •H rH ro I -P C CO • P -p c ■rH CO en rH •H P ro •P H D CO W • ro • -H CD • • (0 D P .H «0 • CD U 2 D PI KDbS ^ CO O >x> 2 D PS 2 < en CO L0 Cfi Ph P P p P Q CO rC rrj rO (0 M 0) CD 0) <1) PI w >i >, >t >i < 2 > H O O o o E-i ■cj< <* ro rH P rH rH P rH rH H rO (0 rO ro ro rO <■ rH •H P rH P P a, s P •H 3 -H •H (0 2 X) P Cn £5 Cn CTi 2 o (0 rO ■H ro -H -H tp E-< Ph a Eh a Q !* ac O & P o o 2 o o o o W o * o o J < < o o Cm o < \ \ -. o O » u 2 2 IT) o Eh o CO , CM » o QJ TJ LTl >. rH O U ro c o 0) - CN Cn p < cd • P U tT> C C CD • •H U Q a, a, u ^ ro -H C 2 JC a +J d) ro o CT cn p o c c cncN ■H Q) in X! 4-1 Ojiri w CD (0 a h IS (A tn W tfi P p P U rO rO rO rO 0) CD (U OJ >! >, >i >! rO P •H Cn rH •H rO ID a P l3 a (0 •H 0) ro 2 Cn a E= 2 •r-t (0 Q a cn P o o o o o o o o o o o * «. o o o o o * » o o o o o - o o w o • • ro g < < ro w CO CO c > • • rO o D D U u >Ji M-t rH P o • p CO P CD • f0 .C d ap w C J Eh T3 M 0) Eh Oh c o -H p rO P 0) P a p CD P C TJ H P o rH O ■H CD Pi CO P o 4-1 C E3 H CD P rH CD -H >i O CO -~ CO CO c < (0 CO ■0-H2 (0 P i< C ro U ro e — u c co ■H a 10 ro rO a P 2 u 1 f0 P o •H 2 CD -H CD x: a Eh x: rH a CD a CO (0 Eh CO ■H (0 U u rO id p CO tji CO < 03 p cn co o a ss •H o a rH a ro Q U CT> a co rH 2 -H •H 2 (0 Eh ^ X3 Q CO Eh E CD a CD CD CO CO CO CO (0 CO a CO CO u < O P U rO O ro <* co m co cn CO Eh CO CQ 5^ D D D D Q 155 # 5 o • o n: U] LD - u in in w C >! X E CN 03 0) O-J CD CD tO 0) H O x U CO o u cn * •H > Q +J rd CM u -H CN •H u CD cn p en P UH c 4-> 10 l*H > -H 3 •H 3 XI >i 3 • CO Cfl CD • P P C > CO P co 4-> ^H -P u p C •H > 4-> -H 4-> U cd cn CO -H CD > -H •P p D Q rH CD rd 3 C • o fO -P C • •P < P ^ ffl -P w +J U CD CD a to -H -P Cd CD CD Q >i cd r0 S cn u c rd •h en P -P 3 in CD 03 CO P e p r> 1-1 -p l^ 4JH CD -rH CD » H D ». LT> U ■H CD 4-> 3 * m co •H C cno cn U C tT> P P c CN •H > - CU U -P C o cn (0 O Eh •HO H < * < CD (0 H o IT) > •H en rd H O r- •H - tH t* >H drl Eh P, •H u a 3 4-> I >-l Q ooa.3 p 1 > C (0 U -S "C p P rd 2 (0 c -h > Q -h C TT co >i E c T <-H U • -P < m CO U Q U > •H 1-1 P ■H <3- a> T3 P -H -=r cn c P C Q 10 X iH CD ■H a) a; • t^x: 1 ro > c c • tnx I P -H rd CO < 1 P • CO 3 r~ P Q u co co < 10 C\! -P j-i o uw< in CN CD > U P 2 ^ CD CO O lh •H o rd o r0 3 u Ifl o cn -h in < >£> cn • a -h Q fc D S CN Q 02 W D S CN D Q fN H U cn DDWtn a cm H CO PI < a o •H T3 >, P CD i M CU ^-) O -P X 4J f0 -P cn -a c •h a o n: 3 e rH P. cd fd 4-> r-l •H 3 tr>xi •h m Q Eh - >1 CD -P 4-> C rd 3 4J O CO U tH CD fd 4-> u fd ■H T3 1-1 o o 4-> 4-) cn •H fi X 3 •H a p fd fd P T3 CD CD C >i c fd O O N H 1+4 -P u < rH CJl Eh fd o < P rH a •H fd CP C e> •H «C 2 Q 1 u H CO TJ -H CO O T3 • C 4H 2 ^4-i C D a fd -H W o td fd 4J CO O 4-1 E cn, cd o cno cd a w o a) rH >, cn Eh o c o E ro u C O U » c id c g O O O rH P -H CO w * in rH -p 3 0) s rH 4-1 CN -H 4J U TS cn "3 •• 4J fd u p fO rH J3 rH cn 2 fd fd & p u c tn w •h cn h4 CD CD Eh p p M fd b< Ch e * •p < 4J Q cn CO W D CJ CO •rH rO C C ro u fd •H -H 2 13 E rO o c C c ro -H U u ■P ' — ' w fd S E M •H -P CO u >P c ^ C CD i O •H > CO to P cr, •H 3 H > CO r» •H p in Qrl I) td -P (0 MUC -P H O -H (1) - U u tru in^ < •HO H Id Eh >H HJHO 2 P O -P fO O 1) 0) Hfa£ u CO O Q +J X 3 P • CO 3 (1) CO O CO to • a -h DDWW z a a< H CO p m (0 p -a u i C (0 •H T3 O O N O H O rH 4-1 -H O O g +J (0 ■H .H P 1*" C O r- d) O o •H (0 CM o a. p O CO CO o • u ■P T3 •H -P T3 c £ MOM o Pn (0 u - •r-i T3 P c o -a p H x p o x O o a. o enja >-3 •H C § i C H (0 CO CO C -P (0 -H -a -p -H •H -H e -p c o T3 T3 td P P p m O o C -H -POO r0 tH n-t > 03 H to x c (D CO IW id •P fO a -p -h x •H o x CO > l3 iH tr> i CO (0 o C (0 c O P -H CO inn-H 3 o H+J(N-r|+JU'O01 » 3 " *» (d O M B hah nig id a u (0 O -P Cno O O O C o E (8 O O r-\ w CO OP 1 a O m p co g 1 in oo co 3 (0 >,rH 00 O - CO X! w h id i -p x x > P XI 2 3 -P -P "O O o (d O O O P C O u o>Hin.QiH to to id ZOoo< o O H *W T3 o P CN o r- > CT\ O H o o O O O -P X « O O P O a. - Id H H CO u o o o to trim *o tr> o O O E OH Om o oo i-i O T3 > o o O -P P H O I ■P c O CO O E C -H O O co >i-H -H CO •p -a -p -h •h to to 2 O OS o td -hp a. ct>h o td c a p u -h &, o Q* < r-\ ■po. a. to td c x o 2 o w S3 157 CLIMATE INVENTORY Working Group 7 Chairman: Members : Kenneth Hadeen Carmen Benkovitz Bernard C. Diesen David Drury David Durica Douglas Edwards J. Patrick Gary Eugene Hoppe William Klint Christopher Noe Douglas Troop SUMMARY The Climate Inventory Working Group reviewed requirements for data inventory and data infomation, reviewed some existing data inventory/information systems, reviewed existing technology, and developed a feasible approach to meet known user requirements. The Climate Inventory Working Group recommends that the National Climate Program Office take active responsibility for coordination and implementation of a climate data inventory/ information program to: • Establish a core implementation team. • Allocate appropriate dedicated resources for the development of the program. • Include interagency participation in a steering group capactiy. Working Group 7 identified 10 major tasks that must be addressed to fulfill the above recommendations. In reviewing pertinent sections of the draft 5-year plan of the National Climate Program Office, Working Group 7 recommends 159 that the Plan be strengthened in the data management and informa- tion services section. The following recommendations are made: • Existing resources be utilized during the remainder of fiscal year 1979 and fiscal year 1980 to develop a management implementation plan for the organizational structure and inventory service activities. • Dedicated resources be made available in fiscal year 1981 for initial implementation activities with a full implementation of the program within the 5-year time frame. • An interagency steering group should be established at the earliest possible date to guide, review, and advise a small, fully dedicated core implementation team charged with the responsibility for developing the implementation plan and specifications for required resources . BACKGROUND Inventory systems have undergone considerable change in the last decade to meet the requirements brought about by an ever increasing volume of data, types of data, and user demand for data. Automated systems have evolved - sometimes haphazardly as a way to get the job done. Initial inventories were oriented toward single files, or even frequently used subsets of single files, were fragmented, were updated randomly, and had differing levels of information. Historically, these inventories were created in response to user demands for data from individual agencies . Only in the last 10 years have serious efforts been made to design higher level, user-oriented inventory systems. The explosive growth of users 1 demands for accurate, timely infor- mation has followed the great advances in hardware technology and software design. Today there exist a limited number of sophisticated data inventory/information systems. The creators of these systems found it necessary to combine like data inventories, in differing forms and stages of completion, from many agencies involved in similar data collection/processing efforts. With these various agencies cooperating in the design and development of these high level inventory/information systems, and using their many existing individual inventories, they implemented a unified inventory 160 system from which multi-disciplined information is now available. Some examples of such systems are National Water Data Exchange (NAWDEX) , Remote Control Bibliographic Data Service (RECON) , Environmental Data Index (ENDEX) , Ohio College Library Center (OCLC) , etc. These are not the only high-level systems in existence but are presented as representative of usable inventory systems. There are many other high-level inventory systems, that have the same or even greater scope and magnitude, in the planning/ development stages. An example of such a system is the present Environmental Data and Information Service (EDIS) effort to combine its five centers 1 data collection/processing/archival resources to produce a single Data Dictionary/Directory system. This system will contain data inventories, software inventories, definitions of software variables, abstracts, history, and location/ media information. Other examples include similar efforts by the U.S. Department of Agriculture (USDA) , National Aeronautics and Space Administration (NASA), State governments, and others. On the international scene, most major international organi- zations, World Meteorological Organization (WMO) , Intergovernmental Oceanographic Commission (IOC) , Food and Agriculture Organization (FAO) , United Nations Environment Program (UNEP) , etc., concerned with climate or climate-related data have or are planning data surveys and inventories. These are at various stages of develop- ment and contain varying levels of detail and scope of coverage. In addition to existing agency contacts, access to and interaction with these organizations will be important, if not mandatory, in developing information on global data sources. TECHNOLOGY CONSIDERATIONS Any recommendations for implementation of an inventory system of climatic information must logically consider the current technology available to support such a system. Computer Hardware During the past two decades, there has been a dramatic decrease in the costs associated with computer hardware. At the same time, there has been a substantial increase in the pro- cessing power of computer systems for each dollar invested. The advent of the personal computer market makes possible access to digital systems by unsophisticated users for under $3,000. This places on-line capability (including hardcopy output) in the hands of consulting meteorologists, local agencies, and others. 161 Computer Software Computer software technology has also made great strides over the past decade. These improvements include more efficient operating systems and more powerful basic-level and higher-level programming languages. In addition, software to support inter- active processing, on-line program development and data entry, and sophisticated telecommunications permit the user maximum flexibility. One of the more recent significant developments in data processing are Data Base Management Systems (DBMS) . These soft- ware systems have proliferated over the past several years to meet the more complex information handling requirements of the user. Typically with the development of computer systems, the early DBMS ' s were not very cost effective, nor did they always satisfy the wide spectrum of demands placed on them. Today, these less efficient systems are giving way to more sophisticated systems which provide the user greater flexibility, and provide lower computer operating costs for a relatively low capital investment . Computerized Reference Services Computerized reference services are another resource to an automated inventory of climatic information. These systems provide access to information about data being collected, data available at a particular location on a particular subject, or abstracts and bibliographic reference data on published reports. Such a service is provided by the Department of Agriculture's Current Research Information System (CRIS) . CRIS provides brief abstracts on research in progress within the Department or at various institutions being funded by the Department. The system is resident on the Lockheed Electronics Company DIALOG system. Retrieval requests can be submitted via the USDA or, if the user has an account with Lockheed, directly via a computer terminal. Telecommunications Developments in digital telecommunications continue to reduce costs and improve the accuracy and speed of transmission for large volumes of data. Leased telecommunication services through the various communication companies have expanded in recent years. These services range from low-speed dial up circuits to very high speed, high-volume satellite communications links between major cities. 162 Mass Storage Systems Mass Storage Systems (MSS) of today allow quasi-on-line storage of massive amounts of data at relatively low cost and with a much improved retrieval rate over conventional magnetic tape storage. Several such systems are presently available and others are currently being developed. USER REQUIREMENTS Based on workshop comments and the experience of various data centers, consideration must be given to the variety (multi- disciplines) of users and the attendant multiple data sources in order to satisfy the users' requirements. The assessment of user requirements for data and information is an important consideration in the design of an inventory/ information system. Due to the highly variable nature of user requirements, the system must be designed to meet a continually changing community of users and their demands. Some of these demands include: • Data description Parameters available - Time/space description Indication of data quality • Indications of available formats • Rapid response to inventory requests PROGRAM CONCEPTS The climate data inventory/information program should con- sist of a centralized facility with the mission of identifying and improving access to existing climatic data resources. It should have the responsibility for: • Coordinating and fostering maximum use of existing climate-related inventory and service systems. • Developing additional inventories and systems necessary to describe and index new and additional climate data. 163 INVENTORY PROGRAM CONCEPT CLIMATE DATA INVENTORY/ INFORMATION PROGRAM (CDIIP) GLOBAL NATIONAL EDIS REGIONAL STATE NASA 164 • Developing a viable program of user services to provide access to its own and participating data and information resources. A structured or hierarchical data information system (not necessarily digital) complementary to the level of information needed would best suit these needs. A hierarchical structure with each level increasing in data information content is con- ceptionalized (see figure) . The inventory/information program would, in a broad way, point the user to the source of the particular data that relates to the user needs. It would be an "inventory of inventories." Once the what, where, and how has been established relative to particular needs, the user might then elect to access the next level of information. This lower level would describe specific data related to the user needs. Ever increasing detail could then be called to the point of ordering the data or products from the data source. RECOMMENDATIONS A. Climate Inventory Working Group Recommendations The Climate Inventory Working Groups recommend that the National Climate Program Office take active responsibility for coordination and implementation of a climate data inventory/ information program to: • Establish a core implementation team. • Allocate appropriate dedicated resources to the development of the program. • Include interagency participation in a steering group capacity. In order to accomplish the above recommendations, the following tasks must be addressed. 1. Encourage (possibly mandate) individual Federal agencies to conduct and maintain detailed survey/compilations of their climate data holdings . (Circular A-62, 1963) In the current scenario of multiplicity of data sources and usage, accurate, up-to-date inventories of a Federal agency's data holdings become indispensable to the efficient management of such data. In order to make a national climate 165 data inventory/information program feasible, these in- dividual agency inventories become imperative in the area of climate related data (USGS 1976). The NCPO should ensure that all participating Federal agencies are aware of the needs of the Climate Plan in this inventory area and should encourage such inventories when needed. 2 . Conduct a survey of non-Federal agency data producers/ managers to identify their climate related data holdings. Not all climate related data holdings reside within the domain of Federal agencies. In order to ensure complete- ness, the climate data inventory/information program must include these non-Federal holdings. The NCPO should assume responsibility to ensure that a survey of these climate data is accomplished and maintained, and that information on this survey is available as part of the data inventory/information program. 3. Develop a categorized but detailed list of user require- ments for different levels/capabilities of climate data inventory. User requirements are expected to be greater than the capabilities of existing inventory systems. An exhaustive listing of existing user requirements will provide a base- line for the climate data inventory/information program. Existing user requirements, when compared with the current capabilities of existing inventory systems, will indicate the magnitude of additional resources necessary. 4. Conduct a detailed comparative study of the capabilities of existing inventory systems. In order to create an adequate inventory system, it will be necessary to conduct such a study. The study will pro- vide the basic information required to assess the capacity and unique handling requirements associated with climatic information. 5 . Conduct a design analysis to identify and evaluate a number of inventory alternatives ! This recommendation addresses the need to evaluate alterna- tive approaches to establishing an inventory system. The 166 advantages of a centralized versus a decentralized com- puter system must be considered. The methodology for capturing inventory information from a variety of diverse organizations must also be considered in recommending the desired approach. 6 . Develop an implementation plan for the phased develop- ment of a national climate data inventory suitable for meeting climate data user requirements^ With the results of the previous analyses, the NCPO should require that the core implementation team develop a re- viewable document describing the method for developing the required national climate data inventory. This document should include a description of milestones commitments of funding and people resources, and assignments of imple- mentation-related managerial responsibilities. This docu- ment would serve as an implementation plan to guide actual implementation activities. 7. Develop proper feedback mechanisms to increase user involvement both in the development of the evolving climate data inventory/information program and in identifying other climate data management areas needing further attention . Increased user involvement during each phase of planned implementation must be encouraged through responsive feed- back mechanisms to allow the data inventory/information program to evolve to satisfy the user community needs. Other candidates for climate data management may be identi- fied for inclusion into the inventory program. Some of these may include inventories of derived products, climate related models, etc. 8 . Implement the inventory and develop procedures for continued maintenance . Implementation of the plan developed under recommendation 6 should begin as soon as possible. The value of an inventory system is directly proportional to how up-to-date it is. Therefore, guidelines and procedures should be developed for the continued maintenance of the inventory. The NCPO should ensure that these procedures are implemented. 9. Devise a mechanism to increase public awareness of the existence of, and the services provided by, the climate data and inventory/information program~ 167 Users and potential users of the inventory must be made aware of the capabilities and services available through the inventory program. An educational process must also be established to interested users in the use of services and product. This will include improving the awareness of all other systems and services participating in the program. 10. The Climate Data Inventory Program should , when opera- tional , provide a service to answer queries from user s who do not have access to a remote terminal" This could be accomplished by: • Periodically publishing a catalog of the inventory. • Establishing an operational office or delegate to a line organization the responsibility to answer users' queries via commercial communication lines to an extent that this service cannot be met by existing archive centers. B. National Climate Program Five-Year Plan Recommendation The previous recommendations of the Inventory Working Group respond to the requirements of the 5-year plan of the National Climate Program, Section IIC on Climate Data, Information, and Services. The plan recognizes the need that climate data users must know which data exist and how they can be obtained. The plan specifically addresses the goal of the climate infor- mation services component to establish an active program to facilitate the use of climate and climate-related information for improved planning, policy-making, natural resources, manage- ment, design, and operation in both the public and private sector of our national activities. In support of the plan, the Inventory working group gave careful consideration to the need for flexi- bility to improve public awareness, and to inventory a wide diversity of climate observations made in support of a wide array of data-collection programs active in a multi-disciplinary program, muliagency environment; accounting for the diagnosis and projection capabilities of inventories sources; maximum utilization of, and close coordination with, existing inventory and data systems in climate-related programs; and designed flexibility to respond to the changing needs of a diverse and changing user community. 168 It is therefore recommended that the 5-year plan be strength- ened to display a more assertive program of response to the stated needs relating to data management and information services, It is recommended that: • Existing resources be utilized during the remainder of fiscal year 1979 and fiscal year 1980 to develop a management implementation plan for the organizational structure and inventory service activities. • Dedicated resources be made available in fiscal year 1981 for initial implementation activities with a nearly full implementation of the program within the 5-year frame. • An interagency steering group should be established at the earliest possible date to guide, review, and advise a small, fully dedicated core implementation team charged with the responsibility for developing the implementation plan and specifications for required resources. 169 CLIMATE INVENTORY ACRONYMS AGENCIES SO FTWARE SYST1 ms A NAWDEX S RECON S ENDEX S OCLC A USDA A NASA A A A EDIS CRIS DIALOG NCPO WMO FAO IOC UNEP ACRONYM LIST - National Water Data Exchange - Remote Control Bibliographic Data Service - Environmental Data Index - Ohio College Library Center - U. S. Department of Agriculture - National Aeronautics and Space Administration - Environmental Data and Information Service - Current Research Information System - Lockheed's software package - National Climate Program Office - World Meteorological Organization - Food and Agriculture Organization - Intergovernmental Oceanographic Commission - United Nations Environment Program TNRIS - Texas Natural Resources Information System GLIRC - Great Lakes Information Referral Center 170 REFERENCES Circular A-62, 1963. Policies and procedures for the coordination of federal meteorological services, Office of Management and Budget, 4 p. Report 1976. Development of a catalog of information on surface meteorological data, U.S. Geological Survey, Office of Water Data Coordination, Interagency Advisory Committee on Water Data, 65 p. 171 II. SPECIAL TOPICS OPEN LETTER TO THE U.S. DEPARTMENT OF AGRICULTURE Kenneth D. Hadeen, Acting Director Center for Environmental Assessment Services National Oceanic and Atmospheric Administration U.S. DEPARTMENT OF COMMERCE National Oceanic and Atmospheric Administration Hockville. Md. 20852 June 13, 1979 Dr. Galen Hart of the U.S. Department of Agriculture (USDA) asked the National Climate Program (NCP) Data Management Work- shop to provide guidance on how the USDA data cataloging and indexing system should be designed to be most compatible with relevant data bases elsewhere. It was not possible, within the context of the Workshop, to address specific agency institutional requirements for cataloging and inventory design criteria. However, there exist several review studies of the configuration and status of major existing inventory/information systems in the U.S. No comprehensive or cohesive study has been made to date which would serve as the basis for the design of a compatible agency system. USDA and other agencies contemplating development of such a system should coordinate their development plan and design criteria with their major data interfaces (suppliers and archives) in cooperation with the NCPO Data Management Staff and advisory of Climate Data Inventory groups designated by that office. Experience gained through the Workshop data survey and back- ground developed by Working Group 7 members indicates that there is overall similarity of information content between existing inventories. At a given inventory level, disciplines or agency standards (where such exist) , specific special area requirements and mode of development have resulted in a multiplicity of formats, nomenclatures and codes. These differences can present serious difficulties in deriving compatible information from many systems for multidiscipline climate program uses. When developing new systems, we strongly urge that agencies coordinate their design specifications with existing interdisciplinary systems such as the 173 National Water Data Exchange (NAWDEX) to gain compatibility among codes and nomenclatures. Regarding cataloguing and retrieval systems: NASA and the Department of Energy both use the REmote CONsole Bibliographic Information System (RECON) . In addition, the CIA is also considering RECON. There may be some advantage to USDA adopting a system which is in the public domain and with which other agencies have had long experience. Mr. Van Wente of NASA's Scientific and Technical Information Office can provide detailed information on RECON. That system, and variants marketed by Lockheed Corporation are not unique. Other systems (e.g., at Department of Justice and National Library of Medicine) accomplish similar tasks. There are many avenues available to meet USDA requirements. We hope that whatever actions the agency takes will be coordinated with existing data bases elsewhere and with those having important needs for USDA data. 174 OPEN LETTER TO THE DEPARTMENT OF DEFENSE Roy L. Jenne, Manager Data Management Activities National Climate Program Office U.S. DEPARTMENT OF COMMERCE National Oceanic and Atmospheric Administration Rockvilla. Md. 20852 Lt. Col. E. W. Friday raised the question: how should DoD climate data and analyses be made more readily available to potential users in the civil sector. DoD data are, of course, collected and analyzed to meet requirements of the various services. However, those data can be of value more broadly, so methods have been set up to make those data available to the public. Often, the problem is on the user side of the interface. The user needs to know how to access the data for DoD facilities often do not deal directly with general civil sector users. The problem then is one of publicizing the sources of DoD data and analyses. Here is a listing of DoD data sources. Basic Atmospheric Data Conventional data are prepared for global military and civilian stations by USAF-ETAC and by Navy-FNWF (Monterey) . Satellite IR sounding data are prepared from Defense Meteorological Satellites Platforms (DMSP) . The USAF and Navy data primarily are available through NCC and some through NCAR and NORPAX administrative office at Scripps Institute of Oceanography. Some Navy data is only available at Monterey, but arrangements may be made to store additional Navy data files at NCC, if necessary. Precipitation, Hydrologic, Snow and Ice Data Precipitation (rain and snow) is available in the USAF weather data file and can be obtained through NCC. The Navy unit at Suitland prepares and publishes ice largely from satellite data. 175 The Army Corps of Engineers gathers some data on river stages, river discharge, and water quality. Data gathered for short periods don't get into the USGS files; they remain with the 37 Army Corps Districts. There may even be some long-term observing sites that aren't available through USGS. This is now being resolved. The Army Corps of Engineer's Cold Regions Research and Engineering Laboratory (CRREL) has prepared ice and snow data and compiled bibliographies. Selections of the data will soon be available through EDIS/WDC-A for Ice and Snow in Boulder, Colorado. Oceanographic Data The Navy and Air Force each decode and archive data from the global military and civilian surface ships. These data are available through NCC or NORPAX. The Navy decodes XBT data which are available through NODC . Radiation, Physics, Chemistry Data are collected by the DMSP satellites. The IR soundings are stored at NCC. Data for parameters such as protons are saved at EDIS/NGSDC in Boulder. The DMSP satellite ozone data are processed at the Satellite Ozone Analysis Center (SOAC) at Lawrence Livermore Laboratories. Satellite Imagery Data The DMSP images along satellite swaths are archived at the University of Wisconsin. The rectified polar images (composites from different swaths placed on standard map projections) are archived at WDC-A (Ice and Snow) in Boulder, Colorado. The maps for tropical areas will probably be kept at Colorado State University, Department of Meteorology. The satellite pictures are on film which can be borrowed, but not copied. .5^ s 1 176 I '^V^ofcO DISPOSITION OF DATA INVENTORY INFORMATION Dudley G. McConnell, Acting Associate Director Information Services National Climate Program Office U.S. DEPARTMENT OF COMMERCE National Oceanic and Atmospheric Administration Rockville. Md. 20852 In preparation for the Workshop, the Working Group Chairmen sought to compile inventories for their data regimes. This was done by circulating Data Set Information Sheets (i.e., data sheets) to data repositories and data managers. The format for the data sheets was developed by the Climate Program Office; mailings were made by the Chairmen and Working Group Members; responses were received, logged, and tabulated by NOAA/CEAS (Mr. David Drury) . The table below summarizes the responses. The completeness of the responses varied over a wide range. So it is necessary to review the data sheets in detail, amend those sheets which are incomplete, and apply uniform criteria in assessing the data sets. It was originally hoped that much of this could be achieved at the Workshop; however, there was insufficient time. Nevertheless, we should not squander all of the effort expended in collecting the data sheets. Therefore, the National Climate Program Office and NOAA/CEAS will work together to see that the collection of data sheets is finally published. There are funds in the NOAA FY80 budget to start work on an inventory project. The Workshop agreed that publication of this collection of data sheets would be a useful step toward a comprehensive inventory. 177 NDP DATA MANAGEMENT WORKSHOP DATA SET SURVEY SURVEY SOURCE LIST June 6, 1979 L. David Drury NOAA/EDIS/CEAS SURVEY SOURCE AES/LOG (07-04) AES /MCKAY AES/SURVEY BENKOVITZ DARLING DOD/OCEAN DOD/SURVEY GMCC JENNE KLINT NASA NCC/EDIS NCC/SDSD NOAA NORPAX TROOP WDC - A/GLAC I OLOG Y WG1 WG2 WG3 WE 4 WG5 WG6 ENTRIES TOTAL ENTRIES 44 1 249 1 17 62 109 13 203 1 1 22 16 1 41 1 42 9 14 27 10 2 27 918 SUMMARY BY GROUP GROUP NUMBER 1 278 2 150 3 197 4 174 5 25 6 64 7 32 178 jgr-fe Center for Climatic Research UNIVERSITY OF WISCONSIN 122S West Dayton Street Madison, Wisconsin 53706 608-262-2839 19 April 1979 Dr. Roy L. Jenne National Climate Program Office U.S. Dept . of Commerce NOAA Rockville, Maryland 20852 Dear Roy, I'm sorry I can't attend the meeting on the Climate Data Base. I would like, however, to make a few comments regarding this most important aspect of the overall research strategy. My own involvement in efforts to define data base require- ments started with the international GARP climate conference held in Stockholm in 1974 — published in GARP 16. That report attempted to estimate the scope and accuracy of climate data sets for research purposes. Many of these estimates have since been modified, but it remains important to produce an initial data set and a plan for improving it. I think that the experience of NCAR is worth recalling. The NCAR effort to gather together certain atmospheric data sets and make them available in compact form has provided a tremendous stimulus to studies of the general circulation and interannual variability . The draft report of the JOC Climate Board's Working Group on Data for Climate Research (1976-1977) represents in my estimation the best guidelines on how to proceed. As you know, the report of the Climate Dynamics Panel to the US Committee for the GARP (Elements of the Research Strategy for the US Climate Program) noted that the JOC report could be used as a basis for further development of data base plans. I've attached a few sheets extracted from "Elements" that suggest broad guidelines. It is my judgement that both "basic" and "applied" research would benefit greatly from easier access to atmospheric, oceanic and hydrologic data sets. If I can answer more specific questions, please call. Sincerely, JEK:mw Enclosures : a) Elements - Recs b) Elements - Sec. Elements - Sec. 1,2 3.2.1 4.1 INSTITUTE FOR ENVIRONMENTAL STUDIES 179 /s/ John John E. Kutzbach, Professor Department of Meteorology Director, CCR NOTE: Extracted from "ELEMENTS." » RESEARCH STRATEGY FOR THE UNITED STATES CLIMATE PROGRAM Fulfillment of objectives 2 and 3 will depend to a certain extent on progress toward objective 1. However, knowledge of how the climate system operates can be very useful in sensitivity and predictability studies and need not require a complete understanding of the system. Thus, research in these three areas can proceed in parallel. The research strategy for the climate program builds from a base of considerable knowledge, summarized briefly in Chapter 2. The degree to which climate is predictable is not yet known, nor can the degree to which climate is sensitive to man-made changes be assessed. However, there is sufficient information to define the elements of an expanded research program that will address these questions. Two lines of research activities are anticipated: (1) efforts of individual scientists covering a range of theoretical and empirical studies and, concurrently, (2) joint efforts of many scientists concentrating on the following four major tasks: 1. Development and maintenance of an accessible data base. 2. Development and maintenance of an observing/monitoring system. 3. Development and conduct of climate process experiments. 4. Development of climate models. Chapter 3 provides a long-term strategy for these four major activities. Chapter 4 provides a detailed listing of major activities to be continued or initiated during the five-year period from 1978 to 1983. It concludes with a discussion of how these activities should contribute to the program objectives in climate diagnosis, climate sensitivity, and climate predictability. 1.2 PRINCIPAL RECOMMENDATIONS The principal recommendations are as follows: 1. It is recommended that the instrumental record climate data base be properly assembled, checked, synthesized, and made easily accessible to scientists engaged in climate research (see Sections 3.2.1 and 4.1). Many of the data sets needed for climate research will also provide information for users. Therefore, efforts invested in this area should yield immediate and long-term benefits . 2. It is recommended that the noninstrumental record climate data base be further developed (see Sections 3.2.1 and 4.1). The instrumental record is relatively short and geographically incomplete. Heavy reliance must be placed on the analysis of noninstrumental records to determine patterns and processes of natural climate variation at time scales longer than a decade." 180 Extracted from "ELEMENTS" (Continued) (General Climate Research Strategy (Long Term) ) "3.2 STRATEGY FOR FOCUSED EFFORTS The strategy for focused work in the four areas mentioned in Section 3.1 is based on our present level of understanding and the present state of the data base and the observing/monitoring system. These, to a large extent, will dictate the timetable for achieving various objectives. The following itmes will be elaborated in Chapter 4 . 3.2.1 Climate Data Base Observations are fundamental to the study of climate. Most of the advances in our knowledge and understanding of the atmosphere and the ocean have been based on observations. Improvement of the present instrumental record data base is possible in the next few years and is, therefore, a high priority for immediate attention. Many of the data sets needed for climate research will also provide climate information for user applications. Therefore, efforts in- vested in this area should yield immediate as well as long-term benefits . The length of instrumental records (on the order of 100 years) limits studies of climate variation to time scales up to the order of decades. Furthermore, the instrumental records are not necessarily representative of other climate regimes (glacial, as an extreme example) and may not be of sufficient length or spatial coverage to reflect extreme events adequately. Efforts must be made to develop a noninstrumental record data base. Significant improve- ment in this data base can be expected in the next few years . However, a long-term effort will also be required to realize the full potential of past climate studies." 181 NOTE: Extracted from "ELEMENTS." "4 SPECIFIC RESEARCH NEEDS FOR THE NEXT FIVE YEARS (1978-1983) Within the framework of the long-term research strategy described in Chapter 3, an outline is now given of many of the specific research needs for the next five years (1978-1983) . This refers only to large, collaborative efforts involving the development of the climate data base, the climate observing/monitoring system, the process/regional experiment, and the climate modeling capability and does not mention the broad range of research projects that need to be undertaken by individual scientists. 4.1 DEVELOPMENT OF A CLIMATE DATA BASE 4.1.1 Development of a Climate Data Base from Existing Instrumental Records The development of climate data sets from existing instrumental records is an important immediate task. A major effort is required to assemble the data, assess data quality, and make the data sets readily available in appropriate form to scientists. For certain purposes, it will also be important to synthesize the data from different sensing systems (space, land, and ocean based and air- borne) into a consistent and optimal form. First, many of the data are not available on magnetic tape or similar storage. Quality control is important yet sometimes absent or inadequate. Second, the data sets often have not been gathered together and structured for easy use. For example, an early 12-year set of all U.S. rawinsonde observations was scattered across several hundred tapes. The National Climatic Center has recently gathered these data into logical order on about 50 tapes. The cost of gathering and preparing data is usually high compared with the cost of exchanging it. Thus, a tape costing $15 can hold the equivalent of 200,000 to 500,000 punched cards, while the keypunch- ing of 500,000 cards may have cost $75,000. A climate study of the southern hemisphere required several man-years and about $300,000 to prepare and publish, yet the digital values are now contained on a portion of one magnetic tape. Over the past several years, the National Center for Atmospheric Research (NCAR) has assembled, checked, and compacted a variety of climate data sets. These are now available on magnetic tape to the research community. Although these data sets are only a first step toward development of a comprehensive climate data base, they have been most useful and form a point of departure for further development. The inclusion of more climate variables, 182 Extract from "ELEMENTS" (Continued) continued work on data quality control, and extension of the data set in terms of global coverage and record are now needed. In summary, data may be available in principle; in actuality, data are often not available for easy use and exchange. The main priority, therefore, should be to obtain or create the various basic data sets. The Report of the JOC Ad Hoc Working Group on the Global Data Base for Climate Research (GARP, 1976b) can be used as a basis for further development. Three subtasks can be identified as follows: 1. Assemble "1950-present" data set. This set would include most of the observations of three-dimensional atmospheric structure from the radiosonde network as well as more recent satellite-based observations (target date for completion: 1983). 2. Assemble "long-term instrumental record" data set. In selected locations, instrumental records from the atmosphere, ocean, and land surface go back to the 1800 's and, in a few cases, the 1700' s. While not global, these records are valuable for estimating time and space scales of natural variability and for exploratory studies of processes (target date for completion: 1983). 3. Assemble "compressed" FGGE data set. The FGGE data set will provide the most detailed picture to date of one annual cycle. The above-mentioned JOC document defines the content of the com- pressed set as consisting of the time- and space-averaged statistical properties of all the FGGE data sets (target date for completion: 1980-1981) . A working-level group (or groups) should be constituted as soon^ as possible to advise the government on priorities for specific tasks within these three categories. Certain data sets can serve both research needs and the needs of users of climate information. It is, therefore, important to be aware of the needs of both research and user communities in order to optimize the data proc- essing. 4.1.2 Applications of Technology to More Efficient Acquisition, Processing, and Retrieval of Present and Future Climate Data There is an immediate need to apply available technology to problems of data acquisition, processing, and retrieval. The tremendous volume of data for climate research is currently overwhelming the data system. For example, certain meteorological satellite signals are not being routinely processed or stored, and large numbers of bathythermograph recordings have never been processed. Without significant improvements in the data-management area, the increased data flow will not be handled properly. The following possible development are noted: 1. Use of modern technology such as microprocessors to extract climate information from the real-time data flow. For example, certain space- and time-averaged statistical properties of the 183 Extract from "ELEMENTS" (Continued) satellite ratiation measurements could be automatically acquired. The climate information could be derived from the composite observing system as well as from individual components of the observing system. 2. Increased use of satellite communication links to acquire data from remote locations. 3. Use of modern data storage, retrieval, and display systems. 4 . Use of multiterminal links to these systems and to large computers. In all of these possible developments it will be important to link the technological aspects of data management with scientific uses so that data sets of high quality and utility can be obtained. This will require investment in scientific manpower as well as electronic hardware (target date for implementation: 1983). 4.1.3 Development of Climate Data Sets from Noninstrumental Records The general rationale for this work is developed in Section 3.2. Currently, there is considerable research in this area; but the task is large, and long-term support will be required. Some of this work will be done by individual scientists, but some will involve large collaborative efforts. The CLIMAP project is an example of a large, interdisciplinary effort to define the global climate at specific times in the past. Highest priority should be given to the assembly of the following past climate data subsets: 1. The past 1000 years (for studies of climate variation on time scales of 10 to 100 years) . This interval provides the possibility of year-by-year chronology in certain parts of the climate system over a time span that overlaps the instrumental record. 2. The past 30,000 years (for studies of climate variation on time scales of 100 to 1000 years) . This interval is chronologi- cally controlled by lk C dating. The response of many parts of the climate system during a major climate change (glacial-inter- glacial) can be studied. 3. The past 1,000,000 years (for studies of climate variation on time scales of 1000 to 100,000 years). This interval has a reasonably good chronological control and a well-developed global record of changes in the ocean. Changes in global climate that appear to have a significant cyclic component can be studied. These climate periods and time scales are most relevant for the program objectives in climate diagnosis, predictability, and sensitivity. The climate history of planet earth prior to several million years ago is certainly of interest and should be supported on its own merits, but several reasons suggest that it should not be given high priority now for purposes of this program. First, with increasing age, the climate record becomes progressively more fragmented, more difficult to read and to quantify, and loses 184 Extract from "ELEMENTS (Continued) temporal resolution. Second, the older records provide insights mainly into the climate effects of changes in boundary conditions that occur very slowly, for example, changes in a continent's position. Third, the opportunities for paleoclimatologists to interact with physical climatologists and to employ conceptual and numerical models of modern climate are more numerous in dealing with the recent past. The immediate task is to develop a more detailed plan for the gradual assembly of the data set most useful for meeting objectives of this program (Section 1.1). For each of the data subsets defined above, guidelines must be established for time and space sampling rates and accuracies. It will also be necessary to assess the resources required for field work, isotopic dating capabilities, and chemical and biological analysis capabilities. Long-term support of this work is required because the task cannot be accomplished in a few years. International collabora- tion will be essential (target date for a detailed plan: 19 79)." ., 185 III. WORKSHOP PROCEEDINGS LIST OF PARTICIPANTS AND OBSERVERS Dr. James A. Almazan National Advisory Committee on Oceans and Atmosphere Page Building 1 3300 Whitehaven Street, N. W. Washington, D.C. 20235 (202) 254-8412 Dr. James R. Anderson Chief Geographer U. S. Geological Survey Code 710 Reston, Virginia 22092 (703) 860-6344 Dr. Albert Arking NASA/Goddard Space Flight Center Code 915 Greenbelt, Maryland 20771 (301) 344-7208 Mr. Norman L. Canfield NOAA Headquarters - OA22 Center Building 6010 Executive Boulevard Rockville, Maryland 20854 (301) 443-8951 Mr. James Churgin National Oceanographic Data Center Page Building 1 - Room 404, D76 2001 Wisconsin Avenue, N. W. Washington, D. C. 20235 (202) 634-7500 Mr. William K. Coblentz Department of Energy Energy Information Administration 2000 "M" Street, N. W. Washington, D. C. 20461 (202) 254-3364 Dr. David Atlas Chief, Laboratory for Atmospheric Sciences NASA/Goddard Space Flight Center Code 910 Greenbelt, Maryland 20771 (301) 344-6925 Dr. Thomas S. Austin Director, Environmental Data and Information Service Department of Commerce - NOAA Page Building 2 3300 Whitehaven Street, N.W. Washington, D.C. -20235 (202) 634-7318 Dr. Robert G. Bailey U. S. Forest Service Rocky Mountain Forest and Range Experimental Station 240 West Prospect Street Ft. Collins, Colorado 80526 (303) 221-4390 Dr. Roger G. Barry World Data Center-A-Glaciology Institute of Artie and Alpine Research University of Colorado Boulder, Colorado 80309 (303) 492-6387 FTS 323-4311 Mrs. Carmen M. Benkovitz Department of Energy Brookhaven National Laboratory Upton, New York 1197 3 (516) 345-4135 Dr. Chris Bernabo Congressional Science Fellow Office of Congressman George Brown The United States House of Representatives 2342 Rayburn House Office Building Washington, D. C. 20515 (202) 225-6161 Mr. Philip M. Cohen National Ocean Survey - NOAA Code C2XS 6010 Executive Boulevard Rockville, Maryland 20852 (301) 443-8241 Dr. Harold Crutcher Private Consulting Meteorologist 3 5 Westall Avenue Asheville, North Carolina 28804 (704) 253-2539 Mr. James F. Cunningham NASA/Goddard Space Flight Center Meteorology Program Office Code 901 Greenbelt, Maryland 20770 (301) 344-7624 Mr. David C. Curtis National Weather Service Hydrologic Research Laboratory (W23) 8060 Thirteenth Street Grammax Building Silver Spring, Maryland 21701 (301) 427-7640 Dr. Paul C. Dalrymple Engineering Topographic Laboratories Department of Defense Ft. Belvoir, Virginia 22060 (703) 664-1561 Dr. Raymond B. Daniels Soil Conservation Service U. S. Department of Agriculture Post Office Box 2890 Washington, D. C. 20013 (202) 447-4991 Mr. Wayne L. Darnell Langley Research Center - NASA Hampton, Virginia 23362 (804) 827-2977 187 Dr. Wayne L. Decker Professor, Department of Atmospheric Sciences University of Missouri Columbia, Missouri 65201 (314) 882-6591 Dr. L. Denis Delorme Government of Canada Canada Centre for Inland Waters Post Office Box 5050 Burlington, Ontario CANADA L7R4A6 (416) 637-4227 Mr. Henry F. Diaz National Climatic Center NOAA/EDIS Federal Building Asheville, North Carolina 28801 (704) 258-2850 (ext. 765) Captain Bernard C. Diesen, III United States Air Force AFGWC/NASA Liaison Goddard Space Flight Center Greenbelt, Maryland 20022 (301) 344-8556 Mr. L. David Drury NOAA/EDIS/CEAS Page Building 2 3300 Whitehaven Street, N. W. Washington, D. C. 20235 (202) 634-7324 Mr. David Durica Crop Condition Assessment Division U. S. Department of Agriculture FAS/CCAD 3200 Auditors Building Washington, D. C. 20250 (202) 447-6277 Dr. Amos Eddy State Climatologist for Oklahoma The University of Oklahoma School of Meteorology 200 Felgar Street - Room 219 Norman, Oklahoma 73019 (405) 325-2998 Mr. M. Douglas Edwards U. S. Geological Survey NAWDEX 421 National Center Reston, Virginia 22092 (703) 860-6031 Dr. Edward S. Epstein Director, Climate Program Office National Oceanic and Atmospheric Administration EDIS/CEAS Page Building 2 3300 Whitehaven Street, N.W. Washington, D. C. 20235 (202) 634-7324 Dr. John E. Estes NASA Headquarters Code ERL-2 600 Independence Avenue Washington, D. C. 20546 (202) 755-1201 Dr. Richard E. Felch World Food and Agricultural Outlook and Situation Board U. S. Department of Agriculture Room 1248 - South Building Washington, D. C. 20250 (202) 447-7917 Dr. Albert J. Fleig NASA Goddard Space Flight Center Code 910 Greenbelt, Maryland 20771 (301) 344-5249 Col. Robert J. Fox, USAF USAFETAC/CC Scott Air Force Base, Illinois 62225 (618) 256-5066 FTS 255-5066 Dr. Elbert W. Friday Director, Environmental and Life Sciences Department of Defense Pentagon - Room 3D129 Washington, D. C. 20301 (202) 695-9604 Mr. J. Patrick Gary NASA Goddard Space Flight Center Code 933 Greenbelt, Maryland 20771 (301) 344-6079 Dr. Donald L. Gilman NOAA Climate Analysis Center NOAA National Weather Service W35 National Meteorological Center Washington, D. C. 20233 (301) 763-8155 Dr. Richard L. Guthrie Soil Conservation Service U. S. Department of Agriculture Matland Building #2 10000 Aerospace Road Lanham, Maryland 20801 (301) 436-7333 Dr. Kenneth D. Hadeen 'NOAA/EDIS/CEAS Page Building 2 - D2X1 3300 Whitehaven Street, N. W. Washington, D. C. 20235 (202) 634-7324 Mr. Galen F. Hart Chief, Research Development Branch U. S. Department of Agriculture ESCS/SRD South Building - Room 4833 Washington, D. C. 20250 (202) 447-8121 188 Dr. Alan Hecht Climate Dynamics Research Section National Science Foundation 1800 "G" Street, N. W. Washington, D. C. 20550 (202) 634-1547 FTS 634-1517 Mr. Gary A. Herbert NOAA Research Laboratories National Oceanic and Atmospheric Administration R329 Boulder, Colorado 80303 (303) 499-1000 (ext. 6842) Mr. Steacy D. Hicks National Ocean Survey 6001 Executive Boulevard Rockville, Maryland 20852 (301) 443-8254 Dr. Joshua Z. Holland Director, CEAS/EDIS National Oceanic and Atmospheric Administration 3300 Whitehaven Street, N.W. Washington, D.c. 20235 (202) 634-7251 Mr. Eugene R. Hoppe NOAA/EDIS/NCC World Weather Building Washington, D. C. 20233 (301) 763-8111 Dr. Frederick B. House Department of Physics and Atmospheric Sciences Drexel University 32nd and Chestnut Streets Philadelphia, Pennsylvania 19104 (215) 328-9527 Dr. Kent'H. Hughes NOAA/EDIS 2001 Wisconsin Avenue, N. W. Washington, D. C. 20235 (202) 634-7510 Dr. Merton C. Ingham National Marine Fisheries Service - NOAA South Ferry Road Narragansett, Rhode Island 02882 (401) 789-9326 Mr. Roy L. Jenne NOAA/NCPO/CP Room 705 6010 Executive Boulevard Rockville, Maryland 20852 (301) 443-8981 Mr. William E. Klint NOAA/EDIS/NCC Federal Building D5X52 Asheville, North Carolina 28804 (704) 258-2850 (ext. 376) Mr. Charles R. Laughlin NASA Goddard Space Flight Center Code 9 01 Greenbelt, Maryland 20771 (301) 344-6291 Mr. James D. Laver NOAA, National Meteorological Center World Weather Building Room 2 01 - 522 Auth Road Camp Springs, Maryland 20233 (301) 763-8071 Mr. Sydney Levitus Geophysical Fluid Dynamics Laboratory NOAA Post Office Box 308 Princeton, New Jersey 08540 (609) 452-6519 Mr. Harry F. Lins U. S. Geological Survey MS-710, National Center Reston, Virginia 22092 (703) 860-7791 Dr. Julius London Department of Astrogeophysics The University of Colorado Boulder, Colorado 80309 (303) 492-8759 Dr. Dudley G. McConnell Associate Director Data and Information Services National Climate Program Office, NOAA 6010 Executive Boulevard Rockville, Maryland 20852 (301) 443-8981 Dr. M. Patrick McCormick NASA Langley Research Center MS-234 Hampton, Virginia 23665 (804) 827-2466 Mr. Gordon A. McKay Director, Climatological Applications Branch Canadian Climate Center Atmospheric Environment Service, Canada 122 Brooke Street Thornhill, ONTARIO M3H5T4 (416) 667-4803 189 Dr. Douglas R. McLain National Marine Fisheries Service, NOAA c/o Fleet Numerical Weather Central Monterey, California 93940 (408) 373-3331 Mr. Daniel B. Mitchell Director, National Climatic Center NOAA/EDIS D5 Federal Building Asheville, North Carolina 28801 (704) 258-2850 Dr. Harry Moses Department of Energy EV-302 MS E-201 Washington, D. C. 20545 (301) 353-5572 Mr. Ernest A. Neil Chief, Meteorology Program Office Goddard Space Flight Center Code 901 Greenbelt, Maryland 20771 (301) 344-6291 Mr. Christopher D. Noe NOAA/EDIS/NOPC 2001 Wisconsin Avenue D782 Washington, D. C. 20235 (202) 634-7298 Dr. James P. Ormsby NASA Goddard Space Flight Center Code 913 Greenbelt, Maryland 20771 (301) 344-6507 Mr. Marion L. Parker Forintek Canada Corporation Western Forest Products Laboratory 6620 N. W. Marine Drive Vancouver, British Columbia M552C6 (604) 224-3221 Dr. Stephen E. Pazan Scripps Institution of Oceanography A-030, SIO San Diego, California 92093 (714) 452-3988 Ms. Karen W. Posey NASA Goddard Space Flight Center Code 933 Greenbelt, Maryland 20771 (301) 344-5083 Dr. Thomas D. Potter Acting Director, EDIS - NOAA Page Building 2 3300 Whitehaven Street, N. W. Washington, D. C. 20235 (301) 443-8110 Mr. Robert G. Quayle National Climatic Center, NOAA Federal Building Asheville, North Carolina 28801 (704) 258-2850 (ext. 765) FTS 672-0765 Mr. Frank T. Quinlan National Climatic Center, NOAA Federal Building Asheville, North Carolina 28801 (704) 258-2850 (ext. 245) FTS 672-0245 Dr. P. Krishna Rao U. S. Department of Commerce NOAA/NESS/MSL c/o NOAA - NESS World Weather Building - Room 701B Washington, D. C. 20233 (301) 763-8031 Dr. Eugene M. Rasmusson Climate Analysis Center NOAA, National Weather Service 5200 Auth Road Room 6 04 Washington, D. C. 20233 Mr. Michael R. Riches U. S. Department of Energy OER 400 First Street, N. W. Washington, D. C. 20585 (202) 376-9364 Dr. Ward R. Seguin National Climatic Center NOAA/EDIS Federal Building Asheville, North Carolina 28801 (704) 258-2850 (ext. 733) Ms. Marilyn J. Shartran World Data Center-A-Glaciology (Snow and Ice) INSTAAR University of Colorado Boulder, Colorado 80309 (303) 492-5171 FTS 323-4311 190 Mr. Charles R. Showen U. S. Geological Survey Automatic Data Section Water Resources Division MS 437 Reston, Virginia 22092 (703) 860-6879 Mr. August L. Shumbera NOAA/EDIS/NCC D5X5 Federal Building Asheville, North Carolina 28801 (704) 258-2850 (ext. 376) Dr. George I. Smith U. S. Geological Survey MS 26 345 Middlefield Road Menlo Park, California 94025 (415) 323-8111 (ext. 2286) Dr. Robert A. Schiffer NASA Headquarters EBT-8 Washington, D. C. 20546 (202) 755-8596 Dr. Morris Tepper Head for Special Projects Office NASA Goddard Space Flight Center Code 910.2 Building 22 Greenbelt, Maryland 20771 (301) 344-7624 Mr. Otto W. Thiele NASA Goddard Space Flight Center Code 910.1 Greenbelt, Maryland 20771 (301) 344-7624 Mr. Douglas R. Troop DOE/EIA 2000 "M" -Street, N. W. Room 7108J MS 461 Washington, D. C. 20461 (202) 254-7304 Mr. John L. Vogel Illinois State Water Survey Box 232 Urbana, Illinois 61801 (217) 333-4261 Mr. Fred B. Warren United States Department of Agriculture Room 4825 South Building Washington, D. C. 20250 (202) 447-4795 Dr. Thompson Webb Department of Geological Sciences Brown University Box 1846 Providence, Rhode Island 02912 (401) 863-3338 Mr. Louis A. Westphal USAFECTAC/OL-A Federal Building Asheville, North Carolina 28801 (704) 258-2850 (ext. 201) Mr. Donald R. Wiesnet NOAA/NESS/ESG World Weather Building Room 810 Washington, D. C. 20233 (301) 763-8036 Dr. Jay S. Winston NOAA/NESS World Weather Building Room 712C Washington, D. C. 20233 (301) 763-8167 Mr. Arthur M. Woll U. S. Geological Survey 514 National Center 12201 Sunrise Valley Drive Reston, Virginia 22092 (703) 860-6741 Mr. Peter B. Woodbury Deep Sea Drilling Project University of California at San Diego A-031 La Jolla, California 92093 (714) 452-3526 Mr. Vincent J. Zegowitz COMM/NOAA/EDIS 2001 Wisconsin Avenue, N. W. Page Building 1 - Room 427 Washington, D. C. 20235 (202) 634-7441 191 CLIMATE DATA MANAGEMENT WORKSHOP AGENDA MONDAY May 7, 1979 5:00 - .9:00 P.M. Registration 8:35 A.M. 8:40 A.M. 8:42 A.M. 9:00 A.M. 9:15 A.M. 9:45 A.M. 9:45 - 10:00 A.M. 10:00 11:30 11:30 A.M. 1:00 P.M. 1:00 - 1:20 P.M. 1:20 - 1:40 P.M. 1:40 - 2:00 P.M. 2:00 - 2:20 P.M 2:20 - 2:40 P.M 2:40 - 3:0. /.M 3:00 - 3:20 P.M 3:20 - 3:40 P.M 3:40 - 4:00 P.M 4:00 - 7:30 P.M (6:30 - 7:30 P.M * * * * * TUESDAY - May 9, 1979 SESSION I Call to Order - Objectives of Workshop Welcoming Remarks Introduction of Dr. Austin Opening Address U. S. Climate Program - Data Management Data Requirements for Research and Factors Inhibiting the Use of Climate Data * B R E A K * SESSION II Initial Organization of Working Groups * L U K C H * SESSION III PLENARY SESSION - DATA USES AND NEEDS Department of AGRICULTURE Department of COMMERCE Department of DEFENSE Department of ENERGY * B R E A K * Environmental Protection Agency Natural Resources/Non-Urban Land Use Consulting Climatology State Climatology *DINNER BREAK* Chairmen's Oversight Group)** SESSION IV Poolside (Indoor) PARLOR "A" D. McConnell E. Epstein J. Holland T. S. Austin D. McConnell R. Jenne Poolside PARLOR "A" Conference Rooms Dining Rooms PARLOR "A" G. Hart J. Holland Elbert W. Friday H. Moses Poolside H. Wiser G. Smith H. Crutcher W. Decker 7:30 - 9:00 P.M. Working Group Sessions Conference Rooms ** Oversight Group consists of the Working Group Chairmen, Workshop Support Staff Coordinator, Mr. Neil (COTR) , and Dr. McConnell. 193 WELCOMING ADDRESS May 8, 1979 Edward S. Epstein, Director National Climate Program Office National Oceanic and Atmospheric Administration Mr. Chairman, Dr. Austin, Dr. Potter, Workshop Participants, friends, it is a distinct honor and pleasure to welcome you to this Climate Data Management Workshop. This is among the first real programmatic activities planned and implemented under the new National Climate Program. The Workshop is jointly sponsored by the National Climate Program Office and NOAA ' s Environmental Data and Information Service. And from the outset, we have conducted planning and preparations in an interagency, partici- pative manner. This reflects one mode of how the Program Office will operate, but I will return to this point later on. In just a few minutes this morning, I would like to give a list of the background that motivated the Workshop and then indicate some of the hoped for results of the Workshop from the National Program Office point of view. I understand that Tom Austin will discuss hoped for results from the EDIS per- spective. The U.S. National Climate Program formally came into being on September 17, 1978, when the President approved PL 95-367, the National Climate Program Act. The stated purpose of the Act is: "to establish a national climate program that will assist the Nation and the world to understand and respond to natural and man-induced climate processes and their implications." Let me emphasize the Congress in setting the policy basis for the program included both understanding and response. That might be termed research and application . In arriving at its position, Congress held a series of hearings and among other things found that "Information regard- ing climate is not being fully disseminated or used, and Federal efforts have given insufficient attention to assessing or applying this information." And further, that: "the United States lacks a well-defined and coordinated program of climate-related research, monitoring, assessment of effects, and information utilization. " Thus, Congress mandated a program which emphasizes inte- grated planning among Federal agencies and authorizes an Inter- governmental Program with State agencies. Specifically regarding climate data and information, the list calls for: 195 • global data collection, and monitoring and analysis activities to provide reliable, useful, and readily available information on a continuing basis; • systems for the management and active dissemination of climatological data, information, and assessments, including mechanisms for consultation with current and potential users; and • measures for increasing international cooperation in climate research, monitoring, analysis, and data dissemination. I have here only quoted some relevant highlights of the Act. A comprehensive review of the Act and Conference Report will go even further to show the Congress 1 strong desire to see reliable data and information being actively used to increase the effectiveness of planning operations in both the public and private sector of our activities. The Academy's Climate Research Board also recognized the vital role of climatic data. The CRB recommended emphasis on the use of climate knowledge to serve the economy. The CRB states explicitly that for climatic data to be useful, they must be accessible and effectively disseminated to users. Thus, the goals of this Workshop are right in line with the National Program. We interpret the goals as follows: 1. To further progress on an inventory of climate data. We recognize that agencies have begun the tasks of inventorying and cataloguing their data bases. Yet, comprehensive data sets may call for contributions from several agencies — for example, say precipitation from NOAA, streamflow from USGS, snow cover from USDA. Thus, an effective access system would require coordinated access strategies among the agencies. 2. To make a preliminary assessment of the status of data sets in relation to requirements. This is recognized as an elusive goal in that there are myriad require- ments. The feeling is that there are enough stated and known requirements that certain general assessments can be made. To that end, there are many "users" represented here at the Workshop. 3. To recommend steps to increase the accessibility of the data. This is very key. NASA's experience in compiling the Nimbus Ozone Data Set is ample evidence that simply having the data is a long way from having 196 the data in an accessible, usable condition. This also raises the point that there is often a lot of expense and effort required to being data holdings to the most effective condition. Difficult trade- offs may have to be made and priorities need to be set. 4. To recommend steps and activities following the Workshop to achieve the goals. Finally, we recognize that this Workshop is a step in a process that began earlier and will continue. There may need to be periodic workshops of this type. Nevertheless, we hope this gathering will give us its views on our planning and additional efforts that may be needed to assure a successful, fully effective Data Manage- ment Program. I will conclude now by thanking you for taking time away from your very busy schedules to take part in the Workshop. It is gratifying to me, that the Office and the National Program are receiving so much support from the community --, the agencies, universities, private sector. I look forward to hearing your recommendations. 197 OPENING ADDRESS May 8, 1979 Thomas S. Austin, Director Environmental Data and Information Service National Oceanic and Atmospheric Administration Good morning : Objective premise — need for national and global climate data bases— meet needs evergrowing number and sophisticated users. I can think of no better way to express and underline the reason for the importance of this workshop with respect to the National Climate Program than to quote Bishop Epstein on the subject. In a recent article in EDIS magazine, in discussing the various elements comprising the climate program, he notes that: "...although it does include research to increase our understanding of climate processes, the emphasis is clearly on providing more and better climate information, on identifying and quantifying the impact of climate and climate fluctuations , and on delivering climate informa- tion and services to users." And, in discussing elements which deal specifically with data collection and management, he states: "Again, the emphasis is on end products: 'reliable, useful, and readily available information, and active dissemination of ... information and assessments. 1 There also is an added caveat that there be consultation with current and potential users. The message is very clear. The purposes of the program are to provide infor- mation, to assure that there are means to determine what information is most needed, and to stimulate its wide application . " These comments on the NCP Act by Dr. Epstein provide the orienta- tion and conceptual framework for your work. Unfortunately, although similar sentiments have been voiced in the past, it is a concept seldom implemented. For example, a few years ago, a National Science Foundation con- tractor asked more than 200 top Federal executives where they go their science information. Their answers? Mostly from newspapers And many of these "decisionmakers" controlled Federal information systems . 199 I think that anyone who has been involved in or followed the public and Congressional debates that preceded the passage of the National Climate Program Act would agree that this unfortunate condition applies to some degree to climatic data and information. Certainly the Congress feels that way, for the Act itself states: "Information regarding climate is not being fully disseminated or used, and Federal efforts have given insufficient attention to assessing and applying this information." We have not been making use of existing data and information and mounting expensive programs to generate more for a specific program of purpose. We have to get away from the all-too-familiar "not- invented-here" syndrome and, through interchange and cooperation, make maximum use of existing resources. The inventory you have been asked to begin is a critical step in this process. Another real-life parable for us to contemplate concerns the European Nuclear Documentation Service (ENDS) in Luxemburg. In 1974, it was the largest automated information system in Europe and the third largest in the world. ENDS computer files then contained nearly 1.5 million articles on nuclear energy. Some 45 to 50 people worked on file maintenance and user searches. The service was free. Yet, despite this tremendous resource commitment, there were, on the average, only three users a day. In commenting on the situation, Dr. George Anderla of the Sorbonne concluded : "...The system does not provide the information the users want or need because we know the users requirements only superficially." I would ask you to keep this statement in mind during your workshop, so that we do not repeat this unfortunate, extremely expensive, but very common mistake. For, as Dr. Anderla also concluded: "...This kind of situation cannot go on very long, because no government. . .will allow millions of dollars to be spent for such limited use and for such a restricted number of users . " 200 I am sure that this also is true of our Government and of any limited and restricted data management system we might develop to implement and support the National Climate Program Act. Who is the user? Who are the users? It is clear from our own experiences at EDIS that the climate-related data and infor- mation user audience encompass a much larger group than the scientist, engineer, and government manager. It includes public policy makers and politicians at the national, state, and local levels; attorneys, regional planners, businessmen, citizen's groups and individual taxpayers. At the same time, increasing public concern over environmental, ecological, energy, and other issues with a scientific base has been shifting environmental, particularly climatic, data applications emphasis from the advancement of scientific know- ledge to using that knowledge to solve global and national problems. These include potential world food shortages the energy crisis, environmental pollution, climatic anomalies such as the recent run of cold winters, planning for the develop- ment of the coastal zone, and even the ill effects of weather and climate on human health. Decisionmaking processes — aid (4 0,000) — CEAS climate/food/energy/health — GNP. Many of these new users require data and ' information in different forms than those traditionally provided to the scientist and engineer, and their applications frequently require multi- disciplinary assessments and interpretations. This, in turn, requires greater emphasis on packaging data to meet user needs. Again, to echo Dr. Epstein, the emphasis of the National Climate Program Act is on end products — it is not on building yet another governmental — intra-government — data and information system. Many of .the products will have to be understandable to the layman. In addition, there must be emphasis on packaging information to meet user needs that cut across traditional (and convenient) divisions by discipline, mission, program, and agency. To quote a phrase being heard more and more frequently in the halls of science, particularly at budget time, our products must have "social relevancy." Son of IDOE. During the coming decade, we expect the volume of user requests for multidisciplinary environmental data to continue increasing, many new data applications to be developed, and user demands for shorter response time to accelerate. To meet these and other user needs, and to achieve the degree of data integration needed to meet national needs, EDIS is integrating its own formerly separate data base operations under a single data archive manage- ment and users services system. 201 In implementing the new system, the data bases of the National Oceanographic Data Center, the Center for Environmental Assess- ment Services (Washington, D.C.), and National Geophysical and Solar-Terrestrial Data Center (Boulder, Colo.) with the data base of the National Climatic Center in Asheville, N.C., the largest climatic data base in the world, will, through referral and merging processes as appropriate, provide for multidisciplinary data and product services. Telephone lines and communications equipment already have been installed at the Washington and Boulder-based centers to provide a direct link to NCC. The installation of a high-density mass-storage system at NCC will increase the amount of data accessible on-line, and further reduce costs and response time. What will all this mean to the user? It will mean that multi- disciplinary requests that today might take several weeks to answer, will be answered in an hour or two. And relative to present costs, the quicker answer will cost less; it will be cheaper for EDIS to provide the answer, and the savings will be passed on to the user. Moreover, the sophisticated user with his own terminal will be able to plug directly into the system. And the user will get better data products; instead of three separate products in three different formats, the output will be an integrated, designed product. Within the last two years, EDIS has sponsored or cosponsored a series of data management workshops ranging in subjects from marine geology to climate and health. Some of these workshops such as the paleoclimate workshop cosponsored with the National Science Foundation in November, were designed to provide a basis for further deliberations at this workshop. The range and variety of skills and activities represented here today attest to recognition of both the importance of data manage- ment to the National Climate program and to the potential for success of this workshop. We need your guidance to develop the desired referral capability, the national and global data bases and to implement an optimum data management system, one that will be responsive to all the users of climate data and information. Without detailed, firm guidance, data management systems and centers are prone to turn out products that may be unwanted and thus unused, either because the users do not understand or relate to the product, or they do not know that the product is available Remember the example of ENDS that I cited at the beginning? I would hope that the acronym will prove prophetic, at least as far as the data management system we are developing is con- cerned, and that we will not, as so many before us have, reinvent a socially (or politically) irrelevant data management system. 202 The degree of cooperation called for if the National Climate Program is to be successful is far beyond any demonstrated to date by agenices, institutions/ and individuals working in this field. I hope that this Workshop will provide proof that such cooperation is not only possible, but productive. Thank you. 203 THE NATIONAL CLIMATIC CENTER CURRENT STATUS AND FUTURE PLANS Daniel B. Mitchell, Director National Climatic Center NOAA/EDIS INTRODUCTION The National Climatic Center (NCC) is the national archive for climatological data. Data observed by the National Weather Service network, cooperative network, satellite systems, and other national sources are archived. Data are both in manuscript or hard copy form and in digital form. However, digital data comprise only a portion of the total NCC data archive. The National Climatic Center also provides user services. Approxi- mately 66,000 customer requests were answered last year. The number of customers serviced each year has been steadily increasing for the last seven years. The type of customer requests include requests for data and information which has been tailored to meet the customer's request. As a result of the emphasis placed on services by the Climate Act, we anticipate that the National Climatic Center's services to the customer will increase dramatically over the next several years. The NCC is located in Asheville, North Carolina. The Center is divided into six Divisions and two Staff Functions. One of its Divisions, the Satellite Data Services Division, is located in Camp Springs, Maryland; and is a special Center within itself. The Administrative and Technical Services Division is responsible for the administrative and logistical support for the NCC. Their functions include printing plant operation, personnel services, procurement, budget and fiscal functions, and space allocations. The Automated Data Processing Services Division is responsible for ADP support to the Center. Their functions include computer opera- tions, data translation, and systems design and operations. The Data Operations Division is responsible for processing of the climatological data. All climatological data received by the NCC are processed by the Data Operations Division. Their functions include the receipt, inventory, edit and verification, and the preparation of special publications of the data. 205 The Climatological Applications Division is responsible for the development of climatic applications tailored to the customers 1 requests. Their functions include the preparation of summaries, statistical analyses, and assessments. The Information Services Division (ISD) is the NCC interface between the Center and its customers. The ISD basic function is to provide user services and manage the manuscript archives. The Satellite Data Services Division, located in Camp Springs, is responsible for archiving satellite data from NOAA operational satellites and for providing user services from these archives. Since the Satellite Data Services Division is not collocated with the NCC, they are also responsible for their data processing, user services, and ADP support functions unique to the satellite data services . In addition to the six Divisions I have just described, we have the Information Management Center and the Data Administrator assigned to the Director's Staff. The Information Management Center is responsible for the receipt of all correspondence and making dis- tribution of this correspondence within the Center. It is also responsible for word processing. We currently utilize up-to-date word processors and a Center-wide dictation system to allow for general correspondence preparation within the Center. The Data Administrator with the Data Base Administration Staff is responsible for the design, development, maintenance, and manage- ment of the NCC digital data base. This staff function has just recently been added to the Director's Staff. GOALS AND OBJECTIVES In the process of serving thousands of customers over the last several years and through the participation in numerous user workshops held over the past three years, we have identified firm user requirements for climatological data. By analyzing these data requirements and reviewing our data management practices, we have identified several principal problems with the NCC data base. In an effort to provide better customer services, we have established four basic goals, which we hope to achieve over the next 7 to 10 years. These goals are: (1) modernize NCC by automating manual data processing functions and by providing automated support to user services; (2) improve the data base by providing better quality control, more comprehensive inventory of data, and by providing a more complete digital data base; (3) develop an interactive digital 206 data base by making more climatological data available on-line to a computer system under a data base management system; (4) pro- vide user access to NCC data base by providing a capability for State Climatologists and certain other customers access to the NCC data base by remote terminals. In an effort to meet these specific goals, NCC has developed a short- and long-term plan. In subsequent paragraphs, I shall discuss objectives we have established, actions we are planning, and actions we have already completed in an effort to meet our goals . In our processing of climatological data for archival and user services, we still employ many manual procedures. With the current ADP technology available today, we have an opportunity to more completely automate our data processing. For example, we manually reduce many data recorded on strip charts onto a data entry form and then key entry the data into digital form by using a key-entry device. This function can be more completely automated by utilizing a digitizer with pen and tablet and going directly from a chart to digital information. Additionally, the NCC receives weather observation data in manuscript form from the National Weather Service observing stations. These data are collected and converted into digital form via an Inforex key-entry system. In the future, the National Weather Service plans to implement the AFOS system. The operational date is scheduled for 1979. The NCC is a spur on this system, and as a result, will receive all the weather observation data on this system in digital form. Thus, this will alleviate the manual key-entry of data from manuscript form which we do today. In our contact with the user community, we have learned new requirements for data quality control. The user wants better quality control of data, and that data which is quality controlled, the user wants information about the quality of the data. Hence, we plan to develop new edit and validation programs for the surface- upper air, cooperative, solar, and marine surface data sets. The development of new programs has already begun and is planned to be completed by 1981. Along with the development of new edit and validation programs, NCC has initiated a project to develop a data base inventory. The project has been initiated this year and shall be completed by 1984. Once this project is completed, NCC and NCC users shall know what data are available in both manuscript and digital form at NCC. In an effort to develop a better quality data base, NCC has also initiated the procurement of a Quality Assurance Graphical Subsystem This system, when installed, shall give the data validator an inter- active capability whereby he can graphically review data that have 207 been identified as suspect in the process of editing and vali- dating of the particular data sets previously mentioned. This system, when implemented, shall give NCC a capability to more readily examine and to more quickly complete validation of data. This system is planned to be implemented in 198 0. The NCC is also in the process of procuring a new computer output to microfiche system. The NCC produces many publications, summaries, and tabulated data each year in support of customer requirements. The capability of the new computer output to microfiche system will provide NCC additional capability to more readily tailor products to the user requirements. This new system is scheduled to be installed and implemented late 1979. Today NCC has access to some 75,000 magnetic tapes which contain digital climatological data. This large, cumbersome tape resource is difficult to manage and often difficult to access economically. As a result, NCC is in the process of procuring a mass store sub- system. With the installation of the mass store subsystem, NCC plans most of the digital data contained by the magnetic tapes onto the mass store subsystem to make it available on-line to the computer system. Once data are loaded into the mass store subsystem, quicker access and more cost effective access to the data will be a reality. Another objective within our plans is to implement an interactive data base between 1981 and 1984. I shall discuss this further in subsequent sections under the 1981 initiative. Current and planned satellites in the Geostationary Operational Earth Satellite (GOES) series of NOAA operational geostationary satellites are collecting and will collect unprecedented volumes of data. These data in their original digital forms, or processed imagery data, represent a valuable national asset with significant retrospective application potential in the areas of oceanography, marine biology, coastal management, deep water port planning climatology, solar insolation climatology, severe weather meteorology, hydrology, and agriculture. Presently, NCC archives only processed imagery and only a small amount of digital GOES data. This limited archive is inadequate to meet the user needs because users are presently limited to copies of this imagery only, and cannot specify the coverage, enhancement, and resolution required to best meet their needs. The potential benefit of GOES data in the areas cited above can only be met by archiving the full-resolution digital data, and by providing the capability of producing from the archived data the digital and image products required by the user community. Hence, NCC has as an objective to procure and implement a GOES data archive/service system. Our goal is to implement this system by 1981. Once implemented, this system would provide us with the capability to archive GOES satellite data for a period of 5 years and to pro- vide services from this archive. 208 Our long-term plans are to upgrade our computer system in the 1984 period. With the Climate Act stressing customer services, we anticipate that our user requests will continue to climb at an even higher rate than they have in the past several years. Hence, we believe that we will need to upgrade our current system to meet the future demands . Also in the same time frame, we are planning to provide access to the NCC data base to the State Climatologists and limited other users who will require a near real-time access to a climatological data base. Such access would be through a remote job entry terminal or a time share terminal. We are currently looking in the 1985 time frame for this type of interface. Today we do not possess the capability to have users other than internal to NCC access to the NCC data base. First, before we can provide such access, NCC must develop the interactive data base, which I will address later, and also enhance its ADP system capabilities. Many actions have already been initiated and some have been com- pleted in our effort to meet our goals. NCC is in the process of microfilming some 80 million manuscript weather records. The microfilmed data are being inventoried and indexed in the process of microfilming. This is being done so that the recently established Micrographics Service Center can readily retrieve microfilmed data in the servicing of customers. In an effort to modernize NCC, a Univac 1100/10 was recently installed and is operational at NCC. We are in the process of pro- gram conversion, and hope to have this particular effort completed by late summer. A data digitizer has been purchased and implemented in our Cooperative Data Branch. The digitizer is being used to automate the input of cooperative data into digital form. An additional three digitizers will be added later this year to com- plete the automation of that function. In the satellite data archive and services area, NCC has recently completed action to procure an Image Display and Hardcopy Sub- system. Once implemented, this subsystem will enable the Satellite Data Services Division to more efficiently tailor satellite data products to the customer needs. Additionally, a mass store sub- system was installed in our Satellite Data Services Division to archive satellite data from TIROS-N. Due to our major effort in the development of our data base, I have established a Data Base Administration Staff to manage the NCC Data Base. This staff is responsible for the design, develop- ment, and implementation of the data base management system and to restructure the data base so that it can be more easily accessible for servicing customers. 209 NCC 1981 INITIATIVE As I previously mentioned, one of the objectives for NCC is to develop an interactive digital data base. For NCC to complete the development and implementation of this digital data base would take approximately 7 to 10 years. However, with the emphasis that the Climate Act places on services, we feel that we must develop this data base much sooner than we could normally do it with the existing resources. Therefore, we have submitted an initiative for 1981 requesting additional resources which will enable us to accelerate the development of this data base. We are planning to complete the development of this interactive data base in approximately 5 years if additional resources are made available over the entire period. Over the history of NCC, hundreds of data sets have been established in response to, and in a mode dictated by, then existing observation systems and data needs. These data exist in digital form on thousands of computer tapes in various forms and documents. The large digital files are cumbersome, extremely expensive to use, and contain numerous duplicates and specialized products. The data vary in quality due to changes in existing procedures dictated by changes in regular/special observing systems, technology, and data require- ments throughout the NCC history. There is a serious lack of comprehensive documentation including inventory information. Gaps and inconsistencies in and between the digital sets require specialized editing techniques in processing data from hard copy records to make the data available in forms which are cost and time effective, relative to existing and projected national needs. These problems make it difficult and expensive to provide users with required data and information. In the development of the interactive data base, we plan to complete the development of data sets by filling in missing observations and extending the period of record. Also, data will be merged and restructured to minimize use of cumbersome media. Duplicate data sets will be eliminated. To provide better quality of data, NCC will more completely auto- mate and standardize data editing and validation. Additional validation will be performed on data as necessary, and the data quality will be flagged so that this information can be passed on to the users. As the interactive data base is being developed, it will be loaded under a data base management system. In an effort to improve the access and reduce the cost to the users, NCC will restructure the data base based on user statistics so that the access becomes more economical . 210 In summary, the initiative will provide: • Edited, validated, and compacted historical data files. • Complete data files. • Ultra long periods of record for selected stations. • Improved digital station library. • Reduced data costs to users. • Enable quicker response to users. • Enable NCC to tailor data products to users. The principal data sets that NCC is planning to put into the interactive data base are: U. S. hourly airway surface observa- tions, global marine surface observations, U. S. summary of day and month observations, global rawinsonde observations, hourly precipitation data, and selected global surface observations. SUMMARY The NCC is presently embarked on a massive rehabilitation of all its processing systems. New hardware, new software, and new ideas and procedures will enable NCC to process the data in a more auto- mated fashion. We anticipate that much of these upgraded processing systems will be implemented within 2 years. These new processing systems, coupled with current development of a modern data base management system and an interactive data base, will provide the facility to rapidly and inexpensively access and service the large amounts of data required by the Climate Program and other national interests including our current customer base. The development of the interactive data base is a long-term project considering our current level of resources. However, with the support of the 1981 initiative, and subsequent additional resources during the period through 1984, we would hope that we can complete the develop- ment of the data sets and the interactive data base by the end of 198 4 The following graphs and information were presented at the Climate Data Management Workshop, May 8-11, 197 9, at Harpers Ferry, West Virginia . 211 330 » • wrv 515 • NW 53a wnr • »' • •> \ . * ^ H*r 33a • mnr p e 1» « /• « * o Nvr o 330 1 z > m Q wnr « « UJ \ " • H- f UJ -J 0. 2 k*r 330 j o a z X o h- •> ^ 8- o Z o Y wnr * «% < 2 en >■ ffi MV \> 330 > e UJ „ to MDf e a: » UJ en Z> — s < Nvr *#. r-~^^ 330 I- \jr o H Nvr C V> 330 unr 1 ! 1 1 1 1 1 1 wvr o 1 o s 8 S | § 8 a • e o e z a « « " » « * * *| «■ an.j-u*i02 twoaav I > 212 DC mi J- HO O -J g < N < (3 CC O - o X T o z UJ g CO < 00 £ ' < CO »— ^^ < 2E Q < © o < 3 2 CD UJ < < > Z •- u-i — < o < a CO CO oe UJ CO uj S3 CO 1^ = s " 1 »— SB = o «» ^ CJ «^> 213 GOALS AND OBJECTIVES MODERNIZE NCC • DATA PROCESSING . USER SERVICES IMPROVE DATA BASE • MORE COMPLETE DIGITAL DATA BASE • BETTER QUALITY CONTROL DEVELOP INTERACTIVE DATA BASE USER ACCESS - NCC DATA BASE 214 >- IS) o X < © c/3 cr ujtr c — Q-< 3 CD C-5 C-3 C/5 V) o < 5 5 UJ f - 08 < < Q i I 215 ^£3 Z ^ U UJ CO eztmm oo oo go GO UJ w 216 SHORT AND LONG TERM PLANS • COMPLETE AUTOMATION DATA INPUT DIGITIZERS— 1979 AFOS— 1979-81 GTS— • IMPLEMENT DBMS/TEST DATA SET-1979 • NEW DATA EDIT PROGRAMS SURFACE UPPER-AIR CO-OP ) 1979-81 SOLAR MARINE • NCC DATA BASE INVENTORY 1979-1984 • NEW COM - 1979 COMPUTER OUTPUT TO MICROFICHE • QUALITY ASSURANCE SUBSYSTEM - 1980 • MASS STORE SUBSYSTEM -1981 • INTERACTIVE DATA BASE -1981-1984 • GOES DATA ARCHIVE/SERVICE SYSTEM - 1981 • COMPUTER UPGRADE -1984 • ACCESS-NCC DATA BASE -1985 STATE CLIMATOLOGIST AND LIMITED OTHER USERS 217 ACTIONS COMPLETED WORD PROCESSING SYSTEM IMPLEMENTED MASS STORE SUB SYSTEM - SATELLITE DATA BASE UNIVAC 1100/10 INSTALLED DATA DIGITIZERS (COOP DATA) IMAGE DISPLAY/HARDCOPY SUB SYSTEM PROCURED - SATELLITE DATA MICROGRAPHICS SERVICE CENTER DATA BASE ADMINISTRATION STAFF MICROFILM OF MANUSCRIPT DATA INVENTORY/INDEX OF MICROFILM DATA 218 THE NCC 1981 INITIATIVE )ESIGN AND INITIATE AN INTERACTIVE DATA BASE (DIGITAL) I COMPLETE • FILL IN MISSING OBSERVATIONS • EXTEND PERIOD OF RECORD II COMPACT • ELIMINATE DUPLICATE DATA SETS • MINIMIZE USE OF CUMBERSOME MEDIA • MERGE DATA III EDIT AND VALIDATE DATA • AUTOMATE AND STANDARDIZE EDITING • PERFORM ADDITIONAL VALIDATION • INTERACTIVELY REVIEW • ADD QUALITY FLAGS RESTRUCTURE DATA BASE • REFORMAT • IMPROVE ACCESSIBILITY 219 LU O O < CC Q O > LU LU or O < Q. CO Q < DC CQ o o o LU 2 o O Q CO Li. h- S LU CO o < < — 1 LU H "» Q^ _J Q C/5 CO £Z oz _j LIDATE . DATA < < G PERI STATIO < Q ED, VA ORICAI LU LU Q. {A-LON CTED Q LU O hh 52 o h- —J -J LU Q. LU X o Z> CO 220 cam < o CO co UJ co H tO O O < < O UJ o Z> Q LU q: < CO UJ CO O I- UJ CO z O a, CO UJ en a: UJ o o- UJ CQ < z UJ CO O Q O CC CL < < or O O O O UJ CQ < z UJ 2 < -J <9 H > o o ujto D- I— 221 (f> CO ^ p^Bft o < yj O ^ h™ o y. co W3*SJBB o < H- ■ < CO Z cc LU Q£ o CO ■MM LU CO < CQ o CO z cq X o O cr H H LU LU z < o CO o < < CQ n or li- Z> o o o z LU CO CQ 1 < CO < < o z CO Li. > LU o q: < Q H < ID Q Z < ^ CO u_ o H or LU o CO Q_ < z or < >■ < Z < o LU cr qz :> 2E C£ CL z> _j :§ _j > o < 3 < __l X CQ CO CQ cc * 4 4^ O • o ^ CO a CO • o o 3 o 1 222 > H GL < 01 LU H O CO O UJ CO CO ^ LU o oJ5 CO o > > < o > LU z LU CO z < < O LU 2 CO C/) 08 O Z O co H CO > CO LU O O < CQ X Ul -J Ul H ID CL Z Z> O O a: CL a CO UJ 01 o O • • • • • • 223 DATA REQUIREMENTS FOR RESEARCH AND WAYS TO PROMOTE THE USE OF CLIMATE DATA May 8, 19 7 9 Roy L. Jenne, Manager Data Management Activities National Climate Program Office We want to welcome you to the Data Management Workshop. We will first discuss the data requirements for research. John Kutzbach has prepared some written material about science requirements but wasn't able to attend this meeting. Then we will discuss considerations that may be useful in your working group deliberations about what is needed in order to make climate data in each discipline area more accessible. History First, we will present a very sketchy history of the climate activity that has led us up to this point. The Joint Organizing Activity (JOC) was organized by agreement between WMO (World Meteorological Organization) and ICSU (International Council of Scientific Unions) in 1967. Its charter called for research to promote understanding of both weather and climate problems. In 1972 there were a number of important anomalies in the world's weather including a drought in the USSR; the USSR purchased significant amounts of U.S. wheat, leading to sharp increases in grain prices. In 1973 a report was published showing relationships between growing season weather and the yield of crops. In 1973 Mitchell and Fletcher prepared a report on the ending of the interglacial . In the last million years, the world's climate has normally been cooler than it is now. Until about 1978, there were many articles reaching the public that implied a rather large chance of a rapid shift in the world's climate toward cooler mid-latitude conditions, a shorter growing season, and more droughts. In 1974 there was an international meeting of climate experts in Sweden to discuss the physical basis of climate and climate modeling. The product of the meeting was the often quoted "JOC-16" report (GARP publication number 16.) In 1973-74 many people were becoming more active in Federal climate planning through separate activities undertaken by ICAS and a subcommittee of the President's Domestic Council; locally this included R. White, J. Fletcher, E. Radok, E. Epstein, W. Sprigg, 225 R. Lavoie, and others. Fletcher and others encouraged me to accelerate the collection ot data information, which led to a publication concerning data for meteorological research in 1975. Between about 1973 and 1977, many were involved in the preparation of the Domestic Council Report: "A United States Climate Program" in 1974; and the ICAS plan: "A United States Climate Program Plan" published in 1977. The GARP Activities Office of the JOC called a working group meeting in Princeton in April 1976 to prepare a small report "The Global Data Base for Climatic Research." I was asked to prepare further information about climate data. This led to the report that you have seen: "The Global Data Base for Climate Research"; it still has not been completed. From about 1975-78, NASA appointed a "NASA Science Working Group" to help refine climate requirements. NASA used this and other information to prepare the NASA Climate Plan published in 1977. In 1978 NASA (at Goddard) printed "Candidate NASA Data Sets Applicable to the Climate Program." Karen Posey gathered most of the material for it. In 1978 the NAS report on the "Research Strategy for the U.S. Climate Program" was published. In the summer of 1978, the NAS Climate Research Board met to discuss the U.S. Climate Program. The political interest in climate was enhanced by the cold winters in the eastern USA of 1976-77 and 1977-78. President Carter signed the Climate Act in September 1978. There was a WMO working group meeting on climate data management in November 1978. The National Five-Year Preliminary Climate Plan was published (first version) in April 1979. Data Requirements for Research Figure 1 shows many of the components of the climate system. It is our job to make sure that the data necessary to monitor and study the different components is as available as possible. In some cases this will require improvements in observing systems. The observing strategy is not our main concern in this workshop. The 1978 NAS research strategy report, listed data needs as follows: 1. Prepare data from the instrumental record. • Assemble " 1950-Present" data set. -- this includes: surface synoptic data and upper air data (rawinsonde, satellite, and aircraft) 226 z o 1- co o Li. Q. OS co LU O 22 . c w W o «5 D ill iC o Q) aj a> +3 O 227 • Assemble long-term instrumental data. — data from stations starting in the 1700s and 1800s • Assemble compressed FGGE data set. — time and space averaged statistics 2. Develop a non-instrumental data base. o Recent 1,000 years. — to study changes for time scales 10-100 years • Recent 30,000 years. — to study time scales of 100-1,000 years • Recent 1,000,000 years. — for understandings of the time scales - 1,000-100,000 years — World Weather Watch: pressure, temperature, humidity, wind -- Commercial aircraft observations — Satellite soundings — Satellite cloud motion winds — Earth radiation budget from satellites — Ocean surface temperature -- Precipitation — Land temperature, soil moisture, surface albedo — Runoff from major rivers — CO , O , and aerosols The NASA climate planning activity found it useful to define four categories of climate as follows: • Climate A -- the status of the current climate 228 • Climate B -- climate changes from one month to one decade GOAL: to document and predict changes • Climate C -- climate changes for a decade and longer GOAL: to understand and observe the effects of the sun, volcanoes, clouds, ocean, and ice sheets • Climate X -- changes due to man's activities. This includes the effects of CO2 changes due to man, overgrazing, forest cutting, atmospheric particles, etc. Table 1 shows the Climate B data requirements from the NASA plan. Table 2 is for Climate C and X. Table 3 gives a listing of some of the planned climate data to be prepared during the FGGE year. It is taken from the report of the JOC Princeton meeting in February 1978. To decide on what data sets are needed, we need to analyze the types of data that may be of use in determining the ne-essary parameters. Figure 2 shows an example of the infor- mation flow of a number of data types in oceanography. When a requirement indicates a need for a one-week or one-month average, we must consider the data necessary to make (and perhaps to remake) the average, as well as the particular data composite itself. There are also important data requirements for climate impact assessments, and for climate services. A number of these needs Should be covered by other talks at this workshop. Considerations to Promote Data Use We are behind in our efforts to prepare climate data. So far as is possible, our plans to prepare or improve data should be modular so that the user can obtain early benefit from the projects. In considering the organization of data sets, we should note that data may be so scattered that they are hard to locate. Thus a gathering effort may be needed for a particular type such as rainfall data. On the other hand, data is sometimes merged too much. If many types of data are merged, a user has to sort through all of the data in order to use any one type. The plans should also provide for ease of updating of the data. 229 Table 1 Climate B Requirements (NASA Plan, 1977) Desired Base Horizontal Vertical Temporal Index Parameter Accuracy Requirement Resolution Resolution Resolution No. • Temp Profile l°C 2°C 500 km 200 mb 12-24 Hrs. 1 Weather • Surface Pres. 1 mb 3 mb 500 km - 12-24 Hrs. "> Variables • Wind Velocity 3 m/sec 3 m/sec 500 km 200 mb 12-24 Hrs. 3 (• Basic -«-" • Sea Sfc. Temp. 0.2°C l°C 500 km - 3 Days 4 FGCE • Humidity 77- 307 500 km 400 mb 12-24 Hrs. 5 Meas.) Precipitation 107r 257 500 km - 12-24 Hrs. 6 Clouds 100 km - 1 Day 7* a. cloud cover 57 207 b. cloud top temp. 2°C 4°C c albedo 002 0.04 d. total hq. I0mg/cm 2 50 mg/cm- H 2 Content Sea Sfc. Temp. 02°C l°C 500 km _ 1 Month 4a Ocean Parameters Evaporation Sfc Sens. Heat Flux 107- 257 IOW/m 2 25 W/nr 500 km 500 km 1 Month 1 Month 9 10 Wind Stress 1 0.3 500 km - 1 Month 1 1 Dyne/cm- Dynes/cm 2 Clouds (Effect on Radiation) 500 km 1 Month 7a ■ a. -cloud cover 57 207 Radiation b. cloud top temp 2°C 4°C Budget c. albedo d. total liq. H 2 Content Regional Net Rad. 0.02 0.04 10 mg/cm - 50 mg/cm* Components IOW/m 2 25 W/m 2 500 km - 1 Month 1(3 Eq.-Pole Grad 2 W/nr 4 W/nr 1000 km Zones 1 Month 17 Sfc. Albedo 0.02 004 50 km - 1 Month 18 Sfc. Rad Budget IOW/m 2 25 W/m 2 500 km - 1 Month 19 Solar Constant 1.5 W/m 2 1.5 W/m 2 - - 1 Day 20 Solar UV Flux 107 per 50 A Interval - - 1 Day 21 Precipitation 107 257 500 km _ 1 Month 6a Sfc. Albedo 002 0.04 500 km - 1 Month 18a Land Sfc. Soil Moist. 0.05 gm 4 levels 500 km - 1 Month 22 Hydrology. H,0/ccSoil and Soil Moisture 0.05 gm 4 levels 500 km - 1 Month 23 Vegetation (Root Zone) H,0/ccSoil Vegetation Cover 57 57 500 km - 1 Month 24 Evapotranspiration I07r 257 500 km - 1 Month 25 Plant Water Stress 4 levels/2 levels 500 km - 1 Month 2ti Cryosphere Parameters Sea Ice (7 Open Water) Snow (7 Coverage) 3% 3% 5% 57 50 km 50 km - 3 Days I Week 27 28 Snow (Water Content) ±1 cm ±3«m 50 km - 1 Week 29 •NOTE: Under "Weather Variables" (Index No. 7), histograms of all fo jr cloud parameters will b<. generated f or 100 km x 100 km boxes. 230 r- 0\ td H < W < s ' (N w -p cu c rH (IJ A e rd a; Eh ^ •h p tr 00 km ariable ariable u X) X) F V. E ra ca eg 03 c 5 c« co > «o>> > — a- «— 1 cao c as 8 o •* e ys e — — r^J o o 55-5 c .§ cj ca t*_ 3 8 £ T o a-2 E S. • C r i ~> ** ' V - M O «co 8S gco T 5 i Oc T > TO ua. J J E id 1 1 m m ro CO £ £ £ CO \ CO ' CO •, ~L UJ ^. UJ >-L UJ Z E E J2 E Z e z " eJ eJ -*o^ ^o •*© o o o o o o o , >/-> o o W> O to O 1 (N — m CM — cm — E 2*j E cm .2 *s c - q o s > "». O w— *3 O o ^8 o o o o E a a. 3 O x: s J= c x ><£-<£• 5 m -q n -ci ^- 2 — * ° — ' c _£0 ra C S" C o co co r- O X o L^TT S « 2 ■si I E E ci. E X) c_ D. cn o O — — o o o X u E E X) E SB D. c_ c_ b CU a. D. a. Cl CO »n o o in o O o O o o X o '5 cr o z c u u u ^ E T df o X>"0 S X) "XJ OS X Q. c ■ c o „ ca a JD .2 X5 ca = 1 re 7; o " o a X Evaporation pan X X X evapotranspiration X X X X X Temperature, dewpoint (humidity) X X X Radiation, net X x x Air Quality X X X Visibility X X X , , — appropriate normals. 246 Freguency DATA ELEMENT (VARIABLE) >i .-( u o u 3 a: i VD i—i ■H TO D o >, 4J c c tfi (0 QJ w Precipitation total X X X X X X intensity X X duration X X Temperature, air, max-min and average X X X X X X Snow depth water content X X X X X X aerial coverage X X Wind velocity X X X direction X X X Soil Moisture (derived variable) X X Temperature, soil X X X Cloud Cover X X X X Special Phenomena Radiation, solar X X X Temperature, dewpoint (humidity) X X X Evaporation pan X X evapo transpiration X X X Air Quality X X Growing Degree Days (derived variable) X X Radiation, net X X X Palmer Index (derived variable) X X Visibility X X Crop Moisture Index (derived variable) X X Heating Degree days (derived variable) X X Pressure surface X X X upper air X X ■ 1 ■ X 247 Frequency - >i DATA ELEMENT >i H A 4J to • H 03 53 (VARIABLE) r-\ o •H ■H >i C -id c rd 1 rd 0) c a r-« 13 s w < Precipitation, total X X X X X Temperature, air, max-min and average X X X X X Special Events monsoon activity X hurricane, typhoon X blizzards, other severe weather X • Wind velocity X X X direction X X X Snow depth X x X X water content X X X Temperature, dewpoint (humidity) X X Evaporation pan X X X evapotranspiration X X X Temperature, soil X X X X Air Quality X X Cloud Cover X X X X Radiation, solar X X X Visibility X X X Radiation, net X X X Pressure surface X X upper air X X X X 248 ATTACHMENT I RATING PROCEDURE FOR SETTING PRIORITIES DESCRIPTION OF RATING FACTORS Agencies Requesting : Enter 1 for each variable requested by Agency. The USDA rating will be derived from agency ratings by Committee as follows: 1 or 2 agencies requesting 1 3 to 5 agencies requesting 2 More than 5 agencies requesting 3 Geographic area to be covered : To be assigned by agencies as follows : Sub-U.S. regions 1 U.S. 2 Hemisphere or Worldwide 3 Frequency of Requirement Need : To be assigned by agency as follows: Weekly or more frequently 3 Monthly 2 Semi-annually or annually 1 Importance of Variable or Service to Agency : To be assigned by agency as follows: Most important variable for agency 5 Second most important variable for agency 4 Third most important variable for agency 3 Fourth most important variable for agency 2 All other variables 1 Impact on Food, Fiber or Forestry Production : To be assigned by agency as follows: Minor 1 Intermediate 3 Major 5 Impact on Environment and Natural Resources : To be assigned by agency as follows: Minor 1 Intermediate 3 Major 5 Total of all Ratings : To be obtained by summing all entries made by agency. 249 s 1) U] 0) u m M fl 0) 0) (/i T> g 5 ana Total of all ratings Impact on en- vironment or natural resources Impact on food fiber, or for- estry production Importance of variable or service TJ 0) 0) <4-l C O 4-1 >i c O 0> c e m aj 3 w ct-h 0) 3 i-i a 1 h o> u Geographic area to be covered 01 c 01 w ■H 4J O 0) C 0) 0) 3 en cr < 0) S-l TJ lu M N •H rH J CO u P m in 0) TJ C 3 01 01 '-I .tJ s < 01 0) ■P (0 -i-t e > ■^ M rH 0) o o e M lu -1 < 1 1 i u TJ 3 O o >1 4J -H rH •H -Q •H (0 H > c ■H u « hi a id > 4J <0 3 o ki -H < TJ C ■H 2 0) 3 4J u) 9 E X U TJ B —i i-i 01 B c a. X 01 "O c —4 0) Sri 3 ■u 0) >H E 0, u ID >i ■a 1> 01 a* u "O a> c 4-1 .0 11 X to >. .0 TJ 11 D U Ji 1) TJ tT •-4 3 u 250 ATTACHMENT II DEFINITIONS Definitions of meterological factors are based on the Glossary of Meteorology published by the American Meteorological Society, 1959. Exceptions are recent terms originating in environmental concerns and those terms referring to specialized agricultural applications . • Air quality - As yet there is no generalized definition for air quality. In the absence of a standard definition, it is proposed that terminology used in U.S. Statutes and U.S. regulations be used where practical. The term includes both airborne particulates and gases orginating in nature and through human actions. • Air temperature - Measured according to existing standards (ongoing conversion to Celsius noted) . Includes maximum and minimum measurements for specified periods, as well as means for period of interest. • Anomalies - See "Special Phenomena." • Cloud Cover - That portion of the sky cover which is attributable to clouds, usually measured in tenths of sky covered. • Crop Moisture Index - A product of the Palmer Index calculations, this index is a measure of the amounts of moisture available to support crop growth and development. • Evaporation - The physical process by which a liquid is trans- formed to a gaseous state. Usually direct measurement is by means of standard evaporation pan; however, other types of measurements are not excluded. • Growing Degree Days - A biological transformation relating temperature to crop growth and development. May be based on different critical temperatures appropriate to the crop of concern. • Heating Degree Days - A measure of heating fuel consumption based on the difference in daily mean temperature from 65°F. • Humidity - A measure of the water vapor content of air. Unless otherwise specified, absolute humidity or vapor pressure is desired. • Net Radiation - A measure of the energy remaining in a layer, normally just above the plant canopy when all fluxes of energy through the layer are considered in total. 251 • Palmer Drought Index - An index which measures the water balance of an area and evaluates the overall moisture situation to climatological normals. • Precipitation - Any or all of the forms of water particles whether liquid or solid, that fall from the atmosphere and reach the ground. Measured according to existing standards. Includes rainfall, snow, hail and other forms. Reporting format will vary with the specific requirement, but will include means and cumulative data. • Soil Moisture - Moisture in the rooting zone of the soil. This is typically the first 4 to 6 feet of the soil profile. • Soil Temperature - Temperature of the soil at varying depth depending upon the specific application. • Solar Radiation - The total electromagnetic radiation emitted by the sun which reaches the Earth's surface. Measured in langleys according to standard procedures, specifically the radiation reaching a horizontal surface from the sun (direct solar radiation) . • Special Phenomena - The description or measurement of specific events which are not measured by standard procedures. • Visibility - The greatest distance in a given direction in which it is just possible to see and identify with the unaided eye (a) in the day time, a prominent dark object against the sky at the horizon, and (b) at night, a known preferably unfocused, moderately intense light source. • Wind - Air in motion relative to the surface of the Earth; the terms refer to the horizontal component. 252 INTERAGENCY AGREEMENT BETWEEN THE U.S. Department of Commerce AND THE U.S. Department of Agriculture (USDA) I. General Information WHEREAS, the Department of Commerce, through the National Oceanic and Atmospheric Administration, has responsibility within the Federal Government for the monitoring and reporting of weather conditions, the forecasting of weather, the issuance of storm warnings, and the collection and dissemination of global weather information, and the distribution of meteorological and climatological information of interest to agriculture; and WHEREAS, the Department of Agriculture has responsibility within the Federal Government to acquire, analyze and interpret information for the purpose of assessing the impacts on national and international food supplies, and to provide appropriate information related to the impacts of weather and climate on agricultural production to the people of the United States; NOW, THEREFORE, the Department of Commerce and the Depart- ment of Agriculture enter into this interagency agreement to cooperate in the application of weather and climate information to requirements of agricultural policy planners and producers, and in the issuance of certain specialized weather, crop, and soil publications. II. References and Authorities Commerce enters into this agreement pursuant to the authority vested in it by 15 U.S.C. 313; Agriculture enters into this agreement pursuant to the authority vested in it by 7 U.S.C. 2201. This agreement supersedes all previous agreements between the two Departments relating to coordination and cooperation in climate and weather matters. III . Purpose The purpose of this agreement is to provide a framework for all cooperative efforts to meet the weather and climate information needs of the Department of Agriculture and the agricultural community. As appropriate, subsidiary agreements will provide 253 details on specific collaborative activities. When such agree- ments are developed they will become addenda to this master agree- ment. These subsidiary agreements will cover activities such as: • providing weather data, interpretation and briefings, and climatic data and information, to Federal agriculture decision makers in USDA headquarters; • joint editing and publishing the Weekly Weather and Crop Bulletion ; • cooperating in the execution of the National Agricultural Weather Service Program, in particular the making and disseminating to agricultural producers relevant weather observations and specialized agricultural weather forecasts and advisories; • developing and testing weather/crop-yield models; • providing climatic summaries for County Soil Surveys and other USDA publications; • participating in cooperative regional research projects and other agrometeorological research; IV. Definitions For the purpose of this agreement, analysis refers to products expressed in meteorological or other geophysical units, while assessment refers to evaluation expressed in agricultural or economic units. V. Responsibilities of the Departments Commerce, NOAA is responsible for preparing and making available meteorological and climatic analyses, summaries, fore- casts and outlooks. Agriculture is responsible for economic and operatonal assessments, and for assisting in dissemination and application of NOAA-prepared agricultural weather information. Details concerning these responsibilities, and joint publication activities, will be set out in subsidiary agreements. VI. Programming, Budgeting, Funding and Reimbursement Arrangements Within the terms of this agreement, budgeting, funding and reimbursements will be accomplished by the respective Departments in accordance with arrangement details established in subsidiary agreements. The expenses of the cooperating Departments shall be limited to those necessary to perform the activities stated above. 254 VII. Procedures for Review and Amendment The activities under this agreement and its subsidiary agreements shall be jointly reviewed at least annually by appropriate policy-level officials such as the NOAA Assistant Administrator for Oceanic and Atmospheric Services and the USDA Director of Economics, Policy Analysis and Budget, and such changes as amendments to operating procedures that are required will be developed and implemented at that time. VIII. Publications The results of these cooperative activities may be published jointly or by either of the Departments separately. Manuscripts resulting from these activities should be submitted to the other Department for suggestions and approval prior to publication. IX. Publication Affairs/Press Liaison Data and information shall be released simultaneously by both Departments or with concurrence by the other when a release is by one Department. X. Subsidiary Agreements Additional working agreements between the Departments covering areas in paragraph III shall be effected in writing and become addenda to this agreement. XI. Third Party Liability Each Department will be responsible for processing any claims arising out of the negligent or wrongful acts of omissions of its employees. XII . Amendments and Review This agreement may be amended by writing at any time by the mutual consent of the Departments concerned. This agreement will be reviewed periodically but not less than annually and may be subject to reconsideration at such times as may be required and as agreed to by the parties entering into the agreement. Each Department agrees that prior to major changes in policy, budgets, or procedural practices affecting this or subsidiary agreements, such information will be communicated to the other Department. XIII. Other Provisions Nothing herein is intended to conflict with current Commerce or Agriculture directives. If the terms of this agreement are inconsistent with existing directives of either of the Departments 255 entering into this agreement then those portions of this agree- ment which are determined to be inconsistent shall be invalid; but the remaining terms and conditions of this agreement not affected by an inconsistency shall remain in full force and effect. Should disagreement arise as to the interpretation of the provisions of this agreement, or amendments and/or revisions thereto, that cannot be resolved at the operating level, the area of disagreement should be reduced to writing by such party and presented to the other party for consideration at least ten days prior to forwarding to respective higher authority for appropriate resolution. XIV. Terms of Agreement The terms of this agreement will become effective upon signature of both approving officials of the respective Departments entering into this agreement. The terms of this agreement will remain in effect until termination by mutual agreement or by 30 days advance written notice by either party. APPROVED : Department of Commerce Department of Agriculture /s/ Juanita M. Kreps /s/ Bob Bergland Name Name Secretary of Commerce Secretary of Agriculture Title Title Date Date 256 U.S. DEPARTMENT OF COMMERCE USES AND SOURCES OF CLIMATE DATA AND INFORMATION IN THE DEPARTMENT OF COMMERCE Joshua Z . Holland Director, Center for Environmental Assessment Services Environmental Data and Information Service Office of Oceanic and Atmospheric Services National Oceanic and Atmospheric Administration An outline of an Oral Presentation delivered at the Climate Data Management Workshop follows. 1. Agencies and activities within the Department which are users, producers or managers of climate data and infor- mation. 2. Examples of climate data and information products. 3. Requirements for climate data and information. Agencies and activities within the Department of Commerce which are users, producers, or managers of climate data and information. Code Known Suspected or Potential User U (U) Producer P (P) Manager M (M) U.S. Department of Commerce National Oceanic and Atmospheric Administration U P M Industry and Trade Administration (U) U.S. Travel Service U National Telecommunications and Information Administration U Maritime Administration U 257 U.S. Department of Commerce (Continued) Economic Development Administration U Assistant Secretary for Science & Technology National Bureau of Standards U P Patent and Trademark Office U National Technical Information Service M Chief Economist Bureau of the Census U Bureau of Economic Analysis U U.S. Department of Commerce National Oceanic and Atmospheric Administration National Climate Program Office M National Marine Fisheries Service Office of Coastal Zone Management Office of Research and Development Office of Oceanic & Atmospheric Services u p u u P M u P M NOAA National Marine Fisheries Service Office of Resource Conservation and Management U Office of International Fisheries Affairs (U) Office of Science and Environment U Office of Marine Mammals & Endangered Species U Office of Habitat Protection U Five Regions U Four Fisheries Centers U P NOAA Office of Coastal Zone Management Sanctuary Program Office U Resource Use Assessment and Coordination Office U 258 NOAA Office of Research and Development Office of Marine Pollution Assessment U P Environmental Research Laboratories U P M Office of Sea Grant U P ' Office of Ocean Engineering U P NOAA Office of Research and Development Environmental Research Laboratories Outer Continental Shelf Environmental Assessment Program Office U P Marine Ecosystems Analysis (MESA) U P Weather Modification Program Office U P (M) Research Facilities Center P National Hurricane & Experimental Meteorology Lab U P M Atlantic Oceanographic & Meteorological Laboratories Great Lakes Environmental Research Lab Pacific Marine Environmental Lab Atmospheric Physics & Chemistry Lab Air Resources Lab Geophysical Fluid Dynamics Lab National Severe Storms Lab Space Environment Lab Aeronomy Lab Wave Propagation Lab P (M) u p u P u P u P u P M u p u p M p p p M ( NOAA Office of Oceanic and Atmospheric Services National Weather Service U P National Ocean Survey U P M National Environmental Satellite Service U P Environmental Data & Information Service U P M 259 NOAA Office of Oceanic and Atmospheric Services National Weather Service Office of Meteorology and Oceanography U P Office of Hydrology U P National Meteorological Center U P Systems Development Office U Six Regions U P NOAA Office of Oceanic and Atmospheric Services National Ocean Survey National Geodetic Survey P Office of Oceanography U P M Office of Marine Surveys and Maps U P M Office of Aeronautical Charting and Cartography P Office of Marine Technology P Office of Fleet Operations U P NOAA Office of Oceanic and Atmospheric Services Environmental Data and Information Service National Climatic Center National Oceanographic Data Center National Geophysical and Solar-Terrestrial Data Center Environmental Science Information Center Center for Environmental Assessment Services U P M u P M u M p M M 260 Examples of Climate Data and Information Products Available from NOAA Examples of Climate Data and Information Products from Environmental Data and Information Service, NOAA Basic weather and climate products Local Climatological Data Climatological Data - National Summary Climatological Data Comparative Climatic Data Monthly Climatic Data for the World Daily Northern Hemisphere Data Tabulations Daily Series Synoptic Weather Maps Northern Hemisphere Sea-Level and 500-millibar Charts Storm Data High Altitude Meteorological Data Global Monitoring of the Environment for Selected Atmospheric Constituents State, Regional, and National Monthly and Seasonal Heating Degree Days Weighted by Population Environmental Satellite Imagery Global Daily Max/Min Temperatures and Weekly, Monthly Summary* Global Daily, Weekly, Monthly Precipitation* Global Daily, Weekly, Monthly Snow Depth* SOLRAD Data** Crops and weather, precipitation, and drought Weekly Weather and Crop Bulletin Major Abnormal Conditions Affecting World Agriculture Palmer Drought Index Hourly Precipitation Data Marine weather Mariners' Weather Log Marine Climatological Summaries Miscellaneous products Solar Terrestrial Physics and Meteorology - A Working Document NOAA Scientific and Technical Publications Announcement Environmental/Resource Assessment and Information EDIS Magazine *"NOAA Products and Services of the National Weather Service, National Environmental Satellite Service, Environmental Data and Information Service, and the Environmental Research Laboratories," NOAA, Rockville, Maryland, 1977. **Not in Catalog. 261 Examples of Climate Data and Information Products from National Weather Service, NOAA Observed data without analysis 24-hr precip amounts Winds aloft chart Observed snow cover chart Max and min temperature chart Surface (sea level) data with analysis North American surface analysis Northern Hemisphere surface analysis Tropical surface analysis 12-hr mean sea-level pressure change analysis Radar analysis (MKC) Weather depiction analysis Degree-day charts Upper air analysis NWP upper air analysis MIA upper air analysis Stability index - moisture chart Public Weather Service and Emergency Warnings World Weather Watch participation Temperature and precip tables Road condition report Winter sport area report Daily weather maps Severe weather statement Storm report Tropical cyclone position estimate Surface long-range forecasts Average Monthly Weather Outlook Average Seasonal Temperature Outlook Aviation Weather In-flight reports from pilots Aviation weather publications Weather support for transportation safety hearings, litigations, investigations 262 Examples of Climate Data and Information Products from National Weather Service, NOAA (Continued) Marine Weather Great Lakes Weather Observations Great Lakes Weather Synopsis Great Lakes Ice Program Agricultural Weather Service NWS/USDA Agricultural Weather Observations 30-Day Agricultural Weather Outlook 90-Day Outlook Environmental Study Service Centers Fire Weather Extended Fire Weather Outlook Annual Summary of the Fire Season Air Pollution Air Stagnation Narratives Air Stagnation Data Hydrology Water Supply Forecast for Western United States Headwater statement Flood Potential Outlook Water Supply Outlook Greatest Known Areal Storm Rainfall Depths for Contiguous U.S 5 to 60 Minutes Precipitation Frequency for Eastern and Central U.S. Precipitation-Frequency Atlas of the Western U.S. Meteorological Criteria for Extreme Floods, Probable Maximum Precipitation, Extreme rainfall and flood case studies for various watersheds Storm Tide Frequency Analysis for various coastal areas Precipitable Water Over the U.S. Some Climatological Characteristics of Hurricanes and Tropical Storms, Gulf and East Coasts Hydrometeorological data tapes for modelling Hydrologic modesl Snow Research Data 263 Examples of Climate Data and Information Products from National Environmental Satellite Service, NOAA Image products Geostationary satellites Full-disk and sector displays Direct readout and GOES-Tap Movie loops U.S. Cloud Cover Depiction Polar-orbiting satellites Stretched, gridded pass-by-pass SR images SR hemispheric polar-stereographic mosaics SR polar-stereographic quadrant mosaics SR Mercator mosaics VHRR basic images 10-day minimum composite brightness Meteorological products Satellite winds: low- , middle-, and high-level cloud motion vector field messages Atmospheric soundings (VTPR) Weather summary and bulletins Satellite interpretation message Satellite weather bulletin Tropical distrubance summary Two-layer moisture analysis Satellite Input to Numerical Analysis and Prediction Hydrological products Basin snow cover observations Northern Hemisphere Snow and Ice Chart Oceanographic products Sea Surface Temperature products Global operational SST observations Great Lakes surface temperatue analysis Ice charts and ocean current analyses Great Lakes and Alaskan ice charts Gulf Stream Wall bulletin West Coast thermal front analysis 264 Examples of Climate Data and Information Products from Environmental Research Laboratories, NOAA Atmospheric Data Global Atmospheric Carbon Dioxide Data Ozone Data for the World Geophysical Monitoring for Climatic Change Annual Summary Report Data Sets on Selected Hurricanes and Tropical Distrubances Obtained by Specially Instrumented Research Aircraft Data Sets for Severe Local Storm Weather Situations Solar Data Preliminary Report and Forecast of Solar Geophysical Data Preliminary Report of Solar and Geophysical Activity (Joint with USAF GWC) Solar-Geophysical Indices Solar-Geophysical Data Scientific and Technical Publications of the Environmental Research Laboratories 265 >- z cr < CO co < Q •— > -J a z 2: < • • < LU 1- OS CO X Z LU < LU O LU LU Q. H u 2: CO >- < »— « oS X 3 K- Q > CO cr a «J O Q- co z =J •— *— » «— i z < < cr LU O LU CO >- J _j CO LU X £T X z "V z cr Q. Q. Q_ _l LU ►— • CO z < 2: Z =5 O ^ < «\ CO O O 2 > h- z Q •— I CO M LU h- O < z X LU H en - - O < rr LU LU Z PQ -XI O Q£ D_ CD LU U •— « ££L LU O 21 Z SI < '// x O c_; < LL < Q \- LU z h- LU < S 21 h- •— • QS _J < <_> Q. LU LU O < CO x •-« «— « LU Q S a a. H- •— Q h- •• Q J H- < Z CO 2: CO cr LU <— > O < Z O CO z ~ co LU < _J Q£ > SI •— 1 •— LU •-• LU Q O M CX CO LL «— h- CO O z CJ CO O _S LL, < to CO z —~ LL. < < z LU O O O < H- < > h- LL •— • z z —* O cr r> LU LL Z LL Q < CO LU H- a 00 cr •— * CO O •— • z a. < CO Z D z LU < !~ z »— i Q < 2: LU LJ z < CO ^ O Z 1- O —• cr f- SI < X> < »— « »M) O < 21 < z X h- Q h- »-^ z z LU CO CO < z *- LU < »- «— 1 LU »— • — z S LU ►— •— LJ z cr X < LL > Q O 21 CJ 21 < < LU LL 266 Notes on the "Requirements" Diagram Each of the six "data types" contains many sub-types. The pre-workshop survey yielded descriptions of more than 800 different data sets. Each of the five "uses" represents many "users" working on different time-space scales, regions, applications. Interconnections of specific data types with specific data users are more numerous than shown, by several orders of magnitude. The National Climatic Center, holding only a fraction of the data sets, responds to 70,000 data requests and distribures 1 1/2 million publications per year. The pathways from producer to user, including processing steps and storage stops along the way, are the subject matter of "climate data management." Summary 1. Qualitatively it is clear that the Department needs essentially all types of climate data and information and, in most cases, could benefit tangibly from improvements in accuracy and space-time coverage. 2. Quantitative requirements, in terms of time-space coverage and accuracy for each parameter, have been defined for some data types and uses, but remain to be defined for many others. 3. An efficient management system will be required to get the many types of climate data and information from producers to users in appropriate form and in a timely and economical manner. 4. The Department has, within the Environmental Data and Information Service of NOAA, some of the elements of such a system in place and others under development. We welcome the joint efforts with other agencies under the leadership of the National Climate Program Office to plan and implement a national program to meet present and future needs. 267 UNITED STATES AIR FORCE ENVIRONMENTAL TECHNICAL APPLICATIONS CENTER Elbert W. Friday, Director Environmental and Life Sciences Department of Defense USAFETAC is a named Air Force organization assigned to Headquarters Air Force Global Weather Central (AFCWC) , a unit of the Air Weather Service under the Military Airlift Command. The Center's mission is first to provide, within allocated resources, assess- ments and advice through tailored studies on environmental problems to the Air Force, Army, and other agencies as directed. Secondly, USAFETAC is to provide the capability to receive, store, retrieve, and process environmental data necessary to support customer requirements. This unit is comprised of 142 authorized personnel (120 Military) at Scott AFB and 85 personnel (all civilians) at Asheville, North Carolina. The majority of USAFETAC ' s project work is descriptive. The Center is organized into six branches at Scott AFB plus an operating location at Asheville, North Carolina. The operations and adminis- trative branches perform traditional support duties of these functions. The Aerospace Sciences Branch has the responsibility of evaluating, developing and adapting models, techniques, and data bases necessary to solve customer requests. Some of this branch's work includes climate prediction investigation, review of electro-magnetic propagation problems caused by atmospheric variations, investigation of probabilistic and statistical techniques and models to provide decision assistance, weather simulation efforts using such recently developed techniques as a dynamic cloud-free-line-of -sight model, and problems relating to future data requirements plus the testing of the present data sets. The Global Environmental Applications Branch has the primary responsibility for a major portion of the Center's customer con- sultations. In this branch, the models and techniques developed by the Aerospace Sciences Branch, as well as other sources, are adapted to answer questions on such areas as, but not limited to, atmospheric ref ractivity , upper air winds application, air quality assessments of air pollution, mixing heights and stability classes studies, engineering design cirteria for the tri-services, operational decision assistance information, weather analysis to assist estimates of crop production (in the USSR and Peoples' Republic of China) and tailored support to on-going special programs. The Center's Climatological Services Branch provides data and technical information retrieval services, as well as editorial and cartographic support for both internal requirements and authorized outside customers. They maintain the Air Weather Service Technical Library and have a secure drop on the DDC Automated Information Retrieval System. The Automation Branch performs systems and applications programming as well as processing and monitoring of 269 certain portions of the Center's data base. The Scott AFB locations ' s automation resources include an in-house IBM 360/44 and a PDP 11/45, plus the use of several remote computers on the Defense Advanced Research Project Agency Network (DARPANET) . The Center's operating location (OL-A) at Asheville is co-located with the National Climatic Center (NCC) . OL-A provides the handling of large data sets including the preparation and processing of standardized climatic summaries using a UNIVAC 1100/10. The unit is responsible for managing and maintaining the data base, including all data restoration. OL-A receives 80,000 daily observations from over 9,000 locations worldwide after processing thru AFGWC and the Scott AFB location. In addition to surface data it is also responsible for processing and monitoring several large gridded data sets of upper air variables. USAFETAC ' s workload is driven entirely by requests from authorized customers (see Attachment 2) . After a request is in-processed by the Operations Branch, an initial cost estimate is made and forwarded to the Center's Project Management Group (PMG) . This group is composed of the Command Section and all the branch chiefs (except Administration) and it reviews new requests to bring corporate memory into effective use and to ensure that the Center takes an appropriate, efficient course of action on each and every request. The PMG assigns resources to a project and reviews active projects to pinpoint and correct problem areas. After a request has been reviewed by the PMG, it is assigned to a project manager. The project manager is the single point of contact for this project although it may require inputs from several areas of the organization. The analyst may use several techniques to satisfy the request such as consultation with the customer, library research for applicable information, a model or technique, programming, and/or computer processing. A project may require a report before final transmission. If so, the Center's editorial and/or cartographic sections will assist the project manager. The project and/or report is reviewed at the section and branch level for technical content prior to shipment to the customer. After the customer has received a USAFETAC project, he has the opportunity to provide the Center with his evaluation, including cost savings of the product. USAFETAC ' s Climatic Data Management role is to receive, process, and store worldwide data from multiple sources. This global data base is required in order to produce the wide variety of customer requested assistance. The Center expends approximately 30 percent of its direct support resources in the production and retention of the data base. The primary data sets that presently reside at USAFETAC are: a. Data Save (DATSAV) files which contain worldwide standard surface and upper air station observations. These data are collected by the Automated Weather Network and comprise observations 270 from all WMO members including military participants. Then the data are sent via high speed circuits to AFGWC . It is processed at AFGWC for real time applications and retained until the completion of a 24 hour cycle. The following day these observations are sent to USAFETAC via the DARPANET. the DATSAV data is then checked for potential errors and flagged as necessary. The data sets are then copied and mailed to OL-A for necessary quality control, corrections and storage. b. The 3-Dimensional Nephanalysis (3DNEPH) data set which contains worldwide cloud information. This is a unique integration of surface observations, aircraft reports, upper air observations and satellite data into a single 15 level, worldwide gridded cloud analysis on a 25 nautical mile horizontal resolution. The cloud analysis includes information on cloud bases and tops, cloud coverage (total and by level) , cloud types, and present weather for each point six times a day. The data set is produced by AFGWC and mailed to OL-A for processing and storage. c. The Analysis Data Set (ADS) which is a re-formatted collection of gridded global data derived from AFGWC ' s course mesh (200NM resolution) 15 level constant pressure upper air analysis. This 00Z and 12Z data set is transmitted every other day from AFGWC. It consists of pressures, D values, temperatures, dew- point depressions, and u and v wind components of the upper atmosphere. d. Summarized Analysis Data Set (SADS) . SADS is a statistical summarized version of the ADS minus the pressure parameter. The data are stored by pressure level, but a vertical stack by grid is also available. In addition to these, the Center possesses many other data sets which are not used as extensively as the primary ones. These data sets include: a. Wind Tower data from the instrumented towers at Vandenburg AFB, Patrick AFB, and Edwards AFB. In addition to winds, this data set includes some temperature and dewpoint information from 6 to 300 feet above ground level. It is customer quality controlled and used primarily for tailored studies supporting programs at these locations. b. The Snow Analysis Data, provides a single, daily analysis of snow age, snow depth, and snow depth climatology. This data set is the output of AFGWC ' s Snow Cover Analysis Model. c. Period of record extracts for individual stations, both surface and upper air, are produced for applications work when required by the analyst for time periods of ten years or more. 271 d. The Terrain/Geography Data Set is derived from AFGWC ' s High Resolution Topography and Terrain Data Base. It is a world- wide gridded analysis of terrain height and geography set on the 3DNEPH grid system. Presently, USAFETAC is initiating an aggressive data quality assurance program which provides both a subjective and a quantitative data analysis. Initiatives taken include a Problem Identification Program (PIP) for surface observations which flag possible data errors for future restoration by OL-A. The primary aspect of the PIP is utilization of the Station Data Information File (SDIF) . The SDIF is a monthly, yearly, and a combined yearly set of climatological statistics which are used for incoming data comparison data. OL-A is the USAFETAC single manager for USAFETAC ' s data base. Also planned is the use of an upper air PIP for additional data quality assurance. USAFETAC * s future data base needs are only roughly known at this time. Increasing requirements for upper air studies and summaries will provide the opportunity to produce better and expanded atmospheric data sets. The increasing complexity of Air Force communications and weapons systems, especially in the electro-optical, infrared, and visual guided munitions, will increase the Center's need for improved tailored data sets. Finally, more and more emphasis today is being placed on the ability to simulate certain meteorological conditions, through actual or statistically generated data. Such simulation provides a technique for examining atmospheric processes and their effects upon modern weapons system. Data are currently being provided for simulator scenarios and some numerical simulation of atmospheric parameters is being done also, e.g., dynamic cloud-free-line-of-sight and wind factors. Expanded investigation of simulation techniques conditioned by real or synthetic data is envisioned for future customer support. In summary, USAFETAC is involved in a wide variety of applied clima- tology to support DoD programs. It is engaged in techniques development and adaptation to ensure the latest products of the research community are available to the national security mission of DoD. 272 Attachment 1 r o »— < 2T < 2: <£ o o a: >— CO g o u CD o s r> o y to to 2 o w a. o to s s g *3 a < o »— 2: uj <: to o O CO o o C_> o »-« 2: o 0: c2 )— UJ CO <£ tO O M EC a. Q W CO W CD o w u 2 O CD u D CD o en a to 01 UJ CD a. o L f- ►3 u 2 H 2 2 H 2 2 u 2 2 UJ CO O to U $ M y C_> UJ i-i to £ H < w H < 0- b < 2 H a. 2: H O CQ O c O §2 CO UJ 2 •■h •J H < •-• O < w U a M D a: Ul to M •J 2 H > •J a £s «t O P. Q Q w Cw W CO 2: L) 2 6 w 2 CO 2 £= > D a a a »— « t_> h- 2C ■a: a> UJ _j _1 _J l-x 2: >- c^ UJ »— as < to 0: < UJ o_ —J CO < w < 2: H O >* .3 2 K— O 2 2 O < O _J 2: •-« § »H O 2 1-1 < uj i— rc c J O H s: < 10 H 2 a O •H < 2 2: 2: 2: OS O fcj en E- 1 U _j «-« < CJ O ft 2 S M M oa jq: t> Cu 1 M h-l w t-3 U •— • D_ CD to a. £> O H u Cm U > Q- > D M 2 2 g a Cm Dm 2:< to to CD W W < to UJ 5 < 2 | 2 2 2 w (4 W W « -*f 273 Attachment 2 Ld O > o u Lu _] OC H U o 2E c u H ■^ (/) §c D O "0 V o £ Z o < D < > a to ~ t! T3 cn U O Jh t 1 i- c o +-> ro 4-> , — fO i- _CT 3 o 4-> i- i — rO Z5 CD O CO •r— QJ s- CC en S_ C O) QJ O I -C -r- CL +-> +-> E fD ro O (U-P U2W CO GO QJ 3 S- -O 2T -r- S_ u_ o CO 0) U M P O CO 03 4J Q U ■H nj a ■H H U 0J > •r- QJ CO -t-> -M C ro ro O S- E •<— QJ •«- 4-> Q. i — ro OU-P O CO CJ> , ro $- co 13 c: -l-> o i — •i — rs +-> o ro •r— 4-> &_ CO en >, O ro zc i — +-> CO 288 +-> H- c O CD OO E +-> +-> c o C CD CD ro CD CL rO co oo qj +-> « =3 oo ■a +-> CU c •i — • 1— (O 4-> +-> OO •1— OO C 00 a; i— i s_ CD CU CU -C > i — o ■r— i — c c: o m =D CJ CU 00 CD C£ 4- O r T3 ro c qj C oo o N o j*i •r— •1— -,— s_ OO 1 — 4-> o •1— ■f— > 4-> S_ u s_ aj QJ CD cu -z. CL ex OO =3 CO JD CO O £Z o •r- 00 +-> CU rO O E T- s- > o s- M- CU C CO OO o 4-> • 1— OO 1 — CU _Q rs Z3 CT Q- CU CC oo CU •I — o 00 c E CU rO CD i- Qu rO +J co c o CU o; ■r— cj +-> 00 • I— rO c > +J QJ $~ CO 4-> cu X CO UJ ■a s- c QJ rO o < .d •r— a; "O CD ro ■I— CC ZS. z< 00 E C ro O i- ■r- CD ■+-> o IOC l- C-PD- "1 *l -a 2 <— J- ro O &- O CU CD "O CU Lu , — (O CU g C +J n3 o ro S- • 1— E cn 4-> ■ i— O rO i — - S- ^ CD Q_ , — fO S- CU C QJ o O _C • 1 — •r- +-> > +-> rO S- <0 QJ cu 2: 3 CO OJ • ,C ft rd +■ U B M n 0) «. iM T £ c rd H 3 M QJ tn T3 £ rd M £ in t7> Cm rH rH rd u .(■! C ft -p c ■H QJ 0) -P 4-> e rd n3 c N E n •H •H Q) C rH > rd U CT rji M 0) M O +■ CU rd 4-> C -M £ — ■ o CD CL. O - > co c-> UJ CD >- ,00 Q- a. 00 00 C5 o CL. 00 CD ^ CD cd o CL o CO UJ C_> C£ ZD O CO cd CO UJ CXl > UJ CD CD C_> CXl s o >- c_> o z UJ UJ CXl ZD :=> C2» CO UJ - CD <*: Cl. X CD c_> o CO Z OC O UJ — • O P— 9— •>- CU uj 3 k- a: n c_> «♦ o CO Z CD UJ o CD CXI or CO Cc CO ZI> CO 2= CD O Cc CD ZD CO CD CD ZD cxz z CD CD C_> CQ CD a: ZD CD 294 o UJ >- en => <: i/> o ►—• h- _j z O- UJ O. 3E O »— <: a. § UJ Bfc. ►—< CT O _i o e> U_ «C *- UJ i O <: 3: UJ o ►- ! 2 «C o 1 g *o l~ «* O z o o 3 o 4j o D. E •4-> u o © u V - > © 4-> « 5 •t— JZ -»-> en or W «o »> a. • « *o O *J E u oc o °»— UJ C 3 «« UJ r— ♦— O T- »o ►- » T- $ ■»- -o o UJ >> JD S8 »— « >» 3E u »^- »—t UJ 1- %. rt c *> _j 21 1-^ *o or © •n * o r> © •o ^^ r— I < 3 © _J O. W> UJ or © a. cr o UJ o o <-> > UJ C o © o 31 > u 1— » ►- C o> o t~ - «_> « en »— • *~ ^v» o c -o e> s JC u_ >> o ^ ^- o ku u ♦- »— « m ♦J H- «c M S_ O < c •» ►- M C «»- M CX CD 3C >> I « z s. o r — >- •o © O e> u 4J UJ T— ^~ ►- ►- W "O cz © «3 •— t •*> •♦-> >>x »!— 1 o o >- *_> *- • %. %. o i 3 •»-> «o -J" => s- ^— r— »~« o 0.-0 E UJ u_ UJ cr «rt >«o z ►- r> ft. © « UJ «9 © 3 > «_> o. « +-> o UJ z oc U- 2 > CLQ »~t *£ *S% > or UJ >- •-* -J -c UJ »— 1 •o • 1 • or ■ • • 1 Sk. ►- • * 1 U- «-> z UJ tx «c UJ «o »-» o < -43 or 295 9 o I— to c_> or «/» "D C ra _J v_ O > c O > o ?3 O q 5 z _ o c o a UJ i Pilot Study Successful Pilot Study Successful ■ ■ ■c o CI c o o v» < rs > in Z o o uu O :< co O o o Z CO >- z CO O o a: < O CO 5 < (r < a. UJ O Q 00 O < CC UJ z o rsi O CO O o 2 D O CO I CO _J o CO O tr UJ < O oc UJ X < UJ Ol a D CO O z UJ < u. cc CO < UJ o > CO a z i o a UJ 5 < UJ CO CO cr UJ > O o o O _/ _l < z < cc u z < UJ L> O = 3 O UJ ec ►- cc O UJ UJ Z ■ CC O u. u. UJ < Z O z t X O o cc a. < < O z ■ o 296 Parameter vs. Source Cross-Reference for Candidate NASA Data Seta Xam*u«4.S.a A£M 1 SEASAT A KaoauaC ACM • SMV GOtS > at 5 at a 1 X at 1 | < at 2 at -4 i M > an ut at X 2 ri 3 at as > 5 > r*~~tu** M j Sea Surface Teeape ratare 44) i V < Sea ke CoacwiMiea (27# V j V V, SnewC»«r"rSj v 1 1 I V ■ Solar Cmumi { 2©t V" i sL fcadtauoa Imapet fwiMtrn f la-IPt V" 1 ^_ OtaaclISt >/ 1 1 >/ V V T,HSo»»A»»lU5) I >f 01 Suaaakpaenc Aeroaofe ( JJ6 j V V «««* *<») V V 1 1 V v> V >l V Oom4 To* Temperature i7i V V Cloud Albedo <7i ^L V Taut LtouaJ »jwC«iiwH7| >f i V V Soil Htm*** «;: 4 ;ii n/ V V EtapoKampiratioa t- is V Ptaat fester St*mi_b) V Snow ftn»fCo«t«flH>l V V Sea Surface Ltevatton 1 1 !» V ice Sheet Surface Elevation « *0) v War UV Flux 8 2 1) yL ^ Stx»iaaahtrkH,0<16» J V NO, tJ7» V ro"v,o«;»j 7 j t'H 4 «i0» i V VI Thermal iKrtu i J»/ *a*« begat. dwectwn. tenet* t I i ! V Stratoapacnc icutperatuM i 1 1 7" V >4e*o*phenc lempcratwpc ! i , | 5 V Suatoaphenc KM), * t * ... V 1 Straio»pacnc A. Uetcpherv CO i " *> ' Tool Aaae«phetic Water Vapor | v v ( j V L and Surface Tea rperataat . 3 y 0f«« Water «" » Over Land * v « Sea ice Surface Temperature . j V" i Saaur Surface Temperature ' 1 VI j Snow Sub- Surface Temperature j j V - Ice Sheet Sunset Temperature : y s MulitA ear ice F racttoa s V I Thin Firm War kc Fraction i V ' 1 I t ■ nt tv v vi ■ burial priority I parameter identified during the Climate Soeacc Advnery Panel Mrrtmg • laitial Priority II parameter identified dunag rae Onmalr Sorree Aiiwury P**rt Meeuag • Initial ptionry Ut parameter identifWc' dunag rttc Claaaai Scwncr AAvwory Paad Meetasg • iintial aoditwaj) parameter identr'Wd aunag tne Cam a t e Science AeHaory Paad Meetiaf ■ lAenlified M Pru/Hsted \ ASA Ct*H:ninn*om to tUr Ci im m tt »uf nam. not iaent a fted at »m Climate Science AJmb«> Panel Meeting • Not htemifiea as a c§ mate parameatr in rV> » o W >ASA C»ttrA^»tm ay aW ( ane r rVafwaa). auf AerKaAAr Irata 297 £ 2 8 *— OH *— UJ S co CO 00 CO CO CO £ o f^ __ €K _ <_ __ _ o ^ o «> o 00 o 00 o 00 o 8 UJ cc *~ i,n 00 »■» — ^— #■— ^— <— »= 32 t © i 00 1 ao 00 oo 1 oo 00 tt ^ it c* # >j-> I $ S S ac '~5* -D 5^'^* s > E JE mi m 8 Z 2 OC 2 3 2 X « *- M « */■> *»■* at 3 ■s» oo UJ f • **. a. > *^ ^m • m v> X O I to f 1 UJ tf j a UJ */> c • f • f ■ f • f 1 e t t • s£S I s UJ UJ s s 2 o UJ 298 >- CO o 1 1 LxJ 9 t— « lil rv «£ a. Q_ h- - ►—• e> ¥- felJ 1*— o i , t 2E z: cc: OO a. o LlT < Q- H- o f— tu »— C3 -J 1— _J «x. «i_ O oc - LxJ > en -=> CO 00 oi — * >» .O v» O O O ■* CO 0* c/i O O fN» r* CO ro Tf TJ- rr r- O CD es <<* ^r -r c% c* G\ "<* rx Q 1 °. O z 1 «• ) i *■» •— • 73 CO O V3 NO vO I £ ^ OO CV c\ ro CO en •"■" ; tT ^ * 1 1 * 1 i' O -^> 1 2 ; *- -> i? c ~^ O O «!i- CO OO NO SO O* OO v> 5 1 z " 1 O O O O O. «»— C cr G en- C3 c C c. O "3 "3 .2 ^— N 3 's B 3£ *~* •A O 5 on 4*» 3> e3 a. C3 Q 2; § M R £ ^-N E ^ S 2. *# O O ^ ® K. O «-> <3 CO 0© 1— ■O a £ -a „ ^ c: ■u. S3 3 s 3 E '3 "S co °- CO «=• *^ S3 C3 v. 5 v.. >» 3 C3 #* C3 r3 — C3 a. — Z S Z E Q S c ' 1 d — £ , - ^ O CO >■ O 5 * < -J • < to » IS < < «o CO < O -J < * CJ ^ . c CU S S s ?S r» rs f- H 299 .">: (J K u E- < CO a, >■< ftj CO W H < 2 En i— i < a UJ u ui M En >