NOAA Technical Report EDS 22 <** TO ' c Q Sr ATES O* * U.S. National Processing Center for GATE: B-Scale Surface Meteorological and Radiation System, Including Instrumentation, Processing, and Archived Data Washington, D. C. April 1977 U.S. DEPARTMENT OF COMMERCE National Oceanic and Atmospheric Administration Environmental Data Service NOAA TECHNICAL REPORTS Environmental Data Service Series The Environmental Data Service (EDS) archives and disseminates a broad spectrum of environmental data gathered by the various components of NOAA and by the various cooperating agencies and activities throughout the world. The EDS is a "bank" of worldwide environmental data upon which the researcher may draw to study and analyze environmental phenomena and their impact upon commerce, agriculture, industry, aviation, and other activities of man. The EDS also conducts studies to put environmental phenomena and relations into proper historical and statistical perspective and to provide a basis for assessing changes in the natural environment brought about by man's activities. The EDS series of NOAA Technical Reports is a continuation of the former series, the Environmental Science Services Administration (ESSA) Technical Report, EDS. Reports in the series are available from the National Technical Information Service, U.S. Department of Commerce, Sills Bldg., 5285 Port Royal Road, Springfield, Va. 22151. Price: $3.00 paper copy; $1.45 microfiche. When available, order by accession number shown in parentheses. ESSA Technical Reports EDS 1 Upper Wind Statistics of the Northern Western Hemisphere. Harold L. Crutcher and Don K. Halli- gan, April 1967. (PB-174-921) EDS 2 Direct and Inverse Tables of the Gamma Distribution. H. C. S. Thorn, April 1968. (PB-178-320) EDS 3 Standard Deviation of Monthly Average Temperature. H. C. S. Thorn, April 1968. (PB-178-309) EDS 4 Prediction of Movement and Intensity of Tropical Storms Over the Indian Seas During the' October to December Season. P. Jagannathan and H. L. Crutcher, May 1968. (PB-178-497) EDS 5 An Application of the Gamma Distribution Function to Indian Rainfall. D. A. Mooley and H. L. Crutcher, August 1968. (PB-180-056) EDS 6 Quantiles of Monthly Precipitation for Selected Stations in the Contiguous United States. H. C. S. Thorn and Ida B. Vestal, August 1968. (PB- 180-057) EDS 7 A Comparison of Radiosonde Temperatures at the 100- , 80-, 50-, and 30-mb Levels. Harold L. Crutcher and Frank T. Quinlan, August 1968. (PB- 180-058) EDS 8 Characteristics and Probabilities of Precipitation in China. Augustine Y. M. Yao, September 1969. (PB-188-420) EDS 9 Markov Chain Models for Probabilities of Hot and Cool Days Sequences and Hot Spells in Nevada. Clarence M. Sakamoto, March 1970. (PB-193-221) NOAA Technical Reports EDS 10 BOMEX Temporary Archive Description of Available Data. Terry de la Moriniere, January 1972. (COM-72-50289) EDS 11 A Note on a Gamma Distribution Computer Program and Graph Paper. Harold L. Crutcher, Gerald L. Barger, and Grady F. McKay, April 1973. (COM-73-11401) EDS 12 BOMEX Permanent Archive: Description of Data. Center for Experiment Design and Data Analysis, May 1975. EDS 13 Precipitation Analysis for BOMEX Period III. M. D. Hudlow and W. D. Scherer, September 1975. (PB-246-870) EDS 14 IFYGL Rawinsonde System: Description of Archived Data. Sandra M. Hoexter, May 1976. (PB-258-0S7) EDS 15 IFYGL Physical Data Collection System: Description of Archived Data. Jack Foreman, September 1976. (PB-261-829) (Continued on inside back cover) NOAA Technical Report EDS 22 ^gSBl& ^e^Tof^ U.S. National Processing Center for GATE: B-Scale Surface Meteorological and Radiation System, Including Instrumentation, Processing, and Archived Data Center for Experiment Design and Data Analysis Ward R. Seguin Paul Sabol Raymond Crayton Richard S. Cram Kenneth L. Echternacht Monte Poindexter Washington, D. C. April 1977 a o o ;a 0) P U.S. DEPARTMENT OF COMMERCE Juanita M. Kreps, Secretary National Oceanic and Atmospheric Administration Robert M. White, Administrator Environmental Data Service Thomas S. Austin, Director CONTENTS Page 1. Introduction \ 1.1 Shipboard data system 6 1.2 The bow sensor acquisition module and sensors 8 1.3 The central sensor acquisition module and sensors 8 1.4 The boom platform and hardware 8 2. Sensor characteristics 11 2.1 Temperature 11 2.2 Thermistor signal conditioner 12 2.3 Wind speed 12 2.4 Wind direction 12 2.5 Atmospheric pressure 13 2.6 Radiation 13 2.7 Ship speed and heading 13 2.8 Precipitation 13 3. Pre- and post-GATE calibration 14 3.1 Temperature sensors 14 3.2 Wind sensors . 16 3.3 Pressure sensors 19 3.4 Ship heading and speed sensors 19 4. Procedures for assuring quality control at sea 20 5. Data processing 23 5.1 Preprocessing 23 5.2 Central processing 23 5.3 Data review and validation 27 5.4 Peculiarities in the meteorological data 28 5.4.1 Temperature 28 5.4.2 Wind direction and speed 28 5.4.3 Ship speed and heading 30 5.4.4 Pressure 32 5.5 Final processing 33 1x1 CONTENTS (continued) Page Analysis and validation of surface radiation data 37 6.1 Recording system error limitations 37 6.2 Incident and reflected solar radiation (K4- and Kf) 38 6.2.1 Instrumentation and transfer equations 38 6.2.2 Data validation 38 6.3 Net radiation (Q*) 47 6.3.1 Derivation of transfer equations 47 6.3.2 Data validation 49 6.3.3 Instrument malfunctions 55 6.3.4 Anomalous net radiation data 55 6.3.5 Radiometer sensitivity and transfer coefficients . . 55 6.4 Yanishevsky pyranometer data (K4-) 57 6.5 Vanguard pyranometer data 58 Format and inventory of the archived data sets 59 7.1 Digital data on magnetic tape 60 7.2 Microfilm time-series plots 65 References 68 Appendix A. Detailed tape formats and examples 70 Appendix B. Pre-GATE intercomparison of pyranometers 82 Appendix C. Translocation of thermistor probes and bridges ... 86 Appendix D. Sensor and SAM transfer equations used in data processing 88 IV U.S. NATIONAL PROCESSING CENTER FOR GATE: B-SCALE SURFACE METEOROLOGICAL AND RADIATION SYSTEM, INCLUDING INSTRUMENTATION, PROCESSING, AND ARCHIVED DATA Ward R. Seguin, Paul Sabol, and Raymond Cray ton Center for Experiment Design and Data Analysis Environmental Data Service, NOAA Washington, D.C. Richard S. Cram Air Resources Laboratories Environmental Research Laboratories, NOAA Boulder, Colorado Kenneth L. Echternacht Institute for Acoustical Research Palisades Geophysical Institute Miami, Florida Monte Poindexter Air Resources Laboratories Environmental Research Laboratories, NOAA Silver Spring, Maryland ABSTRACT . This report documents the surface meteorological and radiation data collected on the four U.S. B-Scale ships Researcher , Gilliss , Dallas , and Oceanographer . The report describes the sensors and recording systems on board the ships, pre-and post-calibration procedures, and data processing and validation methods. A description and inventory of the archived products are included. 1 . INTRODUCTION The Global Atmospheric Research Program (GARP) Atlantic Tropical Experi- ment (GATE) was conducted during the summer of 1974 in an area centered 1000 km southwest of Dakar, Senegal. The central program of the experiment focused on the effects of smaller scale tropical weather systems on larger scale circulations. Some 70 nations supplied the needed equipment and manpower, including 39 research ships, 13 aircraft, 10 satellites, and nearly 4000 scientists and technicians, to meet the goals of GATE. The United States supplied five ships for the B-scale array: the NOAA Researcher ; the University of Miami R/V Gilliss ; the USCG Dallas ; the NOAA Oceanographer , and the NASA Vanguard . The acquisition of continuous, high-quality surface meteorological data was an important facet of GATE, yet making precise measurements of surface meteorological variables at sea aboard large research vessels is difficult at best. In his textbook, Physics of the Marine Atmosphere , Roll (1965) discusses many of the problems and limitations. For example, the obstacle effect of the ship distorts the wind flow, making accurate wind velocity measurements difficult. This same effect can often influence rainfall measurements so that twice as much rainfall is measured at one place on the ship as another. The ship heating affects air temperature measurements, and the engine cooling water modifies the sea surface temperatures. To minimize these problems, four of the GATE B-scale ships (the Researcher , the Gilliss , the Dallas , and the Oceanographe r) were equipped with instrumentation mounted on bow booms and the forward mast. Experience and previous research (Seguin and Garstang, 1971; Ching, 1975) have shown that instrumentation mounted on bow booms suffers the least from the above ship effects. The Vanguard did not have a bow boom, nor the sophisticated automatic recording system carried by the other U.S. ships. However, this ship measured downward total solar radiation, K+, and direct solar radiation, I, and recorded these data on an analog recorder. All the U.S. ships made standard WMO surface marine meteorological observations, which were used in the validation of the automatically acquired data. GATE was conducted from June 17 through September 23, 1974. The experiment included three formal ship Intercomparisons , each lasting 3 days, and three Phases, each lasting 20 days. Figures 1, 2, and 3 show the locations of the A/B-, B- , and C-scale ship arrays. Table 1 gives the dates of these observation periods. Table 1. — Dates of GATE observation periods Periods Date Intercomparison 1 June 17, 18, 19 Intercomparison 2 August 16, 17, 18 Intercomparison 3 September 21, 22, 23 Phase I June 26 - July 16 Phase II July 28 - August 16 Phase III August 30 - September 19 This report has been prepared to familiarize the scientific community with the U.S. B-scale ship automated surface meteorological and radiation data. It describes the instrumentation used to observe and record the data, data quality assurance procedures, data processing programs and methods, and the nature and format of the data that have been placed in the archive. The surface meteorological and radiation data were processed with com- puter programs at the Center for Experiment Design and Data Analysis (CEDDA) . Processing included preprocessing of original acquisition tapes, central processing, and final processing. Preprocessing consisted of converting analog data to digital data and separating these multiplexed data into sub- system files. Central processing included converting new data from recording units to scientific units, automatic editing, and correcting relative wind measurements for ship heading and velocity. Final processing included re- scaling certain variables with new transfer equations, deleting certain data, and calculating low-resolution averages. 18° l r i r t r 16° Cape Verde Islands © Africa Dakar 14 c 12 c Priboy 10 c 8° — 6° — Okean North Atlantic Ocean * IC 1 Acad. Korolov A/B-Scale Gilliss Dallas Oceanographer & o Vanguard Quadra Prof. Zubov Poryv E. Krenkel 4° J L J L J L 27° 24 c 2V 18° Figure 1. — Ship array during Phase I. 18° 1 r t r t r 16 c Cape Verde Islands Africa Dakar 14 c North Atlantic Ocean 12° A/B-Scale Priboy 10° Acad. Korolov 8° 6° Okean 4° IIissm T)Q Oceanographer & Dallas Meteor * IC 2 Prof. Zubov J L Poryv E. Krenkel J L 27° 24° 21° 18° Figure 2. — Ship array during Phase II. 18° t r i r l r 16° o. Cape Verde Islands €> Africa Dakar 14 c 12° A/B-Scale Priboy 10° North Atlantic Ocean Acad. Korolov *IC 3A * IC 3B 8° — Okean 6°h- Vanguard B-Scale V Gilliss Planet Quadra Bidassoa Dallas & Fay Oceanographer Researcher Poryv E. Krenkel Prof. Zubov J L J L 27° 24 c 21° 18 c Figure 3. — Ship array during Phase III, Between the central processing and final processing, manual data review and validation was carried out. Meteorological variables were reviewed at CEDDA, while radiation variables were reviewed at the Air Resources Labora- tories in Boulder, Colorado. As a result of the review and validation, manual edit information and revised transfer equations were developed and used in the final processing. Sections 1.1 through 1.4 present an overview of the data acquisition system, sometimes referred to as the shipboard data system (SDS) . Section 2 gives, by instrument type, the sensor specifications and characteristics as defined by the manufacturer. Section 3 discusses pre- and post-GATE calibrations of meteorological and related sensors. The calibration of radiation sensors and the derivation of radiation transfer equations are discussed in section 6. Section 4 describes the procedures for assuring quality control aboard ship. Section 5 presents the details of the data processing by the U.S. National Processing center (NPC) . Section 6 discusses the validation of the radiation data as carried out by the Air Resources Laboratories in Boulder. Section 7 gives an inventory of the digital data in the archives and the format of the data. Section 7.2 discusses the data available on microfilm. Appendices in this report describe tape formats, the pre-GATE inter- comparison of pyranometers , and the translocation of thermistor probes and bridges. Transfer coefficients used in the data processing for both the meteorological sensors and the data acquisition modules are also given. 1.1 Shipboard Data System The surface meteorological subsystem (SMSS) consisted of the pulse code modulation (PCM) tape recording system, two analog-to-digital converters used as the sensor acquisition modules (SAMs) , and meteorological and radiation sensors. Figure 4 shows the total system concept in schematic form. The GATE shipboard data system (SDS) equipment was designed to gather 8 hr of data from any or all of its subsystems onto one analog tape. The data were written in PCM form onto 1 to 7 tracks of analog tape at a real- time write speed of 1 7/8 in./s. In addition to the recording system, a 16K DEC PDP 11/20 computer was used on board all ships except the USCGC Dallas . Each SAM (bow and central) interrogated and multiplexed a maximum of 15 sensor channels every 0.5 s. The PCM output from each SAM was then routed to the time division multiplexer. This unit combined the individual PCM signals into a single data train and converted this input into a PCM/FM output, which was then transferred to the data recording/processing patch panel for distribution to analog magnetic recorders within the data recording subsystem. The computer, when used, converted the raw analog PCM acquisition tape to a digital "data base" tape containing all or part of the data acquired during the 8-hr acquisition period. Software also enabled some data to be reviewed by listing them in engineering units and/or plotter displays. SENSORS ANALOG RECORDING TO SYSTEMS BOOM AND FORWARD DECK MOUNTED SENSORS DIGITAL CONVERTER BOW SAM SHIP HEADING & SPEED SENSORS PCM TAPE RECORDING SYSTEM AND PDP-11 COMPUTER MAST SENSORS CENTRAL SAM NON-SDS RECORDED SENSORS STRIP CHARTS AND/OR H.P. DATA LOGGER Figure 4. — Schematic of shipboard data system. -Hinge End Plate YSI Wet Bulb, Tw 2 - Gill Microvane 3 Cup Anemometer y Radiation Boom 2" O.D. 1/4-Wall Raingauge Hoffman Junction Boa YSI Wet Bulb Tw YSI Dry Bulb T D Pressure Radiation Boom End Plate Eppley Mod. 2 Incident Pyranometer Swissteco Net Radiometer Eppley Mod. 848 Reflected Pyranometer Sea sfc temp T S2 ,ie -in to Hoffman Box Figure 5. — Surface meteorological and radiation boom. Simple analysis routines were available for on-board processing as well as quality control assessments. 1.2 The Bow Sensor Acquisition Module and Sensors The bow system consisted of the 10-m bow-boom platform, SMSS sensors, and the bow SAM. The surface parameters were measured by sensors physically mounted on the boom (fig. 5) and remoted with the data signal routed to the bow SAM. Table 2 lists the sensors by SAM channel and the physical location of each. In general, the boom-mounted instruments were between 7 and 10 m above the water. 1.3 The Central Sensor Acquisition Module and Sensors Three channels on the central SAM were devoted to SMSS sensors (table 2) and included wind speed and direction, precipitation, and atmospheric pressure. The mast instrumentation was mounted on the forward masts of the ships and ranged in height from 18 to 30 m depending on the ship. 1.4 The Boom Platform and Hardware The bow boom system was selected as a best alternative to a meteorolog- ical instrumented buoy. The following criteria were used in the hardware design: 1. Provide the best sensor exposure with the least influence from the ship and mounting platform. 2. Minimize electronic noise from internal and external sources. 3. Sensor accessibility. 4. Incorporation of on-line and off-line standards to routinely assess sensor performance. With the above objectives in mind, the boom (fig. 5) was constructed of lightweight tubular aluminum tower sections„with a total length of 9.1 m and a triangular cross-sectional area of 550 cm . The tubular construction pro- vides maximum structural strength and minimizes platform cross-section. To reduce ship vibration, the boom was shock mounted at the hinge end plate, while lateral stays and uphaul and downhaul guys were used as supports to reduce horizontal and vertical flexing of the boom. The boom was attached to a moveable davit at the hinge plate so that the sensors were readily accessible for servicing. During baseline and maintenance operations, the boom was swung inboard over the ship foredeck. During the GATE International Sea Trials (GIST) conducted in August 1973, signal problems were encountered due to radio frequency interference (RFI) and to drift of the SAM high-gain amplifiers with temperature. In an attempt to minimize RFI during GATE, all exterior cables were routed through electrical conduits to provide weather protection as well as RF shielding. The a.c. power cable to the boom was routed through a separate conduit from CO u o CO C 0) co cu !-i O O Cu ID fa, x! ■rl CU ■H CU JS 4-1 X 4-1 e S en 3 co 3 o o X) O X) O n •H S-l •rl (-1 m M S ^ £ '-' cu M 4-1 u 13 U) CO •H m CO M E B ^ a p, & B ■H •H •H X! * ,£! o co CO to cu CU 0) QJ 1 J-4 U M QJ 3 3 3 t£ 4-J 4-1 4-1 * CO CO CO CO u cu 4-1 M U H u CO M H QJ 0) CU 3 4-1 CO H a H 3 o cu Cu & T-l H o c iH o o ai 4-1 > e e e cu CU ■H 60 cu O c CO a •rl c CO cu cu QJ c c 4-1 3 B CO o o 4-1 c M c 4-J 4-1 4-J c 3 CJ XI ■rl X cu •H X) •H cfl CO cu o CO cfl QJ QJ X QJ S-i 4J 4-1 QJ 4-1 •H 43 a •H -42 ,43 -Q 43 43 U QJ CO QJ 3 c CO 4-1 CO X a B 4-J H H H a O •rl tx QJ D- CO cu ■H o ■H CO cu CO c 3 3 3 13 CO 43 co CO x; X) OJ X M CD 4-1 4-J ■H 43 43 43 QJ CU CU ■H CO ^H CO u •H CO 1 1 1 U Sj X X a P- !E! o U IH !h 4-J CO CO u u >> 4-1 4-1 CO CO C c ■H •H c CU cu a cu X M CU QJ O- ex ■H •H J3 X M Pi IS CO CO w Q & 3 w co 3 [3; CO CO cu QJ S-I M 3 3 CO CO ca CO 1 aj cu 3 u u O a ^— V Cu •rl 4-1 4-1 o c O ^v CJ X) -H 3 •H 3 QJ CU (-1 S-l CO u CU QJ B QJ B •rl Cu X 01 43 CO X) 3 CO Cu CO CU i-H •H CO CO rH XI m X) O cc O C U 3 B n*-' e m •H •rl 4J 4-J — ' S s N00O>OHMcH CU CU W T3 U H CU bO < CO CU O 3 Tj I ■P o 4-1 CO 4-1 TJ rH 3 O CO > a) CD CO O O a ,d ■U cu 14 CO I CU CD S-t CU a 3 CJ 14-1 •H M-4 CD <4H CU CU O O 3 O •H 4-1 03 > -a cu c (-4 -H 3 rH 4-1 5-1 CO CU 5-1 T3 CU d ex 3 6 ^ co cu 3 4-1 cr CU 4-t d M cu CU rH LM CO CO > ,o m 3 h u cr co H cu o l I co cu rH rO CO H co d o •H 4-1 CO u CUO U '*- 4-1 CO M CU CX cfl e CU H o > CM 4-J CM d cu 3 • cr ,£) <4_| CO -H 4-1 O H (1) r x, CO O O a u cuo 5-i 3 4-1 CO !-i 0) CX CO U > 4_) CM 3 CU e CU H ■H n£> 3 • CT rH CU w o H -H CO CD d 4-1 d o CU O co cj I 5-1 -H (U £> U-i M tI >M (X, rH 0) r- cO O CJ a cj 5h O CO T3 d d cu CO cu cO • ■H O V4 55 en CO •H en 00 00 00 CM CM CM st rx m m lo m LO LO rH I I LO I rH CM /~n " CM ,J=> .Q ,£2 rH rH rH rH 3 3 3 1 >■> >^ 4-1 CO 5-i 5-4 CU CU TJ X) S CO H H H H 5-1 CU ,3 U 5-i CO cu CO cu Pi CM LO CM rx LO OX ix C-J CXI O rH CM rH rH • • • • • • • co oo oo 00 00 co co CM CM CM CM CM CM n CM vD iO • • • • • • • st «* rQ ,jO >X 4-1 4-1 5h CU CU ■O S S >, 4J 4-1 CO CO CO U 0) CU CU > CM LO CO St LO UO v£> LO CO ON rH OX CO ON LO co CO CO n£> 00 r-~ CO o ON ON ON d ON o ON OX ON OX OX d d °^ ON ON OX o • • • 3 • • • • • • • 3 3 • • • • • LO LO LO LO ^D LO LO LO LO LO LO LO LO LO LO CM LO CM st CO CM CO CM 00 00 00 00 CM CM CM CM OiAHnT ix o r~ oo St IX o o OX CO rH H CM ON d o O o ON 3 • • • • vo ic ^o in |x 00 LO LO I I rH CM =tfc =fc ON CM CM LO ,£) ^ rH 3 3 rO rO rO 4-) 4-1 CO 5-1 CU CU CU T3 & S CO H H H H 5-1 cx CO U CjC O a CO CU CJ o CJ) o CX CU a u cu 4-J C •H CM > CU rx o rH CO II i — C_> 16 consisted of measuring the output voltage, V OJ as a function of resistance, A standard precision resistance decade box was used for this test. The res- istance was varied over the range of 6500 to 11,999 ohms in 500-ohm intervals. Following the resistance test, the output voltage of each bridge for the two calibration positions and the output of the bridge excitation voltage, V , u i J oex were checked. 3.2 Wind Sensors The 13 R. M. Young cup anemometers were calibrated as stand-alone sensors in the wind tunnel facility of the Department of Environmental Sciences, University of Virginia, before and after the experiment. The wind tunnel used was a large-volume, low-speed unit. The individual sensors were cali- brated over the range of approximately 0.5 to 16.0 m/s. The calibration consisted of measuring the voltage output, V , of the anemometer generator vs. the tunnel wind speed corrected for atmospheric pressure, temperature, and humidity. Two runs were made for each sensor: increasing wind speed from approximately to 16 m/s and decreasing from 16 to m/s in 21 incremental steps. Each step corresponded to an increase or decrease of 0.5 in the tunnel drive motor speed. In practice, the anemometer was allowed to settle down at each motor setting for about 5 min. Table 4 shows the transfer equation coefficients resulting from the pre- and post-GATE calibrations of the stand-alone sensors. Pre-GATE trans- fer equation were used in the final processing. With the exception of the boom wind speed sensor on the Gilliss , differences between pre- and post- GATE transfer equations were generally less than 0.3 m/s. An 1800-rpm constant-speed, synchronous drive motor was used for all baseline checks at sea. The initial calibration consisted of checking the anemometer output voltage corresponding to a rotation of 1800 rpm. All sub- sequent baseline values obtained at sea were referenced to the initial value. Before GATE, the wind vanes were calibrated in position on the boom and/ or mast from the sensor through the SAMs . In the case of the boom vanes, the dead-band was oriented parallel to the boom facing aft (180 relative to the boom outboard). The initial sighting in of the vane relative to the boom and the boom relative to the centerline of the ship was done with a theodolite. The vane was rotated first clockwise, then counterclockwise, taking readings of output voltage, V , vs . relative position in degrees rotation. The calibration positions corresponded to vane headings of 10, 60, 120, 180, 240, 300, and 357 . Second, the vane was locked in position with a pin and collar device, and V Q readings were taken at the four baseline check positions of 0, 90, 180, and 270°. The full calibration (above proce- dure) was carried out only for the four boom-mounted vanes. In the case of the mast vanes, the deadband was located 180 relative to the longitu- dinal axis of the ship. Following the experiment, the outputs of the sensors were checked for linearity but were not recalibrated. 17 ■u 1 CU G 5-i cu ex H ^^ co > CO •H C 3 o ct-h 0) 4-1 CO »"> T3 M 60 c J3 a CO •H •H rH CO /— \ CO CO ■T3 CJ cu C o O 5-J o CJ CU 5-i CD 4-1 ex CO e rH M o cO 0) s c ex CU •H CO S m u 0) cu w £1 +j H 4-1 cu < B O 1 •H o 4-1 u CO T3 O cu CO ex CO 4-1 3 iH -a O > § cu ^^ i cu 5 CO CD e 5-1 CD o ex 4-1 •H fi 4J 5-1 cu CO o •H 3 m CJ CT •H CU T3 M-l CU <4-l 5-1 CU CU CD a. o r>. LO O 00 ON r^~ 00 13 4-1 cu a rH o O CT. CT> <3> 0) cu H H CXrH CO CO cd > Is > X) •H 00 C 3 rH CN CN O CN CN CN H rH r^ •H cr 3 cu O rH rH CO rH rH rH rH rH w H < O CO CO CO a a CU i CO O rH Pi CO a H CJ 5-1 -H 42 M-j H <4-l CU O r O CJ CO O 1 — r^« ~d" co CJ> 6 > X) -H rH r-> C 3 rH ON (T* o O^ O o o a-x a> •h cr £3 cu O O o rH rH rH rH CO CO S 4-1 w o H -H CJ\ 5j o c CO 0J o o CN CM LO CO cu ex o CJ 18 u o 4-1 c OJ H > •H 3 cr OJ 0) M co en 0) M ex s < a co CO en I H 0) H U O I in CO OJ a c CD /-s 5-1 ,a oj e 4-1 •H P E C QJ CD > •H cr cu OJ w co OJ 5-1 Ph O 5-4 O CO C 0) C/0 w H < O I u CO O Pm o ! OJ 5-i P-, 4-1 CO 3 4-J 3 O o a •H CO CO vO HHH o o o o o On ^o ro cti O H i— I OHH O O O O O O iH rH ^H H H rH CN v£> CN O ON 00 CN CN ON O H H o o o O oo o O rH CN O O O cr> oo m N (N 0> v£> CN r-» CN H On i— I H O o o o H O 00 O O On ON On 00 o -X m 5-i 0) CJ 5-4 co 0) co 0) Pi •H CJ CO a v£> i — I r~- o o o + + + CO O 00 H O oo vD d ON vO P~) On O H H OHH O O O O O O r~-- ON H on oo r^- in CN ON OHH o o o o > 5-i CO OJ G co CO & Pi o •H 4-1 CO 5-i £i •H H CO cj H O I a) 5-i CX OJ ja H 19 3.3 Pressure Sensors Three of the four Kollsman pressure sensors were factory calibrated immediately before GATE, and the fourth, used on the Dallas , nearly 1 year before. Following the experiment, all units were calibrated by the National Bureau of Standards (NBS) . Table 5 gives the pressure equivalents for the Kollsman output (counts) derived from the pre- and post-GATE calibrations. The largest difference, a change of 0.24 mb , was shown by sensor 454, which was used on the Dallas . 3.4 Ship Heading and Speed Sensors The ship heading sensors were precision potentiometers interfaced to one of the ship's gyro repeaters. All units were calibrated statically by manually rotating the output potentiometer in incremental steps and record- ing the output vs. the ship headings. The ship speed sensors were also calibrated by placing the sensor in a "simulate" mode and recording the out- put in incremental steps. 20 4. PROCEDURES FOR ASSURING QUALITY CONTROL AT SEA The operating procedures for the surface meteorological system were designed to assure data of high quality. Systematic procedures were set up to check the SAMs and sensors individually. Five operating modes for the bow and central SAMs were defined and used operationally at sea: 1. Acquisition (bow and central SAM). 2. Simulate (bow and central SAM). 3. Baseline substitution (bow SAM only). 4. Baseline full (bow SAM - central SAM). 5. Maintenance (bow SAM - central SAM). Figure 7 schematically illustrates the operational modes and the components they included relative to the bow-boom hardware. The normal mode was the data acquisition mode. The boom was in an out- board position and sensors were acquiring real data. The simulate mode was designed for checking the SAM calibration and stability. A switch on the SAM enabled the operators to take the sensors off line and input precision fixed voltages of 0, 4, 8, and 40 mV, and 2 and 4 V. The simulation was performed three times per day or once for every PCM acquisition tape. The SAM outputs corresponding to these precision input voltages were checked against strip charts, computer readouts, binary values on the voltage digitizer box and on the decommutation display units. These test data were compared with established "tolerance windows" for each SAM channel. The baseline substitution was designed for checking the YSI thermistor bridges and the boom rain-gage electronics. To check the thermistor bridges, precision resistances corresponding to ambient temperatures of 22 and 32 C were switched in, replacing the YSI thermistors. This test together with a simultaneous reading of the bridge excitation voltage, V , was used to assess the performance of each individual bridge, and was conducted routinely. The outputs were checked in the same way as for the simulate mode. Table 6 gives the maximum possible temperature deviations due to the temperature bridges and the SAM instabilities based upon baseline substitut- ions. As seen, the hardware associated with the thermistor bridges and the SAM were extremely stable, with maximum deviations of less than 0.05 C. The baseline full testing was used for checking the entire system (sensor through to SDS) . During the test, the sensors were on-line in a 'controlled environment and compared against independent standards. At sea, these tests were performed on the boom temperature, precipitation, and wind sensors. The temperature sensors were placed in a stirred, water-filled 21 2 3 CD w 0) CD cc 43 ca c c o o ■H 4-1 Tj 03 CD CD ,D CO CD CD •H 4-J •H 5-i 3 •u c0 H U -H CD 43 a CO e w cd co •H ft •H CO CN C~\ CN CN O O O O O O O O I I I I i— I CN i— I iH O O O O O O O O I I I I CN t— I CM CN O O O O CN CN CN CO O O O O O O O O I I I I o o o o I I I I rH O iH iH O O O O O O O O O O O O I I I I o o o o I I I I CN CN r-\ i— I O O O O O O O O I I I I CN 03 D t 1 :* i- LU i g s. >*; r i i T «D <■ CM O CM CM CM CM '.1 oc '< I 1 CD 1 1 CM 1 O CM CM CM CM n i r CO "* CM CM CM CM .1 ..'I ;» ."I 1 I I CO >fr CM CM CM CM CO LU fc- D CO X) 01 u 4-1 CO cu §• 0) ■U 0) c •H 0) CO •H O & M m o ! CO w I I cu u 3 bO •H I 1 i rn^ 00 CD «» CM CM CM CM CM CU 5-1 4-) CO >-l CU u CO M-l CO cO CD co C o CU CJ c cu H C •H OJ 4-1 cfl s bO C •H rH o o o cu c •H M ti CU a •H CO LM O CO w I I 0) 60 30 The wind speed and direction sensors yielded very good data. The exter- ior shell of the Researcher' s mast wind direction sensor rotated 47 hefore the start of IC 1. The sensor was readjusted after IC 1, and no problems were encountered during the remainder of GATE. The data for IC 1 were corrected based upon the result of a comparative analysis of the data from the Researcher and other international ships. The Dallas ' boom wind speed data, which have been archived, are question- ably low for the time periods of Phase III shown in table 7. The problem is believed to be linked to a faulty sensor. Table 7. — Questionable Dallas bow-boom wind speed data Fr om To Date Time (GMT) Date Time (GMT) Sept. 8 Sept. 9 Sept. 11 Sept. 12 Sept. 16 0600 0900 2200 1300 1400 Sept. 8 Sept. 10 Sept. 12 Sept. 15 Sept. 18 2200 0700 0000 0600 0100 The Dallas ship heading data required special processing because of sensor abnormalities discussed in the next section. The Gilliss data exhibited two other peculiarities. First, the ship would steam occasionally with the wind and sometimes at approximately the same speed as the mean wind speed. In such instances, the boom anemometer cups would stop. Yet the indicated wind speed was greater than zero. The velocity of the ship was not zero, however, and it is this velocity that caused the non-zero wind speed. Second, the rocking motion of the Gilliss frequently shows up in the wind speeds and directions as measured by both mast and boom sensors. This feature is most noticeable during periods of light wind speed. Figure 11 shows this effect in some 0.5-s data. Note that amplitude of the oscillations in the mast wind directions exceeds that of the boom. 5.4.3 Ship Speed and Heading The ship speeds as measured by underwater sensors are of poor-to-useless quality for all the U.S. ships. Relative changes in the speeds are discern- ible for the Oceanographer and Dallas , but the magnitudes are not correct. The Gilliss sensor apparently failed after Phase I and the Researcher 's sensor was not recorded after Intercomparison 2. None of these ships' speeds were used as part of the ship velocity corrections to the wind speeds. MAST WIND SPEED MAST WIND DIRECTION en 300 LU 111 240 CE 180 UJ 120 Q 60 •^AwVw^w — , — vV^JVV^A rN> V\A''VV^-^^ BOOM WIND DIRECTION MINUTES Figure 11. — Example of oscillations in Gilliss 0.5-s wind speed data caused by ship's roll. CLOCKWISE TURN 360° TIME COUNTER-CLOCKWISE TURN 31 360- 300- B 240- § 180- uu Q 120- 60 360° 313° jL TRUE DEAD BAND POTENTIAL TIME' Figure 12. — Dallas indicated ship headings for clockwise and counterclockwise turns 32 The Dallas was the only ship that had problems with its ship heading sensing system. The principal problem originated with the potentiometer, which was mounted on a repeater of the main gyro compass and had abnormally large contact strips at each end. The net result was a loss of continuous o o direction information between 313 and 360 , since the voltage put out by the potentiometer was constant for the directional width of the strip. Figure 12 shows a time-series plot of the ship heading for a clockwise and for a counterclockwise turn. For a clockwise turn, the indicated ship headings in passing from 313 to 360 would stall for a while on 313 and then exhibit an exponential decay and finally jump to 360 , remaining there until the ship heading had reached 01 . This exponential decay was due to the falloff in the electrical potential through the potentiometer as shio headings passed through the true dead-band of the potentiometer. To determine the width of the contact strips in terms of ship heading in degrees, an analysis was made of the Dallas heading data for periods when o o the ship made a continuous and uniform turn through 313 to 360 . Only cases where the rate of turn was held constant were used. It was found that while the ship heading was indicated as 313 , the direction could be any value between and including 313 and 334 . For an indicated direction of 360 , the ship heading could have been 338 to 360 . The true dead-band o o of the potentiometer was 335 to 337 . In an attempt to recover what amount- ed to lost ship heading data and hence wind velocity data, a value of 324 was assigned to the ship heading whenever the output indicated 313 , a value of 349 was assigned whenever the output indicated 360 , and a value of 336 was assigned whenever the ship reading was in the dead-band of the potentio- meter. This assured that the maximum error in ship heading would be 11 . A computer program was designed to automatically assign the above ship headings whenever the indicated ship heading fell between 313 and 360 . The time periods when the ship headings were in the dead-band were determined by examining microfilm time-series plots. This information was entered into the program that corrceted the ship headings. The ship heading data for the remaining three ships presented few if any problems. In general, the few problem areas that did exist were removed from the archived data. 5.4.4 Pressure There were two pressure sensing systems on each of the U.S. ships, the Kollsman and Rosemount. Generally speaking, the Kollsman is more accurate and more stable. The Rosemount pressure sensor drifted by 0.5 to 1.0 mb in both the long and short term (Godshall et al., 1976). The Oceanographer ' s Kollsman sensor functioned erratically during the experiment and much of these data are missing for Phase II. In addition, the Oceanographer ' s Kollsman data contain high-frequency fluctuations that are not contained in the Researcher , Gilliss , or Dallas Kollsman data. Late in Phase II, the Dallas Kollsman sensor drifted almost 1 mb . 33 5.5 Final Processing The objectives of the final processing programs were to rescale some of the data using revised transfer equations, to delete bad data, and to calcu- late averages for the 3-, 10-, 30-, and 60-min averages, and the hourly observations. The input to these programs were the 4-s average data processed by the SMDSS Program. Three computer programs were used in the reprocessing: DALSH, OCESH, and RADREPRO. The first two were special processing programs that were used on the Dallas and Oceanographer ship heading data, respective- ly. DALSH was used to recover the ship heading data between 313 and 360 as discussed in section 5.4.3. OCESH was used to rescale the Oceanographer ' s ship heading data. Because ship heading data were used to calculate absolute wind velocity from relative velocity, reprocessing meant correcting the wind velocities as well as the ship headings. This was accomplished by reversing the steps used to calculate the absolute wind velocity (sec. 5.2). The Oceanographer ' s original ship headings were then rescaled, and fixed headings of 324 , 336 , and 349 were assigned to the Dallas ship heading data as discussed in section 5.4.3. With the adjusted ship headings, the absolute wind velocities were then recalculated using the scheme described in section 5.2. Program RADREPRO was used to rescale some of the temperature data and all of the radiation data, to delete bad data that were not removed in the automatic editing of the central processing steps, and to compute the average data sets. The processing flow is shown in figure 13. The program first read the processing constants and control features into the computer. For the first run of a series of tapes containing contin- ous data, the program set up a directory to keep track of incoming data. On continuation runs, the program read in a partially processed file that provided the necessary information for a restart. Next, the program read a record of data and loaded it into the process- ing array. Some of the temperature and radiation variables were rescaled using revised linear transfer equations, and the program then deleted the bad data in the record. The ship heading data of each record were divided into north-south (SH ) and east-west (SH , components in order to compute the average ship headings. Assuming the speed was unity, and SH Ng = 1 * sin (0), SH NS = 1 * C ° S (G) ' where 9 is the ship heading in degrees. These components were than averaged, and were used to compute the average ship headings. 34 f START 1 SPECIAL PROCESSING FOR THE OCEANOGRAPHER AND DALLAS PROGRAMS OCESH AND DALSH o RE SCALE SHIP HEADING DATA o RECALCULATE WINDS AND COMPONENTS REPROCESSING PROGRAM PROGRAM RADREPRO o RESCALE RADIATION AND TEMPERATURE DATA o DELETE ERRONEOUS DATA o CALCULATE DERIVED VARIABLES o COMPUTE LOW RESOLUTION AVERAGES o OUTPUT AVERAGES 4-s AVG. 3 -min AVG. 10-min AVG. 60-min AVG. HOURLY - OBSERVATIONS Figure 13. — RADREPRO program processing flow. 35 The program also computed the dew-point and specific-humidity values for each wet-bulb temperature. The specific humidity (q) was computed from 622 * e q - (p - 0.378 e ) ; P = I""- 2« mb, where e is the ambient vapor pressure, and p is the atmospheric pressure, both in millibars. The vapor pressure was computed from e = e , - A • p ' (T, - T ), wb dry wet where A = 0.00066 (1 + 0.00115 T ), wet T, = the dry-bulb temperature, dry T = the wet-bulb temperature, and wet e , = the saturation vapor pressure at the wet-bulb temperature, wb The saturation vapor pressure at the wet-bulb temperature was computed by using the Goff-Gratch formulation ( Smithsonian Meteorological Tables , 1951), 1Q (B+C+D) wb where and B = -7.90298 (373.16/T - 1) wet +5.02808 ALOG (373.16/T ), 10 wet C = (-1.3816 x 10~ 7 ) • (10 F -1.0), D = 0.0081328 • (10 G -1.0) + 3.0057, T = wet-bulb temperature in absolute degress, wet OS F = 11.334 * (1.0 - T /373.16), wet G = -3.49149 • (373.16/T -1.0). wet ' 36 The dew-point temperature was calculated using Teten's equation in the following form: 237.3 • ALOG nrt 7^— m _ 10 6.11 T DP "" 7.5 - ALOG 10 6.11 where e is the ambient vapor pressure. As each record of data was processed, each record containing a 4-s average of each variable, it was placed on an output file. At the same time, the 4-s average samples were added to accumulations for the low-resolution averages. Sums and averages were calculated for the following time intervals (in minutes) of each hour: 3-min average: 58.5 to 01.49, 01.5 to 04.40, etc. 10-min average: 55.0 to 4.99, 05.0 to 14.99, etc. 30-min average: 45.0 to 14.99 and 15.0 to 44.99. 60-min average: 00.0 to 59.9. Hourly ovservations : 50.0 to 59.9 for all variables except pressure; 55.0 to 04.9 for pressure. Note that the hourly observations of pressure were 10-min averages centered on the hour. The average wind directions were calculated from absolute average wind components (u , v ) . The 4-s average wind speeds were used in calculat- ing the above scalar wind speeds, i.e., the average wind speeds were not computed from the wind velocity components. The average ship headings were calculated from the ship heading compon- ents (SH , SH ) with the unit speed as discussed above. All other average scalar values, including the derived variables of specific humidity, dew point, and the u - and v -components of the absolute wind velocity were a a calculated by normal summing and averaging. As the individual averages (3 min, 10 min, etc.) were generated, these were placed on output files, which were subsequently copied on to the archive tapes. At the same time, these data were used to generate microfilm time-series plots that correspond to the digital data. In addition to the average values given in each data set, the number of 4-s averages used to generate this average is given. 37 6. ANALYSIS AND VALIDATION OF SURFACE RADIATION DATA The incident solar radiation data, the reflected solar radiation, and the net radiation data were all subjected to careful intercomparison and validation. As a result of these analyses, transfer equations for the net radiation data were developed for the first time, and revised transfer equations were derived for the incident and reflected solar radiation data. The following sections describe how the data were validated, and the transfer equations used. * 6.1 Recording System Error Limitations The SAM, as described in section 1.1, had a voltage range of to 5 V, which was divided into 65,535 counts. The millivolt input signal from each radiometer was fed through an amplifier with a gain of 1000 to accommodate the range of the recording system. The amplifiers showed small variations with time and were sensitive to temperature changes. To check the stability of the recording system, a set of known voltages was put through the system. For the high-gain radiation channels, these voltages were mV, 4 mV, 8 mV, and 40 mV. This procedure, called simulate, was completed three times each day as discussed in section 4. The simulate records for the radiation channels were analyzed to determine the magnitude of the error produced by variations in the amplifiers. All radiation data were recorded as counts. To determine the transfer equations between the recorded counts and the equivalent radiation in W/m~ , the counts-to-millivolt relationship for the SAM had to be developed. The millivolt-per-count ratio defined the slope of the transfer equation, and the instrument zero offset determined the intercept. Variations in the amplifier caused a change in the ratio of millivolts to counts (the slope of the transfer equation) , while variations in the 0-mV count value changed to intercept. An analysis of the simulate records showed that the change in slope was not related to the change in the 0-mV count. The linearity of the slope was checked by comparing the millivolt-count calculated for each simulate voltage interval (4-0 mV, 8-0 mV, and 40-0 mV) . The results showed that each channel was linear over the range of the record- ing system. Because the sensors and recording system had different impedances, the instrument zero was not the same as the recording system zero. The mean of the nighttime values measured by the pyranometers throughout GATE was taken to be the instrument zero offset, and the intercept of the transfer equation was adjusted accordingly. Variations in the sensor zero offset can be attrib- uted, in part, to changes in the amplifiers of the recording system, and the effect of other factors can be determined by calculating the variation in the mean offset and removing the known effect of the amplifiers. Errors due to the change in the ratio of millivolts to counts are less than 0.2 percent. Variations in the 0-mV count value and the instrument zero offset caused an error of 1 to 3 W/m , or 0.8 percent of the mean net radia- tion integrated over daylight hours. The total error attributed to the 38 recording system and the instrument zero is less than 1 percent of the daily integrated net readiation. 6.2 Incident and Reflected Solar Radiation, 6.2.1 Instrumentation and Transfer Equations The pyranometers mounted on the U.S. ships during GATE were the Eppley models 2 and 8-48. The model 2 was used to measure the global solar radia- tion on the four U.S. B-scale ships and the reflected solar radiation on the Oceanographer only. The reflected solar radiation was measured by the model 8-48 on the Gilliss , Researcher, and Dallas . The sensitivity (calibration factor) of the pyranometers, as given in table 8, was determined during the Miami intercomparison in April 1974 (see appendix B) . Transfer equations were determined from the instrument sensit- ivities and the simulate records. The intercept of the transfer equation was adjusted in accordance with the mean instrument zero offset discussed in the previous section. The 0-mV count value for both Dallas pyranometers was not offset. Since the recording system did not record negative voltages, the sensor zeros were suppressed. Intercomparison data show that the Dallas data are in agreement with data from the other ships within the limits of error. This indicates that the zero suppression of the Dallas instruments was small, and the recording system zero was therefore taken as the intercept of the transfer equation. The simulate records of the Gilliss show that the zero of the model 2 pyranometer measuring solar radiation was suppressed until July 4 (Julian day 185) when the simulate zero was adjusted to a positive count value. Intercomparison 1 data show the Gilliss solar radiation to be approximately 15 W/m~2 lower than the other ships. This is due to the instrument zero suppression during Intercomparison 1. Because of the change in the 0-mV count value, two transfer equations were developed for the Gilliss upfacing pyranometer. As with the Dallas , the intercept was taken to be zero until July 5, and approximately -15 W/m~^ after that time. This adjustment brings the Gilliss Intercomparison data into agreement with the other ships. 6.2.2 Data Validation Figures 14 to 18 show the hourly integrated solar radiation data (K^, Kt) for the U.S. ships and the Canadian Ship Quadra during the Intercompari- son periods. The Canadian data were used as an independent check for both solar and net radiation to further verify the quality of the data. The Gilliss did not participate in Intercomparison 2, and the Dallas was absent from Intercomparison 3. Also, the Oceanographer and Quadra took part in Intercomparison 3A while the Researcher and Gilliss took part in Intercomp- ison 3B (fig. 3, sec. 1). The solar radiation data from all these ships agree to well within the 5 to 6 percent recommended accuracy. Differences between ships in the hourly integrated radiation is due primarily to variations in cloud cover over each ship during a particular hour. The curves for Intercomparison 2 39 CO cu co H 0) •H W -> CO C o CO •H S-i cfl a e o cj M 0) Ui 00 CO en vO m oo oo CN m CN O o CO ■K CO O* •H rH CO H CO •H en o vO 00 CN CO m oo i— i r»» o co 00 CN CO SO 00 CN o r^ 00 -sT 00 ■$% \o o r-- s cr i • 4J £ cu e o •H a ^^ 4-' ■z > — ^ r-l |3 CX •H rH CU w CU /-> rH 4-1 CO iw OCN CO •H u CO cu S-i 1 •H CO s, c a cu B M C > CO o 4-J "*>. a) cu 6 U rH c ^ CO CO H CO H ^ cyv cu rC a cO f-i 00 lo CN O . CO r- in r^ O CN CN . • . tH vO o en 1 O r~~ oo ro -d- ^C r- CN m CN o -* rH • • • rH I s * o rH 1 •H £ I — ^ CN 4-J >. C . 4-) E P O •rl U o 4J 53 > *■ o ex ■H H "»v. CU r~> rH 4-1 CO CN UCN cO •H CJ CUI S-I 1 •H CO ~-^ a. E cu E ^ C > o 4-1 ~--^ CU CU s rH £S c 12 co CO >w ' CO *«-' H w 40 -a 01 3 c •H u a o a CO u 0J 4-J 0J E o •H CC u W H < 03 4J c QJ 4H m QJ O a 0) m en C CI] cd >^ 4-J ■H > •H 4-1 •H CO c QJ en I l co CD X) crj H CN ON vO> co co CN CN m CN) o CO CD co ca .c a. 0) •H o w }-i QJ rC * a o- u T-\ cd CN QJ CTi CO ^£> a) Pi a o co •H CO e o u n QJ t*i m cn nj co m C-N in O ON o o I m CN o LO LO o cjn 00 CO m o rH «sT r^ O ON vo on CN co rH on CN o O co CO N I C r^ 1 C • 4-J £ 3 • 4J e 3 o •H a o 4-1 o •H O O 4-) 22 > •H <-* CJ a, QJ /-v 3 > *-- •H rH a QJ ^ <-{ 4-J cd C-J OCN rH 4-1 cd CN ocn co •H CJ QJ U 1 CO •H O QJ U 1 •H CO ^^ a E oj e •H en -^ Ci E 0J g u G > o 4-1 — -~ Sh fi > O 4-J *«-. QJ QJ e rH 3 a rs QJ QJ E r-\ Bs c ts CO CO v — ' CO ^-^ H v -' CO CO ^ CO ^w' M v - / 41 T3 G T3 CO S-i W H < CJ5 w D 0) O CJ }-i •H ■U •H co pi QJ CO I 00 QJ .-I ■s 4-1 •H > CO C Q) CO P- ■, •H cO >H TO CD -o- 4-i r^ CO ON Q H en 1-1 1—1 CNI CO iH co CN 1-1 o O iH OJ o CN O i—l on 00 co o o LJ~| oo co C3N on co O CON o o o en OJ lo o o . CO > a) CO •H C cd O oo <]■ 00 o CNI r- oo O o o l-> o o OJ o OJ r^- CON CO \D H 1 1-1 1 OJ 1 CN 1 1 co 1 ON 1 CO 1 H M0 r~- CON CO rH 1—1 i-H CN co CN co OJ • • 4-) M • PL, 3 00 0) < 1 3 < co 1 oo 1 ON OJ CO OJ QJ . . C 00 00 3 3 3 1-3 < < G i— 1 oo 0, 3 3 OJ 1-3 i 1 CO 1 I 1 oo OJ 1 C3N rH OJ r-H H 0) CD >. . c G H 00 o 3 3 ►-3 *-> >"3 < MO co o 4-1 CO >-. CO -a G CO 3 >■. 3 >-3 O 4-> QJ G '-o O u-i o o co a cu o cu 4-1 c CO co ■H cu ?N 4-1 •H > •H 4-1 •H CO 03 co C o ■H 4-1 CO •H H CO > TO QJ 13 O rG 03 U QJ CO- CO u 00 o G CO QJ CJ o QJ X. G o QJ 4-1 QJ g O G CO a CO > QJ J=. CO •H c CO >- X! C CO !-i QJ 4-J OJ B o •H T3 CO !^ QJ 00 C C •H QJ M (El 3 H T3 42 00 L- to u 0) t/> o 03 aj U —> O Q T 1 CL a i i i o o O o CN CN CN CM O CN CO CD S c cfl "* C •H UJ cfl C CN 3 cfl o U • — ' T3 • O H C /-N ' — ^ CO CTN O • H CO CO Cfl CD Cfl 00 4-1 *D cfl H -3- X) Cfl O Cfl CN radiati Julian d ■^ u ^ CN cfl r-i 00 O r-\ CN Cfl CN T3 0) cfl O 4-1 CN cfl r-. U rH 60 no CU CO 4-> C •H ►n CD >-, - , „ rH CO en H O. <3" C 3 -H •D O X C K cfl UJ 1 1 CM D O X H O 1- CU o 3 60 00 •H Q O o o O O O o o O Oi oo r-- CD ID S 2 o o *fr CO o o CN O O 43 00 CM CN Q c ti ■ — * r O O l- T3 c 53 CO o • 1 — co C/3 H £3 CO CD „ CO CO T3 4-1 TO c ^r T3 CO •H C ^H o D CN •H >"0 J-> V CO •H vXD -a ^H CO S-J 4-) CN S-i CO 01 6B rH 3 CM o <3 CM LO HO -a c o a) X CM ■U CO CT\ u H CJ3 CO aj CD ' 4-J c C 3 •H *-} CD zr c H en U D- ^r ■5 c LU 3 O •H HI CD 1 rsi - c 1 1 I " o h- '— s CJ o S-i 3 CO o "3- o o CM o o CJ) o o CO o o o o o o o o o o o o CO LO «fr 00 CN o o 44 o CN CD Q a> CN CN t 5 co g ro ro $ "2 ? 01 ~ 5) (D c u (o o 3 m o q cc a > I T I i : III!: 00 co CO CN CD • -n cx> co d • CO fa o> CD •> CN CO CN 4-1 CO CO •** T3 ^ CO d "B o CN radiati (Julian ^ M 00 CN CO iH O T3 CN CO c CN CO X) a) r^ CN CO J-< 4J 00 CO 00 CD 4-1 d 60 d d •H <: CD >, * . iH CO O) S-i ex <3" c 3 •H "O O 43 c HI CO LU I fc- i CN D «"" o ^D X T~\ O 0) d 60 00 CD •H o o en o CN -*■ p ° > en o o o o o o oo r» co o o o o o o in ^ en o o CN o o 45 ID CO CN > Q ^r L. CO X (M -C Q. > CD m i_ Q O ro r C "- CD CD T3 d) CO 3 -J O D o a o >3- o co o CN CN CO co C LU c o co CD — c T3 «* "C c 1 — c cO LU • J_ }-i /-^ CN r: 0) LTl o J3 ^o T a CN co O r- 2 BO O T3 C CO CJ c CO c CNI co co O tH CN CO CN CN 13 U CD (D n U X> CN CO e Vj CD no CD CX J-i 0) C CO >, CO H u H -d 3 CO O w c1 CD S-i GO •H 46 -a •~\ g LO rt vO CM M CD 13 4= u u M cO ■U id 03 0) !-i 4J 00 a- CD CD •U en C ■H «\ co >•, CO H •H )-i H 3 H O •H pa i o oo CD •H _>. E o o o O o o O o o V o o o o o o o o o s: CD oo r^ CO m ^r CO CM *~ 47 (figs. 15 and 16), an excessively cloudy period, show much larger anomalies in the hourly values than are seen in the other figures. Data collected during clear periods agree to within 2 percent. Table 9 shows the ratio of the simultaneous radiation measurements on each ship to the measurements on the Oceanographer . Ratios are given for each day of the Intercomparison periods when simultaneous hourly values were available. For Intercomparison 3B, the ratios of the Gilliss data to the Researcher data are shown. The large ratios of K+ for individual days are primarily due to the hourly variation in cloudiness over each ship. The very large difference in reflected radiation can be attributed to the location of the downfacing pyranometer as well as to instrument charact- eristics. The pyranometers were mounted on booms extending beyond the bow of the ships. They were at different heights above the ocean surface and the bow of each ship was within the instrument's field of view. At low solar elevation angles, the shadow of the ship may have fallen below the sensor. The response of a pyranometer is highly variable when receiving energy from low angles and the model 2 pyranometer has a different cosine response than the model 8-48 ( see appendix B) . Comparison of the simultaneous values of reflected radiation obtained from the model 8-48 pyranometer and averaged over all three Intercomparison periods for the Dallas , Gilliss , and Researcher shows that these data are in agreement within 3 percent. The model 2 pyranometer on the Oceanographer indicated much lower reflected energy, except during Intercomparison 2 cloudy periods. This does not imply that the model 8-48 produces a more accurate value, but only that pyranometers of the same type have similar characteristics, The GATE Intercomparison data show that the relative accuracy of the reflect- ed solar radiation data collected from shipboard is marginal and can be estimated only to approximately ±10 W/m~ . 6.3 Net Radiation (Q*) 6.3.1 Derivation of Transfer Equations The net radiometers mounted aboard the U.S. ships were not calibrated prior to GATE. The sensitivity (calibration factor) of the instruments was determined from an analysis of the data collected during the experiment. The net radiation is defined as the difference between the incoming and out- going radiation, given by the following equation: Q* = (K++L4-) - (K++L + ) where Q* = net flux of total radiation, Kl = downward direct and reflected solar radiation, Kt = upward solar radiation, 48 cu o CO 4-1 T3 o 4-1 5-1 CO 01 -x Pi c/ o CO 3 !-J CO CO - E <- o M CJ 5-i •> 0) H* 4-1 W G -a 5-1 CO o •H -C CD CO 3 O CU 3 CO 4-1 w H CUD c •H 5-1 3 -d CO 4-1 CO 3 5-i B cu •H 4= co Cu CO 4-4 5-4 o GO O o c •H CO 4-1 CU CO CJ Pi o o-n QJ i— I XI CO H CU PQ faC CO CO U h H cu > CO CU 4-J CN O co 00 o OJ OJ o oo O o 1 o o O oc o OO o O OO o CO H O rH H ^H H rH Ol O CO rH o rH UO rH OJ Ol O 00 O oo OO o I o H co a U **— ' CO 5-i to cO CO •H co CO 5-i 3 • rH cu <-\ T3 00 o rH CO H cO 3 a) •H cu cO 3 cO a o etf Q CD- > o CN co O o uO CN OO OO LO *£> o ^ UO rH o 1 1 1 5-i • id rH o v-\ CM cu CO C^ rC OO 1 1 oo 4-J o O O 4-) v£> ^£> CU CO CN 3 rH 1 1 C^ T3 rH o >o rH •H 5-1 cO e oo CN •H 1 1 oo oo 5h • • Cu o o CU U uO r-- 00 cO •H C 00 00 CO •H 00 UO CN 1 o o o 5h • • • O rH rH rH >4H o o g UO CO O 1 OO 1 oo 4= • • CO o o -> OO 00 CO o 00 MH Cu 1 o> 1 o\ O -H • • X o o CO CO o •H rH 4-1 CO CO 3 5-i T3 •H 5-1 r4 > CU CU -H rC MX) co CJ 5h fi CO U CO CO CO -H •H cO CO 5-1 rH rH cu rH T3 5h rH en rH CO CU 0) •H cu CO 3 X > o Pi O O* H O 49 L4- = downward atmospheric radiation, and Lt = upward terrestrial surface and atmospheric radiation. The incoming solar radiation (K+) is measured accurately by the upfac- ing pyranometer, and the upward terrestrial radiation (Lf) is calculated from the sea-surface temperature. An estimate of the reflected solar radia- tion (Kf) is known from the downfacing pyranometer. Since the reflected radiation is more than an order of magnitude smaller than the other radiation terms, it is unnecessary to know its precise value. The error in the deter- mination of the net radiometer sensitivity will be insignificant as long as K-l- and Li and accurately determined. The equation defining the net radiation can be solved for the downward atmospheric radiation term (L4-) , which for daylight hours is given by L4- = q* _ K4 + K+ + L+, (2) and at night, when the solar terms are zero, becomes L+ = Q* + L+. (3) Variations in the downward atmospheric radiation are primarily due to changes in the water vapor content of the atmosphere and the amount of cloud- iness. The tropical atmosphere over ocean areas and away from the continen- tal land masses has the unique characteristics of a high and relatively constant moisture content and little or no diurnal variation in moisture or cloudiness. We may therefore assume that, over the many days of GATE, the mean atmospheric radiation is the same during daylight and nighttime hours, and we can set eqs. (2) and (3) equal to each other ^ (Q* + L+) . . = (Q* - K+ + K+ + L+) , • (4) night day By adjusting the sensitivity of the net radiometer until equality in eq . (4) is reached, the correct sensitivity is derived. 6.3.2 Data Validation To determine their accuracy, the hourly integrated net radiation data were compared in the same manner as the solar radiation data. Figures 19 to 23 show the hourly net radiation data for each day during the Intercomparison periods^and table 9 (sec. 6.2) shows the ratio with respect to the Oceano- grapher . The Canadian Quadra data are also included as an independent check and to verify the method used in determining the net radiometer sensitivities, The Canadians verified their net radiation data by an entirely different approach. They took simultaneous measurements with similar instruments and also measured the four radiation components directly. They found good agree- ment between the measured and calculated quantities. 50 0) CD 00 co O P as g 8 s -5 o to vO cd rH > cd CN tf 13 4-1 TO Cd -rH •H H 13 3 cd 1) ^ J-i ^ CM 4-1 00 (U rH CM ti rsi 13 13 C . - CJ) 1— 1 CO c U p. 2 c LU 3 -H O ,£3 ffi CO l_ 1 « 8 1 X o^ rH 1- Q) - ^ u 3 60 00 ■H Pn CD a O O O O O O O O O O O} 00 r^ CD LD "* cr> CM 51 CnI CN O CN 00 CD en •** c '■a c • c cO ^ LU •H CO CN U_ T3 CN 13 CO CN O fl _c CO T3 O 1- U cB 5 'O O S ° cO r^ CO rH CO CO to D co CO C •^ dat ulia CN adiatlon st 16 (J ■<3- n =i CN 60 CN d CN t3 0) CO O 4J CN CO CTn U H 60 00 CU cu i — •i- 1 C •H i-, CD >> •> ■ — - H CO M & 3 -H ^- c LU O 4= M co 1 l CM 3 " O I o CN o H CU <■■ ^ ^ a 3 00 •H Cr, O O o o O O o o o O o o o O o o o o o> 00 r^ CD LO "3- co CN 52 o co CM "3" CN CD 05 CM CN > TO a CD x: Q. (0 -, cO CO T3 03 CM adiation (Julian ^r U CO CM H CM c g CM CO O ■u H CM cd 5-1 4-1 toO co QJ 3 UU •u too C 3 •h <; CD ^ •» H w O) J-4 &, c 3 -H ^3- "D O X c LU W co 1 1 CN 3 • t — o rH X CN O cu u O toO 00 •H a o o o O O o O o o o o a o o o o o o CO CO r^ CD m ^r co CM r- 53 CD CM Q > m m i— a O m c CO c CI) T3 10 U 3 — ) o a CM CM CN CO CD CD c CN g si (J CO CD CD Si c CN 3 <- O X CO CD CN T3 IT) M >£J CD CN ^ ex T3 a) C 'H CO 00 O ■> cfl * r O CO ■M c cO cd T3 -H .H ti 3 O i-) •H ^ 4-> CO cn •H CN T3 CO T3 h u ■u CU H C CN T3 S-i CD CD 4-1 ^ CO e S-i CU 00 4-1 a) ex 4-1 CD a cn •H *\ !>> CO .H )-i S-i T3 3 CO O 3 1 CN CN CU M a 00 •H [X a o O o o o o o o o o O o o o o o o o 05 00 1^ CD ID ^r co CN 54 a Td m d ^d cd CN u T3 a) d ,d cd o V4 cu CN Cfi 0) cn Pi >1 cd - T3 cd ti § -d -H r-\ d d O •-) •H v-' 4-1 cd cn •H CM T3 Cd T3 u s 4-> QJ iH d cn T3 >-i CU cd e M QJ bO 4-i >-i CX CU CU 4J cn d •H - cn >> CO iH -H H iH 3 -H O -H PC O 1 cn CN cu s^ d M •H pK o O o O o o o o O o o o o o o o o o CD co r~ CD ID «* 00 CM «- 55 The net radiation traces in figures 19 to 23 show that the hourly values are in agreement. Variations in cloud conditions over individual ships cause the largest differences in the hourly values, as is the case with the incoming solar radiation. Intercomparison 2, the period of heaviest cloudiness, shows the largest fluctuations in the hourly integrated net radiation. The mean of the hourly averages over all three Intercomparison periods is well within the recommended 6-percent accuracy. 6.3.3 Instrument Malfunctions During Intercomparison 1, the Gilliss net radiometer was found to have a ripped polyethlene dome. It was removed on June 18 (Julian day 169) at 2109 GMT, and the open channel of the recording system affected the other high-gain radiation channels. The Gilliss net radiation data for the 3 days of Intercomparison 1 were deleted because of the instrument damage. The solar radiation data (Kl and K+ ) for June 19 (Julian day 170) were also deleted because of excessive noise caused by the open net radiometer channel. These problems were corrected before the start of Phase I of GATE. The Researcher net radiometer was removed on July 10 (Julian day 191) because of malfunction, and remained off line until the end of Phase I. Figure 24 shows three curves of the 4-s average radiation data for the Researcher , which indicate random drops of from 100 to 450 W/m in the energy received by the ship's sensor. Two such anomalies can be seen in the net radiation trace in figure 24. Note that a similar drop does not occur in the incoming solar trace. These net radiometer anomalies occur randomly throughout GATE, and are most likely due to a malfunction of the recording system. However, no evidence of this can be found in the simulate records. Net radiation for the Researcher has been eliminated from the archived data set for these periods. 6.3.4 Anomalous Net Radiation Data An analysis of the downward atmospheric radiation calculated from the Canadian GATE data shows fluctuations of less than +10 percent of the mean value. Since Li is derived from eq. (2) for the U.S. shipboard data, any large fluctuations in L4- can be attributed to a malfunction of the net radio- meter. The terms K4-, K+ , and Lt are well known within reasonable limits, so that anomalies in LI must be the result of anomalies in the net radiation measurement, Q*. Hourly downward atmospheric radiation data were calculated for the four U.S. ships, and net radiation data were deleted for periods when the downward atmospheric radiation exceeded the +10 percent criterion determined from the Canadian data. In most cases these anomalies are caused by the suspected malfunction of the Researcher ' s recording system. However, a few bad net radiation data were determined for the other ships, and these data were not included in the final data set. 6.3.5 Radiometer Sensitivity and Transfer Coefficients The requirements for data validation and international data exchange are defined in GATE Report 13 (de la Moriniere, 1974), which specified that surface data from ship platforms for the Intercomparison periods were to be 56 ^ c o +_* ro r-W, "O O B3 LO IX 00 +J ' — i** in co *~ J-l cd • H H O M CO CU T3 (/I cu rt 4J x: U Ph CU tH r- m R) H > 4-1 crj cd •H ^3 ""d d cd 5 n a cd -u co cu o d Ph cd i I I— < a> 1—1 0^ < On < 0^ o o^ CNJ 00 2 On — t LU •» on ON ON o OO a> ON ON < a (T< ON ON a. LU o-> ON LU ► Z o On h- > < o On ON < < Q oO on ON oo Q 2 o ON \ a LU < a> ON > 2 o x on ON ON O z> O * o ON CO LU —> en ON ON in ON •■ 00 1 • o ON ON CNJ cr ON LL < »- ON O ON CNJ o i—i (J> on ON *■ ►—i 2- On o ON ON ON o o id on & ON ON ON • o On & ON ON ON «o _l CJN on ON ON ON u_ o o- ON ON ON r- DC cr ON ON ON — < o on ON ON ON • LU o ON ON ON o h- ao o ON ON ON LL UJ CM a- r«- r~ ON •» s: co 0^ LO .—I ON CO O o r^ xt- ON • LU ri 0^ — a LO ON h- i < r*- o r— 1 <— 1 ON LL < on On C\J CM ON CNJ LL o> Qt o> CM CNJ ON •» 0m cy On LU o^ 1 1 ON I s - — » r> on x a^ ON ^ i—i X OO a> a. on m co ON o ^ 00 -H < 0* CM o ON • sO .^ -z. O CC on in on o o ON ON m 00 m o 0^ r» CO ON 4" LU ON o CM ON — H -H X O f- ON o r-^ ON i— i • CM 3: ON CO o ON •* >o O CL ON 00 ON ON «t LL i— t ON o o ON m •— » CM LU X ON }• ON CM •» h- 00 ON r- r» ON >H < —1 ON 0N ON ON o i—i O ro ON r- 1 r-» ON >— mNj-msoi^couN _ ( cMcovr oooooooooooo a LU OO LU CC a. lu OO h~ LU a. s: x a oo < • OO Q LU Z -J O CO o < LU •— ■ oo a: I < *t > LU Q O CC o o LU CC LU X OO O z. o < < X < O O I- X O Q < LU LU Z o z t-l »— OO 00 LU < Q 3 LU _j oo o »- CC X a i- LU I- 2 LU •— O LU Q OO _l LU _l Q < < Z> QmIO ZZ LU < o o 2 < lu ct: O LU LU 3 X > h- \~ < < a o i- y- \- o z z < < o oo z: t-< o i— LU a LU x z o x »--i a: I- •-> 2 I- h- < a •- < LU OO Q lu or o 2 X < Q. < a) ro 0) CD H O CJ a) •H m Li a) a) M) •H Lu i— ir\iro>TLnvO(^ooy>0'-ic\jrn>i-LnvOf > *-ooo N 0'— t oj rn **■ OoOOOOOOOi-li-ti-lf-li-l.-4i-li- 1 <—*<—• CNJ CNJ CNJ OsJ (NJ oooooooooooooooooooooooo 73 1/0 OO < < (— K < < Q Q cr« o> 0^ ON ON ON c* ON ON 0^ CJN on CJN CT> Z ON Z ON ON On 0> ON ON cr> 0> • o ON o O ON o ON ON ON a* 0^ 0^ ON 0^ T3 C* ON 0n ►— 1 O 1— 1 0> ON ON a> 0^ cc cr ct en cc ON cc cu ON o^ C* K ON 1- ON ON ON ON ON o ON o ON O ON c 3 On on en < ON < O 0> ON ON ON t- ON h- ON 1- ON H- •H c CT> ON on O ON CJN CT> 0^ ON ON ON oo ON oo ON OO ON oo O o o »— I o t— 1 CM o (NJ o o t— 1 o [— ( o 1—1 o r— 1 O o o > o > o- o ^ o cc o 5T o 5: o s: o s: O o o < o < O o o o a: o o o ct: o cc o ct o or a • • • Z • z ■ o I • o t oo • LU t LU • LU • LU N — • o o o o o _l o o oo O 2 o X o X o X o X CO 51 s: z LU 1- h- H- 1- O o Q LU to bC a: a: LU 00 • • • • rO LL LL LU oo LU \- t- r- 1- L< o o o o o a o 00 o LU o cc o 00 o 00 o 00 o oo 0) > CO • • • O • a • O0 • < • cc • 3 • z • z • z • z o it o ^. o LU > •—i o LU > o cc LU o Cl o o 3 00 00 o OO oo LU o 1— 1 o 1—1 o o 1 — 1 (J o r— 1 a] 1 o o cc cc _l o LU CC z z z z -d - o o LU LU Cl o a Cl F-< r-t 1— 1 t— < ^ _j _J Q Q Cl cc Cl DC ct ct o ct a: o o o u o o o O o > o o (- o Cl o Cl o CL • o T3 t • • z • z e Q • o • z • z • oo • oO • O0 f- < oo u .— 1 _J r^ -J — 1 o -^ o -^ r^ r-H < r-( 3 l-t .— t — 1 o CJ cu < < 1— 1 ►— 1 > > X o 2 ^ -^ 3 1- K 1- t- CC ct 00 2: o o o o r-l •— 1 *— « k-H ►— 1 CC cc _J UJ _l _J _J _J o o O0 oo LU LU _J 00 _J _i _J _l cu t— • ►— " a o a. Cl o o LU LU LU LU ■H m Q Q Q- Ou • OO oo ^. cc oo > oo > oo > oo > o 3 3 3 3 o O m m LU r-* o oo r-( oo r-* 1— < (NJ ►— t _l —J o o o o 1 > X LU LU LU in LU X Q a s: i— i Q r-l s: o 51 ro Q (NJ Q (NJ o CO o r^ CM (NJ in UN o — i CO Q. CL _) LU K O 5 ON H- o 2 ON ►— < ON t— 1 ON -J ON 00 ON < ON > 0> > ON t- ON _J on _J on oo 0> oo ON X. ON cc ON OO ON Q ON a ON 2 ON O O* o ON O ON o ON O 0^ o ON o ON o 0^ o ON o ON o ON o 0> O On o Q> O ON o ON ON 0^ r-l ON o ON O ON m ON t-i o> —1 ON CM o .— I ON o c\l ON m on a> ON -^ ON o a* O ON o 0^ o 0^ o ON o ON o & o on t— i ON »-< ON (NJ 0^ — 1 ON sf ON v* 0> LT\ ON LO QN lh ON UN ON CO ON CD ON CD cr CD ON ON ON O 0^ ON CJN 0^ ON ON ON ON c^ ON O- ON a O^ O CJN O • O a> H ON K ON (~ CJn (- ON CC ON Q£ ON ON 0> z O^ Z ON z O CC o LU ON LU 0> LU ON LU ON LU ON LU ON 0> ON CJN 3 ON h- a^ CJn O* 0* ON LU o> LU ON 0^ 0^ a 0^ a CJN C O co on Q CJN Q ON Q 0> Q cr ^ ON ^ ON ON 0^ LU a> LU CJN LU •H (JN H- * o 2T 0> Z o 2" ON Z O O LU LU O LU 4-1 0> 2" o < O < o < O* < 51 ST a. a. O Ql o ex o o o LU LU LU LU O CO CO O CO • LU • r-^ t -^ • r— 1 * ^ • z • ■z • z • z t • • N-^ O X o o o < > z z z cn CO _J _) _1 _J a a 1 ! ►—i l-H • 3 3 3 3 3 3 CL ex 3 3: 3 cd J- CD CD CD CD O M O CO o o o l 1 t— t 1— 1 51 2: O 1— 0) • z * > • > • > • >- • CO • ro • ST • 2; • O • • CO ^ O «-H o ac o CX o CC CC c O < Q a a Q _i _i _J _J CD CD 2: Cfi o _i _i _J _J 1 z 2: 2: 2: 2: »—i 1— 1 »— t ►— ( 21 2! 2: > > > >- >- > > > > > CL CC CL CL 5: ^ 51 51 CL or CC 0) UJ LU LU LU LU t • • • LU LU LU H •H <4H co > Q Q CO O CO O • CL • CL CL CL • O • O • O 3 3 3 O O O •-H C\J • CM • (M »— 1 CNJ 1— 1 CM LU —4 LU T-A O O LU p-l LU .— 1 LU ^ f-l o o o O o O u O O O O0 O CO O O co O 00 O CO O 0) _i co ^ CO ^ CO _J CO _J CO "s CO \ CO CO CO co CO V. CO \ CO \ CO T3 CO 0) LU \ V. LU LU CO co LU LU co co CO o co CO <_> O (X CL LU LU CL CL CL • 2: 2: • • LU LU ex CL LU LU LU o < < o O h- H O t~ h- \- 1 LU ex ex LU LU LU CM LU h- LU CM LU r» LU LU LU H Q 00 o O o a ST LT\ s: p- a O O CM 21 SI 2" O LT> o O o O H ^^ CM O CM O O O ■< O CM cnj o .—4 ^H O m m m LO Z LT\ in in LT\ CM Z CM z CM Z CM z CM CM CM a) CM CO i— ( CNJ 21 CO 1— co co CO 21 2: \- •H • ^H CNJ O CO s: H- O O co 0_ > > O < 00 O O < 2: \r- i- • • CD 51 < CD CO 2: PH LU i — i t— 4 o_ Q. 1 1 CO ST •* •> » h- a a 2: LU 21 LU Q LU a LU 1 • • • 51 O 2: 21 O CD 2: 2: t- h- LU LU CL CL O O _J 3 3 • • o_ a 1 — t 1— « 1 1 1 3 X X (— h- oo CO a Q 3 > 3 CD • • a D_ a Q a a O a a Q H on LU cr* LU o 3 ON ^ & 2 2 o> z ON Z 0> z o> z 0> Z CJN LU On CL CJn a cjn LU CJN LU ON t— 1 0^ ►— 1 ON •— 1 ON 1 • cr t—4 ON »— * CJN 1— » 0> 2 CJn on 0> 00 V- Q ON Q ON 3 o> ^ en 3: CJN 3 0^ S 0> 3 CJN 3: CJN O O O o o o o O 0^ ON ON O 0^ O ON CJN O a> CM CT* O ON o ON ^ o> N* ON ON 0> en ■— 1 0> — A CJN nO 0^ r- ON nO O •4- ON sl- ON o cr CJn LTv 0> LO on cr> O o> ^ 0> sfr o^ <1- On LT\ CJN LO 0^ LO CJn LTv en LO & CM c> CM 0> CM 0* CM cr CM O CM CJN CM C^ CM (M OJ CM CNJ CNJ C\J CNJ CNJ CNJ CM CM CM CM CM CM CM CM CM CM CM CM CM CM 75 •j-mmmmmmLnmLnm>Ov0>OvO^Ov0s0-O N ON0r'-r^r , ~- oooooooooooooooooooooooo cc o cr> o o» o> o o o o o a CO < 2: O a. u_ Q LU > ►—I a: LU a o LU 00 00 LU T3 RJ QJ 1 m c\j h- oo < 5T »■ 2: 1 > «■ z- 0> >— 1 O 2 Cn cr r- CT^ cr > O X cr cr O *■ cr CT 3 — . • CT CT cn s0 -5 cr cr in »— i cr cr CT •• i<^ • cr ff» cr ^-t -^ OO ^ cr cr cr . X LU CT cr cr o o 1- • en o (NJ u_ ^ < cc cr • • *MT •» h- Q ON ON m v0 -1 00 CT> cr cr r^ »■ • cr s <\| o s0 o Q cr 0> 1 -^ a> <-i LL LU CT cr cr - ** H- CT 0" a> o •• i— i o> cr a* • «■* Z CT CT CT a> cr vO vO =5 CT CT cr a^ cr LL •-< CT CT cr a- cr> »■ ♦■ CT cr cr C7^ cn vO o cr cr o o »— 1 • cr cr CT^ (7- •" vO cr cr (7> cr fH LL CT cr Cr> cr • *•> 00 s cr <^ cr vO (M .— ! o r- r- cr LL * CO cr r- ro cr *■ —* o CT LH in cr m so ^ CT ^-H CO a-' • HH r- cr i— t .—i cr r<- •» CT 0" CM (M Cr LL rH o a: cr (M N Cr <\J • (J. LU cr 1 1 Cr ► v0 cr I cr CT 5: r» ll CT Q. cr CT r» cr o >— i —^ -H < cr CM o cr • - CM O a: cr m ro cr o vO * o o cr ^ •4" cr o •— < c\| a: O cr r» r*- cr o *~0 * X Z CT cr ro v0 sO o < cr cr «- LL LU LU (^ o o a^ X >t oo o CT o o cr •& ► CO O CT o o cr —* $■ LU cr o ^ cr -H vO o t- cr o a> cr —< LL •4" ■• (_ oo CT r- r- cr —i < ^ cr cr 0^ cr o i — i O CO cr 1— 1 ^-» CT —» Hrgrn«Tinvor*oou»o>H(Njm>j- OOOOOOOOOOOOOO LU o^ •* < OO LU oo LU oo i— i »— < X _) LU h- 1— _) O ►— • LU < < CC 1- 2 < LU z 3 LU LL o < o > LU Q < IT) < Q a: CC LU LU Q o LU O X LU > LU o > < (- > _J _J LU < CL 1— 1 a: o cc LU • CC 3 > > > z LU a o V- _J < o Q X < CL t— 1 < X ■D > h- cc X 1- (- O _J < o o < X =3 •— c Q LL < LU Q o 1- *■ O < t- X < X a: Cl LU o Q 1— LU t- < LU < LU O o 3 LU Z a r» 2 X CL o or i— i LU SL z 1—1 _i t— Q O 1—4 00 < =) »■ o 00 o LU _l Q LU < i— ' X O Z o a o \- 2 < (- 3 LU o •— t- s: _l U~l an Q i— i i— » o Z •■ CC LU O i— Qt t— i OO ■D OO • 2 o < LU O 3 LU O _i LU t- _l X _J _l < y- z CD 1 OO CO 2 • LU o < LL LU < Q »— i OO s: o *— t -J -J ►— 1 Z s: LU cc < o_ ct < o o LU 1- < X SL < > LU CC < 00 LU OO a LU LU LL _i X X 3 X > or < < 1— LL <_> h- h- < Z) 1- o O < »— t U~) < »— < z LU H- 1- 1- Q o o CC < < o o LU CC _J 2 00 _l Q CD O 2 *— 1 o LU SL LL LU O LU X DC CC 3 X »— • CC' (— o LU LU Z LU 1- »— i < (— 2" LU > oo Q LU LU X t— ' O Z 5" o \~ o Q < CO Li I H O CO o u I •H m 5-1 OJ CO I I CN cu 5-1 3 M •H -HC\jro>Tir\Nor«-aou>Oi-ic>sJrn^'Lnvt)r-ao(-r'0'HCNjrn^- 000000000^^-t^H-^'-lr-|'-*rHr-li-((NJfMC\J(\JC\J oooooooooooooooooooooooo 77 • • • t t~ h- O O O O LU LU > > > > CO CO < < < < < < LU LU LU LU h- \- 1- h- 1- h- < < -) ~5 D r> • O Q a. a. o_ a_ o> a* O OS Os 2: ON ON 2: O cr> 2: OS ON 21 o> a> O z cr Z 0> O ON ON O 0^ ON O OS ON O 13 os O o> O CT> ON ON ON O OS ON 0) o O 1— t t-H a> ON ON ON ON fjs ON 3 Os On O f- 0^ 1- o> ON O ON ON O OS OS O a •H 4-1 on os O < <. o> h- ON O* H- ON ON (- OS ON 1— Os ON O Os O OS On 0> ON ON ON OS ON c o O t— 1 O t— 1 CM Q O Q O QC O Q o O > O > • • r> • LU • r> t • =) CD o O 2! O CC O 2: or O LU CO O O CO Q cc O 2T O 2! < O CO a ct O CO 1 ST CO CC CD CO CU > LL LL LU CO or o O O O CL CO O O > O l_ _J • • • O • O • CO • LU t < • LU • Q. t LU • LU • LU o O LU O LU O CC a. O CC O O CC a CC >, > > a: 2: CM 3 M ^ 1— * »— 1 LU t O t » 2: • O C£ or _l O O O _J O O O LU LU o_ LU O LU LU LU 1— 1 IU -C _) _J Q O CL CO at CO Q CO ^ CO o O O O c 1 > O 1 O O O 1 co 1 * • • • 2T • 2T • Q • 4" • O t t v* ^ _J -H _i f-t O -H C 1—1 —1 — ) ^ ^ .H —1 LU i— 1 5-i < < t— • t— ( >- LL >- LL > LL X LL O 1— H K 1- CC O a: O LU O CO O O 1— I 1—' »— 1 •— < cc a: _J t — 1 0) O O CO CO LU • LU • Cu • Z • 1—1 ►— » CL O o_ O o_ O < O CU a a a. CL t CO z CO Z • 2" LU Z t 2: > z •H co CO LU — t O O • ^ O • rH CO O O O O O O a J-l 00 1- co CO CO CO co *v CO 0L co CO co CC CO CO CO QC CO CO CO CC CO 01 Q CO LU LU CO LU LU LU ^ LU \ LU T3 Q CO LU LU cc O LU O co O CO CT3 s: 2: CC cc LU LU CC LU t— CO LU H- LU cu 1 2: 2; O O t- 1- O 1- t- CO ^- \- \- > X LU LU LU m z LU •4" z < in zr < z l > X O O 21 .^ 1— 1 O r-t t— * IS r^ t-H ^ CM y— 1 cm CM in LO O ^ c O >t rH O O CM -^ CM C\J CM f-H ^ <}■ rH H ~4 -^ — H % O X> a. X LU X _) Z! _) Z 13 1— co CO X CO i—t 1—1 H- »-^ ^ •• CO » CO •• LU LLI •— 1 O Q. CO CL CO • CO • CO K as ST en K os z cr> 1—1 ON 1- ON 1— 1 ON h- ON ON »- 0^ OS t- ON < on ►— 4 < O os X z X ON z ON 2T ON 2 1 ON z OS Z OS Q »- o> —I _J CO o> CO ON ON 1— 1 O O ON t— 1 OS O OS O o> O os O O o> ON ON O ON ON O ON O ON OS O OS O O o> O O 0* 0> <—i ON ON CM ON O ON CM OS O ON r-w On O on CM CO ON os ON On >cf ON ON 0^ CO ON OS OS CO OS ON ON O O On 1—1 ^ a> CM O OS ON -^ ON On ON >£ ON ON ON vO OS OS ON > > > > > < < < < < < LU LU LU LU LU LU K t- t- h- »- »- 3 3 -> r> 3 3 Q. Q. a. Q. a. CL cr a« s CT CT s: CT CT s: CT CT T. CT CT 2" CT CT s: ^^ o> cr o CT cr o O CT o CT CT O CT CT O CT CT o X) 0> cr a CT CT o cr CT o CT CT O CT CT O CT CT o 0) cr cr CT CT CT CT CT CT CT a: CT 0> CC CT 3 CT CC 0> o O CT> o 0> CT o CT CT o CT o o O (^ O CT o •H 4-J CT LU CT t- CT CT 1— CT CT 1— CT CT f- CT »— CT 1— CT H- CT 1- CT H- cr CT CT» CT CT CT CT CT co CT CT CO CT C O LU o Q o o a (M o a o o Q O i — » O Q O i — » O Q o o s: o LU o o LU ■* o UJ o cc o LU o s: o LU O S O LU CJ o o o CO o o co O cc o CO o o o oo o a: O CO o a: O OO ^^ • z t 3 * • 3 • O • D • oo • 3 • LU • 3 • LU • 3 Cfi o < o o o o oo o o z o o X o O X o dJ a: CO CO Z CO LU O0 H- CO h- O0 OC > a a LU o CO O Q a c^ Q- ar r* or co QC CC * CC • OCT O LPl a O LU o t- O h- a o 00 o o o o a o LU o o o CC o o o 00 o c CO o o • *t * LU • _l • LU • CC • LU * Z) • LU • z • LU • z: • LU O | o CC o LU o a£ o 3 o C^ o oo o CC o 1— 1 O a i — i o a: >-> oo a CO to 3 • o • CO • LU • CD • CD • _J O s: u LU o a: o z o Z o O LU LU LU CL LU a. LU t— 4 LU >— i LU ,£ Q CO o CO (X OO oo C£ CO CC CO O O o I o h- o | o o 1 o i— o 1 o o. o 1 o a. o 1 Cfl • s: • «*■ • U • .* • Z • •J- t z • nC • CO • -d- • CO • •4- — i -4 —t LU — 1 — 1 < r-l —* 3 —4 f— « -4 i-H —t L| > Li- h- LL s: LL O LL 3 LL 3T LL o LU ci co O CO O s: O o O a O O _l co _) LU _l _! CD a. • t— t * _J t CO » _J • _J • Q. o 2 o O O o o LU O LU o <0J • LU z * CO z ^ 2 CC 2: 00 > Z CO > z i-H 2: •z r> 3 • —1 o • r-4 o CO r-4 O oo <-4 o ►— < CM o *-* rg o ct o o CT O o cc o o oc o o o O o O O o L| CO CO cc ao CO 00 CC 00 < CO cc CO < CO QC 00 _l 00 cc CO -) oo cc 00 a) V, LU *v LU CD LU co LU LU LU LU LU t3 OO o CO o c— ) o *— « CD O CD o CD CO a) t- o LU 1- LU _l LU _l LU • LU • LU y- o S- 1- >4- 1- _J 1— _) h- CD h- CD H- < LO Z < s0 Z 1— 1 CM z t— 1 r>- z LU CT 2 LU Z l 3 CM HH ^ a- t— 3 ST O 1— » s: m t— ' a CM 1— 1 o CM H^ • O >-< o o -c O O .— » o o — i o o i— i o o LO o CN t— { — 1 r-^ <~4 O r-H o -H o — 1 o r— 1 < o z CT o z <* 2 CT O LU Q LU O- O CC O CC 3 _j cc CO cc < z • ^ Q. cc 3 CO 3 Q • »• or »» to •■ « *■ 00 »> CO •i _J co CO _J CO LU CO oo oo LL CT t- CT (- cr K CT _l CT t- CT oo CT h- o < CT h- CT > CT ^- c^ LU CT Z CT LU a> 2 a> O CT- Z CT O CT z CT LU CT Z CT CC CT z CT a: o> O cr z a- O CT ^ CT o CT cc CT o CT oo CT O CT Q o o CT o cr O 0> o a* o CT o CT o CT o CT o CT o CT o CT O CT o CT ro cr o CT< f. cr o 0> o CT- o CT o CT o CT ro CT o CT -t CT o CT co cr CT. CT CO cr CT- CT o CT CT CT o CT CT CT O CT CT CT O CT CT CT O CT CT CT O CT CT o> >•* CT CT a> <4r CT CT CT LO CT CT CT LO CT CT CT CM CM CM k o> oo O HI o 2: O CC • LU O X < LU H- Q. ON S (JX On O on ON O ON ON ON O 0> (- o w # 2 o 00 3 LU O • o LU o o o in H-. o or o a. o • CO • a LU OO ID 00 Q a: o o LU a: cm o oo LU CO _j 3 CD >■ O or o> Q On o o^ ^ O 1 O ON in on LU ON (NJ 00 C\J CQ _l 3 CQ > or Q 00 I— z <_> o o ON 0^ fM cm cm t O > < ON ON o> or on o ON 1- o> OO on HH O 5: o □T o LU • X o 00 O z • H-. O 3 a 2: on O On O On ON O ON I- ON 0> Q On LU ON 00 o 00 Q CC O o LU OT ON CT" ON a: o> O o^ i- a> 00 On h-h On ST ON OT O LU • x o > < LU \- 3 ST ON o o^ O ON ON O ON h- ON a* a o LU o 00 o o or oao • 00 • o LU 00 I 00 O Z • H-< O o CL a. 00 O O LU O O0 LU z 00 > z 3 > 3 w O h-i (\J OON OOO O — I O cD-icDoraoiiiooa: o LU o o o in LU LU a LU !""- H-. in o r-t Z On 00 ^H co o LU o o o LT> CO _l 3 CO ON t- ON LU ON 3 ON O ON CM ON O on in CM CM 3 co LU LU CO 00 co _J 3 CO On ON I- ON LU On IS ON 0> O On z & < ON o CM CO —I CM co co 3 3 3 CO On CO On CO & On (— 0> K On \— On LU 0^ LU O^ ON ON o o o o 00 Q OT o o LU cc 00 I o CO _J 3 co o > • cc O Q O cC u. o • cc O LU z a O • CM t OOOO 00 ^ CO ^ O LU GO •— 1 in o Z & CM OO CM 00 2: < a: o CM in 00 2: < cc o o CM in CM co _j 3 CO 3 3 co ► 00 OI-(M- On LU On z 0^00^0 0> CM On O CMMJ^ 00^0 on o> on m on o> CM CM CM CM CM CM 3 X • U On LU 0> O. ON O0 o> o On O ON V}- on in CM CM 3 X • o On LU ON Q_ ON 00 On O ON O ON Nt on m CM ON On Q o> z On < O O O CO CO CO 3 3 3 CO CO CO o o > • > • > cc O CC O CC OOO O O CC CC o u_ o u. — H Q — t O ^ a: LU o CM O CO or LU Q CM O 00 0> O^ ON CC ON LU 0> I- ON LU o> 2: o o o 5- O LU > < LU h- 3 a on 2: on o On O ON o on 1- ON o o o O < O o. 3 o O I • CO o Q LU 00 3 00 o or o o LU or Q ^ Q LU LU > > OOO t z • "H 3 ^ o 0; LU o CM o 00 LU 00 I 4- O O O 00 or 00 00 o LU O • o LU o o o in LU o LU o o o in CM LU OO or LU o LU LU CM Z 2: in >-. o -* o CM Z On 2: 00 o 2: o CO I o LU LU a. 00 a o CO I 00 O 00 ^S^JOZO^K On IMJJtMJJ^HiO^Z^ OnQOnQO 1 3 o> o o* OnOOnoOnoOnOOn ONvtONv4"ONONO>OON onoonoo^loo^onon ONinoNinoNcMONooN CMCMCMCMCMCMCMCMCM a) d •H a o o Cfi a) u a) > i—l M P O en Td o CJ CD H ■H MH M QJ -d cfl a) W I 1 U bC •H 80 r^^^p^^-p«»r^ooaoooooaocoooo<^o> oooooooooooooooooooooooo * < LU K -) a a* (T> LU 0> H- cr> lu cr 2: o a o 2: O LU O o o> a> o cr> h- o o o LU o o < o a. o O I • m 00 00 Q ex O O LU a: 000 • z • a > • CL O LU -^ 00 CO (X o LU OO I u_ O O LU O • Z t o < o > o cl o o —> o o 000 • z • ^ D ^ o > > < LU -5 Q. 2: 0> o cr o o 0> (- o> 0> Q O LU O 00 o 3 o LU OO I < Q> cr o o LU O • Z t o < o > o CL o O •-. o 2! o o Q LU 00 z> 00 Q or o o LU O OOO • z • -* ID 1— 1 O > o o LU cl •—1 Q Q 0> Z 0> < CT> Q O LU O LU O Q- O to O o o o o 00 co co a: a- LU LU o 0> LU C* LU O CL O co O Q O* Z O < O Q O lu CT* LU O CL O CO O •SLQSL O • O OOO CO CO O I- o • CO • o < o o CL a. o o o CL CL U.OU.O O ^ O LU LU > > CL z • o CL LU Q a: O O UJ O CO CO ^ CO CL LU 5T o m o o CO o LU 00 \ co CL LU I- LU 2! o m C\l 5: o o CO o LU CO ^ co CL LU h- LU 2: o r<~> r\J I- co < OOOQOQOQ CO < O CL LL O LU > CL • O • Q O LU 00 CO ct: LU 21 o (M t- co < O CO 21 SL SL ST OOOO o o o u I I I I r> > 3 •> T3 3 C •H ■U d o o en o CO o u B ■H a; a) rj u bO •H Q Z 0> t— A 0^ 2 0^ O 0> cr> 0^ CO o> z 0^ o 0> o o> o CO & z 0> o ON O C7> O o o z CT^ •— 1 O O 0> ^o Q Q 0> Z C7> Z 0> CT> >—* 0" 1 •— * CT> 0> 3 0> 2 0> 0^0 0^0 0^ CT^ v0 O h- & W ' V • -»»^ V ' ' ' W • ^^ W ' ' 1 W %-^ V' "** W 1 I W ' ^*rf \J I — W LOC^^C^OC^C^O > OC^^CT >, >J"0^^'On^'0 > »4"C7 n CNJCNJCNJ(\i(N)(N|f^rvJt\JC\JC\l(NjC\JC^rgCNJ1 11.50 llTsO 15.50 TRUE S0LRR TIME 17.50 19.51 STflNORRO ID 10036 Figure B.2. — Response of an Eppley 8-48 pyranometer. 85 RRTIO OF SENSOR TO STRNDPRQ 10C36 . m.L UtiOtTtfNt (TwQ MINUTE ftvERPDESl LO LA o ^.50 21 APRIL 1974 7. SO 9. SO 11. SO 19. SO IS. SO TRUE SOLAR TIME 17. SO 19.50 STPNDRRO ID 10038 Figure B.3. — Response of a Yanishevsky pyranometer 86 APPENDIX C Translocation of Thermistor Probes and Bridges The dry-and wet- bulb temperature sensors on the Gilliss , and the sea- surface temperature sensors on the Researcher , Gilliss , and D alla s were not calibrated with the bridges that they were paired with for the pre-GATE calibrations. To establish the correct calibration for the matched thermistor and bridge pairs, it was necessary to subtract the calibration of the partic- ular bridge that was used in calibrating the sensor and build in the calibra- tion of the bridge that was used with the sensor in the field. The following relationship was established relating the bath temperatures (T. . ) to the output voltage (V , ) of the bridge used in the calibration: bath cal T bath ( ° C) " b 2B • V cal + b lB' «•« where b = the slope, b = the intercept, lo The separate bridge calibrations established the relationship between the voltage, V , and the resistance, R, for each of the bridges. Thus, for any given bridge used in the water bath calibration, R " a 2B • V cal + a lB' (C - 2) where a = the slope, and zB a = the intercept. IB For the bridge that was mated with a particular thermistor in the field, a similar resistance-to-voltage relationship was R " a 2F • V field + a lF' (C - 3) where a„ = the slope, and Zr a = the intercept. lr In reality, these relationships were slightly nonlinear. However, for the small resistance ranges associated with the nominal temperatures they represented and because the same two resistance values were used in all translocation operations, the error caused by approximating the nonlinear curve by a linear curve turned out to be negligible. Solving eq. (C.2) for V ., cal V , = R ~ a iB , cal a 2B replacing R with eq. (C.3), v (a 2F ' V field + a iF ) - a iB cal " a 2B S 2F . V-. nj L U 1F " a iB } fxeld + a 2B a 2B and substituting in eq. (C.l), yields T bath ( ° C) " b 2B • af • 'field + b 2B • " a 2B " + b lB - or where and T bath ( ° C) =C 2 ' V field +C .' (C ' 4) c b 2B • a 2F , 2 a 2B b 2B C l = 1^ U 1F " V + b lB ' Equation (C.4) was then used to derive voltage-to-temperature relation- ships for all thermistors that were mated during the GATE field operations , but were not mated during the calibrations. APPENDIX D Sensor and SAM Transfer Equations Used in Data Processing Table D.l. — SAM transfer equations Variable Channel No. Counts to volts Slope xl(T 4 Intercept xlO J Researcher Bow SAM Sea-surface temperature Dry-bulb temperature Wet-bulb temperature #1 Wet-bulb temperature #2 Boom wind direction Boom wind speed 5 7 8 9 12 13 3.0523 3.0597 3.0545 3.0562 3.0565 3.0583 -2.0094 -2.5752 -2.3418 -2.3686 -3.6933 -3.1092 Researcher Central SAM Mast wind speed Pressure (Rosemount 2.9468 2.9467 -0.8104 0.3438 Sea-surface temperature Dry-bulb temperature Wet-bulb temperature #1 Wet-bulb temperature //2 Wind direction Wind speed Gil liss Bow SAM 5 9 12 13 3.0600 3.0522 3.0596 3.0529 3.0550 3.0570 -12.4694 - 0.9919 -12.1620 0.6869 - 5.8808 - 8.4060 Gilliss Mast SAM Mast wind speed Pressure (Rosemount) 2.9639 2.9626 6.7676 6.0980 Dallas Bow SAM Sea-surface temperature Dry-bulb temperature Wet-bulb temperature #1 Wet-bulb temperature //2 Boom wind direction Boom wind speed 10 7 8 9 12 13 3.0539 3.0536 3.0538 3.0555 3.0527 3.0533 -1.9087 -1.9594 -1.6542 -3.5139 -0.1526 -1.3994 Table D.l. — SAM transfer equations (continued) 89 Variable Channel No. Counts to volts Slope xlO -4 Intercept xlO -3 Dallas Mast SAM Mast wind speed Pressure (Rosemount) Sea-surface temperature Dry-bulb temperature Wet-bulb temperature //l Wet-bulb temperature #2 Boom wind direction Boom wind speed 12 13 Oceanographer Bow SAM 5 7 8 9 12 13 Oceanographer Mast SAM 2,9442 2.9441 5.2259 5.5692 3 .0566 -2 6949 3 .0560 -2 5212 3 .0592 -6 2459 3 .0580 -5 8612 3 .0560 -1 9864 3 .0540 -1 .3234 Mast wind speed Pressure (Rosemount) 2.9675 2.9712 -9.8422 -10.1268 90 Table D.2. — Sensor transfer equations for the Researcher Variable Sensor Vo Its to Units serial scientif ic units . No. Slope Intercept Bow Boom Sensors 126 6.021 14.99 °C 51 5.988 15.09 °c 55 5.996 15.12 °c 56 6.006 15.09 °c 201 -68.175 177.21 deg. 151 9.782 -0.05 m/s 114 0.04036 0.21 deg. 115 0.OQ157 -0.02 m/s Mast Sensors 226 -0.02125 +181.77 deg. 176 9.721 0.01 m/s 136 5.8637 1001.02* mb Sea-surface temperature Dry-bulb temperature Wet-bulb temperature #1 Wet-bulb temperature #2 Boom wind direction Boom wind speed Ship heading 1 " 1 Ship speed Mast wind direction" Mast wind speed Pressure (Rosemount) "These sensors were calibrated through the SAM, hence they convert recording counts to scientific units. : Does not include +1.44 mb correction for sensor height above sea level. This correction was applied, however, to the pressure in the archived data. 91 Table D.3. — Sensor transfer equations for the Gllliss Variable Sensor Volts to Units serial scientif ic units No. Slope Intercept Bow Boom Sensors Sea-surface temperature 127 5.981 15.01 °C Dry-bulb temperature 53 5.994 15.03 °c Wet-bulb temperature #1 59 6.013 15.02 °c Wet-bulb temperature #2 60 5.998 14.82 °c Boom wind direction 203 -69.301 172.58 deg. Boom wind speed 153 9.379 0.06 m/s Ship headingl 214 -0.02645 +194.50 deg. Ship speed-'- 96 -0.00041 -1.41 m/s Mast Sensors 2 1 Mast wind direction ' 228 -0.0261 169.2 deg. 3,1 228 -0.0214 169.2 deg. 4,1 228 -0.0209 173.2 deg. Mast wind speed 178 9.356 -0.04 m/s Pressure (Rosemount) 201 5.636 1002.15* mb "These sensors were calibrated through the SAM; hence the conversion is from recording counts to scientific units. 'Transfer equation for IC 1. Transfer equation for Phases I and II. Transfer equation for IC 3 and Phase III. Does not include +0.94-mb correction for sensor height above sea level. This correction was applied, however, to the pressure in the archived data. 92 Table D.4. — Sensor transfer equations for the Dallas Variable Sensor serial No. Volts to scientific 'units Slope Intercepts Units Bow Boom Sensors Sea-surface temperature Dry-bulb temperature Wet-bulb temperature #1 Wet-bulb temperature #2 Boom wind direction Boom wind speed^ M M it 5 Ship heading Ship speed"' ' ii it 6,8 134(7) 134(4) Unknown 54 61 62 204 154 182 314 315 315 063 054 962 995 958 988 -71.188 9.902 9.319 0.02313 0.00135 0.00134 15.00 14.91 15.15 15.06 15.08 15.00 179.00 -0.01 0.03 -0.85 -3.34 -8.71 C deg. m/s m/s deg. m/s m/s Mast Sensors Mast wind direction" Mast wind speed Pressure (Rosemount) 229 179 205 0.02230 9.917 5.7592 -169.3 -0.07 1000.96' deg, m/s mb Sensor 134 was used on bridge 7 from June 17, 0900 GMT, through September 9, 1930 GMT (Julian days 168-252). 2 Sensor 134 was used on bridge 4 from September 9, 2130 GMT, through September 10, 0830 GMT (Julian days 252-253). 3 A sensor of unknown serial number was installed with bridge 7 and used from September 10 (Julian day 253), 1400 GMT, through the end of GATE. 4 Wind speed sensor 154 was used during ICI and from August 31 (Julian day 243) 1950 GMT through the end of GATE. Wind speed sensor 182 was used from June 27 through August 31, 1941 GMT (Julian days 178-243). These sensors were calibrated through the SAM; hence conversion is from recording counts to scientific units. This transfer equation applies from June 17, 0900 GMT, through July 28, 0000 GMT (Julian days 168-209). Q This transfer equation applies from July 29 (Julian day 210) through the end of GATE. Does not include a 1.43-mb correction for sensor height above sea level. This correction was applied, however, to the pressure in the archived data. Table D.5. — Sensor transfer equations for the Oceanographer 93 Variable Sensor serial No. Volts to scientific units Slope Intercepts Units Bow Boom Sensors Sea-surface temperature Dry-bulb temperature Wet-bulb temperature #1 Wet-bulb temperature #2 Boom wind direction Boom wind speed Ship heading^ Ship speed Mast wind direction Mast wind speed Pressure (Rosemount) 129 6.005 14.94 °C 52 5.985 15.02 °c 57 5.978 15.05 u c 58 5.984 15.14 u c 202 -69.733 178.30 deg. 152 9.874 -0.07 m/s 414 0.0264 3.30 deg. 415 0.00152 -0.06 m/s Mast Sensors 227 0.02126 177.00 deg 177 10.241 -0.06 m/s 137 5.705 1001.22* mb GMT (during Phase II) , the Oceanog rapher converted wet-bulb #2 to dry-bulb #2 These sensors were calibrated through the SAM; hence conversion is from recording counts to scientific units. 2 *Does not include a 1.41-mb correction for sensor height above sea level. This correction was applied, however, to the pressure in the archived data, 94 CO c o ■H ■u CO cr 0) U CU <4-l CO c crj O CO c u en cu H CO 'X 0) n a e en o i i H cu CJ 5-4 0) 5-i 01 H o pq o e 5-i T3 CU S3 4-1 O a u H cu X) 5-4 O 5-i •H J3 H O 55 CTJ •H M 0) CO c^ 00 n CO CN m CN o^ ON CN o I in -t- o CM n r-^ co o> CN co m On n O l ON vO «tf o 1 00 1 CN 1 ^0 o in o CN O r^ o CN rH m i— I a> i—i ^D rH H o i-H rH ^D X 00 X in X CO X rC a 5-1 tr- •H 4= CU 4= ee; 5-i CO CJ CO 00 5-i CO CO o CO •H CO c 0) i-H rH CO CO rH rH CU CU •H CO CJ Pi O a o I co M-1 O 0J CO O 4-1 c •H 4J U 01 u U o cj co 0J X) rH CJ c i co '4-4 o S3 o •H 4-J a 01 5-i U o a CO cu X) rH CJ C 43 e CO '4-1 O CJ CU 5-i S-i o o CO 0) T3 rH a S3 •i CN O S3 o •H 4-1 a 0) H H O CJ co cu rH CJ S3 ft U. S. GOVERNMENT PRINTING OFFICE : 1977--240-848/113 (Continued from inside front cover) EDS 16 NGSDC 1 - Data Description and Quality Assessment of Ionospheric Electron Density Profiles for ARPA Modeling Project. Raymond 0. Conkright, in press, 1976. EDS 17 GATE Convection Subprogram Data Center: Analysis of Ship Surface Meteorological Data Obtained During GATE Intercomparison Periods. Fredric A. Godshall, Ward R. Seguin, and Paul Sabol, October 1976. (PB-263-000) EDS 18 GATE Convection Subprogram Data Center: Shipboard Precipitation Data. Ward R. Seguin and Paul Sabol, November 1976. (PB-263-820) EDS 19 Separation of Mixed Data Sets into Homogenous Sets. Harold Crutcher and Raymond L. Joiner, February 1977. EDS 20 GATE Convection Subprogram Data Center--Analysis of Rawinsonde Intercomparison Data. Robert Reeves, Scott Williams, Eugene Rasmusson, Donald Acheson, Thomas Carpenter, and James Rasmussen, November 1976. EDS 21 GATE Convection Subprogram Data Center: Comparison of Ship-Surface, Rawinsonde and Tethered Sonde Wind Measurements. Chester F. Ropelewski and Robert W. Reeves, April 1977. PENN STATE UNIVERSITY LIBRARIES AQDDD7ED173Q2 NOAA SCIENTIFIC AND TECHNICAL PUBLICATIONS NOAA, the National Oceanic and Atmospheric Administration, was established as part of the Depart- ment of Commerce on October 3, 1970. The mission responsibilities of NOAA are to monitor and predict the state of the solid Earth, the oceans and their living resources, the atmosphere, and the space environment of the Earth, and to assess the socioeconomic impact of natural and technological changes in the environment. The six Major Line Components of NOAA regularly produce various types of scientific and technical information in the following kinds of publications. PROFESSIONAL PAPERS — Important definitive research results, major techniques, and special in- vestigations. TECHNICAL REPORTS— Journal quality with ex- tensive details, mathematical developments, or data listings. TECHNICAL MEMORANDUMS — Reports of preliminary, partial, or negative research or tech- nology results, interim instructions, and the like. CONTRACT AND GRANT REPORTS— Reports prepared by contractors or gTantees under NOAA sponsorship. TECHNICAL SERVICE PUBLICATIONS— These are publications containing data, observations, in- structions, etc. A partial listing: Data serials; Pre- diction and outlook periodicals; Technical manuals, training papers, planning reports, and information serials; and Miscellaneous technical publications. ATLAS — Analysed data generally presented in the form of maps showing distribution of rainfall, chem- ical and physical conditions of oceans and atmos- phere, distribution of fishes and marine mammals, ionospheric conditions, etc. ftTMOSp^ ^^FNT Of & Information on availability of NOAA publications can be obtained from: ENVIRONMENTAL SCIENCE INFORMATION CENTER ENVIRONMENTAL DATA SERVICE NATIONAL OCEANIC AND ATMOSPHERIC ADMINISTRATION U.S. DEPARTMENT OF COMMERCE 3300 Whitehaven Street, N.W. Washington, D.C. 20235