NOAA Technical Report ERL 396-ARL 5 c X f <°L% The Ma una Loa » Four Wavelength Nephelometer: P Instrument Details and Three Years >^.ii ^. I .V* St ATCt 0< * of Observations Barry A. Bodhaine April 1978 U.S. DEPARTMENT OF COMMERCE National Oceanic and Atmospheric Administration Environmental Research Laboratories NOAA Technical Report ERL 396-ARL 5 afiSSE The Mauna Loa Four Wavelength Nephelometer: Instrument Details JP and Three Years of Observations a. o Barry A. Bodhame Air Resources Laboratories/GMCC Boulder, Colorado April 1978 Q U.S. DEPARTMENT OF COMMERCE Juanita M Kreps, Secretary National Oceanic and Atmospheric Administration Richard A Frank, Administrator Environmental Research Laboratories Wilmot Hess, Director NOTICE Mention of a commercial company or product does not consti- tute an endorsement by NOAA Environmental Research Labora- tories. Use for publicity or advertising purposes of information from this publication concerning proprietary products or the tests of such products is not authorized. 11 CONTENTS Abstract 1 1. Introduction 1 2. The Instrument and Its Use 3 2.1 Theory of Calibration 4 2.2 Checking Linearity 5 2.3 Resetting Sample Time 5 3. Nephelometer Data 6 3.1 Data in Appendix A 7 3 . 2 Data in Appendix B 8 3.3 Data in Appendix C 9 4. Summary 9 5. Acknowledgments 9 6. References 10 Appendix A 11 Appendix B 31 Appendix C 39 in THE MAUNA LOA FOUR WAVELENGTH NEPHELOMETER : INSTRUMENT DETAILS AND THREE YEARS OF OBSERVATIONS Barry A. Bodhaine Abstract . Three years of high quality total light scattering data from a four-wavelength integrating nephelom- eter have been acquired at Mauna Loa Observatory as part of the Geophysical Monitoring for Climate Change (GMCC) pro- gram. These data may be representative of the upper tropo- sphere under certain weather conditions and may therefore be used to deduce background global aerosol properties and their possible effect on climate. A light scattering value of about b (550 nm) = 10~ 7 m _1 appears to be a good esti- mate of theT.ight scattering of aerosols during the cleanest background conditions at Mauna Loa Observatory. 1. INTRODUCTION This report presents total light scattering data taken with a four wavelength nephelometer at Mauna Loa Observatory from January 1974 to December 1976. A brief description of the principles and operation of the instrument is included to aid the potential user in data interpretation. The user may wish to use this report for a first look at the data and to iden- tify specific events to be examined more closely. A complete set of hourly means of four wavelengths of light scattering data, along with meteorological data, may be obtained from NOAA-ARL-GMCC, Boulder, Colo- rado 80302. The nephelometer approach to the measurement of the integrated light scattering of air was first attemped by Beuttell and Brewer (1949) and later refined by Crosby and Koerber (1963) who used a photo- multiplier tube as the sensing element. Mechanical and electrical design were improved by Ahlquist and Charlson (1967), and the sensitivity of the instrument was improved to the point where the Rayleigh or molecular scattering coefficients of various gases could be measured. In comparisons between the light scat- tering coefficient of ambient air and total suspended particulate mass made by Charlson et al . (1967), the simple linear relationship M = 0.38 b . , where M is mass in g m~ 3 and scat' 3 b . is in m -1 , was found to hold scat true approximately for a variety of natural and anthropogenic aerosol sources. The standard integrating broadband nephelometer manufactured by Meteorology Research Inc. (Charlson et al . , 1969) has gained wide use for the measurement of visual range and air quality, and has been increased in sensitivity by using photon counting techniques in place of the old flashlamp-analog signal technique. A simple relationship between the aerosol size distribution and the Angstrom exponent for aerosol scatter has been developed and is summarized by Butcher and Charlson (1972). Whitby (1975) has shown that particle size distributions are probably best described by a multi- modal log normal curve fit. How- ever, the Junge (1963) power law model for the size distribution of aerosols , dN d log r = Cr Because light scattering is responsible for many chromatic atmospheric effects (blue sky, colorful sunsets, etc.), the wave- length dependence of light scatter- ing by air molecules and aerosol particles has received much atten- tion since the classical works of Rayleigh (1871), Mie (1908), and Angstrom (1929). Angstrom concluded that the wavelength dependence of aerosol scattering could be de- scribed by b , «r in a manner J scat similar to that used for Rayleigh scatter by gases, where b . is the SCd L extinction coefficient due to scat- ter, A is the wavelength, and a is the Angstrom exponent. Typical values of a for aerosol scatter are about 1.5, and pure Rayleigh or molecular scatter gives a = 4. where dN is the number of particles in the radius increment d log r, p is the slope of the size distribu- tion, and C is a constant, is a good approximation over a limited size range. The simple relationship a = p - 2 makes it possible to ob- tain aerosol size distribution in- formation by measuring the Angstrom exponent, i.e., by measuring light scattering at several different wavelengths. An instrument to measure light scattering as a func- tion of wavelength was developed by Ahlquist and Charlson (1969) and used extensively in field experi- ments (Charlson, 1972) to measure the Los Angeles smog aerosol. 2. THE INTRUMENT AND ITS USE The Mauna Loa four-wavelength nephelometer is based on the origi- nal Ahlquist and Charlson (1969) design and was manufactured by Meteorology Research, Inc. Pre- liminary data were reported by Bodhaine and Mendonca (1974). An operational description and calibra- tion procedure is given in detail in the manual supplied with the instru- ment. Briefly, the nephelometer measures the scattering coefficient of atmospheric gases and particles, and the optical geometry of the instrument provides integration of the scattering over all angles from about 7° to 170°. Scattered light is detected by photon counting tech- niques, and a rotating filter wheel just in front of the photomultiplier tube allows the measurement of scattering coefficients in four wavelength bands centered at 450, 550, 700, and 850 nm . Digital counting and multiplexing circuitry along with digital-to-analog conver- sion and logarithmic amplification provide an electrical output pro- portional to the logarithm of the scattering coefficient at each wavelength. Absolute calibration is provided by the known Rayleigh scattering coefficients of certain gases such as air, CO L , , and Freon. In order to separate the aero- sol scattering from the molecular scattering of air and the background of the instrument, two modes are provided. In the NORMAL mode, the counters count up when a filter is passing light and count down for an equal period when a dark disk is in front of the photomultiplier tube. In addition, an electrical signal simulating molecular scattering of air is subtracted from the output signal. In the AIR CHOP mode, the counters count up on ambient air when a filter is passing light. The instrument is then filled with filtered air and the counters count down for an equal period of time, thus subtracting molecular scatter and instrument background as well. Operation in the NORMAL mode gives better time response and may be used under conditions of fairly high aerosol scatter such as those in urban areas. However, operation in the AIR CHOP mode is necessary in a clean or background location such as Mauna Loa and is capable of measur- ing aerosol scatter as low as 1% of the molecular scattering of air. Waggoner et al . (1976) have dis- cussed the theory and measurement of the noise level of a four-wavelength nephelometer and find a value of about 5 x 10 ~ s for a 45-minute averaging time. Furthermore, it has been found that short term fluctua- tions in photomultiplier dark count are primarily responsible for the instrument noise level . 2.1 Theory of Calibration The nephelometer essentially measures the photon count rate present at the photocathode of a photomultiplier tube, because of the light scattering cross section in the sample volume of the instru- ment. Mie scattering theory pre- dicts that count rate is a linear function of the scattering cross section, or R = C b , where R is o sp count rate, b is light scatter sp coefficient, and C is a constant. The nephelometer measures the count rate, takes the logarithm, and presents an output in volts. The output element is a buffer amplifier which incorporates a span and bias adjustment so that V = C 1 log R + C 2 , where V is output voltage, C x is the span, and C 2 is the bias level. We have R = C b o sp and, (2) a bias adjustment, which together correspond to the point- slope determination of a straight line. In practice, the slope is set by the volts-per-decade adjustment on the output buffer by presetting the reversible counters with known counts and then adjusting the output to exactly 2 volts per decade. The instrument is then filled with Freon-12, and the bias adjustments (called gain controls on the nephe- lometer) are set so that Freon scattering coefficients are as shown in Table 1. TABLE 1. FREON SCATTERING COEFFICIENTS FOR DIFFERENT NEPHELOMETER WAVELENGTHS Wavelength Light Scattering Voltage 450 nm 2 8 x lO'-'m- 1 6 8943 550 nm 1 3 x lO" 4 !!!- 1 6 2279 700 nm 4 7 x 10- 5 m-' 5 3442 850 nm 2 2 x 10~ 5 m _1 4 6848 or log R = log b + log C 3 3 sp 3 o and V = Cj log b + (C x log C + C 2 ) or log b where C- sp V + C, 1/Ci depends on the span adjustment, and C 4 = -(log C + Cg/C^ ) depends on instrument geometry, span, and bias so that once the span is set it depends only on bias adjustment. Therefore, the relationship between the logarithm of b and the output voltage may be set by two simple adjustments: (1) a volts-per-decade span adjustment These settings define the simple relationship between light scatter- ing coefficient and output voltage: sp (--7) 10 l 2 ; or log b = % - 1 3 sp 2 After the four wavelengths have been calibrated using Freon-12 gas, the instrument is filled with clean air and a calibration object (thin wire with white tip) is introduced into the sample volume. This object is "white" with high accuracy over the entire visible spectrum and provides a highly stable calibration object for the nephelometer . The light scattering coefficient of this object is "measured" and all chan- nels are trimmed to the 550 nm channel value. This trims the Freon-12 calibration and corrects for the transmission characteristics of the thin film filters used in the instrument. 2.2 Checking the Linearity of the Instrument The instrument is operated over a span of 5 decades, the calibration point is in the upper decade, and most of our data are in the lower 3 decades. It is therefore desirable to check the linearity of the D to A converter, log amplifier, and other output circuitry. The instru- ment calibration for an output of to 10 volts over a 5-decade range is defined by b en = 10*2 '' sp Therefore, the linearity may be checked by presetting the counters to known counts and measuring the output voltage. The instrument is eguipped with the necessary controls so that a linearity check may easily be accomplished over its entire 5- decade range. 2.3 Resetting the Sample Time of the Instrument After the nephelometer is cal- ibrated, it has a 0-10 volt output at 2 volts per decade with a lower limit of b = 10- 7 m- 1 . Since this tap is not sufficiently sensitive for monitoring at Mauna Loa, the sample time of the instrument is increased by a factor of 4, which effectively increases the instrument sensitivity by a factor of 4. The new relation- ship between light scattering and output voltage is V b en = 0.25 x 10 ( 2" 7) sp Because of the logarithmic output, the scale is simply shifted so that the lower limit is b = 2.5 x 10 -8 sp m ', and the span of 5 decades, at 2 volts per decade, is retained. For purposes of data reduction, therefore, the following scale is used: 10 2.5X10" 8 2.5xl0- 7 2.5xl0~ 6 2.5xl0~ 5 2. 5x10-' 2.5x10^ Volts ' ,-i sp NEPHELOMETER DATA sp The Mauna Loa four-wavelength nephelometer has operated continu- ously since January 24, 1974. Reliability has been very good, and only about 5% of the data were lost during the first 3 years. Some of the downtime was caused by the multipoint recorder rather than the nephelometer itself. Data were con- tinuously recorded on a 12-channel Leeds and Northrup multipoint strip chart recorder operated at a chart speed of 2 inches per hour. The chart is 10 inches wide and spans 5 decades of light scattering (b = 2.5 x 10- 8 m" 1 to 2.5 x 10" 3 m" 1 ) The output of the nephelometer has an electrical time constant of 45 minutes, which produces a geometric mean because of the logarithmic form of output. From January to October 1974, the strip charts were analyzed by hand using the usual equal area method to produce hourly averages ending at a given hour. These hourly averages are also essentially geometric means because of the logarithmic form of output. Since there are four channels of output, 96 data points are produced for each day. Finally, three values of Angstrom exponent, a (450, 550), a (550, 700), and a (700, 850) are calculated. In October 1974, the nephelometer was interfaced with the Mauna Loa data acquisition system, programmed to take one sample per second of the nephelometer output voltages, and to calculate hourly means of the four light-scattering values. A comparison showed good agreement between manual strip chart analysis and computer analysis. Since all of the chart reduction was accomplished by the same person, it is believed that the data taken between January 1974 and October 1974 are probably as reliable as the subsequent computer-analyzed data. All data after October 1974 were taken directly from the computer output except for periods when the computer was down or otherwise occupied. During these periods, the data were filled in by manual chart reduction. Mauna Loa data are presented in three ways: (1) hourly means of b sp (550 nm) plotted as a time series for January 1974 through December 1976; (2) monthly means of b sp (450, 550, 700, 850 nm) for each hour of the day for January 1974 through December 1976; (3) monthly means of b (450, 550, 700, 850 nm) sp for 0100 to 0700 hours, plotted as a time series for January 1974 through December 1976. These data are presented in Appendices A-C, along with relative humidity and winds for angstrom exponent. The data are discussed below. 3.1 Data in Appendix A Appendix A gives hourly means of b (550 run), plotted on a log- arithmic scale, along with hourly means of relative humidity, wind direction, and speed. The angle of the wind arrow indicates the direc- tion from which the wind is blowing; the length of the arrow indicates the speed. The length of the barb is equivalent to 5 knots or about 2.5 m sec -1 and may be used for scale. For example, the winds on February 18, 1974 (Fig. Al ) are about 15 knots (7.5 m sec -1 ) from the southeast. The presentation in Appendix A shows the overwhelming influence of the local wind circulation. Often the classic wind regime (Mendonca, 1969) occurred with northerly day- time upslope winds bringing moist contaminated island air, and south- erly nighttime down slope winds bringing dry background tropospheric air to the Observatory. A good example of this diurnal circulation may be seen clearly in the first two weeks of October 1974 (Fig. A5 ) . There appear to be three well- defined climatic regimes at Mauna Loa : (1) The classical upslope- downslope circulation discussed above is usually present during the undisturbed tradewind regime in which the tradewind inversion (Men- donca and Iwaoka, 1969) may be pen- etrated by the thermally induced daytime upslope wind flow. Night- time lightscattering values are very low, of the order of 10 -7 m _1 , and may be representative of the background tropospheric aerosol . (2) Occasionally, the upslope- downslope circulation is over- whelmed by strong southeasterlies (e.g., December 1974, Fig. A6 ) ap- parently associated with a strong high pressure system in the vicinity of the islands. These periods bring the cleanest dry background condi- tions to Mauna Loa and often give light scattering values lower than 10" 7 m" 1 . (3) Often periods of uncertain air mass origins appear (e.g., November 12-18, 1974, Fig. A6 ) in which low relative humidity and high light scattering values are associ- ated with westerly winds, or January 22-30, 1976 (Fig. A13 ) , in which the upslope-downslope winds are so light that the circulation probably does not have a chance to cleanse itself. Of course, there are many exceptions to these three regimes; however, with a little practice it becomes easy to scan the time series and perceive the background level of about 10 -7 m~' . Ordinarily, this level is attained during periods of strong winds with a southerly com- ponent and often is approached during nighttime downslope wind conditions . If we are to attempt to define the background tropospheric aerosol, we must first understand the local wind patterns and determine the air mass present at the site in order to identify the periods when the sur- face measurements at Mauna Loa are representative of the upper tropo- sphere. 3.2 Data in Appendix B Appendix B gives monthly means of b (450, 550, 700, 850 nm) for sp all hours of the day. The upper portion of each graph shows a (450, 550), a (550, 700), a (700, 850), the three most useful independent angstrom exponent values obtainable from 4-wavelength light scattering data. The angstrom exponent is calculated using the formula a(A x , A 2 ) log b (A, ) ■» sp log b sp (^ 2 ) log A x - log \ 2 Since a larger value of angstrom exponent suggests relatively more small particles, most of the data in Figures B1-B7 suggest an absence of larger particles during nighttime clean conditions and a fairly uniform power-law size distribution (over the limited size range seen by the nephelometer ) during the daytime hours. That is, the background aerosol appears to have a steepening size distribution for larger parti- cle sizes. This is seen most clear- ly in the annual means of Figure B7 where a 34 is slightly larger during the night than during the day. One obvious feature of Figures B1-B6 is that the diurnal variation in light scattering is greatest in fall and winter and minimal in the spring. However, the maximum value, occurring at about 1530 hr, is fairly uniform throughout the year whereas the early morning values show a pronounced annual cycle . The diurnal change amounts to about an order of magnitude and this must be caused by an increase in particle concentration rather than particle growth caused by the daytime in- crease in relative humidity. Pues- chel et al. (1973) found that the effects of relative humidity at Mauna Loa could account for only about a 50% increase in light scat- tering. Furthermore, condensation nucleus counts nearly always in- crease markedly during a daytime upslope event (see, e.g., Pueschel and Mendonca, 1972). Another important feature of Figures B1-B7 is that the cleanest time of day at Mauna Loa is during the hour just before sunrise. Although there is some variation throughout the year, the best time of day for sampling background air at Mauna Loa on a regular basis is the period from about 0100 to 0700 hr. 3.3 Data in Appendix C Figure CI gives monthly means of data plotted for all three years. The two most obvious fea- tures are the annual variation in light scattering with a maximum in about May, and the annual varia- tion in angstrom exponent with a minimum at about the same time. atmospheric transmission variations at Mauna Loa are accounted for very well by variations in aerosol mass concentration . SUMMARY It is interesting that many atmospheric variables measured at Mauna Loa show an annual variation similar to that of light scattering. In particular, the atmospheric transmission, as derived from solar radiation data (Ellis and Pueschel, 1972), shows a pronounced minimum in the spring. Bodhaine and Pueschel (1974) suggested that these varia- tions in transmission may very well be caused by variations in the tropospheric aerosol . They con- cluded that a variation in aerosol of only 0.5 pgm m ~ :i at Mauna Loa could account for the seasonal change in transmission. The three- year monthly means given in Figure B-8 show a springtime maximum of b ^ sp (550 nm) = 1.33 x 10 _s m" 1 and a winter minimum of b (550 nm) = sp v ' 1.33 x 10" 7 m _1 . By using Charl- son's relationship, these light scattering values may be converted to the equivalent mass concentra- tions 0.5 ugm m -3 and 0.05 (jgm m~ :x . This amounts to an annual variation of 0.45 fjgm m~ 3 , very near the value suggested by Bodhaine and Pueschel (1974). Therefore, it is seen that both the magnitude and phase of The data show that diurnal upslope-downslope wind circulation is present much of the time, giving background sampling conditions during the night and contaminated conditions during the day. Periods of strong southeasterlies give the cleanest sampling conditions. Occasional uncertain periods occur which may give either low or high scattering values and are difficult to interpret. Variations in atmo- spheric transmission measured at Mauna Loa are closely in phase with light scattering variations suggest- ing that under certain conditions the surface aerosol measurements are representative of the upper tropo- sphere and that aerosols are primar- ily responsible for variations in the Mauna Loa atmospheric transmis- sion measurements. 5 . ACKNOWLEDGMENTS I wish to thank Cris Maeda, Ed Lundin, and Bob Evans for their untiring assistance in data reduc- tion. 6. REFERENCES Ahlquist, N. C, and R. J. Charlson, 1967. A new instrument for measuring the visual quality of air, J. Air Pollut . Control Assoc . , 17, 467-469. Ahlquist, N. C, and R. J. Charlson, 1969. Measurement of the wave- length dependence of atmos- pheric extinction due to scat- ter, Atmos . Environ . , 3, 551- 564. Angstrom, A., 1929. On the atmos- pheric transmission of sun radiation and on dust in the air, Geogr . Ann . , 11, 156. Beuttell, R. G. , and A. W. Brewer, 1949. Instruments for the meas- urement of the visual range, J. Sci . Instrum. , 26, 357-359. Bodhaine, B. A., and B. G. Mendonca, 1974. Preliminary four wave- length nephelometer measure- ments at Mauna Loa Observa- tory, Geophys . Res . Letters , 1, 119-122. Bodhaine, B. A., and R. F. Pueschel, 1974. Source of seasonal vari- ations in solar radiation at Mauna Loa, J. Atmos . Sci . , 31, 840-845. Butcher, S. S., and R. J. Charlson, 1972. An Introduction to Air Chemistry , Academic, New York. Charlson, R. J., 1972. Multiwave- length nephelometer measure- ments in Los Angeles smog aerosol, J. Colloid and In - terface Sci . , 39, 240-265. Charlson, R. J. , H. Horvath, and R. F. Pueschel, 1967. The direct measurement of atmos- pheric light scattering co- efficient of visibility and pollution, Atmos . Environ . , 1, 469-478. Charlson, R. J., N. C. Ahlquist, H. Selvidge, and P. B. Mac- Cready, Jr., 1969. Monitor- ing of atmospheric aerosol parameters with the inte- grating nephelometer, J. Air Pollut . Control Assoc . , 19, 937-942. Crosby, P., and B. W. Koerber, 1963 . Scattering of light in the lower atmosphere, J. Opt . Soc . Am . 53, 358-361. Ellis, H. T., and R. F. Pueschel, 1972. Solar radiation: ab- sence of air pollution trends at Mauna Loa, Science , 172, 845-846. Junge, C. E., 1963. Air Chemistry and Radioactivity , Academic, New York. Mendonca, B. G. , 1969. Local wind circulation on the slopes of Mauna Loa, J. Appl . Meteorol . , 8, 533-541. Mendonca, B. G. , and W. T. Iwaoka, 1969. The trade wind in- version at the slopes of Mauna Loa, Hawaii, J. Appl . Meteorol . , 8, 213-219. Mie, G., 1908. A contribution to the optics of turbid media, especially colloidal metal- lic suspensions, Ann . Physik , 4(25), 377-445. Pueschel, R. F., and B. G. Mendonca, 1972. Sources of atmospheric particulate matter on Hawaii, Tellus , 24, 139-149. Pueschel, R. F., B. A. Bodhaine, and B. G. Mendonca, 1973. The proportion of volatile aero- sols on the island of Hawaii, J. Appl . Meteorol . , 12, SOS- SIS. Rayleigh, Lord, 1871. On the light from the sky, its polarization and colour, Philos . Mag . , 41, 107-120. Whitby, K. T., 1975. Modeling of atmospheric aerosol particle size distributions, Progress report on EPA grant No. R800971, Univ. of Minn., Minneapolis, Minnesota. Waggoner, A. P., N. C. Ahlquist, R. J. Charlson, 1976. Recent developments in nephelometers, In NASA CP 2004. 10 Appendix A Hourly means of 550 nm light scattering, relative humidity, and wind at 3-hour intervals for Mauna Loa Observatory. 11 q 601 °oHU -o .."J q Boi c ' 5 m tr> h- co Si I I I I L -o t .uJ P 601 c g m ^ N OD Ol M Si >H (0 3 •? 0) tin Si m 1-3 0) 3 •H 6h %HH q Boi %HU .q B0| %Hd q Boi %HH s q 6o| 12 < i- uj ''q 6o| a «*: i •c: o 5: a. 13 < X i- i MM ■ s q 601 10 to r- co o o o ui ds q 601 o oHd C 3 ha QJ 3 uj as q 6oi 14 +J n a: tr. a *c i a« ■J U 15 r t w u *q 601 in U3 N CO . . V y s q B0| w as q 601 16 O It Q O 3 •H 17 °oHd w ° s q 601 in CD r- co oHH .lu as q 60| — r^ I /I I ^ \i V % % t .ai as q Bo| in co ^ co r IT) l\ <^i CO CM M & U w in 3 CM (H -q aj E>4 •■cr 1 CM 3i ss (« 3 cr> R CM fa in hi r^ a> CM £• • CM CO -O CD Li. H CM 3 tn •H k, in co I s * co s q 6oi %Hd ai as q 6o| w ds q 6o| %HU s q 6o| 18 °oHH w ds q 601 oHd < $ * A \n (£ r- ! * » d uu "q 601 tf> to h» co ON I ■§ Is 00 CD Is 3 c* 'q 601 oHH o Bo 19 oHb q 601 s q Bo| g i r 5 q 6o| %HH in to N co w as q 6o| g ^ \ q 601 I I L s q 601 IT) t\ (^ M CD q i 3i 3 ■rl &4 %Hd q 6o| %Hd s q 6o| 20 tu q 601 (/) 3 q 6oi 21 °/oHU w as q B0| LO CD P- CO X J- i X s q 6o| 1 ( y i f q 601 Y Y Y Y 't I Y Y Y I L lu u *q BO| s q Bo| %HU q 601 %HU q 601 22 S — o R 1 : t w "q 601 %HH o m ic f-- lO r-- GO lu as q 6oi I y oHU *q 6oi 8 ° uj 0, q 6oi In Q 5: 23 oHH °oHH £ T as c t .UJ q 6o| ) s ai q 6o| [\ M Si fH 3 -Q QJ I is R CD 3 •H &4 , uj q 6o| 24 ,uj q 601 OS I 5: (D 3 •H m "qO< 25 1 2 s q 6o| , uj q 601 in cd h~ c °oHH o ? o 1 5 , lu s q 6o| oHd Q) 3 hi i 26 ,HU t i t : : q 601 L^ 5 f 1 ) } \ > ■ ? I s q 601 ON 4J to 3 Cn 3 I 3i 3 ■"H Q) 3 tn ■H 27 s q 601 s q 6o| .w q 6o| %Hd a> , I) a J .w s q 6o| J> ■^ i y Y I i <* m O •L I I L u CD ■8 o o I cu ho a °oHH § a , uj a *q 601 in uj "*q 6oi 29 Appendix B Monthly Means (means, for each hour of the day, of 450, 550, 700, and 850 nm light scattering and angstrom a 12 (450, 500 run ) , a 23 (550, 700 nm ) , a 34 (700, 850 nm ) for July through December 1975] and Annual Means 31 E _ O QJ £ c ~ 4 c X < LU 2 2- 1 1 - 1 1 1 ,— • ,^**< > ~ 1 l ' January f 1 I - a 12 Legend a 23 a 34 - -6 - I I I - -^!Ztnnrs>wc^^«--~i>- :s * i »e£c==s- February I ! i I l - 00 06 12 Time of Day 24 00 06 12 Time of Day 24 Figure Bl . Monthly means for January-June 1974. 32 a 12 a 23 E £ O QJ < LU CD fa C K o 01 Q. C X < LU Septe 7 - ~^=- "C^- ~_ ^c: 1 October 1 ! Novembei 6 - 7 - December 00 12 1{ • II. l\ .'1 110 06 24 Figure B2 . Monthly means for July-December 1974 33 a 12 Legend a „^ g, S 2 F ^ 4 o S || 2 c X < LU March 24 00 4 2 24 00 - — - — ~~ ™ -^ZSZZe = — *e3KXS t c w o 01 a 5 * < LU tt 4 1 ! 1 November _l 1 1 00 06 12 Time 24 00 06 12 of Day Figure B6 . Monthly means for July-December 1976. 37 a 12 Legend a 03 06 12 Time of Day Figure B7 . Annual means, individually and combined. Appendix C Monthly Means for Three Years «12' Legend O93. a 34 ^1 — n — 1 — 1 — 1 — 1 — 1 — 1 — 1 — 1 — 1 — r "i — 1 — r ~\ — r~r~] — r~i — 1 — r~i — n~] — 1 — mini — rn — n - ! — 1 — m — 1 — 1 — 1 — 1 1 1 1 5 - 6 - 7 - 1 1 1 1 1 1 1 1 1 1 I 1 1 1 1 1 1 1 1 1 1 1 I 1 1 1 1 1 1 1 1 .. i 1974 1975 1976 5 - S- 6 J I L "i r J F M A M 1974. 75. 76 0100 0700 hr. N Figure Cl . Monthly means of 0100-0700 hr light scattering and angstrom exponent for 1974, 1975, 1976, and for all three years combined . 1. IVUIMRNMISII'KIMN.IHUI I •- ?4 > RFC.ION NO S 39 LABOR AT DRIES The mission of the Environmental Research Laboratories (ERL) is to conduct an integrated program ot fundamental research, related technology development, and services to improve understanding and prediction of the geophysical environment comprising the oceans and inland waters, the lower and upper atmosphere, the space environment, and the Earth The following participate in the ERL missions MESA OCSEA W/M Wear gram 0/'/ and iion nent and severe lion, NHEML National Hurricane and Experimental Meteorology Laboratory. De rmiques for i ilus systi beneficial moi RFC Res' arn res' vilh US' mea' logged data, and ri AOML At! PMEL GLERL GFDL APCL NSSL WPL ARL AL SEL Air Resources Laboratories Studies the diffusion, transport, and dissipation of atmos- pheric pollutants, develops methods of predicting and controlling atmospheric pollu- tion, monitors the global physical environment to detect climatic change U.S. DEPARTMENT OF COMMERCE National Oceanic and Atmospheric Administration BOULDER. COLORADO 80302 PENN STATE UNIVERSITY LIBRARIFC HIIIIIIHII A0DDD7EQEnD3