Solar Irradiance Measurements from a Research Aircraft
Post on 05-Oct-2016
Solar Irradiance Measurements from a Research Aircraft
M. P. Thekaekara, R. Kruger, and C. H. Duncan
Measurements of the solar constant and solar spectrum were made from a research aircraft flying at11.58 km, above almost all of the highly variable and absorbing constituents of the atmosphere. Awide range of solar zenith angles was covered during six flights for over 14 h of observation. Results arepresented from nine different instruments which complemented each other in measuring techniquesand wavelength range and were calibrated and operated by different experimenters. A new value of thesolar constant, 135.1 mW cm-', has been derived, as well as a revised solar spectral irradiance curve forzero air mass.
A group of experimenters from Goddard SpaceFlight Center, National Aeronautics and Space Ad-ministration, undertook a project in August 1967 tomeasure the solar constant and solar spectral irradiancefrom a research airplane flying at 11.58 km (38 000ft). The flights were made from Moffett Field, Cali-fornia. The airplane was fitted out as a high-altitudesolar irradiance laboratory, with twelve different in-struments, seven for spectral irradiance and five fortotal irradiance. The instruments represented a widevariety of measuring techniques and incorporated someof the best available features of precision radiometry.
The solar constant is the amount of total solarenergy of all wavelengths received per unit time perunit area exposed normally to the sun's rays at theaverage distance of the earth in the absence of theearth's atmosphere. The solar spectral irradiance isthe distribution of the same energy as a function ofwavelength. Our present knowledge of these physicalquantities is based mainly on ground-based measure-ments.
Most authors agree that the irradiance of the sunat the average sun-earth distance is in itself highlyconstant. But there is little agreement as to what thatvalue is. Some of the more significant revised valueswhich have been proposed during the last forty yearsare presented in Table I (Refs. 1-9). Most authorsgive the value in units of calories per min per cm2 .Conversion to units of mW cm-2 has been made on theassumption of 4.186 J/cal for the mechanical equivalentof heat. Table I shows that the solar constant hasfrequently been revised and that up to the advent of
The authors (as are the additional authors named in Sec. III)are with Goddard Space Flight Center, NASA, Greenbelt, Mary-land, 20771.
Received 30 September 1968.
space-borne instrumentation most revisions yieldedvalues higher than those previously accepted. Thedisagreements between authors are due largely to thedifficulties of malting measurements through the highlyvariable smoke, dust, haze, and cloud cover of theearth's atmosphere.
The discrepancies between different investigatorsare even greater for the spectral distribution of thesolar energy. Dunkelman and Scolnik'0 have pub-lished charts which show these differences in a strikingmanner. A detailed report on the state of our knowl-edge of the solar constant and solar spectral irradiancehas been published by M. P. Thekaekara."1 In thewavelength range beyond 1.2 A where the atmosphericwater vapor bands are strong, most authors, followingP. Moon, have assumed the solar spectrum to be that ofa 6000 K blackbody. The wide differences, as muchas 40%, between authors in the uv and visible ranges,and the almost complete lack of experimental data inthe longer wavelength range strongly indicated theneed of a new method which is free of the uncertaintiesof ground-based measurements.
II. Description of the High-Altitude SolarIrradiance Observatory
At the flight altitude of 11.58 km the observers wereabove nearly 80% of the permanent gases of the at-mosphere, and more importantly above nearly 99.9%of the water vapor and all the dust and smoke whichform the highly variable and absorbent constituents ofthe atmosphere. The combination of 12 different in-struments complementing each other in wavelengthrange, spectral resolution, and measuring techniques,operated by different experimenters, calibrated withreference to several independent standards and carriedaloft in an airborne laboratory at 11.58 km, practicallyfree of atmospheric absorption, is considered a valuablefeature in enhancing the level of confidence in the finalresult.
August 1969 / Vol. 8, No. 8 / APPLIED OPTICS 1713
Table I. Significant Revisions of the Solar Constant
Solar constantAuthor Year (mW cm-2)
P. Moon' 1940 132.3L. B. Aldrich and G. C. Abbot 2 1948 132.6W. Schuepp3 1949 136.7-141.6C. W. Allen4 1950 137.4L. B. Aldrich and W. H. Hoover5 1952 134.9F. S. Johnson 6 1954 139.5R. Stair and R. G. Johnston7 1956 143.0C. W. Allen 1958 138.0E. G. Laue and A. J. Drummond9 1968 136.1M. P. Thekaekara, R. Kruger, and
C. H. Duncan 1969 135.1
The aircraft was a four-engine jet, NASA 711, ap-propriately named Galileo, commercially known underits model name Convair CV-990. The team for thesolar irradiance measurements consisted of 18 ex-perimenters, scientists, and technicians to operate theinstruments and nine aircraft personnel. An in-flightview of the aircraft is shown in Fig. 1. Prominentamong the special features which make it an airbornescience laboratory are the large extra windows on theroof of the aircraft and the replacement of some of thepassenger windows by optical quality material. Otherspecial features are electrical power outlets at eachexperimenter's station, defrosters for observation win-dows, time code generator, and precise navigationalequipment for the determination of latitude, longitude,and solar zenith angle.
A listing of the instruments in the order in whichthey were located, starting from the front of the air-plane, is given in Table II, as also some basic summarydata on each. The five total irradiance instrumentsare indicated by the word "total" in Column 3, typeof instrument. The names given in the last columnare of those who manned the stations during the flights,though the contributions of many of them extended toseveral other instruments as well, especially during thepreparation for the flights and the data analysis afterthe flights. These instruments were selected inpreference to others, partly because they had all beentested and believed to be sufficiently reliable at levelsof solar irradiance, and partly because they could bereadily mounted and operated on an airplane almostas well as in a laboratory.
The instruments were mounted in the aircraft onaluminum frames specially designed to meet the ratherstrict aircraft specifications for loading and rigidity.For all but two of the instruments the mounting fixtureswere such that the instruments could rotate about ahorizontal axis parallel to the length of the aircraft,and thus track the sun in elevation. The two excep-tions were the Perkin-Elmer monochromator and theCary 14 monochromator for which movable externaloptics were provided to direct the sunlight to the en-trance slit. Tracking the sun in azimuth was ensuredby the aircraft itself flying along a path computed priorto the flights, such that the solar rays were constantlyat right angles to the length of the aircraft.
Six flights were made between 3 August and 19August, 1967. The plan followed for each flight wasto start from Mfoffett Field, go outward to a distantpoint where observations were scheduled to begin,make a U-turn, and return along a flight path, with allthe instruments constantly tracking the sun. The out-bound leg of the flight gave the experimenters time totest the equipment and to make calibration scans whereneeded. It also enabled the aircraft to use up part ofthe fuel and become sufficiently light to maintain a con-stant high altitude for the observations. The altitudeof 11.58 km (38 000 ft) was chosen as a compromisebetween being as high in the atmosphere as possible and
Fig. 1. NASA 711 Galileo, Convair 990A, the airborne labora-tory in flight.
130'W 120W 101W 1OO'W 90W
AU.GAS 10 CANADA
5ON ____1 \ a \r 94AUGUS16 \ SIOUX F40
= 5 93 p mod Z p m f \ UNITED STATESo FRANCISCO~~~~~~~. IT K
14- ~ ~ ~ f
LONGITUDE DEGREES WEST
Fig. 2. Flight paths of NASA 711 Galileo, 3-19 August 1967.
1714 APPLIED OPTICS / Vol. 8, No. 8 / August 1969
Table II. Data on Instruments Used for Solar Irradiance Measurements Onboard NASA 711 Galileo, August 1967
AircraftType of window Wavelength
Instruments Energy detector instrument material range Experimenters
I Hy-Cal pyrheli- Thermoelectric emf Total Infrasil 0.3-4 p McNutt, Rileyometer
2 Angstrom pyr- Resistance strip Total Infrasil 0.3-4 u McNutt, Rileyheliometer 6618
3 Eppley normal Thermoelectric emf Total Infrasil 0.3-4 p McNutt, Rileyincidence radi-ometer
4 P4 interfer- 1P28 or R136 PbS Soleil prism Infrasil 0.3-0.7 A Rogers, Wardometer tube Soleil prism Infrasil 0.7-2.5 p
5 14 interfer- Thermistor Michelson's Irtran 4 2.6-15 p Ward, Rogersometer bolometer bi-mirror
6 Perkin-Elmer 1P28 tube thermo- LiF prism Sapphire 0.3-0.7 a Thekaekara,monochromator couple LiF prism Sapphire 0.7-4 p Winker, Stair
7 Cone radiometer Resistance wire Total Infrasil 0.3-4 p Kruger, Winker,Ward
8 Filter radiometer Phototube Dielectric thin Dynasil 0. 3-1 .2 p Lesterfilm filters
9 Angstrom pyr- Resistance strip Total Dynasil 0.3-4 u Duncanheliometer 7635
10 Electronic scan- Image dissector Grating Dynasil 0.3-0.48 p Webbning spec-trometer
11 Leiss mono- EMI 9558 Q.A. tube Quartz double Dynasil 0.3-0.7 a McIntosh,chromator PbS tube prism 0. 7-1. 6 A Park
12 Cary 14 mono- 1P28 tube Pbs tube Grating and Dynasil 0.3-0.7 p Arvesen, Griffin,chromator prism 0.7-2.5 a Kinney
making observations for as long a time interval as pos-sible.
During each flight, at intervals of 10 min, the flightnavigator made measurements of the latitude andlongitude of the aircraft location. Figure' 2 is a mapof the geographical area covered in the six flights, withthe curved lines showing the route taken during thedata taking portion of each flight. Each flight pathis indicated by the date and the time in Pacific daylightsaving time of the start and finish of data acquisition.
The position data provided by the navigator usingLoran A and doppler radar are considered accurate towithin 10 nautical miles. These data were used in acomputer program to calculate the zenith angle of thesun and the air mass above the aircraft for 1-minintervals. Air mass data are needed to determine thesolar irradiance for zero air mass, that is, in the absenceof the earth's atmosphere. Air mass is defined as theratio of the distance traveled by the rays of the sun inthe atmosphere for a given zenith angle of the sun andaltitude of the observer to the distance for the samealtitude and zenith angle zero. Air mass is unity whenthe sun is vertically above the observer. For otherpositions of the sun, air mass is the secant of the zenithangle in the first degree of approximation. Twoformulas are available for obtaining a more accuratevalue of the air mass for ground-based measurements,one due to Bemporad 2 published in 1907 and a more re-cent one due to Kasten 3 published in 1964. However,for measurements at a high altitude, above nearly 80%of the atmosphere, Kasten's formula is overcorrected.
The value for air mass at 11.58 km which we have usedas being most satisfactory is a weighted average of thesecant of the zenith angle and the value given byKasten's formula, with relative weights in the ratio of4 to 1. Figure 3 shows the variation of air mass withtime on the six days of the flight.
Another factor which has to be considered as a cor-rection factor common to all the instruments, is thedistance of the earth from the sun. The solar constantand spectral irradiance are defined for the mean sun-earth distance, which is the semimajor axis of the earth'sorbit or one astronomical unit (a.u.). During the
Fig. 3. Variation of
20O NOON 2:00 P...PACIFIC DAYLIGHT SAVING TIME
air mass with timeNASA 711 Galileo.
for the flight paths of
August 1969 / Vol. 8, No. 8 / APPLIED OPTICS 1715
- WINDOW/ N, LI NWIRE WOUNDCONICAL DETECTOR
~ 0 -- 20.3 cm _ I
Fig. 4. Drawing of the Cone radiometer.
days of the flight, the ratio of the sun-earth distanceto 1 a.u. had the following values: 3 August: 1.0146;8 August: 1.0139; 10 August: 1.0136; 14 August:1.0128; 16 August: 1.0125; 19 August: 1.0119. Itis seen that since the sun was at a distance greater thanaverage during the flight days, the values obtainedfrom our observations have to be increased by per-centages varying between 2.9 and 2.4 to give the solarconstant and spectral irradiance.
A more detailed description of the NASA 711 Galileoproject has been published by the Goddard experi-menters. 4 Copies of the report may be obtained bywriting to the authors of this paper. The report dis-cusses several features of the airborne solar irradianceproject, which for the sake of brevity have been omittedhere: the window material for each instrument, mount-ing of the instruments in the aircraft, flight paths andrelated problems of water vapor above the aircraft,computation of solar zenith angle and air mass, residualgas analysis of water vapor and other absorbents in theaircraft, etc.
Brief discussions of each of the experiments on boardand their results are given below in Sec. III. The sub-sections of Sec. III have been contributed by the ex-perimenters who undertook the main responsibility forpreflight preparation, in-flight measurement, and dataanalysis of each experiment. Of the twelve experimentslisted in Table II, only nine have been chosen for dis-cussion here. The Eppley normal incidence pyrheliom-eter was used only for a short time during one of theflights and the data seemed questionable due to calibra-tion difficulties. The electronic scanning spectrometerwas used on all the flights and yielded extensive data;here again difficulties in calibration made the resultsunreliable. The Cary 14 monochromator was flownby a group of NASA experimenters from Ames Re-search Center, MXoffett Field, California, and forms thetopic of a separate publication.
Ill. The Solar Irradiance Experiments:Instrumentation and Data Analysis
A. Cone Radiometer (R. Kruger)Description of the Equipment
The cone radiometer is designed to operate as anabsolute total radiation detector using an electricalpower substitution method. It is therefore referencedto the scale of absolute electrical units. The radiom-
eter is designed to operate in a vacuum environmentwhere gaseous conduction and convection may beneglected.
The theory behind the operation is relatively simple.If an electric current is passed through a wire wrappedin the shape of a cone with the base open, the energydissipated will be radiated out through the base of thecone and through the slant conical outer surface. Ifthe slant surface is enclosed in a housing whose tempera-ture is maintained constant, the energy incident fromthis enclosure to the slant surface may be consideredconstant. The energy incident on the base of the c...