Embed Size (px)
Citation preview
JSC -08682
NASA TECHNICAL MEMORANDC'M NASA TM X-58129 February 1974
MEASUREMENT OF ATMOSPEERIC PRECIPITABLE WATER
USING A SOLAR RADIOMETER
N7449993
NATIONAL AERONAUTICS AND SPACE ADMINISTRATION
LYNDON B. JOHNSON SPACE CENTER
HOUSTON, TEXAS 77058
https://ntrs.nasa.gov/search.jsp?R=19740011870 2018-08-09T15:14:45+00:00Z
MEASUREMENT OF P.ThIOSPHERIC PRECII'IT in!.t. V.'XTER
CSIKG A SOLAR RADIOhIETER
David E . P i t t s and Will iam A k A i l : ~ : ; ~ Lyndon B. Johilson Space Center
Houston. Tesas 77058
Alyce F:. Dillingcr Lockheed Zlectronics Co
Houstw, Tesas 77958
CONTENTS
Section
S W M Y . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . s Y M 3 o L s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . RADIOMETER E a E n m S AND ACCURACY . . . . . . . . . . . . . . . . . CALIBRATION OF THE SOLAR RADIOMETW . . . . . . . . . . . . . . . . . . CONCLUDING REMARKS . . . . . . . . . . . . . . . . . . . . . . . . . . REFwEncEs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Page
1
1
2
4
c
12
1 3 .
V
TABLE
I SOLAR RADIOMETER CALIBRATION VALUES . . . . . . . . . . . . . . 15
FICURES
Figure
1
2
3
4
5
6
7
PaPC _ .
The seven-channel solar radiometer . . . . . . . . . . . . . . 1; . -
F’ilter function for the water band for solar radiometer. u n i t 4 i . : . . . . . : . . . . i . . . . . . . . . . . . . 17
-_ Calculated transmission through the atmosphere for
Albany, N.Y., Jan. LO, 1972, raciiosonde, 0.202 centimeter precipitable water, 1 centimeter accuracy . . . . . . . . . 18 -1
Calculated transmission through the atmosphere smothed by a
triangular f i l t e r function 20 centimeters-’ wide. Albany, N.Y., Jan. 10, 1972, radiosonde, 0.202 centimeter precipitable water . . . . . . . . . . . . . . . . . . . . . 19
Precipitable water calibration for solar radiorceter u n i t 4 . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
Calibration curves for 17 solar radiorceters . . . . . . . . . . 21
Calibration verification between solar radiometers . . . . . . 22
v i
OF ATMOSPHERIC PRECIPITABLE WATER
USING A SOLAR RADIOMETER
By David E. P i t t s , W i l l i a m &Urn, and Alyce E. Dillinger* -don B. Johnson Space Center
Measurements of t o t a l atmospheric water vapor are accomplished by ratioing a water-vapor absorbing region (0.9435 micrmeter) t a a clear channel (0.8730 micrometer). of wavelength i n the v is ib le and near infrared. inates the aerosol and Rayleigh scattering leaving only the water-vapor absorp- t i o n effects. water bands and a large amount of water vapor may not saturate the weak bands, this r a t i o is neikher a linear nor an exponential func t im of precipitable water.
Aerosol extinction is usually a slowly varying f c x t i o n This rat ioing effectively e l i m -
Because a s m a l l amount of water vapor may saturate the strong
IBTRODUCTIOII
The purpose of this paper is t o describe a method of using a solar radiom-
One channel covers a water-vapor absorption region, and the other, a eter t o deduce precipitable. water i n the atmosphere. channels. nearby region in which water-vapor absorption is absent. The method operates on %he assumption that the aerosol and Rayleigh scattering opt ical depths are approximately the same i n the two chsnnels. This assumption allows the r a t i o of the two channels t o be used t o eliminate the aemsol and Rayleigh scattering and thus gives the precipitable water i n the s lan t path. By dividing the water amount by the air mass? the precipitable water' i n the ver t ica l path may be obtained. Such an instrument was reportedly constructed and calibrated against radibsonde data during the early 1960's, but the de ta i l s of the instrument and
its success have not been verified.
The technique uses two
3
"Lockheed Electronics Co . , Hctrston, Texas 77058. h e air mass i s defined as the secant of the angle from the zenith. %recipitable water (centimeters) is defined as the t o t a l amount of water
vapor (grams) i n a column above a surface area of 1 square centimeter, divided by the density of l iquid water ( re f . 1).
the successful construction and cal ibrat ion of th i s instrument. %urcrw and Barker (ref. 2) accredited Foster, Volz, and Foskett with
SYXBOLS
A = 0.416
B = -1.01 x 10 -3 (pm)-*
constants S C 2
4
Io,v51, 1,2 F2(V)du -
c3 - Io ,uc2 394
F 1 $ ( A ) of channel 1
4(X) of channel 2 F2
2 I intensi ty (w/cm um sr)
T average intensity
I downwelling known intensity 0
J meter reading of solar radiometer for air mass m
meter reading of solar radiometer for a i r mass m = 0 JO
K
K
experimentally determined constant = 0.k22
absorption coefficient of the gas a t frequency. w V
m air mass = sec e (an approximation val id 3 percent for Oo 5 e 5. 80") = l / u
precipitable water (cm) i n the ver t ica l sath 0
P
T atmospheric transmission
T average atmospheric transmission
2
T(A) atmospheric transmission as a function of wavelength
atmospheric transmission for radiometer unit 4 T4
atmospheric transmission for uncalibrated radiometer U T
z altitude in the atmosphere
0 zenith a w e
X or v
Xo or u
wavelength or frequency, respectively
center wavelength (0.9435 wn) or frequency (10 598.83 cm-') , O respectively
P atmospheric density . density of water at standard temperature and pressure "H20
T optical depth
t optical depth of entire atmosphere 0
$ ( A ) least-squares fit of the transnission of the solar radiometer filter
Superscripts :
- downwe!.ling quantity
X experimentally determined constant = 0.574
Subscripts:
192
394
0.9420 quantity at 0.9420 um
0.8730 qvantity at 0.8730
a atmospheric gaseous absorption processes
b
quantity over limits 1 to 2
quantity over limits 3 to 4
black-body radiator at a specified temperature
3
S
t
atnospheric Rayleigh and aerosol extinction
dummy variable
Atmospheric aerosol and molecular scattering and absorption by atmospheric gases are deleterious conditions that distort the upwelling terrestrial radia- tion and cause problems in the automatic classification of multispectral imagery by automatic pattern recognition techniques (refs. 3 and 4). atmospheric conditions have large temporal and spatial variations that may re- quire atmospheric corrections to be made over a grid covering the field of view of the remote sensor. For example, large gradients of precipitable water exist across d r y fronts in the Midwest. These gradients may cause variations in pre- cipitable water of from 1.9 to 3.0 centimeters over a distance of 40 Kilometers (ref. 1). eter for measuring the optical depth of the atmosphere was designed, constructed, and tested in the field for more thaa a yea- (fig. 1). This unit is similar to the Volz Sun photmeter used bj Flowers et al. (ref. 5) and Volz (ref. 6) in 43 cities in the United States to determine atmospheric turbidity for air qual- ity monitoring and is a lso similar to the multiple wavelength radiolcater de- scribed by Shaw et al. (ref. 7). aerosols (e.g., continental and meritime), Reeser (refs. 8 and 91, and Kleen [ref. 10) determined that it was necessary to measure atmospheric extinction at multiple wavelengths.
Moreover, these
To allow for such atmospheric variations, a seven-channel Sun radion-
To distinguish between differect types of
A device w a s needed to adequately measure such atmospheric extinction and to cover the principal visible and near infrared bands of the Earth Resources Technology Satellite (EIITS) multispectral scanner (MSS) (ref. 11) and the Earth Resources Experiment Package (EREP) multispectral scanter (ref. 12). ysis (ref. 13) was made that indicated an optimum set of waveiengths 10.38, 0.50, 0.61, 0.7487, 0.8730, and 1.04 micrometers) that could be used t3 meet this need. A study of the accuracy of the operatz-radiometer-atmosphere sys- tem revealed a 510-percent error in measurements of the aerosol optical depth. An additiocal t a x at 0.9435 micrometer was chosen to measure absorption cause& by atmospheric water vapor.
An anal-
Kleen (ref. 10) showed that by ratioing the meter reading in the 0.9435-micrometer band to the meter reading in the 0.8730-micrometer ban?, the water-vapor absorptior. could be measured. chosen over that at 1.04 micrometers because the former is closer t-, tne peak response of the silicon detector used in the radiometer. radiometers in the field of view of 'he ERTS or EREP multispectral scanners will give a small-scale picture of b,;h aerosol and water-vapor effects on tra- versing signals. However, a more practical operational scheme that uses space- craft data for this purpose will have to be developed to facilitate pattern recognition of large or remote areas of the world.
The band at 0.8730 microceter was
A network of these
4
Cousin e t al. ( ref . 1 4 ) have described the method ( the Prediction of Response of Earth Pointed Sersors-Reconstruction of Target Reflectance (PREPS- ROTAR)) fo r using these data t o correct multispectral scanner data from ERTS a t bands of 0.5 t o 0.6, 0.6 t o 0.7, 0.7 t o 0.8, and 0.6 t o 1.1 rzicrometers fo r aerosol e f fec ts using a table look-up scheme based on a soluticn t o t h e radia- t ive t ransfer equation known as the doubling method ( r e f s . 15 and 16) . t he calibration of t he water channel of one solar radioneter represented by t h i s pape?, t he other 16 radiometers may be cross-cslibrated. The preci-itable water obtained may then be used as an input, and water-vapor absorption erfects m a y be removed from bands a t other wavelengths fo r ERTS, EREP, o r a i rc raf t d a t s with a similar look-up table. compressed line-by-line atmospheric transmission model by Deutschman and Calfee (ref. 17) by using a set of atmospheric soundings t o represent an extreme rang? of atmospheric precipitable water over extremes i n temperature and moisture distributions (ref. 1). These results w i l l then be included i n the PREPS-ROTAR correction program. ( A paper describing t h i s look-up table for the ERTS MSS7 (0.8 t o 1.1 micrometers) and the effect of water vapor on signature extension for corn and soybeans is being prepared by the authors as a cornparlion paper.
With
This look-up table was genera%ed by modifying a
CALIBRATION OF THE SL?LAR RADIOMETER
Atmospheric transmission T(X) fo r the 0.9435-micrometer water band was calculated by using a least-squares f i t of the transmission of the f i l t e r f c r solar radiometer uni t 4 , as determined from a spectrophotometer ( f ig . 2) . Gaussian function $ ( A )
The used fo r the f i t w a s of th? :omi
Because the spectral response of the s i l i con detectcr changes by less than 0.5 percent and the solar irradiance changes by less thar. 2 percent over t h e width of t he f i l t e r function, it w a s not necessary t o include these terms in
calculating the average transmission trausmission T ( 1)
for the band :i-a the atnospheric
5
The integrals i n equation (2 ) were evaluated using the Simpson rule , in- corpoA-ating atmospheric transmissions obtained for a 10-layer atmospheric model, and calculating water-vapor absorption with about 1 centimeter-' accuracy
(fig. 3) smoothed by a triangular f i l t e r functiou 20 centimeters-' wide at the base (f ig . 4). representing a range of 0.166 t o 5.851 centimeters precipitable water ( f ig . 5 ) . This information includes the extreme range of precipitable water observed i n the United States over a 9-month period i n 1972 and 1973.
Values were h l c u l a t e d fo r 10 radiosonde data sets (ref. 1)
The solution for a downwelling known intensity Io, such as the Sun passing through an absorbing atmosphere where no soattering occurs, is
However, i f the atmosphere does not radiate strongly a t the wavelengths consid- ered and neasurements me made at the Earth s-mface simplifies t o
T~~ = 0, equation ( 3)
l o w , i f single scattering without an absorptive process is considered, a similar expression can be derived as a solution t o the radiative t ransfer equations (ref. 18)
I f these two processes are visualized as dis t inc t phenomena with each occurring i n a separate mathematical layer representetive of the r ea l atmosphere where both phenomena occar, then t.he tots inteilsity transversing the layers is
6
3
or
Nm, i f the solar intensi ty is measured i n two bands, one with aerosol and water-vapor absorption and one with only aerosol absorption, if f i n i t e f i l t e r functions F and F are used for each, and i f I is a slowly varying
function between v1 t o v arid v t o vIs then equation ( 5 ) becomes 1 2 ov
2 3
V f r 2 o,v,s
and equation (7 ) becomes
where v t o v2 covers the 0.8730-micrometer band and v t o v4 covers the 0.9420-~1icrometer water band. N o w , i f aerosol opt ical depth T is a
slowly varying f u x t i o n of frequency over the range of frequencies v1 t o v L ,
1 3 0 , V Y S
7
-r then e 09v’* t h a t the r a t i o
can be removed from the integrals in equations (8) and (9) so of eqxations (8) md (9) gives
where F2 is -..e f -
3 1,2 Jv
l t e r function Over the --ster band v t o t4. Now, since 3
2 where F2(v) = A exp[B(X - Xo) I . Furthermore
For wavelergths near 1.0 micrometer, K is approx-mately independent of z (i.e. , not dependent on temperature) so
T = K P p o,v,a v o H20
8
(13)
so
J ~ 0 8 0 where i: - - 3 = u. Numerical evaluations of the preceding integral
show that it is of the form exp(-KPoxmx) A Jo(0.9420)
so tha t
Joo 50 = exp(-XPo"m"> Jo ( 0.9420 ) J( 0.8730 )
(15)
now includes the terms C /A. 3 /J c (0.8730 o( G. 9420) i f t he r a t io .J
The water-vapor transmission i s not an exponential function of precigita- ble water (f ig . 5). B e e r ' s law. This is because the centers of strong bands saturate a t luw water amouts, leavirig the weak a-worbing lir?es t o contribute t o additional absbrp- tion. "he ad hoc functicn
"his fac t indicates that the transmission does not obey
was determined t o represent a good f i t t o the curve i n figwe 5 f o r so l a r ra- diometer unit 4.
Equations (14) and (15) assume the optical depth f c r aerosol and Hayleigh scattering i s approximately the same i n these aaacen t channels. The Rayleigh optical dep4-h changes by about a factor of 2 for these wavelengths, but it is a very s m a l l value of approximately 0.01 ? 10 >ercent, whereas the aerosol op- t i c d . depth i s about 0 .5 (ref. 18).
,
Ra<'Iosonde data from downtown Houston, Texas, end solar rcciiometer Cata from the NASA Lyndon E. Johnson Space Center (JSC) a t Clear Lake City, Texas, Were collected far the same days.
pared t o the precipitable water emowits from the rdiosonde data i n figure 5 and gave a subjective best f i t of 0.44 for the v a u e of J
The ratic of J (c 91,35)/J(o.8730) w a s com-
[Jo(o.8730)/ 1 tha t represented the curve adequately except a t higher water o(0.9435)
amounts. m i r t u r e amounts present at JSC, which is 40 kiloseters southeast of the damtom Houston area where the 0600 1.s.t. radiosonae is launched. radiome&er m@asurea@nts are made at O900 t o UOO 1-s-t. vhen the sea breeze is
These aaaaalous values are very likely attributable t o the higher
Most solar
S w i n g t o PWk@tL%te the COmtd. -88-
Once the C a l i b r a t i o n value is knm for radim-
f r o m radicieter (0 8730 and J (0.9135) eter unit 4, the meter readings J
unit 4 can be used to obtain the transmission in the water band $¶ which ailovs th@ pr@cipitable water in the s lan t path to be calculated using the curve ir. fi&ure 5- Dividing the ppecipitable water ia the slsnt path by the air mass (set 01 w i n give the precipitable water in me vertical path.
Precipitable water measu ren t s using 0th- radiameters s ~ e g be ob ta ined by and multiplying by
( 0 -9435 ( 0 . 8730 taking the r a t io of the measurements J
for the appropriate radiometer o( 0.8730 )jJo( 0.9b35 the salibration value J
frcm table I.
Figure 6 w i l l then relate this calculated value t o precipitable w a t e r in the slant path. diviang the slant path amount by the air mass m.
'Be precipitable :-Rter i n the vertical path may be obtained by
Because each filter has a sl ight ly different tnrnsau 'ssioa fbnction, the response of transfission as caarrap.pd t o precipitable water is different for the O.gk3F.IPicmmeter band cf each radiareter. curves for a l l 17 radiometers. measured filter resposure functions over the a b s m t i a n caused by w a t e r atmos- pheric vapor; five different a;t;lsospsreric prof i les were use5 in the stme manner as just described for radiamzter unit 4. purposes, identical.
Figure 6 shows the calibration The carrres were obta€ned by integrating the
buy of the curves are, for practical
Once the calibration values are knavn for the
r&aueter unit 4, calibration of the other units that were available were un- dertaken by making simultaneous meBsumments with radiometer un i t 4 and the un- known &ou&er. lJo(0. 8731 )/Jo(0.S435)
The t?chniqw used for finding the calibration values for the ancalibrated racliumetcrs is as follows: 1
1. Obtain tm, sets of sinualtaneous readings wLth radiamcter unit 4 and the unbnown inotrarent for the 0.87- and 0.9435-8aicnneter channels. sets should be comared t o elimimte mading or writing errors.
The twcr
2. use fo r the calibrated photometer and the
for the calibrated photcmeter t o cal -
10
3. U s e figure 6 t o calculstc precipitable water in the slant path based on measurements from radiometer un i t 4.
4. U s e the precipitable water fran step 3 t o Find the transmission for the uncalibrated radiometer ?u ming Ffgure 6.
from the uncaUbrated (0.8730) and J (0 -9435) 5. Usemeasuredvalwe, J
radiumter and the FU from step 4 t o calculate the calibration constants
Table I gives cellbration values Obtained in this mariner for all radiometers tha t were atrailable in the summer of 1W3.
6 . Obtain a second set of simul~eous measurements on a different day, perform steps 1 through 5, and cross correlate the values of water obtained With radiaaeter unit 4 and the instrument being calibrated. Figure 7 is an example of such a calibration verification.
If radiameter uni t 4 becanes uncalibrated, units 1, 2, or 3 should be used a8 the 8tand&rd.
In several studies, ~ b r r i s s e y and &ousai~les (ref. 191, hmes et al. (ref. 201, and others discuss temperature-induced errors in humtdity, problems in repetition, and problems with response of radiosondes that served as the
Brousaides and Morrissey (refs. 2Z and 22) discuss an improvement i n the radiosonde that elim- inates the systematic trend t o messufe drier day tin^ profiles than infrared and other watell.-vapor meamaremenix in the Barbados Oceanographic and Meteorological -% pmJect. ?heir redesign of tk ~~Uoacmde that O T B ~ implemented by %he Bat ionah C k d c aad Atmospheric Br3m.Iniration in 1972 shows an increase of 30 percent re la t ive humidity (e.&, From 50 t o 88 percent re la t ive humidity)
for some cases m e 500 x 10 E/m (500 millibars). All the data used i n t h i s report were from the improved radiesonde system.
=n*d for measurement of Water vapor for this technique.
2 2
According t o Brousaides (ref. 23) , most radiosondes have laboratory accu- racies of *3 percent or less for relative humidity at tempe-atures above 298 K; huwever, at temperatures below 253 K, the emor increases t o 26 percent with a mean bias of as much 88 7 percent. However, at present, no assessment of the absolute accursrcies for temperatwe and re lat ive humidity is available t o cal- culate the accuracy of &terminlng total water f2ma radiosonde data.
11
sensor requiring tens of d n u t e s t o nteasure . It is the authors' opinion tha t the relatin radiosonde is accurate t o about 210 percent fo r values of re lat ive humidity greater than 20 percent. I f absolute radiance could be derived for the two channels of the radiometer, the accuracy of the radiosonde would no longer be "the limiting factor i n determini= precipitable water In the atmosphere.
mofile w i l l alwws be limited. midi ty measure!nent from the
The technique of using the solar radianmeter al lows a self-consistent m e a s - 4
urement of precipitable water t o be made that can be used as an input t o a trans- mission look-up table for other spectral bands of water-vapor absoiption. such, precipitable water serves as a convenient, physically meaningful, in te r - mediary quantity.
As
Igndon B. Johnson Space Center National Aeronautics and Space Administration
Houston, Texas, February 26, 1974 951-16-00-00-72
1.
2.
3.
4. .
5.
6.
7.
8.
9-
10.
11.
12.
13.
Jeske, K. W.: Ertreme Atmosphere Models, 1973. BASA TM X-58i12, 1974.
plhxrcray, D. C.; and Barker, D. B.: Balloon Borne Humidity and Aerosol Sensors. HASA CR-6l379, Feb. 1971.
Anon. : Remote Multispectral Sensing in Agriculture. Res. Bull. Bo. 844, Laboratory for Remote Sensing, Vol. 3, Annual Report, Purdue Univ., Sept. 1968.
Bauer, Eaarvin E.; and C i p r a , Jan E.: Identification of Agricultural Crops by Computer Processing of ERTS MSS Data. sults OWained From the Earth Resources Technology Satellite-1, Vol. I: Technical F'resen+,ations, Sec. A, Stanley C. Freden, Ehrico P. Mercanti, and Margaret A. Becker, caanpilers and eds., HASA SP-327, 1973, pp. 205-2l.2.
Symposium on Significant Re-
Flawers, E. C.; McComick, R. A.; and Kurfis, K. R.: Atmospheric Turbidity Over the United States, 1961-1966. Dec. 1965, pp. 955-962.
J. Appl. Meteoro: ., vol. 8, no. 6,
Volz, F. E.: Some Results of Turbidity Networks. T e l l u s , vol. AI, A i r Force Cambridge Research Laboratories, Nov. 1968, pp. 625-629.
Shw, G. E.; Reagan, J. A.; and Heman, B. M.: Investie. i m s of Atmos- pheric Ektinction Using Direct Solar Radiation Measurements Made With a Multiple Wavelength Radiometer. J. Appl. Meteorol., vol. U, no. 2, M a r - 1973, pp. 374-380.
Reeser, W. K.: Aerosol Size Distribusion Detection for PREPS Using Simple Processing Techniques. Inc. (NASA Contract NAS 9-l2200), Mar. 1972.
LEC/HASD 640-TR-091, Lockheed Electronics Co.,
Reeser, W. K.: Aerosol Models Used for PREPS. LE/HASD 640-TR-109, Lockheed Electronics Co. , Inc. (NASA Contract HAS 9-122001, M q 1972.
Kleen, R. H.: PREPS-ER"S-A. M a r . 1972.
Sttn Photometer F i l t e r Selectfon and Gaseous Absorption for LEC/HBSD TM 641-471, Lockheed Electronics Co., Inc.,
Anon.: EIiTs Data Users Handbook. NASA Technology Sa te l l i t e , Goddard Space Flight Center, Document 7lSD4249, 1972.
Anon. : Skylab Experiments. Government Printing Office 3300-0423, AM. 1972.
Reeser, W. K.: A Feasibil i ty Study on the Use of a Sun Photometer i n Gathering Aerosol Optical Depth Data for PREPS. Lockheed Electronics Co., Inc. (NASA Contract NAS g-l22OO), Mar. 1972.
LEC/HASD 640-TR-086,
13
14. Cousln, S. B. ; Anderson, A. C.; Paris, J. F. ; and Potter, J. F. : Signifi- rer.t Techniques in the Processing and Interpretation of' ERTS-1 Data. S:.iposium on Significant Results Obtained From the Earth Resources Tech- logy Satellite-1, Vol. I: Technical Presentations, Sec. E, Stsnley I . Freden, Enrico P. Mercanti, and Margaret A. Becker, compilers and ens., NASA SP-327, 1973, pp. U51-1158.
15. Htmaen, James E.: Radiative Transfer by Doubling Very Thin Layers. Astrophys. J., vol. 155, no. 2, Feb. 1969, pp. 565-573.
16. Pot .or, John F.: Scattering and Absorption i n the Earth's Atmosphere. I-*oceedings of the Sixth International Symposium on Remote Sensing of
$ E:.wironment, Vol. 1, Oct. 13-16, 1969, Willow Run Laboratories, Univ. '! G? Michigan, pp. 415-429.
17. Deutschman, E ldne M.; and Calfee, Robert F.: Two Computer Programs t o Pcoduce Theoretical Absorption Spectra of Water Vapor and Carbon Dioxide. Env. Sci, Services A b . Tech. Rept. 1333 31-ITSA 31, Apr. 1967.
18. Elte-man, Louis: Vertical-Attenuation Model With Eight Surface Meteoro- iod;ical Ranges 2 t o 13 Kilometers. Air Force Cambridge Research Laboratories, M a r . 1970.
Environmental Research Paper 318,
19. .b r r i ssey , James F.; and Brousaides, Frederick J.: Temperature-Induced Errors i n the -476 H-dd i ty Data. act . 1970, pp. 805-808.
J. Appl. Meteorol., v01. 9, no. 5 ,
20. Bar.ies, Stanley L.; Henderson, James H.; and Ketchum, Robert J.: Rawi.,- . sonde Observation and Processing Techniques at the National Severe Storm
Laboratory. NOM TM ERL NSSL-53, Apr. 1971.
2.1. Broaddes , F. J.; and Morrissey, J. F.: Lmproved Humidity Meltsurements With a Reaesigned Radiosonde Humidity Duct. Bull. Am. Meteorol. SOP., vol. 9, 1971, pp. 870-875.
22. Brousaides, Fr.A.arick J.; and Morrissey, James F.: Residual Temperature- Induced Y.--;idity Krrors i n the National Weather Service Radiosonde, P a r t 1 instrumentation Paper No. 184, A i r Force Cambridge Research Lakcratories , TEt-73-0214, Apr. 1973.
23. B:.wsaides, Frederick J. : i n Radiosonde Systems. or1 .tories , 19?1
An Assessment of t h e Carbon Humidity Element TR-73-0423, A i r Force Cambridge Research Lab-
14
TABU I.- SOLAR RADIOMETER CALIBRATION VALUES
Solar radiometer unit number
1
2
3
4
5
6
7
8
9
10
11
i2
13
14
15
~~ ~ ~~
Calibration value
1.24
- 9 1
.67
-44
.27
.381
- 39
.298
- 30
.22
.407
-- .284
330
.271
. 0.5
.4
.3
.2
.1
0 0.9350 .9400 .9450 ,9500 .9550
Wavelength, k, p m
Figure 2.- F i l t e r function f o r the water band f o r so l a r radiometer unit b .
1 .O
.9
-8
-7
.6
e 5
e4
-3
.2
e 1
0 10 530 10 570 10 610 10 650 'LO 690
Wave number Figure 4.- Calculated transmission through t h e atmosphere smoothed by a
triangular f i l t e r funct ion 20 centimeters-' wide. 1972, radiosonde, 0.202 centimeter p rec ip i t ab le water.
Albany, N.Y., Jar,. 10,
10 .o
If-
E 0 ul ul
. .- .- 5 2 1.0 S
.ad
0
a, c P In
.- L
E" z
Photometer measurements
-
Radiosonde data for - theoretical transmissions
0 Houston, Feb. 29, 1973 0 Houston, June 4, 1973 5 Houston, June 2, 1973 0 Houston, May 29, 1973 V Houston, Nov. 27, 1972 Q Houston, Nov. 28, 1972 D Houston, Sept. 18, 1973
1 Key West, Fla., July 17, 1972 2 Lake Charles, La., Sept. 5, 1972 3 Brownsville, Tex., Oct. 20, 1972 4 Boothville, La., Nov. 11, 1972 5 NewYork, N.Y., Dee. 10, 1 9 7 2 6 TUCSOII, Ariz., June 30, 1972 7 Albany, N.Y., Jan. 30, 1973 8 Caribou, Maine, Jan. 10, 1973 9 Ely, Nev., Dee. 5, 1973
10 Rapid City, S. Dak., Jan. 10, 1973
Curve for theoretical calculations using filter functions and atmospheric transmissions from compressed line-
by-line model i = 0.44 J(o .9435)
.1 I I I I I I I 0 1 2 3 4 5 6 7
Precipitable water, cm
Figure 5.- Precipi table water ca l ibra t ion for solar radiometer unit L.
20
9 rl
23373333
004 0 Don0
Y . ' - E > :
Ln // rl
21
6
4
2
0
/ 0 Unit 1 0 Unit 2 A Unit 3
I - 2 4 6
Solar radiometer unit 4 precipitable w !r, cm
Figure 7.- Calibration v e r i f i c a t i o n be.,ween solar ra3io.T.et e r z .
22
