Atmospheric radiation measurements enhanced shortwave experiment (ARESE): Experimental and data details

JOURNAL OF GEOPHYSICAL RESEARCH, VOL. 102, NO. D25, PAGES 29,929-29,937, DECEMBER 27, 1997 Atmospheric Radiation Measurements Enhanced Shortwave Experiment (ARESE): Experimental and data details Francisco P. J. Valero, Anthony Bucholtz, Brett C. Bush, Shelly K. Pope, William D. Collins,  and Piotr Flatau Atmospheric Research Laboratory, Center for Atmospheric Sciences, Scripps Institution of Oceanography, University of California, San Diego, La Jolla Anthony Strawa and Warren J. Y. Gore NASA Ames Research Center, Moffett Field, California Abstract. Atmospheric Radiation Measurements Enhanced Shortwave Experiment (ARESE) was conducted to study the magnitude and spectral characteristics of the absorption of solar radiation by the clear and cloudy atmosphere. Three aircraft platforms, a Grob Egrett, a NASA ER-2, and a Twin Otter, were used during ARESE in conjunction with the Atmospheric Radiation Measurements (ARM) central and extended facilities in north central Oklahoma. The aircraft were coordinated to simultaneously measure solar irradiances in the total spectral broadband (0.224-3.91 /xm), near infrared broadband (0.678-3.3/xm), and in seven narrow band-pass (-10 nm width) channels centered at 0.500, 0.862, 1.064, 1.249, 1.501, 1.651, and 1.750/xm. Instrumental calibration issues are discussed in some detail, in particular radiometric power, angular, and spectral responses. The data discussed in this paper are available at the ARM ARESE data archive via anonymous FTP to ftp.arm.gov. 1. Introduction The amount of solar energy that reaches the top of the Earth's atmosphere has been measured with great accuracy. However, the distribution of this energy in the planetary sys- tem (atmosphere plus surface) is still one of the least quanti- fied parameters required to understand and describe the cli- mate system [Kiehl and Tremberth, 1997]. According to current theoretical understanding the absorp- tion of solar radiation by an atmospheric column containing clouds is about equal to the absorption by the same column in the absence of clouds. In contrast, experimental investigations during the last several decades have shown that the observed magnitude of the absorption of solar radiation by the cloudy sky exceeds model predictions by varying amounts [Stephens and Tsay, 1990]. More recently, Cess et al. [1995], Ramanathan et al. [1995], Pilewskie and Valero [1995], Liet al. [1995], and Waliser et al. [1996] confirm the previous observational findings at tropical latitudes. Kondratiev et al. [1996] and Chou and Zhou [1997] also indicate that clouds absorb more solar radi- ation than predicted by theory. Valero et al. [this issue] and Zender et al. [this issue] report similar findings at midlatitudes over the North American continent. Cess et al. [1995] used collocated satellite-surface measure- ments at four different locations (American Samoa, Barrow, Boulder, and Cape Grim) to evaluate the absorption of solar radiation by clouds. A comparison with the European Centre for Medium-Range Weather Forecasts (ECMWF) Model and Now at National Center for Atmospheric Research, Boulder, Col- orado. Copyright 1997 by the American Geophysical Union. Paper number 97JD02434. 0148-0227/97/97JD-02434509.00 Version 2 of the National Center for Atmospheric Research Community Climate Model (CCM2) shows that the observa- tions and theoretical calculations differ by a substantial amount as discussed in the previous paragraph. Pilewskie and Valero [1995] made measurements of cloud absorption during the Tropical Ocean Global Atmosphere-Coupled Ocean At- mosphere Response Experiment (TOGA-COARE) and the Central Equatorial Pacific Experiment (CEPEX) using direct, simultaneous aircraft observations above and below clouds in major tropical cloud systems. Studies byArking [1996] and Imre et al. [1996] claim that it is enhanced clear-sky absorption that produces the measured absorption in excess of theory predictions. Other studies by Stephens [1996], Li and Moteau [1996], Francis et al. [1997], Ackerman and Toon [1996], and King et al. [1990] claim small or negligible excess absorption by clouds. Stephens [1996] and Francis et al. [1997] base their conclusions upon indirect esti- mates of the absorption by clouds; they compare modeled and measured cloud albedos and transmittance assuming, following the standard understanding, that any excess absorption must exclusively occur in the near-infrared spectral region where water absorption bands are known to exist. Ackerman and Toon [1996] use purely theoretical calculations also assuming that absorption by clouds occurs only in the near-infrared; Li and Moteau [1996] consider the possibility of aerosol effects; King et al. [1990] find, from observations within a marine stratocumulus cloud, comparatively small discrepancies be- tween theory and spectral diffuse domain observations used to determine cloud single scattering albedo. Attempts have been made to explain the difference between measurements and calculations by postulating that horizontal transport of photons by clouds results in an \"apparent\" absorp- tion [Ackerman and Cox, 1981; Stephens, 1996]. Figure 1 illus- trates schematically the possible case of a single, isolated 29,929

29,930 VALERO ET AL.: ARESE EXPERIMENTAL AND DATA DETAILS APPARENT ABSORPTION the net fluxes at each level in the atmosphere. The net radiative flux is the difference between the downwelling and upwelling fluxes at each altitude. Net solar flux must be acquired simul- taneously at both flight altitudes to minimize the uncertainties associated with cloud advection, evolution, three-dimensional effects, etc. The net flux at a given altitude is the absorption by the column and the surface below. The aircraft, stacked at different altitudes, flew tracks over the surface stations. In this manner, it was possible to obtain coeval measurements of radiative fluxes from which the ab- sorption of radiation by the atmosphere was computed [Valero et al., this issue]. Surface observations provided the radiative flux transmitted through the column. Radiative flux measurements at the stratosphere, the upper troposphere, and the lower troposphere were made using the NASA ER-2, a Grob Egrett, and a Twin Otter aircraft, respec- tively. All three aircraft were equipped with identical radio- metric instrumentation, as were the surface sites. Table 1 lists the dates, observational platforms, and atmospheric conditions encountered during the ARESE radiometric observations. SURFACE Figure 1. This sketch illustrates the idea that horizontal scat- tering of photons by clouds can result in an apparent absorp- tion. cloud, for which two fixed hemispherical field of view radiom- eters, located above and below the cloud, could miss photons scattered in nearly horizontal directions. The uncertaimies described above, along with the lack of a possible explanation or a definitive mechanism, led to the Atmospheric Radiation Measurements (ARM) Enhanced Shortwave Experiment (ARESE). The primary goals of AR- ESE were to obtain sets of measurements to determine accu- rately the absorption of solar radiation by the clear and cloudy atmospheric column at midlatitudes, to investigate the spectral regions where the excess absorption occurs, and to answer the question, Is the measured absorption real, or is it the result of \"apparent absorption\" resulting from the horizontal transport of photons by clouds? A secondary but fundamental goal was to begin the process of identifying the physical mechanism responsible for the absorption, or the inability of the models to reproduce the observations. In this paper, the experimental aspects of the research are summarized; a detailed analysis of the data is made in two other papers, Zender et al. [this issue] and Valero et al. [this issue]. 2. Exoeriment Following the previous discussion and to meet the goals described above, the experimental emphasis of ARESE in- volved the acquisition of radiometric data by multiple, coordi- nated aircraft and from surface sites. The aircraft covered the range from the lower stratosphere to the low troposphere. Surface observations were made from the ARM Clouds and Radiation Testbed (CART) central facility located at 97.48 ø longitude and 36.59 ø latitude and from secondary surface sta- tions (extended facilities) which are part of the ARM Southern Great Plains (SGP) site. The absorption by the atmospheric layer between two alti- tudes (the flux divergence) is defined as the difference between 3. Instrumentation The basic instruments, in addition to the CART site facili- ties, required to meet the ARESE objectives are included in the radiation measurement system (RAMS). The RAMS is a multiple array of radiometers covering the spectrum from the near-ultraviolet to the far (thermal) infrared. Components of the RAMS vary depending on the purpose and platform being used. 3.1. Aircraft During ARESE the following aircraft radiometers, pointing in zenith and nadir directions, were utilized: 3.1.1. Twin Otter and Egrett. (1) Total Solar Broadband Radiometers (TSBR) covered the spectral range from 0.224 to 3.91 /m [Valero et al., 1982]. (2) Fractional Solar Broadband Radiometers (FSBR) covered the near-infrared spectral range from 0.678 to 3.3 /xm [Valero et al., 1982]. (3) Total Direct Diffuse Radiometers (TDDR) covered seven spectral channels in the solar spectrum centered at 0.500, 0.862, 1.064, 1.249, 1.501, 1.651, and 1.750/m with a band pass of about 10 nm [Valero et al., 1989]. 3.1.2. ER-2. The ER-2 was equipped with the RAMS, a camera, a lidar [Spinhirne et al., 1996] and the Modis Airborne Simulator (MAS) [King et al., 1996]. 3.2. Surface Zenith pointing RAMS instruments identical to those in the aircraft were installed at the CART central facility and the extended facilities at Byron and Ringwood in Oklahoma. 4. Radiometric Calibrations The RAMS broadband radiometers use electrically cali- brated pyroelectric detectors, which are by themselves absolute devices [Geist and Blevin, 1973; Valero et al., 1982]. Each in- strument is individually built, quality controlled, and charac- terized. Components of the electronic circuits of each radiom- eter are custom mashed to the individual detector elements. The RAMS broadband radiometers are very different from other commonly used instruments. For this reason, we will briefly describe their operational principles since they are im- 21562202d, 1997, D25, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/97JD02434 by University of California San Diego, Wiley Online Library on [24/09/2026]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License

VALERO ET AL.: ARESE EXPERIMENTAL AND DATA DETAILS 29,931 Table 1. ARESE Radiometric Observations Day Date Aircraft Ground Stations Conditions 268 Sept. 25 ER-2, Egrett, Twin Otter broken clouds 269 Sept. 26 ER-2 scattered clouds 270 Sept. 27 ER-2 (satellite calibration, Gulf of Mexico) 271 Sept. 28 ER-2 low clouds and cirrus 272 Sept. 29 ER-2, Egrett, Twin Otter scattered clouds 276 Oct. 3 ER-2, Twin Otter clear sky 279 Oct. 6 ER-2 (survey of hurricane-damaged areas) 284 Oct. 11 ER-2, Egrett, Twin Otter L, R clear sky 285 Oct. 12 ER-2 (satellite calibration, Gulf of Mexico) 286 Oct. 13 Egrett, Twin Otter L, B broken-to-solid clouds 287 Oct. 14 ER-2 (satellite, calibration south central United States) 288 Oct. 15 L, B clear sky 290 Oct. 17 ER-2, Egrett, Twin Otter L, B, R clear sky 291 Oct. 18 L clear sky 292 Oct. 19 ER-2, Egrett, Twin Otter L, B, R clear sky 294 Oct. 21 ER-2 (satellite calibration, south central United States) 295 Oct. 22 ER-2 B, R clear sky 296 Oct. 23 ER-2 scattered clouds (transit across United States) 297 Oct. 24 Egrett, Twin Otter L, B, R thin cirrus 298 Oct. 25 L, B, R scattered clouds 299 Oct. 26 Egrett, Twin Otter L, B, R solid cirrus deck; broken clouds 301 Oct. 28 Twin Otter L, B, R clear sky 303 Oct. 30 Egrett, Twin Otter L, B, R thick uniform stratus deck 304 Oct. 31 L, B, R thick uniform stratus deck 305 Nov. 1 Egrett, Twin Otter L, B, R broken-to-solid stratocumulus Ground station key: L, Lamont; B, Byron; R, Ringwood. portant to follow the calibration procedures. The key to the operation of the RAMS radiometers is the comparison of the radiative signal incident on the detector, with an internally generated electrical signal synchronized 180 ø out of phase with the optical chopper. This signal is conducted through a resis- tive heating coating deposited on the detector surface. The resulting thermal signal is in opposition to that generated by the chopped optical radiation. The electrical signal is then adjusted by a microprocessor until a null is detected. At this point the electrically and radiatively generated signals are equal. The optical power absorbed by the detector is deter- mined by calculating the electrical power required to produce a null output. Optical chopping helps to eliminate drift, and null balanced operation render the measurements insensitive to changes in detector responsivity, to amplifier gain, and to ambient temperature. Because the detectors are absolute devices in the sense ex- plained above, the calibration of the RAMS broadband radi- ometers is in reality a characterization of the optical front end of the instruments. The angular, power, and spectral response of the optical system are the fundamental elements of our calibration procedure. The RAMS was calibrated before, during, and after the experiment. The calibration includes (1) power calibration, (2) angular response calibration, (3) spectral response calibration, and (4) surface and in-flight comparisons. 4.1. Power Calibration The power calibration of the broadband radiometers was done using two different methods; (1) comparison with a Hickey-Frieden absolute cavity radiometer, and (2) high- altitude calibration from the ER-Z aircraft, flying at approxi- mately 20 km altitude, using the Sun as a reference standard. The results of the calibrations using the cavity and high- altitude methods are compared in Table 2 for the two instru- ments that were utilized as reference instruments for the sur- face comparisons described in another section. It was found that the broadband power calibration coefficients obtained with methods 1 and 2 agree to better than 1% which is suffi- cient to meet the precision requirements of the experiment. The spectral differences between the broadband solar radia- tion at ER-2 altitude and at the surface are possibly partly responsible for the small differences noted. A correction for differences in spectral properties was not possible because a detailed spectral characterization of the cavity radiometer was not available. The broadband data were processed using the cavity radiometer calibrations validated by comparison with the high-altitude calibrated instruments. The power calibrations of the TDDR were also done using two app.oaches; first, the National Institute of Standards (NIST) absolute lamp calibrations method and second, an ER-2 high-altitude calibration, similar to the one employed for the broadband instruments but using now the solar irradiance in each TDDR spectral channel. The NIST calibration values were used as preliminary; the high-altitude calibrations were used for the reduction of the TDDR data. In summary, the power calibration of the solar radiometers includes the follow- ing steps: 4.1.1. Cavity radiometer reference calibration at the sur- face. Before the field phase of the experiment, the broadband radiometers were compared with a Hickey-Frieden cavity ra- diometer, accurate to better than 1%, following the established method: point the cavity radiometer to the Sun under clear-sloj, Table 2. RAMS Radiometer Calibration Constants Percent Detector ER-2-Derived Ground-Derived Difference 7A (TSBR) 3.28 X 10 6 3.26 X 10 6 0.6 18A (FSBR) 1.67 x 106 1.66 x 106 0.8 21562202d, 1997, D25, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/97JD02434 by University of California San Diego, Wiley Online Library on [24/09/2026]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License

29,932 VALERO ET AL.: ARESE EXPERIMENTAL AND DATA DETAILS 20% , , ,    15% 1 ,- 10% ß  0.8 o  5% o 0.6   _. - 0% u ,- -5% 0.4  -10% -o 0.2 -15% -20% 0 -80 -40 0 40 80 angle Figure 2. The angular response of a solar broadband radi- ometer is shown as a function of angle of incident light. The scale for the solid line (cosine function) and the dashed line (normalized detector signal) is on the right, while the scale for the symbols (percent deviation from cosine) is on the left. conditions and then determine by comparison the radiometer calibration constant after subtracting the diffuse radiation field seen by the hemispherical field of view instruments and not seen by the relatively narrow field of view of the cavity radi- ometer (-5.5ø). The diffuse field is determined by blocking the direct component of the solar flux using a shadow disk that blocks the hemispherical radiometer's field of view by an angle equal to the field of view of the cavity. Errors in the reference total irradiance depend on the mea- surement of the diffuse flux which is measured as described above. Possible errors include the following: 1. Errors in calibration of the hemispherical field of view radiometer used to determine the diffuse radiation field are minimized using the same radiometer (the one under calibra- tion) to determine total and diffuse fields. The diffuse field is determined by blocking the direct solar radiation with a shadow disc that covers an angle identical to the field of view of the reference cavity; the total flux is the measurement ob- tained with the shadow disc removed. From these measure- ments the fraction of the total flux corresponding to the direct radiation is obtained, and a calibration factor is deduced by comparison with the absolute cavity radiometer value. 2. Mismatching occurs between the blocking of the direct solar radiation and the 5.5 ø field of view of the cavity radiom- eter. The blocking disc size and distance from the radiometer's aperture require careful and accurate adjustment to properly match the field of view observed by the cavity instrument. Typical errors in this adjustment result in calibration errors of less than 0.2%. 3. Deviations from cosine response in the hemispherical radiometer render the diffuse field measurements sensitive to anisotropies in the observed scene (broken clouds, for exam- ple). The angular response of the radiometer is most critical in the evaluation of both the total and diffuse radiation fields. Accurate cosine response is key to accurate flux measure- ments. The angular response calibrations are discussed in sec- tion 5.2 below. 4.1.2. High-altitude, ER-2 calibration using the Sun as a reference. The high-altitude calibration that uses the solar constant as a reference was done during ARESE with data collected from the ER-2 aircraft. The radiometers were flown in the zenith-pointing position in the ER-2, and calibration factors were determined by comparing the known solar con- stant, after making the proper astronomical corrections, with the radiometer signal. Corrections for relative aircraft-Sun po- sitions were made using data from the aircraft's navigational system and the algorithms described by Hammer et al. [1991]. A radiative transfer model (MODTRAN, including the stan- dard stratospheric aerosol) is used to correct for atmospheric absorption and scattering above the ER-2 altitude. As many radiometers as possible were flown in the ER-2 zenith position by rotating them between the Egrett, Twin Otter, and ER-2 aircraft. The high-altitude calibrations were utilized for the calibra- tion of all radiometers, but it was most important for the calibration of the TDDR that requires calibrations under flight conditions, particularly due to the temperature sensitivity of the detectors. The TDDR channels could not be absolutely calibrated to better than about 5% because of the complicated angular response of their asymmetric optical system combined with uncertainties in the aircraft's navigational data. However, the precision of the ARESE TDDRs was about 1%. 4.2. Angular Calibration The complex front end of the optical head of each radiom- eter is composed of reflective light collimating elements, dif- fuser, chopper, windows, and optical filters. These components are individually polished optical elements and have their own angular characteristics. High-quality radiometry demands the accurate knowledge of such angular response, not only as a quality control element but also as an important part of the calibration and data retrieval processes. Each instrument is calibrated for angular response in the laboratory by exposing the radiometers to a collimated radiation field and rotating the instrument in zenith and azimuth to obtain the angular re- sponse in a hemisphere. From our experience, optically sym- metric radiometers (i.e., broadband radiometers) require only a zenith angular response calibration. Multiple channels in- struments possessing azimuthal asymmetries (like the TDDR) require zenith and azimuth calibrations in each spectral channel. Figure 2 shows the angular response of a solar broadband radiometer. The ordinate is the signal, and the abscissa is the angle of incidence of the collimated broad spectral band radi- ation. Note that the deviations from a perfect cosine response are within about _+2% up to almost 70 ø . The angular response varies from unit to unit, thereby the requirement for individual angular calibration of the radiometers. In many cases the de- viations from perfect cosine response are within 1%. During the data processing the angular response calibrations are used, as described in the following section, to correct data to an ideal cosine response after estimating the direct and diffuse compo- nents of the flux. Figure 3 depicts the angular calibration from a single chan- nel of a Total Direct Diffuse Radiometer. The azimuth and zenith angular responses are combined in the figure. The z axis represents the angular response of the detector relative to an ideal cosine response. Just as with the broadband radiometers, the angular response calibrations are used to correct the data for cosine response. The very accurate angular responses shown in Figures 2 and 3 eliminate the need for large corrections to be applied to the measured fluxes. However, in order to achieve the best possi- ble results, we correct the measurements as follows. 4.2.1. Direct field correction. First, the ratio between di- rect and diffuse fluxes is estimated using the TDDR instrument 21562202d, 1997, D25, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/97JD02434 by University of California San Diego, Wiley Online Library on [24/09/2026]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License

VALERO ET AL.: ARESE EXPERIMENTAL AND DATA DETAILS 29,933 and/or model calculations (i.e., MODTRAN). The direct ra- diation flux is corrected by directly applying the angular re- sponse calibration values that correspond to the time varying angles of incidence determined from Sun position calculations (solar zenith and azimuth). The correction to the total flux is less than 1% for the angular response illustrated in Figure 2 and a total to diffuse ratio of 10. Accurate measurements of the level of the radiometer optics are also required. This latter measurement is relatively simple for surface instruments. For aircraft the position of the radi- ometers relative to the Sun depends on the aircraft attitude, and the data are treated as described in a following section. 4.2.2. Diffuse field correction using the angular response function. The radiometer's angular response calibration is also used to correct the diffuse field. This is accomplished by first approximating the diffuse field as an isotropically emitting flat surface. Using this characterization, the radiation field is integrated over a hemispherical field of view, first using the measured angular response function and second, using an ideal cosine response. The ratio of these integrations yields the ap- propriate diffuse field correction which amounts to less than 0.2% for the angular response in Figure 2 and a total/diffuse ratio of 10. 4.3. Spectral Calibration The filters employed in the broadband and spectral radiom- eters were calibrated for wavelength response by the manufac- turer using normal spectroscopic techniques. This calibration, together with the power and angular calibrations, comprises the overall radiometer calibration. Figure 4 depicts the spectral band passes for the broadband radiometers and the center wavelength of the TDDR spectral channels (rectangular 10 nm full width at half maximum transmission). 4.4. Surface and In-Flight Comparison of Radiometers 4.4.1. Surface comparisons. The final step in our quality control and calibration procedures involves the comparison of all instruments with each other, by exposing them simulta- 1.4 1.2 Figure 3. The azimuth and zenith angular responses are combined in this plot of the angular calibration of one of the total-direct-diffuse radiometer channels. 0.8 0.6 i \ \ '- 0.2 0.0 , , I , 0.2 0.4 0.6 0.8 1.0 3.0 wavelength (pm) Figure 4. The spectral response curves are shown for the radiometers. The total solar broadband (solid line) has an effective square band pass of 0.224-3.91 /m, and the near- infrared broadband (dashed line) effective square band pass is 0.678-3.3/m. Vertical lines indicate the center wavelengths of the seven narrowband channels of the total-direct-diffuse ra- diometer. neously to the open atmosphere radiation field. Comparisons at the surface are made for clear and cloudy atmospheres. This step is very useful in detecting errors and differences between instruments and to determine errors useful to estimate preci- sion. If a large error is detected, the whole calibration process is traced back until the source is discovered and corrected. 4.4.2. In-flight comparisons. On clear-sky days during ARESE, when it was logistically possible to fly the Egrett and Twin Otter aircraft at the same altitude and very close to each other \"wing to wing,\" we compared the radiometric instrumen- tation for performance under actual flight conditions. A special flight plan was designed to help correct for offsets in the mounting of the instruments and in the aircraft's inertial nav- igational systems. The flight plan for instruments comparisons consisted of four data legs flown in directions directly toward and away from the Sun, as well as with the Sun perpendicular to the right and the left of the aircraft (Figure 5). The two former legs are useful to determine aircraft pitch offsets, and the two Sun perpendicular legs are useful to determine roll offsets. Figure 6 depicts the radiometer comparison results after applying all aircraft navigational corrections and radiometric calibrations as discussed above. The average flux measurement is indicated for each instrument. Comparison of these values gives an idea of the precision of the instruments. Averaged values, tabulated in the figures, indicate excellent agreement. However, one particular case exhibiting large discrepancies should be noted. On October 17, 1995, the third comparison leg shows a discrepancy of about 70 W/m 2 between the Egrett and the Twin Otter zenith broadband radiometers. Such dis- crepancy is the result of the Egrett's tail partially shadowing the radiometer during the \"away from the Sun\" flight leg. 5. Airborne Measurements The high-altitude Egrett was used as an above cloud plat- form flying in coordination with the low-altitude, below cloud, Twin Otter. The purpose of these aircraft was to provide si- 21562202d, 1997, D25, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/97JD02434 by University of California San Diego, Wiley Online Library on [24/09/2026]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License

29,934 VALERO ET AL.: ARESE EXPERIMENTAL AND DATA DETAILS 36.8 36.7 - _ 36.6 - 36.5 - 36.4 -97.4 i i i i to s from sun \"' ø\"i I -  .-' to sun un . - Iiil -- I I I I -97.2 -97.0 -96.8 -96.6 -96.4 longitude Figure 5. The solid curve in this plot is the track for the in-flight comparison of October 17, 1995. The dashed curve is the track for the October 19 in-flight comparison, displaced by 0.5 in longitude to simplify the plot. The straight segments are not exactly perpendicular or parallel due to changing solar azimuth. multaneous and collocated measurements of net radiative fluxes at specified altitudes for a variety of conditions: cloudy, clear, and partly cloudy atmosphere. The Twin Otter top speed is below the Egrett's slowest operational speed. Such differ- ence in speed was compensated for by periodically introducing 360 ø turns in the Egrett flight track (Figure 7). After each 360 ø turn, the Egrett was behind the Twin Otter and then pro- ceeded past it, performing another data run. Data taken during turns and other maneuvers were eliminated from the data set. The Egrett and Twin Otter aircraft were based at the Black- well-Tonkawa airport, located 20 km to the northeast of the CART site. Data from the onboard instruments were teleme- tered to a receiving station located at the airport. The telem- etry stream was recorded on tape in addition to being piped to a local computer network for real-time, in-flight monitoring. The Twin Otter carried an onboard tape recorder as backup. Data rates were 15.625 Hz for the Egrett and 12.5 Hz for the 5OO 400 300 200 , Egrett Twin Otter b% &(watts) total down 319.6 322.0 0.8 2.4 near-ir down 193,6 189.9 1.9 3,7 total up 63.5 65.2 2.7 1.7 near-Jr up 45.1 46.6 3.3 1.5 total down  . near-ir down 100 total up .... . .... { ............ i ...... near-it up 0 ,, I , , , ...... I ......... I ......... I , , 21.8 21.9 22.0 22.1 time (UT hours) Figure 6. The two aircraft were flown side-by-side at the end of the October 26, 1995, flight in order to compare fluxes measured by the two sets of radiometers. The circles are Egrett values, and the open squares are Twin Otter values. Each symbol is a 10-s average, and overall average values are shown in the inset table. Each of the four \"legs\" of data span about 4-5 min in time and 15 km in distance. The sky was mostly clear during this time. Twin Otter. During preflight operations on the ground the radiometers were covered so that the \"dark\" signals (offsets) could be measured. The aircraft typically took off at midmorn- ing and flew in coordination for 2 to 5 hours. For the longer flights there was an interruption of about an hour while the Twin Otter returned to the airport and refueled. A total of 10 coordinated research flights were carried out by these air- planes. The primary objectives of the ER-2 were to calibrate satel- lites and radiometers and to measure broadband and spectral reflectivity and net radiative fluxes for cloudy and clear skies at the tropopause. Throughout ARESE the ER-2 was based at Bergstrom Air Force Base in Austin, Texas. Fifteen ER-2 flights were carried out in support of this experiment. Eight 37.4 37.2 37.0 36.8 36.6 36.4 RON : 4 ................. ! .................... : ...................... i ............. : .... i-- - LAMONT ' RINGWOOD $ -99.5 -99.0 -98.5 -98.0 -97.5 longitude Figure 7. Flight tracks for the Egrett (solid line) and Twin Otter (open circles) are shown for October 24, 1995. The Twin Otter line is obscured by the Egrett line much of the time in this plot, illustrating the excellent coordination of the aircraft. The ground stations (squares) are also shown. 21562202d, 1997, D25, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/97JD02434 by University of California San Diego, Wiley Online Library on [24/09/2026]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License

VALERO ET AL.: ARESE EXPERIMENTAL AND DATA DETAILS 29,935 x lOOO 800 600 40O 200 -;\ 0 800 600 400 200 16 17 18 19 20 21 UT hours Figure 8. Data from the Egrett and Twin Otter broadband detectors are plotted as a function of UT hour for October 19, 1995. For each aircraft, in order of decreasing magnitude, are shown: downwelling solar flux, downwelling near-infrared flux, upwelling solar flux, and upwelling near-infrared flux. The gap around 1800 is when the Twin Otter went to refuel, and the gap around 2030 is when the aircraft were maneuvering to get into side-by-side (i.e., same altitude) formation. The effects of some broken clouds can be seen in the Twin Otter fluxes, particularly around 19 hours. flights took place over the Oklahoma ground sites with six of the flights in coordination with the lower altitude aircraft. The ER-2, flying at a much higher speed in the stratosphere, but on the same ground track, flew over the other two aircraft numer- ous times during the coordinated missions. Four flights were dedicated to satellite calibration. As described above, all ER-2 flights during ARESE were also used to calibrate the zenith viewing radiometers using the Sun as a standard reference source. 6. Navigational Corrections The data acquired from airborne platforms require correc- tions to determine the equivalent \"horizontal platform\" values of the measured irradiances. This is done following the method described by Hammer et al. [1991]. The basic problem is to position the Sun with coordinates fixed to the aircraft. This is achieved by transforming the vector representation of the Sun position in an Earth surface fixed coordinate system to an aircraft fixed coordinate system. Once the data were reduced to a horizontal platform equiv- alent, the angular and other calibrations are applied, as de- scribed in a previous section. The navigational corrections are used to account for only fairly small deviations, less than 2 ø around the mean values in roll or pitch. The data acquired during airplane attitude exceeding the 2 ø limit were eliminated. Figure 8 depicts an example of the irradiances measured on 1000 800 600 400 00 o 12 .224 - 3 91 0 68 - 3 3 /zm  , , , I , , , I , , , I , , , t , 14 16 18 20 UT (hours) 22 24 Figure 9. Broadband data from the CART site (Lament) ground station are plotted as a function of UT hour for Octo- ber 13, 1995. October 19, 1995, by the Egrett and Twin Otter flying at 13.5 and 0.8-1.5 km altitude, respectively. The data during the period from 2030 to 2100 hours UT correspond to radiometer comparison legs that were flown at about 3.7 km altitude. 7. Surface Measurements During ARESE, three ground stations acquired both broad- band and spectral flux irradiances using instrumentation iden- tical to that on the aircraft platforms with the exception of having only zenith pointing radiometers. The locations of these ground stations were determined by the preexisting network of ARM measurement facilities located in northern Oklahoma as well as by the planned flight trajectories of the aircraft involved in the ARESE, thereby providing measurements at the bottom of the atmospheric column examined in the stacked aircraft configuration. Additionally, being located at one of the ARM facilities provided supplemental information necessary for analysis of the data (i.e., ozone and standard sondes, various meteorological readings, LIDAR and RADAR measurements, etc.). Additionally, Baseline Solar Radiation Network (BSRN) measurements and Solar and InfraRed Observation Stations (SIROS) measurements are available for comparison and cross reference. A comparison of the RAMS, BSRN, and SIROS is discussed by Bush et al. [1997] and Bush and Falero [1997]. The three surface stations were located at the central CART facility in Lament and at the Byron and Ringwood extended facilities along the northern and southern ARESE flight tracks (Figure 7). The data acquisition system was located below the three radiometer mounting boxes in a tent to protect it from the natural elements. The stations were operated during all Oklahoma based aircraft missions between October 11 and November 1, 1995, as well as on several selected nonflight days to supplement the overall data set. Typical operation times spanned sunrise to sunset, minus a few hours on either end. Standard operating procedures included a brief (30 min) pe- riod for instrument warm-up as well as \"dark\" signal baseline determinations. All calibrations were completed prior to de- ployment of the instruments to the field and verified after the experiment. Data were acquired at a nominal rate of 10 Hz which was more than adequate for the broadband instruments but nec- essary for the TDDR in resolving features associated with its 21562202d, 1997, D25, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/97JD02434 by University of California San Diego, Wiley Online Library on [24/09/2026]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License

29,936 VALERO ET AL.: ARESE EXPERIMENTAL AND DATA DETAILS 10 , , , i , , , i , , , i , , ' i , , , i , , , E _ : 6 16.6650 16.6655 16.6660 '--' UT (hours) x E 4 o 2 0 , , , i , , , i , , , ! , , , i , . , ! , , , 12 14 16 18 20 22 24 UT (hours) Figure 10. Data f[om the ]064 m chac] of the tota]- dffcct-diffusc [adiomctc[ ale plotted fo[ the same Jtca] as the p[cvJous figu[c. The Jsct shows a small [cgJo of the data to illust[atc the effect of the shadow ling. An important difference between our calibration methods and others is the introduction of the angular response calibra- tions, not just as a quality control procedure but as a necessary practice in high-quality radiometry. The final retrieved irradi- ances incorporate corrections applied on the basis of the an- gular calibration of each instrument. The power calibration of the broadband radiometers is val- idated against the solar constant using the ER-2 high-altitude observations. It is most important, for purposes of model- measurements comparisons and satellite calibration, that the radiometer measurements and models be referenced to the same standard (the solar constant). High-altitude ER-2 cali- brations are employed for the narrow spectral band pass chan- nels of the TDDR. The in-flight \"wing to wing\" comparisons illustrated in Fig- ure 6 validate the precision of the data in real flight conditions. Analyses of the data presented here are reported by Zender et al. [this issue] and Valero et al. [this issue]. shadow ring. Figures 9 and 10 depict samples of the broadband and TDDR data acquired at the Lamont (CART central facil- ity site) Ground Station on October 13, 1995. During this date, atmospheric conditions changed from clear to broken clouds to heavy overcast. The upper and lower curves in Figure 9 show hemispherical downwelling irradiance for the band passes 0.224-3.91 and 0.678-3.3/xm, respectively. The visible portion of the spectrum (0.224-0.678/xm) is calculated as the difference between these two signals. Effects of scattering off the edges of clouds are seen from 1800 to 2130 hours UT when a broken cloud field prevailed. After 2130, overcast conditions predominated. In Figure 10 one of the seven TDDR spectral channels (1064 nm) is displayed. The effect of the shadow ring can be seen in this figure, particularly during the clear atmosphere period. The maximum envelope of this graph represents the variation of the total hemispherical downwelling flux, whereas the min- imum envelope is essentially the variation of the diffuse por- tion of this flux. The resulting difference of these two compo- nents is thus a measurement of the direct solar signal. The direct component of the flux may then be used to calculate optical depths of the atmospheric column. The inset of this figure expands a small region of the TDDR data showing the amount of data needed to resolve a single dip during a shadow ring transit. Without a large enough data rate the dip may not be adequately resolved, and thus the direct and diffuse com- ponents, and any information derived from them, could not be accurately determined. 8. Discussion The general strategy followed during ARESE was briefly discussed with the objective to inform potential users of the data which is freely available at the ARM ARESE data archive via anonymous FTP to ftp.arm.gov. The radiometric data introduced in this paper provides in- formation necessary to assess the magnitude and overall spec- tral distribution of the excess absorption of solar radiation by clouds and of the absorption by the clear atmosphere. A major consideration in the planning of the experimental strategy for ARESE was related to the sampling issues discussed by Ack- erinart and Cox [ 1981]. Acknowledgments. The authors gratefully acknowledge the De- partment of Energy's ARM program and its Unmanned Aerospace Vehicle (ARM-UAV) component, the Strategic Environmental Re- search and Development Program (SERDP) of the Department of Defense, and NASA. This work was also supported by the Department of Energy through grant DEFG0395ER69183 to the Scripps Institu- tion of Oceanography, and contracts 351574-AQ5 and 352731-AQ5 with Batelle Memorial Institute, Pacific Northwest National Labora- tories. References Ackerman, S. A., and S. K. Cox, Aircraft observations of shortwave fractional absorptance of non-homogeneous clouds, J. Appl. Meteo- rol., 20, 1510-1515, 1981. Ackerman, A. S., and O. B. Toon, Unrealistic desiccation of marine stratocumulus clouds by enhanced solar absorption, Nature, 380, 512-515, 1996. Arking, A., Absorption of solar energy in the atmosphere: Discrepancy between models and observations, Science, 273, 779-782, 1996. Bush, B.C., and F. P. J. Valero, Comparison of ARESE clear sky surface radiation measurements, J. Quant. Spectrosc. Radiat. Trans- fer, in press, 1997. Bush, B.C., S. K. Pope, A. Bucholtz, F. P. J. Valero, and A. Strawa, Surface radiation measurements during the ARESE campaign, J. Quant. Spectrosc. Radiat. Transfer, in press, 1997. Cess, R. D., et al., Absorption of solar radiation by clouds: Observa- tions versus models, Science, 267, 496-499, 1995. Chou, M.D., and W. D. Zhou, Estimation and model validation of the surface solar radiation and cloud radiative forcing using TOGA COARE measurements, J. Clim., 10(N4), 610-620, 1997. Francis, P. N., J.P. Taylor, P. Hignett, and A. Slingo, On the question of enhanced absorption of solar radiation by clouds, Q. J. R. Meteo- rol. Soc., 123, 419-434, 1997. Geist, J., and W. R. Blevin, Appl. Opt., 12, 2532, 1973. Hammer, P. D., F. P. J. Valero, and S. Kinne, The 27-28 October 1986 FIRE cirrus case study: Retrieval of cloud particle sizes and optical depths from comparative analyses of aircraft and satellite-based infrared measurements, Mon. Weather Rev., 119, 1673-1692, 1991. Imre, D. G., E. H. Abramson, and P. H. Daum, Quantifying cloud- induced shortwave absorption: An examination of uncertainties and of recent arguments for large excess absorption, J. Appl. Meteorol., 35, 1991-2010, 1996. Kiehl, J. T., and K. E. Tremberth, Earth's annual global mean energy budget, Bull. Am. Meteorol. Soc., 78(N2), 197-208, 1997. King, M.D., L. F. Radke, and P. V. Hobbs, Determination of spectral absorption of solar radiation by marine stratocumulus clouds from airborne measurements within clouds, J. Atmos. Sci., 47, 894-907, 1990. King, M.D., et al., Airborne scattering spectrometer for remote sens- ing of cloud, aerosol, water vapor, and surface properties, J. Atmos. Oceanic Technol., 13, 777-794, 1996. 21562202d, 1997, D25, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/97JD02434 by University of California San Diego, Wiley Online Library on [24/09/2026]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License

VALERO ET AL.: ARESE EXPERIMENTAL AND DATA DETAILS 29,937 Kondratiev, K. Y., V. I. Binenko, and I. N. Melnikova, Solar radiation absorption by cloudy and cloudless atmospheres, Meteorol. Gidrol., N2, 14-23, 1996. Li, Z., and L. Moreau, Alteration of atmospheric solar absorption by clouds: Simulation and observation, J. Appl. Meteorol., 35, 653-670, 1996. Li, Z., H. W. Barker, and L. Moreau, The variable effect of clouds on atmospheric absorption of solar radiation, Nature, 376, 486-490, 1995. Pilewskie, P., and F. P. J. Valero, Direct observations of excess solar absorption by clouds, Science, 267, 1626-1629, 1995. Ramanathan, V., et al., Warm pool heat budget and shortwave cloud forcing: A missing physics?, Science, 267, 499-503, 1995. Spinhirne, J. D., W. D. Hart, and D. L. Hlavka, J. Atmos. Sci., 53, 1438, 1996. Stephens, G. L., How much solar radiation do clouds absorb?, Science, 271, 1131-1133, 1996. Stephens, G. L., and S.C. Tsay, On the cloud absorption anomaly, Q. J. R. Meteorol. Soc., 116, 671-704, 1990. Valero, F. P. J., W. J. Y. Gore, and L. P.M. Giver, Radiative flux measurements in the troposphere, Appl. Opt., 21, 831-838, 1982. Valero, F. P. J., R. D. Cess, M. Zhang, S. K. Pope, A. Bucholtz, B. Bush, and J. Vitko, Absorption of solar radiation by the cloudy atmosphere: Interpretations of collocated aircraft measurements, J. Geophys. Res., this issue. Waliser, D. E., W. D. Collins, and S. P. Anderson, An estimate of the surface shortwave cloud forcing ove? the western Pacific during TOGA COARE, Geophys. Res. Lett., 23; 519-522, 1996. Zender, C. S., S. K. Pope, B. Bush, A. Bucholtz, W. D. Collins, J. T. Kiehl, F. P. J. Valero, and J. Vitko, Atmospheric absorption during the Atmospheric Radiation Measurements Enhanced Shortwave Experiment (ARESE), J. Geophys. Res., this issue. A. Bucholtz, B.C. Bush, P. Flatau, S. K. Pope, and F. P. J. Valero, Atmospheric Research Laboratory, Canter for Atmospheric Sciences, Scripps Institution of Oceanography, University of California, San Diego, La Jolla, CA 92093-0242. (e-mail: abucholtz@ucsd.edu; brbush@ucsd.edu; pflatau@ucsd.edu; spope@ucsd.edu; fvalero@ ucsd.edu) W. D. Collins, National Center for Atmospheric Research, 1850 Table Mesa Drive, Boulder, CO 80303. (e-mail: wcollins@kavieng.cgd. ucar.edu) W. J. Y. Gore and A. Strawa, NASA Ames Research Center, Mof- fett Field, CA 94035. (e-mail: strawa@ra.arc.nasa.gov) (Received March 20, 1997; revised August 7, 1997; accepted August 26, 1997.) 21562202d, 1997, D25, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/97JD02434 by University of California San Diego, Wiley Online Library on [24/09/2026]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License