Sunphotometer measurements of aerosol optical ...
Sunphotometer measurements of aerosol optical thickness in the Gerlache Strait and Marguerite Bay, Antarctica ROBERT FROuIN Scripps Institution of Oceanography La Jolla, California 92093-0221
MICHEL PANOUSE Observatoire Oceanologique de Banyuls Banyuls-sur-Mer, France
JEAN-CLAUDE DEVAUX Laboratoire d'Optique Atmospherique Liniversité de Lille, France
During the 1991-1992 Research on Antarctic Coastal Ecosysn Rates (RACER) cruise, aerosol optical thickness measuremts were made using a three-channel sunphotometer proled by the Laboratoire d'Optique Atmosphérique of the Unirsity of Lille, France. The objective was to verify aircraft estites of aerosol optical thickness obtained with the Polarization Ld Directionality of the Earth Reflectance (POLDER) instruant, which measured the intensity of reflected sunlight at )624,572 meter altitudes over the RACER study sites. Details out the POLDER instrument and the aircraft missions, includg scientific objectives, are given in Frouin et al. (this issue). The sunphotometer measures solar energy attenuated along e direct sun-to-surface path in spectral bands centered at 450, 0, and 850 nanometers. If E 01 denotes the solar irradiance at the p of the atmosphere in spectral band i, the optical thickness, taj deduced from the measurement of direct solar energy, E . by ing the Lambert-Bouguer law: Esi = E01 exp iere ttm is the optical thickness due to molecular scattering, T 9 is the tical thickness due to gaseous absorption (ozone), and mis the air ass. Corrections for tm and T9 must therefore be performed to rieve ta, which is accomplished using climatological or in situ lues of surface pressure (for 'r" in all bands) and ozone amount ts in the 450- and 650-nanometer bands). From measurements spectral bands 1 and 2, the Angstrom exponent, a, which characrizes the spectral dependence of the aerosol optical thickness suming a power law), can be computed as a=1n[(ta)/ta2)1/ln(?1/A), and ?2 are the equivalent wavelengths of the spectral The sunphotometer measurements are thus limited to aero1 optical thickness and its wavelength dependence, although size distribution index can also be estimated, albeit not acirately. These measurements will therefore not allow verifica-
1992 REVIEW
tion of POLDER retrievals of aerosol type (refractive index) and size distribution (see Deuzé et al. this issue). We are aware of the limitations, but it was out of the question to install aboard R/V Polar Duke a sophisticated, stabilized atmospheric optics station that would have provided measurements of not only direct solar attenuation, but also of sky radiance, including solar aureole, as well as polarization (necessary measurements for a complete description of the aerosols). The sunphotometer measurements were taken whenever possible (sun disk not obscured by clouds) by pointing the hand-held instrument toward the sun and recording the voltage output in the various channels. Air temperature and pressure were also logged at the time of the measurements, and care was exercised to avoid stack fumes. Since clear skies were required, most of the data were collected during the days of the aircraft missions. In general, the sea was calm, facilitating tracking of the sun. The sunphotometer, however, recorded peak voltage, which ensured that energy from the sun disk was actually measured when tracking the sun was difficult. The table summarizes sunphotometer measurements made during the campaign. Data were collected in the Gerlache Strait (about 64 S and 61 W) and Marguerite Bay (about 68 S and 68 W). Except for one instance, on 2 January 1992 when the air mass was 5.05, the optical thickness was high, around or above 0.2, in all spectral bands. The high values were unexpected, since the region was far from any source of anthropogenic aerosols, and since retrievals from the FOLDER data suggested much smaller loadings (Deuzé et al. this issue). Furthermore, horizontal visibility logged by R/V Polar Duke and by the Rothera meterological station was generally high (above 32 kilometers), indicating that aerosols, if any, were likely located aloft. On the other hand, the spectral dependence of the aerosol optical thickness was weak, suggesting the presence of large particles. To explain the anomalously high aerosol optical thickness values, we recalled that Mount Pinatubo in the Philippines erupted on 15 June 1991, about 6 months prior to the 1991-1992 RACER cruise. The resulting stratospheric aerosol layer, as observed from the advanced very-high-resolution radiometer (AVHRR), circled Earth in 21 days, and covered about 40 percent of Earth's surface by the end of August 1991, with patches progressing poleward and extending to latitudes below 30 S (Stowe et al. 1992). Mount Pinatubo's eruption was a major climatic event (according to Stowe etal. 1992, mean optical thickness was 0.31 on 23 August 1991). Furthermore, preliminary assessments indicate that the cooling effect of the stratospheric aerosols may offset the warming caused by anthropogenic greenhouse gases during several years (Hansen et al. 1992). Although the evidence is not conclusive at this point, it is likely that the high aerosol optical thickness values measured during the austral spring of 1991-1992 in the Gerlache Strait and Marguerite Bay were the result of Mount Pinatubo's eruption. In any case, the contrast between the sunphotometer measurements (indicate high aerosol amount along the sun-to-surface path) and the visual observations (suggest low aerosol amount at the surface), provide sufficient evidence that a large fraction of the aerosols was located above the aircraft altitudes of 3,962-4,572 meters. Correcting the POLDER measurements for atmospheric effects will therefore require partitioning total aerosol loadings vertically. This may be accomplished by estimating, from the visual observations, the aerosol optical thickness below the aircraft and deducing, from the sunphotometer measurements, the aerosol optical thickness above the aircraft. Owing to the vertical heterogeneity of the aerosols, however, it will be rather difficult,
193
Sunphotometer measurements of aerosol optical thickness at 450, 650, and 850 nanometers Month Day 18 26 26 26 28 28 28 29 29 29 29 29 30 30 31 2 2 2 2 2 2 2 2
Year 1991 1991 1991 1991 1991 1991 1991 1991 1991 1991 1991 1991 1991 1991 1991 1992 1992 1992 1992 1992 1992 1992 1992
Latitude (°) Longitude (°) -64.3300 -64.1900 -64.1900 -64.1900 -64.2950 -64.3100 -64.3100 -64.6660 -64.5400 -64.4630 -64.5450 -64.4560 -64.5360 -64.5130 -65.1060 -68.1000 -68.0700 -68.0550 -68.0350 -68.0380 -68.0380 -67.9830 -67.9460
-61.6800 -61.2800 -61.3300 -61.3300 -61.8050 -61.7200 -61.7200 -62.0730 -62.2450 -62.2380 -61.8820 -61.8460 -62.5950 -62.7780 -64.6600 -68.0500 -68.0300 -68.0450 -68.0900 -68.0730 -68.0730 -68.0830 -68.1780
even impossible, to verify the POLDER estimates of aerosol optical thickness directly against the sunphotometer measurements. This research was supported in part by the National Aeronautics and Space Administration under Grant NAG W-2774 to Robert Frouin, the National Science Foundation, the British Antarctic Survey, the French Space Agency, and the Centre National de la Recherche Scientifique. We thank the captain and crew members of R/V Polar Duke for their assistance during the cruise.
194
Air mass .t450 1.5442 1.4519 1.3483 1.3231 2.0617 1.8475 1.6376 2.1642 1.8542 1.6428 1.4183 1.3393 2.2261 1.7650 1.7660 1.8430 1.5451 5.0524 1.4204 1.4260 1.4459 1.5420 2.2739
.2167 .2286 .3668 .3714 .2364 .2732 .2572 .2344 .2252 .2196 .2196 .2152 .2061 .2051 .2435 .2804 .2471 .0450 .2556 .2358 .2692 .2797 .2616
t650
.1979 .1991 .4180 .3646 .2104 .2401 .2459 .2158 .2055 .2177 .1936 .2065 .1910 .1975 .2318 .2222 .2319 .0112 .2489 .2314 .2418 .2312 .2235
t850
.2145 .1790 .5436 .3116 .2234 .4853 .3643 .1993 .2376 .3423 .3252 .4426 .2047 .2047 .2374 .1908 .3493 .0478 .5437 .2661 .2507 .2412 .3323
References
Deuze,J. -L., P. Goloub, P. Y. Descamps, M. Herman, and R. Froum. 1992 Retrieval of aerosols over the Gerlache Strait from aircraft photopolarimetric observations. Antarctic Journal ofthe U. S., this issue Frouin,R.,J. Y. Balois,P. Y. Descamps,M. Herman, M. Panouse,J. Priddle and C. Verwaerde. 1992. Aircraft photopolarimetric observations o the ocean, ice/snow, and clouds in coastal regions of the Antarctic Peninsula. Antarctic Journal of the U.S. , this issue; Hansen, J. A., R. Lacis, R. Ruedy, and M. Sato. 1992. Potential climate impact of Mount Pinatubo eruption. Geophysical Research Letters, 19:215-218. Stowe, L. L., R. M. Carey, and P. 0. Pellegrino. 1992. Monitoring the Mount Pinatubo aerosol layer with NOAA-11 AVHRR data. liT Aerospace/communications division poster, available from NOAA/ NESDIS, Washington, D.C.
ANTARCTIC
MICHEL PANOUSE Observatoire Oceanologique de Banyuls Banyuls-sur-Mer, France
JEAN-CLAUDE DEVAUX Laboratoire d'Optique Atmospherique Liniversité de Lille, France
During the 1991-1992 Research on Antarctic Coastal Ecosysn Rates (RACER) cruise, aerosol optical thickness measuremts were made using a three-channel sunphotometer proled by the Laboratoire d'Optique Atmosphérique of the Unirsity of Lille, France. The objective was to verify aircraft estites of aerosol optical thickness obtained with the Polarization Ld Directionality of the Earth Reflectance (POLDER) instruant, which measured the intensity of reflected sunlight at )624,572 meter altitudes over the RACER study sites. Details out the POLDER instrument and the aircraft missions, includg scientific objectives, are given in Frouin et al. (this issue). The sunphotometer measures solar energy attenuated along e direct sun-to-surface path in spectral bands centered at 450, 0, and 850 nanometers. If E 01 denotes the solar irradiance at the p of the atmosphere in spectral band i, the optical thickness, taj deduced from the measurement of direct solar energy, E . by ing the Lambert-Bouguer law: Esi = E01 exp iere ttm is the optical thickness due to molecular scattering, T 9 is the tical thickness due to gaseous absorption (ozone), and mis the air ass. Corrections for tm and T9 must therefore be performed to rieve ta, which is accomplished using climatological or in situ lues of surface pressure (for 'r" in all bands) and ozone amount ts in the 450- and 650-nanometer bands). From measurements spectral bands 1 and 2, the Angstrom exponent, a, which characrizes the spectral dependence of the aerosol optical thickness suming a power law), can be computed as a=1n[(ta)/ta2)1/ln(?1/A), and ?2 are the equivalent wavelengths of the spectral The sunphotometer measurements are thus limited to aero1 optical thickness and its wavelength dependence, although size distribution index can also be estimated, albeit not acirately. These measurements will therefore not allow verifica-
1992 REVIEW
tion of POLDER retrievals of aerosol type (refractive index) and size distribution (see Deuzé et al. this issue). We are aware of the limitations, but it was out of the question to install aboard R/V Polar Duke a sophisticated, stabilized atmospheric optics station that would have provided measurements of not only direct solar attenuation, but also of sky radiance, including solar aureole, as well as polarization (necessary measurements for a complete description of the aerosols). The sunphotometer measurements were taken whenever possible (sun disk not obscured by clouds) by pointing the hand-held instrument toward the sun and recording the voltage output in the various channels. Air temperature and pressure were also logged at the time of the measurements, and care was exercised to avoid stack fumes. Since clear skies were required, most of the data were collected during the days of the aircraft missions. In general, the sea was calm, facilitating tracking of the sun. The sunphotometer, however, recorded peak voltage, which ensured that energy from the sun disk was actually measured when tracking the sun was difficult. The table summarizes sunphotometer measurements made during the campaign. Data were collected in the Gerlache Strait (about 64 S and 61 W) and Marguerite Bay (about 68 S and 68 W). Except for one instance, on 2 January 1992 when the air mass was 5.05, the optical thickness was high, around or above 0.2, in all spectral bands. The high values were unexpected, since the region was far from any source of anthropogenic aerosols, and since retrievals from the FOLDER data suggested much smaller loadings (Deuzé et al. this issue). Furthermore, horizontal visibility logged by R/V Polar Duke and by the Rothera meterological station was generally high (above 32 kilometers), indicating that aerosols, if any, were likely located aloft. On the other hand, the spectral dependence of the aerosol optical thickness was weak, suggesting the presence of large particles. To explain the anomalously high aerosol optical thickness values, we recalled that Mount Pinatubo in the Philippines erupted on 15 June 1991, about 6 months prior to the 1991-1992 RACER cruise. The resulting stratospheric aerosol layer, as observed from the advanced very-high-resolution radiometer (AVHRR), circled Earth in 21 days, and covered about 40 percent of Earth's surface by the end of August 1991, with patches progressing poleward and extending to latitudes below 30 S (Stowe et al. 1992). Mount Pinatubo's eruption was a major climatic event (according to Stowe etal. 1992, mean optical thickness was 0.31 on 23 August 1991). Furthermore, preliminary assessments indicate that the cooling effect of the stratospheric aerosols may offset the warming caused by anthropogenic greenhouse gases during several years (Hansen et al. 1992). Although the evidence is not conclusive at this point, it is likely that the high aerosol optical thickness values measured during the austral spring of 1991-1992 in the Gerlache Strait and Marguerite Bay were the result of Mount Pinatubo's eruption. In any case, the contrast between the sunphotometer measurements (indicate high aerosol amount along the sun-to-surface path) and the visual observations (suggest low aerosol amount at the surface), provide sufficient evidence that a large fraction of the aerosols was located above the aircraft altitudes of 3,962-4,572 meters. Correcting the POLDER measurements for atmospheric effects will therefore require partitioning total aerosol loadings vertically. This may be accomplished by estimating, from the visual observations, the aerosol optical thickness below the aircraft and deducing, from the sunphotometer measurements, the aerosol optical thickness above the aircraft. Owing to the vertical heterogeneity of the aerosols, however, it will be rather difficult,
193
Sunphotometer measurements of aerosol optical thickness at 450, 650, and 850 nanometers Month Day 18 26 26 26 28 28 28 29 29 29 29 29 30 30 31 2 2 2 2 2 2 2 2
Year 1991 1991 1991 1991 1991 1991 1991 1991 1991 1991 1991 1991 1991 1991 1991 1992 1992 1992 1992 1992 1992 1992 1992
Latitude (°) Longitude (°) -64.3300 -64.1900 -64.1900 -64.1900 -64.2950 -64.3100 -64.3100 -64.6660 -64.5400 -64.4630 -64.5450 -64.4560 -64.5360 -64.5130 -65.1060 -68.1000 -68.0700 -68.0550 -68.0350 -68.0380 -68.0380 -67.9830 -67.9460
-61.6800 -61.2800 -61.3300 -61.3300 -61.8050 -61.7200 -61.7200 -62.0730 -62.2450 -62.2380 -61.8820 -61.8460 -62.5950 -62.7780 -64.6600 -68.0500 -68.0300 -68.0450 -68.0900 -68.0730 -68.0730 -68.0830 -68.1780
even impossible, to verify the POLDER estimates of aerosol optical thickness directly against the sunphotometer measurements. This research was supported in part by the National Aeronautics and Space Administration under Grant NAG W-2774 to Robert Frouin, the National Science Foundation, the British Antarctic Survey, the French Space Agency, and the Centre National de la Recherche Scientifique. We thank the captain and crew members of R/V Polar Duke for their assistance during the cruise.
194
Air mass .t450 1.5442 1.4519 1.3483 1.3231 2.0617 1.8475 1.6376 2.1642 1.8542 1.6428 1.4183 1.3393 2.2261 1.7650 1.7660 1.8430 1.5451 5.0524 1.4204 1.4260 1.4459 1.5420 2.2739
.2167 .2286 .3668 .3714 .2364 .2732 .2572 .2344 .2252 .2196 .2196 .2152 .2061 .2051 .2435 .2804 .2471 .0450 .2556 .2358 .2692 .2797 .2616
t650
.1979 .1991 .4180 .3646 .2104 .2401 .2459 .2158 .2055 .2177 .1936 .2065 .1910 .1975 .2318 .2222 .2319 .0112 .2489 .2314 .2418 .2312 .2235
t850
.2145 .1790 .5436 .3116 .2234 .4853 .3643 .1993 .2376 .3423 .3252 .4426 .2047 .2047 .2374 .1908 .3493 .0478 .5437 .2661 .2507 .2412 .3323
References
Deuze,J. -L., P. Goloub, P. Y. Descamps, M. Herman, and R. Froum. 1992 Retrieval of aerosols over the Gerlache Strait from aircraft photopolarimetric observations. Antarctic Journal ofthe U. S., this issue Frouin,R.,J. Y. Balois,P. Y. Descamps,M. Herman, M. Panouse,J. Priddle and C. Verwaerde. 1992. Aircraft photopolarimetric observations o the ocean, ice/snow, and clouds in coastal regions of the Antarctic Peninsula. Antarctic Journal of the U.S. , this issue; Hansen, J. A., R. Lacis, R. Ruedy, and M. Sato. 1992. Potential climate impact of Mount Pinatubo eruption. Geophysical Research Letters, 19:215-218. Stowe, L. L., R. M. Carey, and P. 0. Pellegrino. 1992. Monitoring the Mount Pinatubo aerosol layer with NOAA-11 AVHRR data. liT Aerospace/communications division poster, available from NOAA/ NESDIS, Washington, D.C.
ANTARCTIC
