--Jack Kaye (jkaye@hq.nasa.gov), TOMS Program Scientist, NASA Headquarters, Washington D.C.
The first meeting of the science team for the Total Ozone Mapping Spectrometer (TOMS) series of instruments was held on April 20-22 at the National Wildlife Visitor Center at the Patuxent Environmental Science Center in Laurel, MD. This science team was selected in response to a NASA Research Announcement (NRA) issued in 1997 and includes scientists funded through TOMS-related data analysis funding and also relevant investigators in the Atmospheric Chemistry and Modeling Program (ACMAP) of NASA's Office of Earth Science. The meeting provided an opportunity for the newly selected team members to meet each other as well as the scientists and programmers who have been carrying out TOMS processing and data analysis in recent years. Attention was focused on all of the TOMS observables, including total ozone, tropospheric aerosols, surface ultraviolet radiation, and stratospheric sulfur dioxide, as well as tropospheric ozone information, which can be obtained either from TOMS data alone or by use of TOMS data in conjunction with those from other space-borne sensors.
The meeting began with greetings from Jack Kaye, Program Scientist for TOMS at NASA HQ, and continued with a summary of the TOMS series of missions from P. K. Bhartia, TOMS Project Scientist at NASA's Goddard Space Flight Center (GSFC). A total of four TOMS instruments have flown in space - Nimbus 7 TOMS (1978-1993), Meteor-3 TOMS (1991-94), ADEOS TOMS (1996-1997), and Earth Probe TOMS (1996 - present). A commitment has just been made by the government of the Netherlands to provide an Ozone Monitoring Instrument (OMI) using a similar observing technique to that used by TOMS for the EOS CHEM spacecraft for launch in late 2002 or 2003. NASA plans to issue an NRA for a U.S. Science Team for OMI later this year. Measurements with a TOMS-like technique are also expected to be part of the National Polar Orbiting Environmental Satellite System (NPOESS) through the planned Ozone Mapping and Profiling Suite (OMPS) instrument, although the first such instrument may not be launched until 2010.
The TOMS instruments have provided excellent data for total ozone, although there are some limitations that must be considered in certain applications, especially for studies of tropospheric ozone. In general, data at high solar zenith angles are less reliable than those at lower angles. Currently, no data are released for zenith angles greater than 84°, although data are in principle available for up to 88°. The validation of these high solar zenith angle data is difficult, however. The determination of surface flux of ultraviolet radiation from TOMS data appears to be quite robust, especially if ratios, such as that of UVA/UVB (UV radiation longward and shortward of 320 nm), are considered. The major factor affecting the determination of the surface flux of UVA radiation, which is not attenuated by ozone, is our knowledge of the UV extinction associated with aerosol particles, especially UV attenuating forms such as smoke or mineral desert dust. Results to date suggest that up to 50% of UVA can be attenuated by such aerosols.
The status of the Earth Probe (EP) TOMS instrument was reviewed by Rich McPeters, the EP TOMS principal investigator, of NASA/GSFC. EP TOMS was launched on a Pegasus XL rocket on July 2, 1996 into a 490 km orbit, which provided a 26 km field of view (FOV) at nadir. This is significantly better horizontal resolution than previously obtained by TOMS instruments (typically 50 km at nadir), but did not provide full daily global coverage at all latitudes (this was obtained only poleward of approximately 60 degrees). The EP TOMS instrument, like that of ADEOS TOMS, differed from the previous TOMS instruments (Nimbus 7, Meteor-3) through slight changes in wavelength (309, 313, 318, 322, 331, and 360 nm instead of 312, 317, 332, 340, 360, and 380 nm). The new instruments also include a carousel of three diffuser plates exposed at differing intervals (regularly, periodically, very rarely) so that the effects of diffuser plate degradation can be better characterized.
Data were obtained routinely from July 25, 1996 until December 3, 1997 except for a three-day period in mid-November, 1996. Following the failure of the ADEOS spacecraft in the summer of 1997 as mentioned below, a decision was made to boost EP TOMS to a higher orbit, and in early December 1997 this reboost was carried out, lifting EP TOMS to a 739 km orbit, providing a nadir viewing size of 38 km. This provides full daily global coverage except for small inter-orbit gaps in the tropics. This new orbit minimizes the need for orbital adjustments as we approach the maximum of the current solar cycle; without this boost, the orbit would have dropped by a half km per month, requiring orbital adjustments every few months. The instrument has operated in its normal observing mode except for several sequences designed to focus on regions where high concentrations of sulfur dioxide were expected, including the region of El Popocatepetl in Mexico during May 1997 and later for several large metropolitan areas. Such studies were terminated following the loss of ADEOS TOMS, when EP TOMS became the primary ozone mapping instrument.
The status of data from some of the other TOMS instruments was also reviewed. Jay Herman of NASA/GSFC summarized the three years of data from the Meteor-3 TOMS. The Meteor-3 spacecraft was in a non sun-synchronous orbit, and roughly half the data were obtained at sufficiently large solar zenith angles that the data must be used with great care. The first half of the Meteor-3 TOMS record overlapped that of Nimbus 7 TOMS, so these data have been examined only in a limited way. The second half of the Meteor-3 TOMS record helped extend the TOMS record, and has been more extensively used in long-term trend studies. He also discussed plans to fly a new TOMS instrument on a Russian Meteor 3M spacecraft scheduled for launch in the year 2000.
Arlin Krueger, also of NASA/GSFC, summarized the record of ADEOS TOMS. This instrument obtained data from instrument activation on September 12, 1996 until the end of mission on June 29, 1997. During that period, more than 99% of possible data were obtained, with the largest data gaps being in September and October of 1996. Because of the relatively high orbit of the ADEOS spacecraft, ADEOS TOMS provided full daily coverage of the sunlit Earth. The ADEOS spacecraft contained several other instruments whose data may be used together with those from TOMS both for scientific studies and to better understand the TOMS data. Those of greatest interest are the OCTS, an ocean-color oriented instrument using visible wavelengths whose data will be of great use in studying aerosol and cloud distributions because of their high spatial resolution (~1 km), and AVNIR, an infrared radiometer with very high resolution (6 m, 16 m), which can also provide information for use in radiative transfer studies.
The status of the Earth Probe TOMS ground operations (including processing to Level 0 data) was provided by Ed Macie, leader of the NASA/GSFC flight operations team for Earth Probe and the Upper Atmosphere Research Satellite (UARS), whose operations are collocated. Instrument status at the time of the meeting was excellent. One item that had been of particular concern was the battery (which is a critical non-redundant component). This has been showing no signs of degradation and has, in fact, been providing higher voltage than expected, but this is not expected to be a problem. The attitude control system is within required specifications, although the system is somewhat more susceptible to noise than was expected. The instrument does experience some electromagnetic noise when passing over the South Atlantic Anomaly. The flight software is currently being reviewed for year 2000 issues. The transmitter had been operating continuously during data collection periods in daytime. A failure of the primary transmitter after the meeting (late April, 1998) has caused a backup transmitter to be used, and this is only turned on for scheduled data downlinks so real-time downlink of data no longer takes place.
The status of TOMS data processing was discussed by Charlie Wellemeyer of Raytheon STX, Inc. TOMS level 2 and level 3 total ozone data are available from the Distributed Active Archive Center (DAAC) of the Earth Observing System Data and Information System (EOSDIS) at the Goddard Space Flight Center. Additional data are made available through an internal web site at GSFC; these include "research products" such as tropospheric aerosols. The available total ozone measurements are based on the Version 7 algorithm. Several known factors can contribute to errors, including the presence of UV absorbing tropospheric aerosols, the occurrence of high solar zenith angles, scan angle dependence during the cross-track scanning carried out by the TOMS instrument (this was a particular problem during the post-Pinatubo period when stratospheric aerosol distributions were largest), sun glint, and cloud heights, which are based on climatology rather than actual cloud heights. Some fixes for the last of these may be possible for Nimbus 7 TOMS during the time when the THIR instrument was operating (1979-1987).
Knowledge of the calibration of the TOMS instruments, especially EP TOMS, was reviewed by Glen Jaross, also of Raytheon STX, Inc. No major problems with the calibration for EP TOMS are known, although some small problems may exist. In particular, it is possible that the 360 nm channel center is actually 0.3 nm shorter than originally measured. This will have a small effect on the TOMS aerosol and ozone products. The degradation likely to have occurred in the diffuser plate during the period of operation of EP TOMS is not thought to have impacted TOMS ozone retrievals and to have only minimally affected those of tropospheric aerosols and surface UV radiation. This is in spite of a fairly significant reduction in signal strength of the instrument (~25%). Comparisons of EP TOMS total ozone columns to those measured with the World Standard Dobson instrument suggest that EP TOMS values may be about 1.5% too high.
Given this knowledge of the EP TOMS instrument, there was a discussion of the desirability of doing a reprocessing at this stage as opposed to waiting for an additional time period (e.g. one year) before considering a reprocessing. Based on the expected reduction in error, which is quite small for total ozone and only somewhat larger for aerosols and surface UV radiation, the assembled team felt that there was no need for an immediate reprocessing of the EP TOMS data.
A discussion on the ideal mechanisms for distribution of TOMS data was carried out, with the DAAC represented by George Serafino of NASA/GSFC. A particular issue was the need for availability of the detailed Level 2 data (individual measurements) instead of the Level 3 (gridded) data that have been used in most studies to date. The problem is data volume about 6 Gb per year. The advantages and disadvantages of various media (CD-ROM, 8 mm tape, 4 mm tape), as well as the benefits of data compression, were all discussed.
Results of prior research and plans for current research by members of the TOMS science team were then presented. These were considered largely in three groups research related to the TOMS total ozone product, to TOMS tropospheric ozone products, and to the TOMS aerosol and surface UV flux products (which are considered together because of the crucial role which aerosols have in affecting surface UV flux). Some related work was also discussed; this is included with the total ozone section that follows. Since some of the science team members are just in the process of beginning their investigations, some efforts are in their formative stages, while others are continuations of work with long heritage.
Efforts to create a single long-term data set for total ozone derived from space-based measurements from multiple instruments were summarized by Rich Stolarski of NASA/GSFC. A data set covering the period from late 1978 till February 1998 for zonal mean ozone should be available in early May 1998. A more complete data set, providing some information on longitudinal distributions as well, should be made available in late 1998 or early 1999. One of the key questions to deal with is how to extend the total ozone record through the "TOMS gap" from the time of the Meteor-3 TOMS failure in late 1994 until the launch of EP TOMS in mid-1996.
Comparisons of TOMS total ozone amounts with those determined from ultraviolet and visible radiation measurements made using the Composition and Photochemical Flux Module (CPFM) instrument that flies aboard NASA's ER-2 aircraft were discussed by Steven Lloyd of the Johns Hopkins University Applied Physics Laboratory. The CPFM instrument could look upward, downward, and at the limb, so a combination of upward and downward looking observations provided a total ozone column that could be compared with TOMS observations made at similar times and locations. Comparisons will focus on the region near Fairbanks, AK which was the base for the ER-2 during most of the Photochemistry of Ozone Loss in the Arctic Region in Summer (POLARIS) aircraft campaign in 1997. Given the high latitude of the aircraft observations, the ER-2/satellite comparisons will be particularly valuable in probing the effect of profile shape on the TOMS retrieval at high solar zenith angles. The possibility of combining aircraft- and space-based data will also be considered; an example might be the use of cloud top heights determined from CPFM 762 nm data.
The status of comparisons of TOMS-measured total ozone columns and those measured from Dobson instruments was summarized by Sam Oltmans of the Climate Monitoring and Diagnostics Laboratory of the National Oceanic and Atmospheric Administration. The most detailed comparisons carried out are those with the World Standard Dobson (#83), made each summer at Mauna Loa, Hawaii. Results to date have shown good agreement. Other Dobson instruments, such as that at American Samoa, do show systematic differences in comparison of total column ozone measurements with those made from TOMS. These comparisons are a critical component of the TOMS validation effort and will continue in the future.
The use of space-based ozone measurements as input for a data assimilation system designed to produce accurate and internally consistent ozone fields was described by Ivanka Stajner of the General Sciences Corp., who is working with Lars-Peter Riishojgaard of the University of Maryland at Baltimore County. TOMS total ozone Level 2 data are used along with partial profile information from the SBUV instruments. Chemical production and loss rates are provided at 10 day intervals and are used together with meteorological fields from the GSFC Data Assimilation Office GEOS system as input for the model. Ozone data are assimilated using a global, physical space based statistical analysis scheme. Results to date, which have focused on the winter of 1992, show that the calculated ozone fields compare better with those from ozonesondes (not themselves used in the assimilation process) than with those calculated using the GSFC parameterized chemistry-transport model. Further developments of the assimilation system should lead to improved ozone fields. The ozone assimilation system should become operational in the next several months.
The relationship between TOMS total ozone measurements and small scale tropopause dynamics was reviewed by Mark Olsen, who is working together with TOMS science team member John Stanford of Iowa State University. Analysis of TOMS data along with meteorological data (especially potential vorticity) and satellite-measured water vapor fields is being used to help understand the relationship between ozone distributions and the dynamics of the tropopause region. A particular emphasis is on how intrusions of stratospheric air into the troposphere affect the total ozone columns as measured by TOMS. Comparisons with the chemistry/transport model of NASA/GSFC are being used to help test understanding of these mechanisms. One potential spin-off of this work is improved knowledge of the relationship between the structure in the total ozone field measured by TOMS to the strength of the jet stream in the upper troposphere.
Studies of the interannual variability in total ozone and its long-term trends in mid-latitudes were reviewed by Lon Hood of the University of Arizona. This work was set up to better understand the sources of both interannual and long-term variability in ozone amounts, with a particular emphasis on dynamically-driven contributions. These are looked at through parallel analysis of TOMS ozone data and meteorological fields. Results to date suggest that changes in lower stratospheric temperature and/or tropopause height could be contributing significantly to the mid-latitude ozone decreases observed by TOMS. It was seen that short term meteorological events can significantly affect the zonal mean distributions of total ozone, especially in February when there can be enormous variations from one winter to the next. Changes in the zonal wind that may be occurring over the long term could influence the variability of planetary waves, which in turn may be affecting the forcing of stratospheric wave events, which can help mix mid-and high-latitude air in the winter.
Studies of ultraviolet radiative transfer algorithms for use in satellite measurements of ozone were discussed by Dave Flittner, who is working as part of the TOMS science team investigation of Ben Herman at the University of Arizona. A particular area of emphasis in this work is on the development of the ultraviolet limb scattering technique, which holds promise for being an excellent way to determine the vertical profile of ozone in the stratosphere. As part of this work, attention will be given to determining the effect of aerosols on ozone determination, as well as the sensitivity of the technique to horizontal variations in ozone distributions and the presence of broken clouds. Tests of the ozone algorithm may be made by analyzing data from the Shuttle Ozone Limb Sounding Experiment (SOLSE) and Limb Ozone Retrieval Experiment (LORE) instruments that flew on the Space Shuttle in the fall of 1997.
Measurements of other species with UV absorption in the ultraviolet using the Global Ozone Monitoring Experiment (GOME) instrument aboard the European Space Agency's ERS-2 spacecraft were discussed by Kelly Chance of the Smithsonian Astrophysical Observatory. The most observed of these species has been bromine monoxide (BrO), for which column distributions have been obtained, and interhemispheric differences have been demonstrated. Changes in the BrO column believed to be associated with tropospheric BrO in the springtime were observed. Studies of chlorine monoxide (ClO) are beginning, but the retrieval of ClO is much more difficult than that of BrO because of its overlap with the Hartley band of ozone. Studies of sulfur dioxide (SO2) are underway, but must be considered as a work in progress. In some cases, strong correlations of SO2 and ozone distributions are seen, and it is not clear whether this is a real geophysical signal or whether there is some aliasing of SO2 by ozone. Future work will include derivation of ozone profile information (especially through use of information in the Chappuis bands of ozone) and column distributions of formaldehyde.
A discussion of details of the TOMS retrieval algorithm and how those may be relevant to currently used methods for determination of tropospheric ozone was provided by Bob Hudson of the University of Maryland. The TOMS algorithm starts with a climatological ozone profile that assumes that some 90% of the ozone is in the stratosphere. Thus, if in reality there is an increased ozone column due to a larger amount of ozone in the troposphere, the algorithm will tend to "see" only a fraction of it. Depending on what altitude regions are considered, this fraction can be fairly small (in one study, only 22% of the ozone added in the 1000-891 mb region was observed, while the fraction rose to only 57% if the 1000-501 mb region was considered). This partial observation of tropospheric ozone must be considered in using techniques such as TOMS-SBUV residuals (in which tropospheric ozone is taken as the difference between TOMS total ozone and integrated stratospheric ozone from one of the Solar Backscatter Ultraviolet instruments) or the difference in ozone columns over mountains and nearby sea-level areas (such as have been carried out over the Andes Mountains and nearby Eastern Pacific). Other techniques tried to date should suffer less from such complexities, such as that which looks at the differences between total ozone columns over nearby clear and cloudy regions or that based on "scalloping" (small scan-angle dependence in the total column which may be due to differences in the ozone profile from that assumed in the TOMS algorithm). The latter technique is complicated by the presence of aerosols, with UV-absorbing and non-absorbing aerosols having very different effects.
Two methods of deriving tropospheric ozone information from TOMS data were discussed by Sushil Chandra of NASA/GSFC. One of these is a residual method, in which tropospheric ozone is obtained as the difference between TOMS total ozone and stratospheric ozone determined from a combination of the Microwave Limb Sounder (MLS) and Halogen Occultation (HALOE) instruments aboard UARS. The other, known as the Convective Cloud Differential (CCD) method, derives tropospheric ozone from the average difference between total column ozone over clear regions and over regions of high clouds. A key consideration in the latter method is having correct cloud top height information. The assumption is made that clouds with very high reflectivity are also at high altitudes. During the first part of the Nimbus 7 TOMS data, some information on cloud top pressure was available from the THIR instrument aboard Nimbus 7. The TOMS/MLS/HALOE differential method has been most extensively tested for the period of September 1992, during which the TRACE-A airborne campaign was carried out and the largest amount of data exists for testing the method. Analysis of time series of tropospheric ozone from these methods shows strong evidence of an effect of El Niño. The shift in location of convection during El Niño periods is believed to dominate these changes, although some contribution due to increased fires resulting from droughts in regions affected by El Niño cannot be ruled out.
Studies of tropospheric ozone based on TOMS were reviewed by Jack Fishman of the NASA Langley Research Center. After doing studies using TOMS data only and TOMS/SAGE residuals, more recent work has centered on a residual approach using TOMS and SBUV data. A key element of this method is the accurate placement of the tropopause so that accurate separation of tropospheric and stratospheric amounts can be made. Several different approaches have been tried, with a polynomial fitting routine being used most recently. This has led to some changes in the tropospheric ozone fields with respect to those defined previously. Results of the TOMS/SBUV residual method are being compared with available data, and attempts are being made to understand why the agreement appears to be good in only some cases. The long-range goal of the work is to look at air pollution events in the eastern U.S. and understand the contributions that satellite data can make to their characterization.
A new effort to help improve our knowledge of ozone vertical profiles in the tropical region was presented by Anne Thompson of NASA/GSFC. This effort is known as the Southern Hemisphere Additional Ozonesondes (SHADOZ) project, and will lead to more regular ozonesonde flights from several stations in the 0-20 degree South latitude band. No new stations will be brought on line as part of this effort; the focus will be on providing more regular flights at those stations that already exist. A fundamental goal of this two-year project is to help characterize the nature of the wave one pattern in ozone that is seen in the tropics. Currently available data do not yet provide definitive evidence for the origin of this pattern (although data seem to indicate that it resides in the troposphere). The assumption of a tropospheric wave one pattern is the basis of a new method for deriving tropical tropospheric ozone (TTO) at GSFC and the University of Maryland. Like the CCD technique, the new Modified-Residual Method is a TOMS-only method of obtaining both stratospheric and tropospheric column amount. The 14-year Nimbus 7 TTO maps, provided as 2-week averages at 1 ¥ 2 degree resolution, can be previewed on a homepage. Further evaluation of the Modified-Residual Product will be performed by the ozone processing team and other TOMS science team members before the data set is released to the public.
Efforts to understand tropospheric ozone from TOMS data were presented by Mike Newchurch of the University of Alabama in Huntsville. Previous work with Jae Kim of the Korean National University of Education has emphasized the differential method in which TOMS ozone column amounts over high mountains and nearby sea-level regions were compared, with applications to date being the Andes Mountain region of South America and the mountain regions of New Guinea. These regions differ in that in the Andes the mountains are downwind of the region of biomass burning thought to be responsible for ozone production, while in New Guinea the biomass burning region occurs downwind of the mountains. However, in both locations significant increasing trends in lower tropospheric ozone of 1%/year were found. These ozone increases are attributed to increases in biomass burning. A new effort being undertaken is to study ozone amounts over mesoscale convective cloud complexes (MCCs). By looking at averages of the significant ozone variation between clear and cloudy pixels, information on tropospheric ozone should be retrievable from this technique.
The effects of UV-absorbing and non-absorbing aerosols on surface UV radiation as seen by TOMS were discussed by Jay Herman of NASA/GSFC. One of the more difficult aspects of applying TOMS to this subject is differentiation of cloud- and aerosol-containing regions; to focus on the effect of aerosols, a reflectivity limit was put on scenes studied to select against high reflectivity cloud-dominated scenes. The TOMS surface UV product obtained compares well on the average to ground-based measurements, such as those made with Brewer spectrophotometers, although some systematic disagreements in such comparisons do exist and are not yet well-explained. For instance, the seasonal comparisons show that there is worse agreement between TOMS and Brewer surface UV values in summer than in other time periods. It is felt that TOMS can do a very good job in getting ratios of surface UV flux at different wavelengths.
The validation and interpretation of the TOMS aerosol product was described by Sundar Christopher of the University of Alabama in Huntsville, who emphasized the use of data from the Advanced Very High Resolution Radiometer (AVHRR) instrument that flies aboard the NOAA polar orbiting operational meteorological spacecraft. AVHRR provides aerosol information over water, and efforts using both spectral and textual information (e.g. homogeneity of background) are now being carried out to help determine aerosol amounts over land. AVHRR can also help locate fires that serve as the source of the UV-absorbing smoke detected by TOMS. Particular time periods examined include that of the Smoke, Clouds, Aerosols, and Radiation (SCAR B) mission, as well as the Zambia International Biomass Burning Experiment (ZIBBE) held in the summer of 1997.
Plans for the validation of the TOMS aerosol product through comparison with in situ aerosol measurements to be made from light aircraft were presented by Joseph Prospero of the University of Miami. A payload suitable for use in commercially leased light aircraft will be constructed and used in geophysically interesting regions for which information about aerosol properties is needed. Initial validation studies has focused on ground-based data, especially that taken from the Atmosphere-Ocean Chemistry Experiment (AEROCE) measurement network. This includes stations in Miami, Bermuda, Barbados, and Tenerife, among others. There is clear evidence for transport of material from African deserts to these stations. Preliminary results show good correlations of the TOMS aerosol index with the surface measurements of mineral dust concentrations. Attention should be paid to the fact that the TOMS measurements are at a single time of day (typically 11:30 AM), while the aerosol measurements tend to be longer-term averages (either one hour or nighttime). The location of the focused field campaigns to be carried out over the next few years has yet to be determined, but will probably emphasize aerosol measurements in "hot spots" - regions of expected aerosol sources.
Information on detection of volcanic ash using TOMS was summarized by Nickolay Krotkov of Raytheon STX, Inc., based on work he has done in conjunction with science team member Arlin Krueger. A case study examined was that associated with the eruption of Mt. Spurr in Alaska in August 1992. TOMS and AVHRR made measurements almost simultaneously on August 19, so direct comparisons of their observed aerosol field should be possible. One difficulty with such observations is that the particle size distribution is not known. The satellite systems were able to obtain information on the product of optical depths and the effective particle radius to give a column mass. There was good agreement between that calculated from TOMS and AVHRR, especially considering the significant error bars in the calculations. There appears to be significant potential for TOMS ash measurements to be able to provide information on total ash amounts from large volcanic eruptions. For some large eruptions, especially when plumes are in cloud-free regions, comparisons of TOMS ash amounts and sulfur dioxide columns may be carried out.
The comparison of TOMS measurements of surface UV radiation, total ozone, and cloud properties, with ground-based measurements at a single station in the U.S. Southwest (Socorro, NM), was presented by Ken Minschwaner of the New Mexico Institute of Mining and Technology. A Biospherical Instrument radiometer (GUV-511C), which combines a quartz teflon diffuser plate with 5 detectors (305, 320, 340, 380 nm and one measuring photosynthetically active radiation or PAR) in a temperature controlled jacket, is used at Socorro. This instrument was calibrated at the manufacturer before shipping and is recalibrated annually there against a double monochromator instrument. Its location in New Mexico means that it has cloud-free viewing conditions over much of the year. Comparison of the ground-based and TOMS ozone and UV measurements shows reasonable agreement. The ground-based measurements have also been used to measure the radiative amplification factor, and will be used to determine the direct/diffuse flux in solar radiation. These measurements can be useful in testing the algorithm used by TOMS for its surface UV measurements.
The use of TOMS data to study UV reflectivity of clouds, as well as the use of mountain-surface level differentials to help characterize total ozone distributions, were discussed by Yuk Yung of the California Institute of Technology. A long-term study on cloud optical thickness variations during the period from 1983 to 1991 was carried out using both TOMS UV reflectivity data and monthly mean cloudiness data from the International Satellite Cloud Climatology Project. The region emphasized in these studies was the Inter-Tropical Convergence Zone. The relationship between these two data sets, which are similar but not identical (and are based on different wavelength regions), is in the process of being understood. The origins of the variability observed over time are not yet known, especially the relative contributions from the solar cycle and from periodic El Niño events. The tropospheric ozone studies emphasized the region of Taiwan, which is a mountainous region off the east coast of Asia. TOMS data were used to determine tropospheric ozone through use of the mountain effect technique noted above; results show evidence of significantly higher tropospheric ozone on the west (Asian) side than the east (Pacific) side of Taiwan. Long-term studies suggest a small increase in tropospheric amounts over the length of the TOMS record. Although this increase is quite small in an absolute sense, it is large in a percentage sense (1-2%/year).
The contribution of dust in forcing of climate as inferred by studies of meteorological data assimilation products was discussed by Pinhas Alpert of Tel Aviv University in Israel. The motivating question behind this work is to understand whether the neglect of dust in climate models is an important source of error. By comparing dust distributions over the Atlantic Ocean with the residual errors obtained in GSFC's data assimilation model over the Atlantic, it was shown that dust appears to be a non-negligible source of error in atmospheric climate models (where dust is most often present, there are larger errors in the assimilation system than in dust-free regions; such a relationship is seen to vary seasonally and spatially with dust loading). The initial work did not use TOMS data, but now the TOMS data are beginning to be used, and clear correlations between the TOMS aerosol index and the errors in the assimilation are seen. This is true over Africa, South America, and the region near China, all of which have large amounts of biomass burning. TOMS and surface observations of a large dust storm in the middle east in March, 1998 were also shown.
Analysis of TOMS data taken over South America and adjacent regions, and comparison with ground-based data from Latin America were presented by Ruben Piacentini of the Rosario Observatory in Argentina. The presence of high levels of surface UV flux in South America is well established. In one recent case, the total solar irradiation measured at the Earth's surface was well above the "solar constant" because of the contribution of broken clouds, leading to multiple scattering/reflection. Comparisons of total ozone measured from EP TOMS with measurements made from Dobson stations are typically quite good, although differences at higher latitudes, especially Ushuaia, Argentina and the Argentine Antarctic station in Marambio, can be somewhat larger. An event leading to low ozone amounts in the Austral autumn of 1997 was shown. This is similar to the "mini-hole" events seen in the Northern Hemisphere, and was observed by three different Dobson instruments in South America.