--Gary G. Gibson (g.g.gibson@larc.nasa.gov), NASA Langley Research Center
The 17th Clouds and the Earth's Radiant Energy System (CERES) Science Team meeting was held at the NASA Langley Research Center (LaRC) in Hampton, VA on April 21-23, 1998. The major topics included an assessment of CERES and Visible Infrared Scanner (VIRS) data from the Tropical Rainfall Measuring Mission (TRMM) spacecraft, launched in November 1997; a review of progress on the validation for each subsystem; and CERES launch readiness for the first Earth Observing System (EOS) satellite.
Bruce Wielicki (LaRC), CERES Co-Principal Investigator, opened the meeting with an EOS program status report. The CERES instruments for EOS-AM and EOS-PM are on schedule. A slip in the EOS-AM launch schedule of at least 9 months is likely due to spacecraft control software problems. EOS follow-on mission concepts are due to the EOS Project in June. CERES will propose flying 6-spectral-channel instruments on small spacecraft with a downsized Moderate Resolution Imaging Spectroradiometer (MODIS) instrument in the same orbits as EOS-AM and -PM plus an orbit inclined at 67 degrees.
Bruce Barkstrom (LaRC), CERES Co-Principal Investigator, reviewed the processing strategy for getting CERES data into the archive. CERES will do all processing for TRMM and EOS-AM on the Langley TRMM Information System (LaTIS). Barkstrom outlined two options. Option 1 calls for full production of BDS (BiDirectional Scan), ES8 (ERBE-like S8 data product), ES9, ES4, and SSF (Single Satellite Footprint) products with immediate public access. Option 2 is to put BDS in full production now and provide validation month samples of ES8, ES9, and ES4, and validation-day samples of SSF. The Science Team endorsed the second option because it focuses on getting out the highest quality data.
Jack Cooper (LaRC) presented the instrument status report. The CERES instrument on the TRMM spacecraft is operating as expected with no problems. The two EOS-AM Flight Models (FM1 and FM2) completed thermal vacuum testing in March. FM1 performed nominally throughout the test, but FM2 experienced unexpected watchdog timer resets during the testing. The FM2 instrument was removed from the spacecraft and investigation into the problem continues. Delivery of FM3 and FM4 for EOS-PM is scheduled for December 1, 1998.
Robert B. Lee III (LaRC) confirmed that the Level-1 geolocated filtered radiances from the CERES instrument on TRMM meet accuracy requirements and recommended that these data be archived. Kory Priestley (LaRC) reported that the CERES zero-radiance offsets are an order of magnitude less than those for the Earth Radiation Budget Experiment (ERBE). The numerical filter designed to account for a slow transient found during instrument testing is now operational. The sensors and calibration sources are stable within 0.5% of the TRW ground calibrations.
The CERES instrument Level-1b data are ready to archive, but Level-2 ERBE-like data products are not ready. The ERBE-like-versus-VIRS scene identification results are currently inconsistent and require further validation. Studies have been prioritized to validate ERBE-like products by July 1998.
Bruce Wielicki reported on a recent presentation to the Department of Energy Atmospheric Radiation Measurement (ARM) Science Team on the need for continuous cloud and radiative flux data sets at the ARM sites to determine remote-sensing errors for a complete range of cloud and satellite viewing conditions. ARM surface sites for measurement of cloud properties and radiative fluxes are the most critical source of CERES ground truth data.
Richard Green (LaRC) presented validation results for the ERBE-like ES8 product. The processing uses the same code, Angular Distribution Models (ADMs), and Maximum Likelihood Estimation (MLE) technique as ERBE. The only changes are: the longwave (LW) channel is replaced by a window channel, new spectral correction coefficients, and a smaller field of view. An examination of the data showed that LW at night is significantly higher than from the Earth Radiation Budget Satellite (ERBS). One possible cause for the difference could be the current El Niño event.
Norman Loeb (Hampton University) noted that the CERES ERBE-like MLE appears to overestimate daytime clear-sky frequency of occurrence relative to ERBE and VIRS. A comparison of SW radiance showed good agreement between CERES, ERBS, and the Scanner for Radiation Budget (ScaRaB) over the oceans after spectral correction change for clear scenes and CERES footprint degradation. Over land and desert, CERES was up to 10% lower than ERBS for clear scenes. The CERES spectral correction was shown to be highly sensitive to aerosol optical depth and surface reflectance over ocean due to a decrease in the spectral response function between 0.3 and 0.4 µm. The average SW clear-sky albedo over the tropics changes by about 5% due to differences in CERES and ERBS footprint size.
Kory Priestley showed 3-channel intercomparisons for CERES. He simulated the LW broadband (BB) radiance from the window channel measurements for high, cold cloud conditions. Results of the intercomparisons implied an inconsistency between the SW channel and the SW portion of the total channel. David Kratz (LaRC) used line-by-line and correlated k-distribution models to calculate top-of-atmosphere (TOA) radiances. A comparison of these results with linear fits to unfiltered total and filtered window channel data showed a discrepancy between the theory and measurements that could not be reconciled even by extreme changes to atmospheric ozone and water vapor abundance and temperature. Confidence in the simulated LW BB radiance remains low until this issue is resolved.
David Young (LaRC) validated the CERES ERBE-like monthly mean products using comparisons to other data sources. The monthly mean comparisons are consistent with inversion studies. CERES-derived clear-sky albedos that are 4.5% lower than ERBS over oceans and 6.5% lower over land were attributed, in part, to temporal sampling deficiencies. Differences in other LW and SW radiances were within the ERBE-like uncertainty limits.
Takmeng Wong (LaRC) examined the effects of the El Niño event on the outgoing clear-sky LW flux. The Fu-Liou radiative transfer model did reproduce the basic El Niño signal in the observed 1998 clear-sky LW radiation field using Reynolds sea-surface temperatures and National Centers for Environmental Prediction (NCEP) atmospheric parameters. The observed 1998 clear-sky LW anomaly is consistent with the temperature and moisture anomalies that accompany an El Niño event. Very little correlation was observed between changes in sea-surface temperature and the clear-sky LW El Niño anomaly.
Patrick Minnis (LaRC) gave an overview of the cloud mask and cloud property retrieval algorithms. Cloud retrieval using VIRS is still problematic, but a great deal of progress has been made. The VIRS thermal infrared calibration looks good. Work is continuing to unscramble problems with solar channel calibrations (still inconsistencies at 10% level, negative night time data). A fix to the VIRS 1.6-µm thermal leak is underway. The leak should not be a problem for cloud retrieval, but may be a significant problem for aerosol retrieval. Validation against ARM data in the Tropical Western Pacific (TWP) site has been delayed by ARM cloud lidar failure for January through March.
Robert Seals (LaRC) presented the status of the Langley TRMM Information System (LaTIS) for CERES data processing at the Langley Distributed Active Archive Center (DAAC). He discussed the wide range of DAAC services for producing, archiving, and distributing Earth science data.
Jim Kibler (LaRC) presented the CERES Data Management System (DMS) status. The CERES Release 2 DMS deliveries to the DAAC were completed before the TRMM launch. The TRMM Instrument Support Terminal worked well for monitoring post-launch operations. The instrument and ERBE-like subsystems are running continuously at the DAAC using the LaTIS. A decision has been made to use LaTIS for EOS-AM processing for CERES. Quick-look data from TRMM were posted on the World Wide Web on 12/31/97, only 4 days after the contamination covers were opened. The BDS data were made available to team members on 2/13/98, and the ES8 data were available on 3/13/98.
Bill Collins (National Center for Atmospheric Research, NCAR) compared satellite observations of visible and near-infrared radiation reflected by the ocean and atmosphere to fluxes calculated with a general circulation model. The Nimbus satellite data and model simulation are global and extend from 1979 to 1987. Under clear-sky conditions, the observed and modeled spectral albedos are nearly identical. The observations and simulations diverge with increasing cloud albedo and cloud amount, regardless of cloud type or cloud phase. The differences between the model and satellite data occur for all tropical and mid-latitude ocean regions and exhibit minimal seasonal and interannual variability.
Lin Chambers (LaRC) gave an update on the CERES Students' Cloud Observations On-Line (S'COOL) Project. The S'COOL project has been selected for use in a joint French/U.S. demonstration of science education cooperation during a May visit by Mrs. Clinton to France. Dan Goldin and Bruce Barkstrom will be at a school in the Washington area. Schools in New York and France will also participate. [Ed. note: The demonstration took place as planned and was very successful.]
Instrument Working Group: Robert B. Lee III led the Instrument Working Group meeting in discussions of the accuracy of the CERES instrument on TRMM. Measurement accuracy and precision goals have been satisfied. The Working Group recommended that Level 1 filtered geolocated radiances be archived. On-orbit solar calibrations indicate sensor measurement precisions of the order of ±0.3%. On-orbit internal calibration results indicate measurement precisions of the order of ±0.7%. The flight offset variations are less than 1.5 counts. No coefficient changes will be made unless the validated gain changes are derived above 0.5% for LW radiances and 1.0% for SW.
Cloud Working Group: Patrick Minnis led the discussion of cloud retrieval, archival, data dissemination, and validation issues. He presented a long list of candidate algorithm improvements. The group unanimously agreed that Barkstrom's Option #2 was desirable, i.e., releasing oceanic cloud amounts and heights in July. The Cloud Working Group is still on track for validation of the cloud product by 2 years after launch.
Jim Coakley (Oregon State University) and Ron Welch (University of Alabama-Huntsville) will examine the gross characteristics of both the cloud mask and parameter retrievals when the data sets become available. Coakley is a member of the Indian Ocean Experiment (INDOEX) Science Team and will request that CERES have access to lidar imagery for validation purposes. Jennifer Francis (Rutgers University) initiated discussion about using alternative input data sources for constructing independent validation data. Xiquan Dong of Analytical Services and Materials, Inc. (AS&M) discussed the validation of CERES cloud products using ground-based measurements and retrievals. Jay Mace (University of Utah) presented results from lidar/radar retrievals for cloud properties. A full year of Southern Great Plains (SGP) retrievals is available for validation activities. Ron Welch showed automated regional and global contrail detection results from the Advanced Very High Resolution Radiometer (AVHRR). He also presented Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) validation sites that could be used for CERES.
Surface and Atmospheric Radiation Budget (SARB) Working Group: A joint meeting of the SARB and Surface-only Working Groups co-chaired by Tom Charlock (LaRC) and David Kratz addressed algorithm and validation issues. Several presentations focused on applications of radiative transfer codes to problems related to SARB and its validation. Charlock explained that the SARB system is an application of the Fu-Liou radiative transfer code to the Meteorology, Ozone, and Aerosol (MOA) data in which the computed fluxes are constrained by CERES TOA measurements. The latest version of Subsystem 5.0 was delivered to the DAAC on April 10, 1998, and results from test runs are already available.
Jim Stobie (GSFC/Data Assimilation Office, DAO) discussed the status and quality of Goddard Earth Observing System (GEOS)-2 products. Charlock suggested that DAO consider direct assimilation of Tiros Operational Vertical Sounder (TOVS) water vapor channel radiances into their system to improve the upper tropospheric humidity (UTH). Shi-Keng Yang (NOAA/NCEP) reviewed the status of the Stratospheric Monitoring-group Ozone Blended Analysis (SMOBA), which is used as the primary source of ozone data for CERES processing. Comparisons with other ozone products demonstrate that the vertical distribution, and temporal and horizontal variabilities of ozone are well captured by SMOBA.
V. Ramanathan (Scripps) presented the information available from the Indian Ocean Experiment (INDOEX) which can potentially be used for validation of CERES results. Nitchie Smith (AS&M) outlined the CERES ARM Validation Experiment (CAVE), a continuing program for validation of CERES results in the Oklahoma/Kansas region surrounding the ARM SGP site. CERES retrievals will be validated with site-measured values and those derived from radiative transfer models.
Fred Rose (AS&M) presented results of a study in which radiance computations made with MODTRAN code and European Centre for Medium-Range Weather Forecasts (ECMWF) inputs were used to examine CERES/TRMM results. Differences were calculated between CERES measurements and MODTRAN computations for clear-sky filtered radiances for the window channel, and unfiltered radiances for broadband LW for a 1-hour swath over the eastern tropical Pacific Ocean. These differences were found to be very small for the filtered window radiance, but larger and negative for the unfiltered broadband radiance. Slight cloud contamination of CERES clear scenes was indicated. Rose also showed the dependence of the differences in unfiltered broadband radiances on the UTH.
David Kratz described an effort to revive and document the Staylor SW model for deriving surface albedos from CERES TOA measurements. These albedos will be used to convert the surface net SW retrievals made with the Li-Leighton algorithm into surface insolation. Anne Wilber (AS&M) presented global maps of surface emissivity for the CERES window channel, broadband LW, and the 12 bands of the Fu-Liou radiative transfer code. These maps were developed from available laboratory and field measurements of emissivities of various surface/soil/vegetation types.
Yong Hu (Hampton University) presented a method being developed for computing equivalent albedos of ocean surface as a function of solar zenith angle, wind speed, and cloud optical depth. The method uses the Cox-Munk wind speed-wave pattern relations and the DISORT radiative transfer code for simulating a large database of spectral radiances for a wide range of conditions. The radiance database will be used to develop parameterizations for equivalent surface albedo in terms of solar zenith angle, wind speed, and cloud optical depth.
Time Interpolation and Spatial Averaging (TISA) Working Group: David Young led discussions of software development, current temporal and spatial averaging studies, and ongoing CERES ERBE-like validation efforts. There was general agreement that the ERBE-like TISA algorithm is functioning properly. Currently, only Subsystem 3 (ERBE-like monthly) and Subsystem 11 (Grid Geostationary) have been tested and run at the Langley DAAC. All other Subsystems have been delivered and are awaiting data for testing. Doug Spangenberg (AS&M) presented results from an analytical method for estimating temporal sampling errors developed for testing ERBE non-scanner monthly mean fluxes. A discussion of possible uses of this technique for CERES raised the possibility of adapting the method for estimating the errors from poor temporal sampling of clear-sky fluxes. Lou Smith (Virginia Tech) and Bob Kandel (CNES, France) of the ScaRaB Science Team raised the possibility of producing a combined CERES/ ScaRaB monthly mean product once the second ScaRaB instrument is launched.
Tim Alberta of AS&M (representing Tom Charlock) examined the transmission of SW radiation to the surface. CERES/ARM/GEWEX, Global Energy and Water-Budget Experiment (CAGEX) calculated clear-sky insolation is significantly higher than that measured by ARM instrumentation. A relationship between the bias between calculated and measured surface insolation and the humidity path length was not evident in CAGEX Version 2; however, a strong relationship exists between this bias and aerosol path length. Robert Cess (State University of New York at Stony Brook) compared atmospheric clear-sky SW radiation models to collocated ERBS and surface insolation measurements for 24 stations in southern Canada. The goal was to determine whether or not the results indicate anomalous clear-sky SW absorption. The two models are a satellite-to-surface algorithm for net surface SW and a column radiation model from the NCAR Community Climate Model (CCM3). Both models were in excellent agreement with the surface insolation measurements and there was no evidence that the models incorrectly portray any important physical processes.
Jim Coakley (Oregon State University) examined cloud layer structure from the Lidar In-Space Technology Experiment (LITE) and from AVHRR observations using a spatial coherence analysis method. He determined the frequencies of cloud-free, single-layered, and multilayered cloud systems when viewed at different spatial scales. At an 8-km scale over ocean scenes, LITE results and spatial coherence analysis were in good agreement.
Steven Dewitte of the Royal Meteorological Institute of Belgium (representing Dominique Crommelynck) reported on the real-time radiation budget processing of METEOSAT data and suggested a strategy for comparison with CERES data from TRMM.
Leo Donner (GFDL) addressed cloud-ice sedimentation rates and the implications for cloud-radiative interactions. Current estimates show that the range of ice-terminal speeds is quite large with correspondingly large ranges in microphysical and radiative properties of convection. There is significant uncertainty in estimates of ice fall speeds. High priority should be given to resolving this issue which is important for both cloud-system models and General Circulation Models (GCMs).
Qingyuan Han of University of Alabama-Huntsville (representing Ron Welch) summarized the varying definitions of effective particle size and their effects on satellite retrievals, particularly with respect to ice particles. The effect of assuming different shapes of crystals was also examined including the applicability of various definitions of effective particle size for these shapes.
Alexander Ignatov, UCAR Visiting Scientist (representing Larry Stowe, NOAA/NESDIS) presented results of a preliminary analysis of aerosol retrievals from TRMM/VIRS. They retrieved aerosol optical depth at 0.63 µm using the AVHRR channel-1 algorithm, and at 1.6 µm using an AVHRR channel-1-consistent algorithm with a correction for the 5-µm leak.
Anand Inamdar of Scripps (representing V. Ramanathan) reported on tropical- and global-scale interactions between water vapor, atmospheric greenhouse effect, and surface temperature. Their data do not support the suggestion that increases in tropical or global mean surface temperature would lead to a decrease in water vapor greenhouse effect by drying the mid to upper atmosphere. The global-scale sensitivity derived from the annual cycle is consistent with the magnitude of the positive feedback obtained by GCMs.
Bing Lin of Hampton University (representing Patrick Minnis) used SSM/I data, ARM ground observations, and a microwave radiative transfer model to investigate the temporal variations of land surface microwave emissivities and their relationships with soil moisture.
David Randall (Colorado State University, CSU) presented results from tests of a coupled ocean-atmosphere-land surface model. Problems due to cloud feedbacks developed when the model was coupled with the CSU GCM.
Nitchie Smith of AS&M (representing Lou Smith) showed the effects of randomness of radiation anisotropy on empirical bidirectional reflectance models. Bidirectional reflectance distribution functions (BRDFs) computed by dividing the average grid radiances by average albedo are in error due to a correlation between BRDF and albedo.
Shi-Keng Yang of Research and Data Systems Corporation (representing Jim Miller, NOAA/NCEP) presented recent modifications to the NCEP Reanalysis including a new SW code and surface albedo. Improvements to the NCEP model are still needed in several areas.
The next CERES Science Team meeting is scheduled for September 15-17, 1998 at the State University of New York at Stony Brook, NY. Major topics will include assessment of CERES and VIRS data from TRMM, data validation and archiving, and CERES launch readiness for EOS-AM.