GRP Report to GEWEX SSG-14

Global Satellite Projects

ISCCP: The available data record now covers 1983-2001 (18 years) (various products with resolutions from 30 km, 3 hr to 300 km, monthly). A polar-orbiter-only record extends back to August 1981. A slow variation in global monthly-mean cloud cover, but no other cloud property, has been observed that is consistent with an observed change in top-of-atmosphere radiative fluxes. An analysis of cloud variations associated with different strength midlatitude cyclones suggests a cloud-radiative feedback would be produced by a change of the distribution of storm strengths.

GACP: The available data record now covers 1983-2001 (18 years) (oceans only, monthly mean at 300 km resolution). Time variations in global aerosol optical thickness and average size are consistent with evolution of the stratospheric aerosols between volcanic events. No long-term trend is apparent, but calibration drifts play a significant role in the determining the results.

ERB: The available data record from a combination of the NIMBUS-7 ERB, ERBE, ScaRab and CERES datasets now covers 1979-2002 (23 years) (various products with resolutions from 30-50 km, daily to 300 km, monthly). Notable variations are associated with Pinatubo and ENSO events; a long-term variation in the tropics is consistent with slow cloud cover changes. GERB launched on MSG-1.

SRB: The available data record now covers 1983-1995 (12 years) (various products with resolutions from 100 km, 3 hr to 100 km, monthly). Production will resume later this year to extend the record to 2002. Another radiative flux product has also been produced by the ISCCP team covering 1983-2001. Surface shortwave flux variations are consistent with cloud cover changes; surface longwave flux variations suggest significant "slow" exchanges of energy between the atmosphere and ocean.

GPCP: The available data record now covers 1979-2002 (23 years) (various products with resolutions from 100 km, daily to 300 km, monthly -- the daily product covers 1997-2002). Snow over oceans is not included, however. There appears to be a notable decrease in global mean precipitation during the Pinatubo event but no other significant variations.

GVAP: The NVAP pilot study produced a 11-yr water vapor data record covering 1987-1998 (daily, monthly at 100 km resolution). The NOAA operational sounder analysis provides a water vapor data record covering 1979-2002 (23 years) (resolution is 300 km, daily); an analysis of this record has been done focused on the long-term variations of upper tropospheric humidity, which show trends at low latitudes that may explain part of the observed change of longwave radiative fluxes.

Evaluation Projects

The global satellite analyses are supported by a number of studies to help evaluate their accuracy. The major extrinsic evaluation activity organized under GRP is the BSRN to support the SRB project. Currently, BSRN consists of 35 functioning sites in 19 countries. More than 1800 monthly data records (5 min resolution) are available from the central archives, representing an average of 4 years of data from all stations. Significant clarification and, in some cases, reduction of the sources of measurement error have been achieved. Significant activities in evaluating water vapor measurement accuracies occurred over the Southern Great Plains ARM site and as part of IHOP 2002. TRMM validation activities indicate that there are still major problems in validating precipitation measurements of any sort.

Study Projects

GRP has organized several specific study projects to examine particular problems and to foster work towards their solutions. The most notable study projects are ICRCCM and SeaFlux.

ICRCCM-SW: The study shows that treatments that represent small-scale variability as an "independent column approximation" and calculate areally-averaged fluxes for horizontally varying layers provide fairly accurate results (accounting for absorption pretty well). A test kit containing specifications of inputs and outputs from 3-D and other radiative transfer models is being prepared for publication on the Web. Information about and representations of the radiative transfer in ice clouds remains a significant problem.

ICRCCM-LW: A new test kit for the Web is being prepared to focus on tests in cloudy scenes and over the full range of water vapor abundance.

I3RC: This study has shown that all of the various methods currently used to calculate 3-D radiation agree quite well. A test kit for checking new methods is on-line. An new working group is now being organized under the auspices of the International Radiation Commission to examine how radiation couples to the atmosphere (boundary layer dynamics) and the surface at the scales where the radiation is 3-D. The GRP endorsed this activity and highlighted the role that GCSS/GABLS/GLASS could play in this study.

SeaFlux: A workshop is planned for February 2003 to evaluate progress on several comparison activities: retrievals of skin SST and air temperature/humidity, turbulent surface flux formulations and global flux products. Contact has been made with the GODAE-SST group to exchange ideas and test results (some of their test products are included in the SeaFlux comparison activity). New versions of global products are now being released. This group is also interacting with others to compare retrievals of sea surface winds, especially from scatterometers, and precipitation over oceans.

LandFlux: The chair of the GRP attended the meetings of WEBS/GHP and ISLSCP last year and held discussions with participants in GSWP. Currently, none of these efforts is focused on development or compilation of global, long-term versions of the land surface datasets (albedo, skin temperature, latent and sensible heat fluxes) that are needed to close the global energy and water budgets.

Feedback WS: A workshop on Climate Feedbacks was held in Atlanta on 18-20 November 2002, organized jointly by the GRP and the JSC/CLIVAR Working Group on Coupled Models (see accompanying report).

Objective Analysis WS: A workshop on the Objective Analysis of Precipitation, organized by GPCP, is to be held at ECMWF on 11-13 March 2003. Although focused on the problem of interpolating sparse precipitation measurements, this workshop is expected to address general issues of the interpretation of sampled measurements.

Working Groups

Organizational meetings of the new Column Profiling WG have occurred. The objectives for this group are to discuss common issues, particularly regarding analysis methods, and to establish some common practices and data formats so that the rest of research community can more easily use these datasets.

The several data management working groups for the global data projects have been re-organized into a single Working Group on Data Management and Analysis (WGDMA) to coordinate activities better for the purposes of (1) conducting joint data analyses to describe the global energy and water budgets more completely and consistently, (2) examining the natural variability of these budgets and beginning analyses of feedbacks, (3) stimulating the more complete use of data products from new satellites, (4) cross-connecting better with other GEWEX elements, especially for WEBS and GCSS/GLASS, and (5) connecting with CLIVAR, CLIC and SPARC data management groups to facilitate joint activities.

Highlights of Plans

The highlight of plans for the coming year (see also specific actions/recommendations in more complete report of GRP-13):

1. The review of GCM radiative transfer code features and metrics for testing clouds and radiation in GCMs will be completed.

2. An effort to define and apply a "standard" set of statistical and diagnostic analyses to all of the GEWEX global data products (and possibly a few others) will resume. The purpose of this exercise is to characterize the variability of all the measured quantities in some common way as a prelude to a joint analysis. This activity was delayed until after the Climate Feedback Workshop.

3. More development of feedback analysis methods is needed and the best way to do this is in joint data-model teams. The GRP will work with WGCM to formulate a plan of action.

4. A workshop on polar clouds and precipitation (snow) will be organized jointly with CLIC.

5. Joint activity with SPARC on aerosols (Pinatubo) & upper atmosphere water vapor (convection & cirrus) & radiation will be explored.

Issues

For the GEWEX SSG

1. The general interaction of the GRP projects with WEBS/CEOP and ISLSCP is currently very weak to non-existent. The consolidation of GRP data management activities into a single group will now provide a single point of contact with other GEWEX elements. There is a specific need for an effort within GEWEX that is focused on completing the characterization of the global, long-term energy and water cycles. If GRP is to lead this activity (as recommended at the last SSG meeting), then clear lines of responsibility need to be defined to avoid (the appearance of) duplication with other activities.

2. Reviewing the experiences that GPCP and TRMM have had in trying to validate satellite-measured and (even) ground-based-radar-measured precipitation, the GRP reached three conclusions. (1) The different precipitation measurement communities (using different measurement tools) are not using consistent formulations of the characteristics of precipitation. (2) There is not currently any adequate validation data available, especially over oceans. (3) The latter is true, in part, because there is no measurement system that actually resolves the precipitation process. To resolve these key difficulties and to prepare better for GPM, the GRP proposes that GEWEX (possibly the whole WCRP) organize an International Precipitation Year (IPY) to organize an intense global effort to collect the most complete and detailed set of precipitation measurements possible and to focus analysis efforts on resolving these difficulties. This could (and probably should) include efforts to collect better snow measurements. If TRMM lasts beyond 2004 and CloudSat launches on schedule in late 2004, then the year 2005 would be the best year because, in addition to two precipitation radars in space (TRMM is better suited to heavier rainfall, whereas CloudSat is more sensitive to lighter rainfall and snowfall), there would be two AMSR (together with TMI) and two SSM/Is, three AMSU/MHS systems, probably two scatterometers, together with TERRA, ENVISAT, AQUA and ADEOS-2 plus the constellation of weather satellites. Collections from these could be combined with especially intense collections (continuous coverage) from precipitation and cloud radars, particularly the precipitation radar networks in the US, Europe and Japan, and from rain and snow gauge networks on all continents. Possible limits on the lifetime of TRMM and the delayed launch date of CloudSat might require that this effort start in 2004 and continue through 2005.

3. The Feedback Workshop did not develop any definitive plans for making further progress on this problem. Actions to be taken include posting the material from the Workshop and contributed by the participants on a Web site to begin an outreach and education effort that tries to elucidate the issues and collect ideas for resolving them from a larger group of researchers. The WG on Coupled Models (WGCM) decided to go ahead with its already planned "feedback" experiments, incorporating some suggestions from the WS, but generally following "classical" approaches. Two further activities will be to work on forming ad hoc research teams composed of modelers and data analysts to try testing some of the new analysis ideas on common datasets and from a hierarchy of models, possibly leading to a second workshop in about 18 months. These results suggest two things: that this topic needs to be given more attention throughout WCRP, rather than being limited to GRP or WGCM and that a specific JSC task group might be useful if composed of modelers and data analysts along the lines of GCSS-DIME and GLASS-ALMA. Then specific tests of new analysis ideas could be performed focused a few common problems using common datasets and a range of models. This approach might hasten development and evaluation of new approaches.

For the JSC

1. The climate feedback issue concerns all of WCRP, not just GRP or WGCM.

2. The fact that GCOS plans specifically highlight the key importance of radiation as a measurement for long-term monitoring and yet has expressed reservations about the "quality" of the Baseline Radiation Network because its funding base is not "established", when, in fact, the BSRN is struggling to maintain funding, raises a key issue that the GCOS plans do not contain any procedure for transitioning key research observations to an (operational) long-term observing system. To date much GCOS effort has focused on the state of observing systems that are already operational. The WCRP needs to join with the other partners (IGOS-P) in discussing how best to begin moving well-established research systems to an operational system. This is the climate equivalent of establishing the World Weather Watch. BSRN is an early and primary candidate for this transition.

3. The GRP notes with concern a general lack of plans to cross-calibrate the new experimental sensors on TERRA, AQUA, ENVISAT and ADEOS-2 and to cross-compare the data products that come from these experiments.

4. The GRP notes that the lack of a data management group (or point of contact) within CLIVAR that can provide information on datasets being collected, assembled and analyzed as part of CLIVAR activities has inhibited effective communication and collaboration between the CLIVAR and the GRP global satellite data projects, both for the exchange of datasets and for carrying out studies of mutual interest. In organizing CLIC, providing a data management focus should be a high priority.