Advanced Microwave Scanning Radiometer (AMSR) Science Team Meeting
--R. Spencer (roy.spencer@msfc.nasa.gov), Team Leader, NASA Marshall Space Flight Center
--E. Lobl (elena.lobl@msfc.nasa.gov), Team Coordinator, Earth System Science Laboratory, U. of Alabama in Huntsville
The first U.S. EOS PM-1 AMSR Science Team meeting was held on April 23-24, 1996 at Goddard Space Flight Center. The agenda included welcoming remarks from the EOS PM Project Office, status of the hardware, and other team business. In the remaining time, the Team members discussed their roles and responsibilities in producing the standard product algorithms and additional needed validation experiments. It is expected that all of the Team members will have important roles to play in the successful delivery and operation of standard data product software for the PM-1 AMSR. The Team also recognized the need to include their Japanese counterparts in algorithm development or refinement, to the extent possible, given the short lead time before the Algorithm Theoretical Basis Documents (ATBDs) and other documentation are due in late 1996.
A meeting of the newly selected ADEOS-II AMSR Team members will take place in Tokyo in late June. At this meeting, it is hoped that each of the AMSR product types will be represented by visiting U.S. Team members. It is expected that a useful dialog will be initiated with the Japanese on algorithm issues related to both ADEOS-II and the PM-1 AMSR instruments. Of particular concern to the U.S. Team is the desire to receive ADEOS-II AMSR data soon after launch in mid-1999.
Marty Donohoe, EOS PM Project Manager, welcomed the Team to Goddard and gave the status of the common spacecraft contract: the date of the contract restart was April 15, and the redesign of AMSR (to fit in the PM-1 launch shroud) is complete. Pete Pecori, PM-1 Instrument Manager, and Bernie Graf, AMSR Project Manager, have worked on this interface and are satisfied that the latest proposed design from the Mitsubishi Electronic Company (MELCO, the AMSR instrument contractor) will be successful.
Michael King, EOS Senior Project Scientist, discussed briefly the PM-1 mission status and then concentrated on the Team Member and Team Leader responsibilities. He then reviewed the ATBD contents and schedules. The plan is for our ATBDs to be in to the Project Science Office (PSO) by November 15, 1996, to be reviewed, and the peer review presentation to take place in early 1997.
Frank Wentz will have the lead role for the development and application of a spatial resampling scheme for matching the high-resolution, high-frequency channels of the AMSR to the low-frequency, low-resolution channels. This processing is necessary for most algorithms that require multiple-frequency observations. Much of this development work has already been accomplished, and was presented to the Team at the meeting. One of the major findings of this work was that a regridding of all of the data to a raster-scan geometry or latitude/longitude grid is counter-productive since it smoothes out some of the spatial information content of the raw data.
Roy Spencer, AMSR Science Team Leader, presented the ground rules for an acceptable standard data product algorithm. The Team agreed that these standards should include (but not necessarily be limited to):
He also presented a proposed ATBD index, matching Team members with algorithms:
| PMR-01 | Level 1c algorithm | F. Wentz |
| PMR-02 | Ocean Parameter Suite | F. Wentz, T. Wilheit, J. Alishouse |
| PMR-03 | Sea Ice | D. Cavalieri, J. Comiso |
| PMR-04 | Precipitation | R. Adler/C. Kummerow, T. Wilheit, J. Alishouse/R. Ferraro |
| PMR-05 | Land | TBD (Team member to be selected) |
| PMR-06 | Snow | TBD (Team member to be selected) |
Following is a summary of the discussions between the algorithm authors:
The lead role for production of the "ocean parameter" suite will be assumed by Frank Wentz, with John Alishouse leading the effort on ground-truth validation data sets, and Tom Wilheit providing a regression-based set of ocean suite retrievals as a validation "sanity check." The physical retrieval of the ocean suite of parameters from the DMSP SSM/I has been very successful and has demonstrated the superiority of physically-based algorithms where the physics of the various retrieved parameters are well defined. The use of routinely gathered meteorological data for validation will be critical to validation of the ocean parameters, with buoy and radiosonde data being the most useful types of data. These data will be gathered by NOAA. Tom Wilheit's regression-based ocean retrievals will be tested after launch of ADEOS-II AMSR for the possibility of using them as a first guess in the physical retrieval method.
Agreement could not be reached at the meeting concerning which algorithm should be used for sea ice parameter retrieval. As a result, Don Cavalieri and Joey Comiso agreed to hold a workshop in late July with the goal of having the sea ice community form a consensus about which technique should be implemented as a standard product. The ground rules for standard products (discussed above) will provide the primary basis for selection. Both of these Team members are aware that this will reduce the time available to meet the November due date for the ATBDs, and they will plan their schedules accordingly.
The precipitation retrieval procedures will build upon the Tropical Rainfall Measuring Mission (TRMM) experience. A cloud-model and radiative-transfer- model-based physical retrieval algorithm from GSFC (Chris Kummerow/Bob Adler) will provide Level 2 oceanic rain estimates, while a NOAA/NESDIS (Grody/Ferraro/Alishouse) empirical algorithm will provide rain estimates over land. The Level 2 oceanic rain rates will be input into Tom Wilheit's probability distribution function (pdf) technique to provide Level 3 (monthly gridpoint) rainfall estimates. As a separate validation procedure, Tom Wilheit will separately retrieve Level 2 rain rates and, through his pdf procedure, provide monthly rain rates that can be compared to the GSFC rain totals.
It was also tentatively agreed to retrieve the relative proportions of convective versus stratiform precipitation. This is consistent with the basic information content of the low-frequency observations (rain below the freezing level) and high-frequency observations (large ice condensate above the freezing level).
This Team member has not yet been selected.
This Team member has not yet been selected.
Dawn Conway, the Team Leader's software engineer, presented briefly some of the EOS standards that will have to be followed in the coding of the algorithms.
The last topic for discussion was a Validation plan for the AMSR standard products. This plan will rely heavily on routinely gathered meteorological data. These data have proved valuable in the past for validation and calibration of SSM/I algorithms. The accompanying table, assembled during our Team meeting, illustrates the types of data that will be needed for the validation efforts after launch. In Table 1, "Expected Data" refers to those routinely gathered data that will only need to be coordinated and made available to specific team members, and includes data from other satellite instruments as well, e.g., AVHRR sea surface temperatures. "Needed Data" refers primarily to aircraft deployments of existing radiometer packages.
The next Team meeting is planned for June 12, 1996 at Goddard Space Flight Center, a day before the PM-1 platform science working group meeting.
Table 1. PM-1 AMSR Validation/Field Experiments
| Ocean Suite |
|
| Precipitation |
|
| Sea Ice |
|
| Land/Snow |
|