NUMBER 60
DATE: 08/05/88
Sector Processing Center (SPC):
During June, ESA, CSU and NOA collectively processed greater than 93% of A data into B1 and/or B2 data. AES processed 70% of A data into B1/B2 data.
Normal operations were reported by ESA, CSU and NOA.
AES reported the loss of six days of data by the protracted failure of the front-end receiving system. The April B1 and B2 data were successfully retrieved from the experimental optical disk storage system and delivered to the GPC.
CSU reported that all images, except the 0616 GMT image, are being transmitted from GOES-6 as partial images starting in early June. These images begin, randomly, at scanline 51 to 151 and are no more than 1350 lines in length (this implies a loss of about 300 - 400 lines or about 15-20% of each image).
ESA reported that METEOSAT-P2 is scheduled to become the operational satellite, replacing METEOSAT-2, on 11 August 1988.
The monthly report for May, but not June, was received from JMA.
Special Area Processing Center (SAPC):
No reports were received.
Satellite Calibration Center (SCC):
New BC data were received from the SCC for METEOSAT-2 for January 1988 and for GMS-3 for July 1987
through April 1988. At the WGDM meeting, the SCC reported that the changes in AC data formats and
calibration tables, introduced in April 1987 by CSU, together with a computer system change, will delay
processing of 1987 GOES-6 BC until late this year. Also, the processing of AES BC is delayed for similar
reasons. If the BC for GOES-6 and GOES-7 for 1987 can be processed by the end of this year, then the
backlog of BC will have been eliminated.
Global Processing Center (GPC):
The GPC continued to receive B1 and/or B2 data and correlative data in a routine manner.
The final tests of the new B2 data format from JMA have been completed and no serious problems were found. Only one or two minor inconsistencies between the documentation and the format were discovered, but the format is acceptable. A final report is being prepared for JMA.
NOAA-9 B3 re-calibration has been completed and the data delivered to the ICA through March 86. The October 86 IR calibration anomaly is still being investigated.
Several small clues in the QC statistics and the C1 data, together with the report of a heater failure on GOES-6 by CSU and the discovery of a diurnal IR calibration cycle by JMA, have led to an extensive re-examination of all of the radiance data covering about three years. This study has uncovered numerous short-term calibration changes in both the VIS and IR channels for all of the satellites. These changes include single image anomalies or changes over periods of one to several days that are too brief to be detected by the SCC normalization procedure. In addition, both GMS and GOES satellite exhibit occasional spurious diurnal cycles of IR calibration that vary in amplitude. Although these calibration changes are relatively smaller than those anomalies that have been previously identified, they cause significant errors in the C data.
If the calibration of a particular image increases the VIS reflectances or decreases the IR temperatures, then the cloud amount determined for that image suddenly increases well above the statistical average and variation for that satellite. However, if the calibration change decreases the VIS reflectances or increases the IR temperatures, then the cloud amount for a week or a month can be increased because these anomalous values may get into the clear sky composite. For example, occasional changes of the GOES-6 IR calibration, equivalent to 3 - 5 K changes in sea surface temperature, caused significantly higher cloud amounts in January 84 for this satellite. Therefore, using the QC statistics that identify these anomalous images and time periods, the GPC has started to correct the calibrations to eliminate the spurious diurnal variations and sudden episodic changes in the radiance data. This processing will require about two months to clean up the first 2.5 years of B3 data already delivered to the ICA. Since the corrections must be determined from the QC statistics derived from the B3 data, B3 data production will actually continue; however, deliveries of B3 data for 86 and 87 to the ICA will be delayed until the data for 83 through 85 are corrected.
GOES-6 and GMS-3 B3 production has been completed through December 85; however, deliveries to the ICA have been halted after August 85 for GMS and September 85 for GOES to complete corrections of short-term calibration anomalies, discussed above. METEOSAT-2 B3 production for 85 has been started. Despite the delay necessitated by the spurious calibration changes, all B3 data through 1986 ought to be delivered before the end of this year. If the BC data can be obtained and the format and navigation questions resolved for GOES, then the 1987 B3 should also be completed in early 1989.
The TOVS data for 1985 have been delivered to the ICA. Production of TOVS data for 1986 is about 75% complete.
Production of the final versions of the July 83 and Jan 84 C1 data have been completed; the July 83 data will
be delivered to the ICA once the documentation is revised. The January data cannot be delivered until the
IR calibration of GOES-6 is corrected to eliminate several spurious, one day changes and a diurnal cycle of
about 4 K. Re-processing of the FIRE Cirrus Special data set from NOAA-9 is underway and should be
completed next week. The remainder of August will be spent in producing the rest of the FIRE special data
sets, except GOES-6 for June/July 1987, and the WGNE/ERBE data set for April 85.
ISCCP Central Archive (ICA):
The ICA continued to receive B1 data from JMA, CSU, ESA, and NOA in a routine manner.
The health of GMS-3, METEOSAT-2, GOES-6, GOES-7, NOAA-9 and NOAA-10 remained good.
METEOSAT-2 is scheduled to be replaced by METEOSAT-P2 on 11 August 1988. NOAA-H launch is
scheduled for 24 September 1988.
The seventh meeting of the Working Group on Data Management was held last month in Banff, Canada, hosted by AES. Review of data collection activities indicated no serious problems; all SPC's are now fully operational and delivering B1 to the ICA and B2 to the GPC. However, a number of minor difficulties continue to plague the production and archival of final data products. The most notable problems are adjusting to the change in GOES data transmission format, finding suitable cold and bright clouds in the GOES-WEST area for proper radiance normalization, eliminating some questions about scanline numbering for GOES-EAST, eliminating the spurious diurnal calibration cycles for GOES and GMS, and correcting episodic calibration changes that have been found in all satellite data. Investigations are underway or correction procedures are being implemented for all of these problems. The first five years of B3 data, covering the initial ISCCP data collection period (July 83 through June 88), should still be completely archived in early 1989, only slightly behind the original schedule.
Review of the cloud algorithm and data products showed no major issues. Study of the first C1 data for July 83 and January 84, distributed last year, showed the need for some minor adjustments in the algorithm; these changes have been tested and implemented. However, comments on the data contents and definition of the reported statistics, received from researchers, have been used to refine the C1 data set to improve its usefulness for climate research. Most of the needed changes in C1 data contents were discussed and approved at the WGDM meeting last year; some additional changes were approved this year. All these changes have implemented. A second draft of the C1 data documentation was presented at the meeting for review. Although some changes were recommended at the meeting, no further revisions have been suggested. Delivery of C1 will begin this month as soon as the documentation can be completed.
Although the contents of C2 data were again discussed, no decision on C2 data contents was reached. A
proposal forwarded by the GPC will be reviewed by the Working Group members in the next two months.
Based on comments received from the Working Group members, the GPC will distribute a revised proposal
in October of this year for final comments. First delivery of C2 data is planned to coincide with the completion
of the first year of C1 data in early 1989.
Obtaining NOAA-10 B1/B2 data over the Indian sector starting in December 1986.
Arranging for the delivery of INSAT B1 data.
Completing BC data for 1987.
Delivering all B3 data for 1985 and 1986.
Delivery of special C data sets for FIRE, the WGNE, and ERBE.
Beginning delivery of the final version of C1 data.
Beginning delivery of C2 data.