NUMBER 92

DATE: 04/08/91

Sector Processing Center (SPC):

During February, NOA, ESA, JMA and AES processed greater than 99% of A data into B1/B2 data. AES also reported processing 100% of A data into B1/B2 data for January.

Normal operations were reported by NOA, ESA, JMA and AES.

CSU reported shipment of the INSAT B2 data for 87 to the GPC. A special problem occurred in these data, namely, one bit occasionally appeared to be permanently "on" in every other image pixel. A search of count value frequencies, plus comparisons of neighboring pixels, allowed for a statistical correction of this problem. Since the stuck bit was one of the less significant bits, the effect of this problem was small and, with the correction, is a random error at most.

ESA reported the successful launch of METEOSAT-5 on 2 March 91. Commissioning of this satellite has commenced, but image data will not be collected before the beginning of April. If successfully commissioned, METEOSAT-5 will become the prime operational satellite in May.

Special Area Processing Center (SAPC):

No reports were received.

Satellite Calibration Center (SCC):

MET-4 BC data for October 90 were sent to the GPC. All BC data for 90 have been delivered and the time lag for delivery of BC data is now as originally planned.

Global Processing Center (GPC):

The GPC continued to receive B2 data, BC data and correlative data in a routine manner.

Production of NOAA-11 (preliminary) B3 data is current. NOAA-11 B3 data have been re-calibrated through September 89 and data through June 89 sent to the ICA. A small (1-2%) calibration shift in Channel 4 (IR) radiances occurs in the last few months of 89, coincident with the first time that a problem was encountered with stray sunlight in the instrument during the calibration cycle over the illuminated south pole. NOAA reports that this problem was detected in Channel 3 calibrations, but that no effect on Channel 4 calibration was noticed. However, an exmination of the statistical history through the end of 90 at the GPC shows that a similar, though larger, change also occurs at the end of 90. This repeated occurrence suggests that the stray sunlight, which occurs only with the sun near winter soltice, is causing some calibration shift. The nature of this shift is still being investigated.

Production of NOAA-10 (preliminary) B3 data is current. Further re-calibration of NOAA-10 B3 data awaits determination of whether a week of data from late May through early June 89 can be recovered; the B2 data that were received are filled with a constant count value. Corrections for the changed visible channel calibration coefficients have been determined.

Production of METEOSAT (preliminary) B3 data is current.

Production of GOES-7 (preliminary) B3 data is current. Because of the calibration check provided by the SCC using AC data, which looked completely normal, no one noticed that the GOES-7 B2 visible count values produced by AES were half what they are supposed to be. After investigation by AES, it appears that two errors occurred. First, with the several changes of personnel at AES, the requirement to report the averaged visible count values as 8-bit counts was not understood. They intended to deliver 6-bit count values. Second, an arithmetic error led to the production of 7-bit count values, instead. Since the SCC received Stage A data and performed its own averaging, the SCC normalization results looked very similar to those for GOES-5 and GOES-6. After testing this new version of the calibration in the cloud analysis, all GOES-7 B3 data will be re-processed to multiply all visible counts by two. Since AES cannot re-process the Stage B datasets (they do not archive Stage A data) and since we have about four years of data, no change in the count values produced for GOES-7 will be made; ie, all Stage B1/B2 data will have 7-bit visible counts, while Stage B3 data will have 8-bit count values. The correct averaging procedure will be implemented by AES when GOES-7 is replaced.

Production of GOES-6 B3 data is completed.

Production of GMS-4 (preliminary) B3 data is current. An error in the IR calibration adjustments performed by the GPC was discovered in B3 data for August and September 87. This occurred at a transition between calibration tables that were corrected by JMA after the SCC normalization and corrected tables supplied directly to the SCC. This error has been fixed and the replacement B3 data shipped to the ICA.

INSAT-1 B2 data for 87 were received from CSU. Nominal calibration information, including spectral response functions and pre-launch calibrations, were also obtained.

Production of TOVS correlative data is current; however, a small error was discovered in 89 and 90 data that affects only a few grid cells. These data have been re-processed for January through June 89.

C1 data for September 87 have been completed. With all of the problems reported above, some repair work on the C1 datasets for April - August 87 is required. The re-analysis of the GMS B3 data for August 87 has been completed and the analysis of September 87 B3 data for GOES-7 is being examined to be sure that the visible radiance calibration is correct.

ISCCP Central Archive (ICA):

The ICA continued to receive B1 data from NOA, AES, ESA, and JMA and B3 and C data from the GPC in a routine manner.