NUMBER 66
DATE: 02/06/89
Sector Processing Center (SPC):
During December, NOA, ESA and JMA collectively processed greater than 96% of A data into B1/B2 data. AES processed more than 72% of A data into B1/B2 data and CSU processed more than 67% of A data into B1/B2 data.
Normal operations were reported by NOA, AES and JMA.
ESA reported that another exercise to decontaminate the infrared sensors (remove residual water vapor in the instrument) on METEOSAT-3, was carried out from 0900 GMT on 20 December to 0830 22 December, making the IR and WV channel calibrations unreliable for this time period.
AES reported losses of data on 31 December caused by a protracted power outage.
The loss of about nine days of December A data at CSU was due to high winds that damaged the receiving
antenna; the damage has been repaired. CSU (as well as the press) reported the loss of VISSR data from
GOES-6 on 21 January 1989, after more than five years of operations. The loss was caused by the failure
of the last encoder lamp in the imaging system; all other systems on the satellite appear to be operating.
NOAA will continue to use the satellite to broadcast information to users. GOES-7 is being moved to 108
West for the remainder of the winter season and should arrive on station by 21 February; its position will then
oscillate between 108 West in winter to 98 West in summer. CSU is planning to collect GOES-7 data as a
backup to AES.
Special Area Processing Center (SAPC):
No reports were received.
Satellite Calibration Center (SCC):
No report received.
Global Processing Center (GPC):
The GPC continued to receive B2 data and correlative data in a routine manner.
The first NOAA-10 B3 data was produced for December 86; the results of the calibration monitor will be used to normalize the VIS channel calibration of NOAA-10 to NOAA-9 and, thence, to NOAA-7.
Re-processing of NOAA B3 data to correct the small IR count error has been completed through November 84 for NOAA-7; all of the re-processed data for NOAA-7 and NOAA-8 have been delivered to the ICA.
Delivery of the geostationary satellite B3 data (GMS, METEOSAT, GOES) that have been re-inspected for shorter term calibration anomalies has begun; all of the data for 83 have been delivered to the ICA. All data for the first half of 84 have been re-processed and will be delivered this month. Inspection of data for the second half of 84 is underway.
TOVS data for 88 have been processed into the ISCCP Correlative Data format and delivered to the ICA.
Special pixel-level data for the FIRE Cirrus and Marine Stratus Intensive Field Observation periods (October/November 86 and June/July 87) are being processed into the FIRE standard format (except for GOES-6 data in 87 which is in a new format). The data from NOAA-9 have been delivered to NCDS (NASA Climate Data System); GOES data are being processed and should be completed within one week.
C1 data for December 83 and February 84 have been completed and sent to the ICA. Production of March and April 84 C1 data has started.
In a system as complex as this, IT was bound to happen. We have discovered a small, but important, error in the processing of the C1 data that affects all data in the archives. This error causes the cloud algorithm to compare the retrieved clear VIS radiances over sea ice with those from a model of reflectances for open water. This comparison is done for all water locations to detect residual cloud contamination in VIS only, but was supposed to be ignored for sea ice covered regions. The effect of this error is to increase the number of VIS cloud detections over sea ice (when there is sunlight) and to increase the retrieved value of cloud optical thickness. Comparison of the results with and without the error shows how remarkably hard it is to notice the effect; however, anyone specifically interested in sea ice areas would not like the result. Since the reflectance of open water at very high solar zenith angles is actually >20%, the error in surface reflectance is not that large (especially when converted back to radiance). Moreover, the relation between scene radiance and cloud optical thickness over bright surfaces is double-valued; hence substitution of a brighter surface reflectance in the correct analysis will not alter cloud optical thickness values that much (it is hard to predict this effect exactly since it depends sensitively on illumination and viewing geometry).
We have begun a re-processing of all data affected by the error (only regions covered by sea ice that is
illuminated by sunlight with solar zenith angles < 80 degrees are affected). Since the magnitude of the error
and the fraction of the globe affected are small, we are continuing the scheduled production (the error was
corrected in the most recently delivered data for December 83 and February 84) until we complete the first
year of C1 data; however, we are doing some of the re-processing in parallel (we have completed the
METEOSAT data for August 83). Thus, we expect to replace all of the affected C1 data by May 89. In fact,
we are carrying out another detailed inspection of the C1 data and will take this opportunity to correct other
small problems or inconsistencies in the format.
ISCCP Central Archive (ICA):
The ICA continued to receive B1 data from NOA, ESA, and JMA and B3 and C data from the GPC in a routine manner. Delivery of AES B1 data has begun.
In addition to your own comments about the contents of the C2 data; please send to the GPC any suggestions
for other people to contact, especially those interested in climate diagnosis studies. A new proposal,
incorporating suggestions, for C2 contents is being prepared; it should be sent out next month. First delivery
of C2 data is planned for next May.
The health of GMS-3, METEOSAT-3, GOES-7, NOAA-10 and NOAA-11 remained good. Some minor
engineering problems have appeared on NOAA-10, but imaging data quality has not been affected. Launch
of the next NOAA satellite is still scheduled for December 89. The last encoder lamp in the VISSR on GOES-6
failed on 21 January 89, ending imaging data from that satellite after five years of operation. The position of
GOES-7 is being changed to 108 West for the remainder of the winter season; after that, as with GOES-6 in
85-86, the position of GOES-7 will oscillate between 108 and 98 West. The earliest possible date for launch
of a replacement GOES satellite is summer of 90; however, even this date is in jeopardy. Launch of MOP-1
has slipped to 28 February 89.
Arranging for the delivery of INSAT B1 data.
Completing BC data for 1987 and 1988.
Correcting small, short-term calibration anomalies in the B3 data for 1984, 1985 and 1986.
Delivering all B3 data for 1986, 1987 and 1988 to the ICA.
Delivering snow and sea ice data for 1987 and 1988 to the GPC.
Completing the first year of C1 data.
Beginning delivery of C2 data.
Delivery of some special C data sets for FIRE, ICE and the Surface Radiation Budget project.