MODAPS PCR 10-005
PCR Number | 10-005 |
---|---|
Date | 2010-04-27 |
Initiator | Xu Geng |
Abstract | Level 1B sector rotation correction. |
Problem Statement | L1B products often show spuriously high radiances prior to sector rotations, in some cases so high that the Fire Product would often spuriously report outbreaks of fire, even in places like the Antarctic. Investigation showed that these symptoms were caused by the timing of the instrument command to perform the sector rotation and the recording of the telemetry point that reports the angle of sector rotation. This telemetry point is sampled at a frequency of 8 sec. Scans between samples are written with a copy of the last reported value. As a result, some scans that were collected after the start of a sector rotation cannot be identified as such. As a result, those scans are considered valid by L1B, but have extremely low counts. The b1 parameter is calibrated against a 40-scan average, and as result is anomalously high whenever it tries to compensate for the presences of such scans within that average. As result, scans prior to the sector rotation are calibrated to with anomalously high radiance values. All TEB (except band 21) are affected. |
Description of Change | The algorithm was changed so that when a packet indicating a sector rotation of 0 is preceded or followed by one showing a non-zero sector rotation, all scans of data collected between those two packets are treated as suspect, since there's no way to know which of them were collected before the change, and which of them were collected afterward. They are therefore not included in the 40-scan average used to set the b1 calibration coefficient. |
Products Affected | Level 1B TEB bands except band 21 |
Software Affected | PGE02 |
Processing String to Receive the Change | Terra Forward, Aqua Forward |
Downstream Product Effects | All products that use the TEB bands, but in particular, the Fire Product. |
Data Granules to be Used for Science Testing | Data Granules to be used for Science Testing:Aqua: 2008-343 22:35Terra: 2006-313 17:30 and 2006-314 17:35 |
Justification | We have no way of determining whether or not the scans in question were collected during a sector rotation. If they were, they do not contain radiances from the surface of the Earth, and should therefore not be used for the purpose of calibrating the b1 coefficient. The high radiances created by the incorrect use of those scans for that purpose are entirely spurious. |
Effective Date for Implementation of Change | Terra V6.1.6/Aqua 6.1.7: 2010-03-10, 2009-12-21Terra V5.0.46/Aqua V5.0.39: 2010-03-31, 2010-03-17 |
Status | Steven Platnick approved for Atmospheres, Chris Justice approved for Land,Michael King approved on 05/04/2010. |