Errata
V8 data product errata:
January, 2021: An issue with the radiometric calibration for MOPITT NIR-only data has been identified. This leads to errors in NIR total column CO that grow to around 10% near the annual hot calibration every March. This issue is being investigated and we expect to have an improved NIR calibration method for V9 data.
V7 data product errata:
May 21, 2018: A bug has been identified in the V7 retrieval processing code which results in invalid total column averaging kernels (TCAK) for all retrievals for which the surface pressure is less than 900 hPa. This bug affects the TCAK diagnostics in the V7 Level 2 and Level 3 files, but not the reported V7 total column values themselves. A work-around for this bug is possible by calculating the TCAK from the full retrieval averaging kernel matrix, as described in Appendix 7.4 of the MOPITT Version 4 User's Guide, although a slight modification to that method is necessary because of the pressure level/layer associations used in V7 processing. An example IDL script demonstrating the correct method for calculating the TCAK from the full AK matrix for V7 is available here.
March 23, 2017: Due to an error in the reported PMC (pressure modulated cell) pressure following the February 2016 calibration activity, it was necessary to reprocess and redeliver MOPITT V7 data. Since V7 data are processed with interpolated calibration for the archival product, this required reprocessing for 2015 as well. We are still evaluating the errors in total column and vertical profile CO values, but so far these appear to be relatively small (< 0.5%) and should not affect most users who have already downloaded data from 2015 or 2016.
Units of DryAirColumn are not labled correctly in the L2 product
fldname = 'DryAirColumn'
attrval = 'g/cm^2'
The units of both DryAirColumn and WaterVaporColumn are molecules/cm^2
V6 data product errata:
March 23, 2017: Due to an error in the reported PMC (pressure modulated cell) pressure following the February 2016 calibration activity, it was necessary to reprocess and redeliver MOPITT V6 data. We are still evaluating the errors in total column and vertical profile CO values, but so far these appear to be relatively small (< 0.5%) and should not affect most users who have already downloaded data from 2016.
MOPITT V6N (NIR) and V6J (TIR/NIR) data produced after March 2012 and delivered to the ASDC were generated using forward processing for the calibration, which introduces errors compared to the products that use retrospective processing with time interpolation for the calibration data obtained from yearly "hot cal" events, usually scheduled in March. (See the the V7 User's Guide for more details).
V5 data product errata:
1) MOPITT V5N and V5J data produced after March 2011 and delivered to the ASDC were generated using forward processing for the calibration, which introduces errors compared to the products that use retrospective processing with time interpolation for the calibration data obtained from yearly "hot cal" events. (See the the V7 User's Guide for more details).
2) A small systematic error in the MOPITT geolocation values (latitude and longitude) has been identified. A detailed report analyzing this effect and describing a correction method is available here. This issue affects both current (V4 and V5) and previous (V3) MOPITT operational products. In MOPITT daytime (descending) overpasses, reported longitudes (contained in the Level 1 and Level 2 data files) appear to be systematically biased by approximately 0.35 degrees to the west of MOPITT's actual field of view. Reported latitudes in daytime overpasses do not appear to be significantly biased. The longitudinal error is similar to the size of a single MOPITT field of view (~22 km). Longitudinal errors increase at high latitudes where lines of longitude converge. Geolocation errors for nighttime (ascending) overpasses are different than errors for daytime overpasses. This data issue should be recognized particularly by researchers applying MOPITT products to study CO concentrations on fine spatial scales (e.g., tens of km). A method for eliminating this geolocation error at the initial stage of data processing is described in this report and will be implemented in the next operational MOPITT product (Version 6). For users of current MOPITT Level 2 and Level 3 products, the geolocation error can be corrected to first order using a latitude-dependent table of zonal-mean geolocation bias values included in this report.