OCTS Ocean Color Reprocessing 2014.0

Introduction

The Ocean Biology Processing Group (OBPG) completed a full-mission ocean color reprocessing of the Japanese ADEOS Ocean Color and Temperature Scanner (OCTS) dataset on 26 August 2014. This reprocessing is part of a multi-mission effort to update common algorithms, product suites, and data formats across all supported missions. Sensor-independent changes are detailed in the R2014.0 Ocean Color Reprocessing General Description. Here we describe the OCTS sensor-specific details of the reprocessing, and provide an assessment of data quality and impact relative to the previous reprocessing. This latest reprocessing of OCTS follows closely the calibration approach of the R2006 OCTS reprocessing that was most recently updated in the R2010.0 multi-mission reprocessing.

Sensor-specific Processing Details

Source Data

Level-1A HDF4 source data and geolocation were provided by the Japan Aerospace Exploration Agency (JAXA) in 2001 and are unchanged. This global area coverage (GAC) data is approximately 1-km native resolution, subsampled on-orbit every 4th pixel and every 5th line.

Calibration

Since OCTS lacks any functional onboard calibration capabilities to track sensor degradation, the OBPG developed vicarious techniques to assess and improve the temporal calibration. The temporal calibration is unchanged for R2014.0, but the vicarious gains were updated to account for changes in the Level-2 processing algorithms and ancillary data. A discussion of the temporal calibration approach is provided here. Revised vicarious gains are shown in the table below:

Level-2 Processing

Again, the algorithms employed and products produced from OCTS are as described in the R2014.0 Ocean Color Reprocessing General Description. Sensor-specific center wavelength values used in processing and product naming are as shown in the table above.

The design of the OCTS sensor results in observations that are not co-registered within each band, so the co-registration is done in post-processing using a nearest neighbor algorithm. This results in individual scan lines that constitute a mix of detectors (OCTS has 10 detectors distributed along-track for each band). Imperfections in the detector-to-detector relative calibration give rise to systematic artifacts across-scan and from scan to scan. The co-registration and geolocation is as originally provided by JAXA. To reduce these striping artifacts, a sensor-specific smoothing filter is applied to the Rayleigh-subtracted OCTS radiances. Details are provided in the whitepaper Franz, B.A. and J.E. O'Reilly (2001), "An algorithm for the Reduction of Speckling and Striping Artifacts in OCTS GAC Data".

Impact and Quality Assessment

Impact of Reprocessing on Timeseries

To assess the impact of this R2014.0 reprocessing relative to the previous R2010.0 reprocessing, a comparative timeseries analysis was performed. See the whitepaper Franz, B.A. (2009), "Methods for Assessing the Quality and Consistency of Ocean Color Products" for details on the approach. The impact of all calibration, algorithm, and ancillary data changes on remote sensing reflectance is relatively small, as shown in this comparison of the weekly mean Rrs in various globally-distributed geographic subsets. Impact on the chlorophyll time-series is to increase the clear-water chlorophyll by about 10% on average, which is primarily an effect of the new chlorophyll algorithm employed.

Comparison with SeaWiFS

R2014.0 Ocean Color Reprocessing - A detailed analysis of temporal trends relative to SeaWiFS was also performed. The SeaWiFS data is not contemporaneous (years 2000-2001 versus OCTS 1997-1998), so an exact match is not expected, and the spectral bands at 510 and 555 differ from OCTS bands at 520 and 565, respectively. However, the trends provide some indication of relative bias and temporal stability. Results show some indication of temporal trend in OCTS water-leaving radiances relative SeaWiFS. No effort has been made to remove temporal degradation in the OCTS visible bands, so this may be an instrument artifact, but the onset of the 1997 El Nino may also be a contributing factor. Clear-water (oligotrophic) chlorophyll retrievals are in reasonable agreement between the two missions, but OCTS is biased low in the higher chlorophyll waters, relative to SeaWiFS 2000-2001 retrievals, as it was in the R2010.0 reprocessing.

Comparison with In Situ Measurements

Due to the short duration of the OCTS mission, we have very few in situ water-leaving reflectance measurements for direct validation of satellite retrievals. However, we do have sufficient chlorophyll match-ups to draw some tentative conclusions. The figures below show chlorophyll scatter plots and comparative statistics for the 2010.0 reprocessing and the 2014.0 reprocessing of OCTS relative to field measurements from the SeaBASS archive. Results show a significant improvement for R2014.0, particularly with regards reducing a negative bias over the full dynamic range.

R2010.0 OCTS Chlorophyll Validation

R2014.0 OCTS Chlorophyll Validation