SeaWiFS Reprocessing 5 Validation Analysis

The effect of the proposed SeaWiFS reprocessing changes on the comparison with in situ data are presented here. The difference in N between the operational and reprocessing sets is due to additional data added since the last time the operational set was run. Given the large sample sizes, the statistics should be comparable despite the differnce in N. A description of the methods employed for these analyses, with minor modifications, can be found in Volume 10, Chapter 7 of the SeaWiFS Post Launch Technical Memorandum Series.
There are approximately radiance and chlorophyll matchups which pass all exclusion criteria.
Statistics Summary:
Current Operational Processing Proposed Reprocessing
Band N Mean Ratio* % Difference** Bias rms+
nLw_412 491 0.8748 22.564 -0.1477 0.4425
nLw_443 593 0.931 17.201 -0.0898 0.3546
nLw_490 551 0.955 14.140 -0.0659 0.3059
nLw_510 424 0.9584 13.476 -0.0542 0.3030
nLw_555 593 0.990 14.5190 -0.0132 0.2308
nLw_670 571 1.0764 47.836 0.0012 0.0731
chl 1267 1.106 35.908 0.104 0.7085
N Mean Ratio* % Difference** Bias rms+
523 0.9203 22.7320 -0.0683 0.3384
625 0.9572 17.189 -0.0316 0.2696
625 0.947 14.5180 -0.0443 0.2210
490 0.932 14.684 -0.0546 0.2110
625 0.9410 16.698 -0.0412 0.1785
590 0.9134 50.897 -0.0098 0.0678
1392 0.9610 36.473 -0.0456 0.7303
* arithmetic mean of all satellite / in situ ratios
** median absolute percent difference
+ sqrt ( &sum(satellite - in situ)^2 / n )
chlorophyll data were log transformed prior to statistics calculation

The Plots


Discussion

Overall, there was little change with the in situ comparisons between operational (reprocessing 4) and the proposed reprocessing results. The biases were slightly reduced with the proposed reprocessing, although the median percent difference (absolute) are nearly identical between the two.