Updated MISR Dark-Water Research Aerosol Retrieval Algorithm – Part 1: Empirical Calibration Corrections and Coupled 1.1 km Ocean-Surface Chlorophyll-a Retrievals

Abstract

As aerosol amount and type are key factors in the “atmospheric correction” required for remote-sensing chlorophyll-a concentration (<i>Chl</i>) retrievals, the Multi-Angle Imaging SpectroRadiometer (MISR) can contribute to ocean color analysis despite a lack of spectral channels optimized for this application. Conversely, an improved ocean-surface constraint should also improve MISR aerosol-type products, especially spectral single-scattering albedo retrievals. We develop and apply new calibration corrections to the MISR top-of-atmosphere (TOA) reflectance data, and introduce a self-consistent retrieval of <i>Chl</i> together with aerosol over dark water. The calibration corrections include: a modified stray-light model based on comparison with coincident MODerate-resolution Imaging Spectroradiometer (MODIS) Terra data, and trend analysis using MISR TOA bidirectional reflectance factors (BRFs) over three pseudo-invariant desert sites. The trend analysis shows that MISR radiometric sensitivity decreased by up to 2 percent for MISR spectral bands between January 2002 and December 2014. <br><br> After applying calibration corrections, we run the MISR Research Retrieval Algorithm (RA) to validate the MISR RA-retrieved <i>Chl</i>, and analyze both the MISR and corresponding MODIS-Terra values compared to a set of 49 collocated SeaBASS in situ observations, constrained to <i>Chl</i>_{<i>in situ</i>}&amp;thinsp;<&amp;thinsp;1.5&amp;thinsp;mg&amp;thinsp;m^&amp;minus;3. Statistically, compared to the validation data, MODIS demonstrates a higher correlation coefficient (r) of 0.91 vs. 0.86 for MISR, a lower root-mean-squared-error (RMSE) of 0.25 vs. 0.22, but a higher median absolute error (MAE) of 0.14 vs. 0.10. Because 49 data points are insufficient to draw strong conclusions, we also compare MODIS-Terra and MISR RA <i>Chl</i> statistically, over broader regions. With about 1.5 million MISR-MODIS collocations having MODIS <i>Chl</i>&amp;thinsp;<&amp;thinsp;1.5: r&amp;thinsp;=&amp;thinsp;0.96, MAE&amp;thinsp;=&amp;thinsp;0.09, and RMSE&amp;thinsp;=&amp;thinsp;0.15. MISR-MODIS agreement is substantially better than the 49-data-point MODIS-SeaBASS comparison, indicating that MISR <i>Chl</i> retrievals might complement MODIS, especially after further upgrades are made to the MISR RA ocean color model. <br><br> The new dark water aerosol/<i>Chl</i> RA can retrieve <i>Chl</i> in low-<i>Chl</i> (<&amp;thinsp;1.5), case I waters, independent of other imagers such as MODIS, via a largely physical algorithm, compared to the commonly applied statistical ones. At a minimum, MISR’s unique multi-angular data can better constrain aerosol type, helping reduce uncertainties in the MODIS Terra Ocean color retrieval, and suggesting how a joint MISR-MODIS over-ocean algorithm might exceed the capabilities of either instrument alone.