Evaluating the seasonal and decadal performance of red band difference algorithms for chlorophyll in an optically complex estuary with winter and summer blooms

Standard ocean color chlorophyll algorithms, including normalized fluorescence line height algorithms, over-estimate summer blooms and often miss large winter blooms in mid-latitude estuaries such as Long Island Sound (LIS). Thus, they are unable to accurately determine year-round chlorophyll concentrations. Here, we show that winter phytoplankton blooms cannot be retrieved using MODIS' 667 and 678 nm bands. Although MERIS' equivalent 665 and 681 nm bands are able to better characterize chlorophyll a concentrations at LIS compared to MODIS, we show that it is a red band difference (RBD) algorithm relying on observing the maximum spectral difference between R-rs681nm-R-rs665nm and R-rs709nm-R-rs665nm from the MERIS sensor that is able to successfully capture the seasonal chlorophyll dynamics measured in situ. Such switching algorithms are useful in areas where the shape of the reflectance spectrum changes throughout the seasons associated with changes in the phytoplankton community structure, and provide seamless implementation compared to algorithms based on seasonal tunings. We apply the switching RBD algorithm to the 10 years of available MERIS data, and results reveal important spatial and temporal patterns that cannot be assessed from in situ data alone. These include the observation that the decreasing trend in surface chlorophyll concentrations between 2002 and 2012 is localized, and may be related to changes in the flow of Atlantic waters into the estuary rather than a result of active management against eutrophication. Following from past studies, we demonstrate how high sensitivity red bands are valuable to successfully determine proxies for phytoplankton biomass in coastal areas where optical properties are dominated by dissolved organic matter, and where the fluorescence peak may not accurately capture the dynamics of productive waters. Such switching RBD algorithms applied to data from the Sentinel or PACE missions may allow distribution of near-real time chlorophyll a products for coastal waters across winter and summer seasons.