Global monthly sea surface nitrate fields estimated from remotely sensed sea surface temperature, chlorophyll, and modeled mixed layer depth

Abstract

Information about oceanic nitrate is crucial for making inferences about marine biological production and the efficiency of the biological carbon pump. While there are no optical properties that allow direct estimation of inorganic nitrogen, its correlation with other biogeochemical variables may permit its inference from satellite data. Here we report a new method for estimating monthly mean surface nitrate concentrations employing local multiple linear regressions on a global 1 by 1 resolution grid, using satellite-derived sea surface temperature, chlorophyll, and modeled mixed layer depth. Our method is able to reproduce the interannual variability of independent in situ nitrate observations at the Bermuda Atlantic Time Series, the Hawaii Ocean Time series, the California coast, and the southern New Zealand region. Our new method is shown to be more accurate than previous algorithms and thus can provide improved information on temporal and spatial nutrient variations beyond the climatological mean at regional and global scales. Key Point Global estimation of oceanic nitrate variability through SST, Chl, and MLD