Satellite-derived NPP and SST

Corresponding to the EP measurements described above, there were very limited measurements of in situ ¹⁴C-NPP. In contrast to the EP measurements that represent the average over several weeks, NPP in the euphotic zone was measured instantaneously during incubation, which lasted only several hours. The temporal difference between these two measurements could result in pronounced decoupling^38,72,73. To overcome these problems, we used satellite-derived NPP products with a 1/6° spatial resolution and 8-day temporal resolution for the period 1997–2013, obtained from the Ocean Productivity website at Oregon State University (https://www.science.oregonstate.edu/ocean.productivity/). Three different NPP products were used based on different algorithms: chlorophyll-based vertically generalized productivity model (VGPM; ref.⁷⁴), the Eppley-VGPM (Eppley; ref.⁷⁵), and carbon-based productivity model (CbPM; ref.⁷⁶). Overall, the differences between these models reflect differences in sensitivity of photosynthetic rate to temperature and to different indicators of phytoplankton biomass.

The daily mean Optimum Interpolation Sea Surface Temperature (OISST) dataset, with a horizontal resolution of 0.25°, was obtained online ( https://www.esrl.noaa.gov/psd/data/gridded/data.noaa.oisst.v2.html).