| Method |
To estimate regional Chlorophyll trends, researchers divided the global ocean into ten regions, in which similar variability in phytoplankton biomass was observed in response to seasonality and climate forcing (Behrenfeld, M. J., Boss, E., Siegel, D. A. & Shea, D. M. Carbon-based ocean productivity and phytoplankton physiology from space. Glob. Biogeochem. Cycles 19, 1–14 (2005).) (Fig. 3a). To capture the range of potential Chl trajectories, regional trends were estimated from generalized additive models (GAMs) as linear functions of time on a log scale in three different ways: (1) continuous (linear trend), (2) discrete (mean year by- year estimates), and (3) smooth functions of time (non-monotonic trend). This approach is comprehensive: it allows both the quantitative (magnitude) and qualitative nature (shape) of trends to be estimated. |