Insights and predictions of Phytoplankton Biomass in Equatorial Pacific using Random Forest Modeling

Abstract

Phytoplankton biomass drivers are integral to our understanding of how climate change will impact the carbon cycle as photosynthesis and carbon sequestration play a crucial role in mitigating climate change. The equatorial Pacific is a region with high primary productivity due to the convergence of easterly trade winds and equatorial currents, though few oceanographic cruises take place here. To address this lack of data, this study both assesses the predictive performance of random forest regression models for phytoplankton biomass and identify which environmental variables play the largest role in biomass. I hypothesized that nutrient concentration, salinity, and sea surface temperature are dominant determining factors of phytoplankton biomass because they represent key environmental conditions that influence the growth and distribution of phytoplankton in the world's oceans. Results suggest that Prochlorococcus demonstrates a large correlation between biomass and salinity, while iron plays a lesser role. Nitrate emerges as a vital predictor for Prochlorococcus, reflecting its role as a limiting nutrient. In contrast, Synechococcus relies heavily on phosphate, with nitrate contributing minimally due to specialized assimilation systems. Picoeukaryotes are primarily driven by nitrate, with salinity and temperature exerting less influence. By filling a knowledge gap in picophytoplankton biomass in the equatorial Pacific, this study contributes to our understanding of phytoplankton dynamics in the face of climate change and El Niño events. Further research is necessary to explore additional factors and enhance prediction models for a comprehensive understanding of equatorial Pacific phytoplankton dynamics.