Abstract: 

Dissolved organic matter (DOM) represents one of the largest reduced carbon reservoirs in the ocean, and its cycling is tightly coupled to microbial food web dynamics. Although zooplankton are known to release DOM through grazing, excretion, and fecal pellet production, their role in shaping DOM molecular composition, the physiological mechanisms underlying this process, and the fate of zooplankton-derived DOM under ocean warming remain largely unknown. In this study, we integrate ultra-high-resolution mass spectrometry, metatranscriptomics, and machine learning approaches to systematically investigate the chemical properties, molecular process behind the production, and molecular fate of zooplankton-derived DOM across estuarine-coastal waters and under simulated ocean warming scenarios. Through controlled incubation experiments with zooplankton communities from contrasting trophic regions (mesotrophic and eutrophic), we observe the mechanistic linkage between organismal metabolic response to environmental conditions and the molecular characteristics of their derived DOM. We further develop a machine learning model trained on molecular fingerprints of zooplankton-derived DOM to predict zooplankton contributions to natural DOM pools. Applying the model to estuarine-coastal water DOM samples demonstrates that zooplankton-derived DOM accounts for 16.5–20.1% of total assigned molecular formula, with its vertical distribution exhibiting diel pattern that mirrors zooplankton vertical migration. Extending these investigations to the context of future ocean conditions, we will use microcosm experiments to examine how warming-induced changes in zooplankton gene expression may alter the chemical properties of their derived DOM. Collectively, these works highlight the active role of zooplankton in the oceanic DOM cycle, showing that their contributions are physiologically regulated, behaviorally dynamic, and sensitive to environmental change.