NOVEL INSIGHTS INTO OCEAN TRACE ELEMENT CYCLING FROM BIOGEOCHEMICAL MODELS

Abstract

Ocean biogeochemical models have become critical tools for interpreting trace element and isotope (TEI) distributions observed during the GEOTRACES program and understanding their driving processes. Models stimulate new research questions that cannot be addressed with observations alone, for instance, concerning processes that occur over vast spatial scales and linkages between TEIs and other elemental cycles. A spectrum of modeling approaches has been applied to date, including (1) fully prognostic models that couple TEIs to broader biogeochemical frameworks, (2) simpler element-specific mechanistic models that allow for assimilation of observations, and (3) machine learning models that have no mechanistic underpinning but allow for skillful extrapolation of sparse data. Here, we evaluate the strengths and weaknesses of these approaches and review three sets of novel insights they have facilitated. First, models have advanced our understanding of global-scale micronutrient distributions, and their deviations from macronutrients, in terms of a “ventilation-regeneration-scavenging” balance. Second, models have yielded global-scale estimates of TEI inputs to and losses from the ocean, revealing, for instance, a rapid iron (Fe) cycle with an oceanic residence time on the order of decades. Third, models have identified novel links among various TEI cycling processes and the global ocean carbon cycle, such as tracing the supply of hydrothermally sourced Fe to iron-starved microbial communities in the Southern Ocean. We foresee additional important roles for modeling work in the next stages of trace element research, including synthesizing understanding from the GEOTRACES program in the form of TEI state estimates, and projecting the responses of TEI cycles to global climate change.