Distributional Models of Ocean Carbon Export

Cael Barry, Ph.D., 2019
Michael Follows, Advisor

In the ocean, particle fluxes connect the entire microbiome and are fundamental to biogeochemistry. Particle fluxes are variable and complex, impeding mechanistic or quantitative description. This thesis pairs data compilations with probabilistic mathematical models to explore relationships between particle flux and temperature, net primary production, and depth. First, I develop a simple thermodynamically-based mechanistic model that explains the observed temperature dependence of export efficiency. I then estimate the metabolically-driven change in average export efficiency over the past three decades, and show the underlying mechanism may help explain glacial-interglacial atmospheric carbon dioxide drawdown. Second, I show that measurements of particle flux and production are relatable via their probability distributions, because on large spatiotemporal scales both are lognormally distributed. I derive a relationship between their distributions’ log-moments and find agreement between independent estimates of this relationship, suggesting that upper ocean particle flux is predictable from production on large spatiotemporal scales. Third, I show several particle flux-versus-depth models with different mechanistic and surface export implications capture observations equivalently. I then propose a flux-versus-depth model accounting for measurements of both the flux profile and the settling velocity distribution, and thus more consistent with empirical knowledge.