Trace metal -carbon interactions in marine phytoplankton: Implications on the cellular, regional, and global scale

Abstract

Trace metal availability modulates carbon uptake by phytoplankton, which in turn affects the carbon cycle on regional and global scales. This thesis investigates mechanisms and implications of trace metal-phytoplankton interactions at cellular, regional and global scales.

On the cellular scale, we develop a model to explain carbon isotopic fractionation of photosynthesis (ε_p) as a function of carbon dioxide (CO₂) concentrations, growth rates, cellular carbon source, and active carbon uptake. The model explains published ε _p-observations for cyanobacteria, diatoms and coccolithophores, including the effects of zinc-limitation and variations in the ratio of calcification to photosynthesis. The variability of ε_p due to cellular physiology increases the uncertainty in deducing CO₂ concentrations and phytoplankton growth rates from measurements of ε_p.

On the regional scale, we examine trends in the data collected in the GEOSECS (1970's) and WOCE (1990's) global ocean studies for evidence of decadal changes in the biological carbon pump in the North Pacific. Nitrate concentrations and apparent oxygen utilization decreased in deep waters (below roughly 1000 m, p ≤ 0.05) but increased (p ≥ 0,05) in shallower waters. The deep water tracer trends are too large to be caused by perturbations of the biological carbon pump, but can be explained by changes in water ventilation rates. Trends in oxygen and nitrate concentrations provide relatively poor constraints on decadal scale trends in the marine biological carbon pump because they should occur mostly in highly variable and undersampled shallow waters, and because alternative explanations cannot be firmly rejected.

On the global scale, we explore the link between scientific uncertainty and economic optimal CO₂ abatement. We first incorporate a simple description of a CO₂ dependent environmental threshold (a North Atlantic thermohaline circulation collapse) into an economic optimal growth model. Our analysis suggests that significant CO₂ abatements may be profitable in order to avoid or delay even small (and arguably realistic) damages from such a threshold. A CO₂ dependent threshold increases the sensitivity of optimal CO₂ abatement to uncertainties in the CO₂ sinks. Finally, we argue that scientific uncertainty (with or without learning) can act to decrease optimal CO₂ abatement.