UC Berkeley

This dissertation focuses on two aspects of biogeochemical cycling essential for our understanding of future carbon cycle trajectories and carbon dioxide removal strategies: carbonate crystal growth and soil carbon cycling. In the first two chapters, I develop the use of calcium isotopes as molecular probes that shed light on the complicated suite of processes occurring at the fluid-mineral interface during crystal growth. The solution stoichiometry dependence of calcium isotope fractionation is first investigated as a direct test of classical ion-by-ion models of isotope partitioning during calcite growth (Chapter 2). Fractionations measured through a series of constant-composition calcite growth experiments are well-captured by an ion-by-ion model that incorporates the influence of surface speciation. This provides strong supporting evidence for the model of calcium isotope discrimination driven by Ca exchange at kink sites on the growing crystal surface and yields much-needed constraints on the solution chemistry dependence of Δ44/40Ca, critical for the interpretation of Ca isotopes in natural systems. Model predictions of the relationship between Δ44/40Ca and growth inhibition in the presence of impurity ions then lay the theoretical groundwork for the use of Ca isotopes to probe interfacial processes during carbonate crystal growth.

In Chapter 3, I operationalize this new tool, employing calcium isotope fractionation to help elucidate the mechanism by which two divalent cations with starkly contrasting compatibility, magnesium and manganese, interact with the growing calcite surface and are ultimately incorporated into the mineral lattice. Invariant Δ44/40Ca with increasing {Mn2+}/{Ca2+} and {Mg2+}/{Ca2+}, despite more than an order of magnitude decline in growth rate, is indicative of a dominantly kink blocking inhibition mechanism. For Mg2+, experimental trends are consistent with inhibition driven by slow Mg2+-aquo complex dehydration relative to Ca2+, but large Mn2+ partition coefficients cannot be explained by desolvation rate-limited attachment of Mn2+. Instead, I argue that the dominant Mn2+ species interacting with kink sites is not the free ion in solution but instead an ion pair, hydrated species, or possibly larger polynuclear cluster and that growth kinetics are limited by a carbonate-based kink blocking mechanism. These findings raise questions about the prevalence and broader ramifications of non-monomer trace constituent incorporation during otherwise classical crystal growth.

In the second half of the dissertation, I turn to carbon cycling at the field scale, investigating the controls on soil carbon cycling and soil-atmosphere CO2 exchange in the Mojave Desert. Arid soils can contain significant concentrations of inorganic carbon in the form of pedogenic carbonate, but the short-timescale dynamics of the soil inorganic carbon system and its impact on CO2 fluxes remains poorly constrained. I present results from a multi-year field campaign, including two years of continuous measurements of meteorological and soil conditions from a series of soils along a climate/elevation gradient. In Chapter 4, I focus on unpacking the primary controls on CO2 production in these highly water-limited ecosystems, and develop quantitative models to describe how the sensitivity of CO2 production to environmental conditions varies with depth in the soil profile and spatially on scales of meters to kilometers. Significant nighttime CO2 pulse events observed in near-surface soils of the more densely vegetated, higher elevation, sites are also explored and provisionally linked to microbial activity stimulated by the delivery of non-rainfall moisture to the litter layer and surficial soils.

In Chapter 5, I discuss the causes and consequences of two types of CO2 consumption documented at the lowest elevation, most arid site: periods of frequent negative nighttime surface fluxes during the dry season and acute episodes of CO2 uptake following rain events. The negative surface fluxes are driven by almost continuous nighttime CO2 consumption in shallow soil layers (0-15cm depth), the magnitude of which is strongly dependent on the amplitude of the diurnal soil temperature oscillation. Quantitative evaluation of potential driving mechanisms suggests that thermal impacts on the soil carbonate system alone cannot produce the magnitude of consumption observed, and that temperature-dependent CO2 adsorption to soil minerals may also contribute to an abiotic diurnal cycle of CO2 uptake and release in these desert soils. In contrast, the CO2 uptake observed following rain events is consistent with CO2 consumption due to carbonate mineral dissolution, potentially augmented by near-surface biotic carbon fixation. I end with a discussion of the broader implications of the documented inorganic CO2 fluxes for the interpretation of carbon dynamics in arid ecosystems.