Date of Award

Summer 2001

Document Type


Degree Name

Doctor of Philosophy (PhD)


Ocean & Earth Sciences



Committee Director

Eileen E. Hofmann

Committee Member

Larry P. Atkinson

Committee Member

David M. Karl

Committee Member

John M. Klinck

Committee Member

Curtis D. Mobley


In recent years, a new appreciation for the role of diazotrophy in the oceans has emerged. This dissertation reports on three modeling studies designed to investigate ecological processes associated with Trichodesmium spp., the most conspicuous marine diazotroph: (1) characterization of a generalized model Trichodesmium and issues of macronutrient resource competition; (2) carbohydrate ballasting by Trchodesmium and implications for the formation of surface accumulations; and (3) the vertical distribution of Trichodesmium and implications for detection from space.

The first study focuses on issues of nitrogen and phosphorus competition and ecosystem structure. It utilizes a simple ecosystem model that includes dissolved nitrogen and phosphorus plus two classes each of primary producers, grazers, and particulate detritus. In a monoculture submodel, the Trichodesmium biomass is most sensitive to the nitrogen: phosphorus compositional ratio and the senescence and gross growth rates. In the competitive model, Trichodesmium is adversely affected by competitors for model phosphorus, while the contribution of diazotrophy to fueling non-diazotrophy new production is limited by the concomitant lack of other nutrients. This model's outcome is most sensitive to the Trichodesmium gross growth and senescence rates. Experimental studies that would be particularly useful in this context include determination of the Trichodesmium half-saturation coefficient for phosphate, as well as quantitative co-occurrence data for Trichodesmium and Macrosetella gracilis.

In the second study, an individual-based Lagrangian model is used to explore carbohydrate ballasting and its implications for Trichodesmium vertical distribution in quiescent waters. The model results indicate that mean population depth is controlled primarily by environmental conditions (incident irradiance and its vertical attenuation) and physiological rate parameters for ballast processes. Morphologic parameters have a greater effect on the amplitude of ballast-driven oscillations. Post-mixing quiescence, high incident irradiance, and high water clarity all encourage the formation of surface accumulations. Post-mixing quiescence produces a depth-segregated population, with the proportion ascending to the surface increasing as a function of water-column turbidity. This study provides insight into environmental and biological conditions that encourage Trichodesmium accumulation at the marine boundary layer and identifies key processes for further study.

In the third study, a radiative transfer model is used to quantify the effects of Trichodesmium vertical distribution on remote-sensing reflectance, Rrs(λ). For the detection thresholds employed here, the model results indicate that surface accumulations of Trichodesmium can be detected when chlorophyll ≥1.5 mg m−3. For near-surface populations, R rs is most sensitive to chlorophyll concentration. For populations at 10–20 m depth, Rrs is most sensitive to population depth. Populations deeper than 20 m are not detected. These results, in conjunction with recent field surveys, indicate a detection rate of approximately 25%. These results have implications for ocean-color sensor and algorithm development and may have direct application to satellite estimations of N2-fixation.

In summary, these three modeling studies confirm the importance of Trichodesmium in marine ecosystems. Moreover, these studies identify critical areas in which future research is required for illumination of the role of Trichodesmium in elemental cycling and marine ecosystems.


In Copyright. URI: This Item is protected by copyright and/or related rights. You are free to use this Item in any way that is permitted by the copyright and related rights legislation that applies to your use. For other uses you need to obtain permission from the rights-holder(s).