Process-understanding of the impact of dust deposition on marine primary production

Abstract

This thesis aims to provide a better understanding of some aspects of the impact of at- mospheric iron input on the iron cycling and the biological productivity in the ocean. In seawater, the fate of iron supplied by dust deposition is influenced by various processes such as iron dissolution, speciation, particle surface adsorption and redissolution of particulate forms of iron. Two one-dimensional models of Fe speciation and biogeochemistry for different ocean regions focus on some of these processes in this thesis. The iron cycle in the tropical eastern North Atlantic, a site with high episodic dust fluxes from the Saharan desert, is modelled in the first study. The role of dust particles in removing dissolved iron is studied by a complex description of particle aggregation and sinking. The vertical distribution of different particle classes shows a high sensitivity to changing aggregation rates. The model considers two classes of iron-binding organic ligands, strong and weak ligands, and describes their sources and fate explicitly. The long residence time of weak ligands that is required in the model to obtain realistic profiles indicates that a fraction of weak ligands is more refractory. Colloidal aggregation is present as the main iron removal process below the mixed layer in the model and organic colloids could play an important role in regulating the complexation and the removal of iron. The other model for the Mediterranean Sea simulates a mesocosm dust addition exper- iment, in which a significant decrease of dissolved iron in seawater has been observed after dust addition. This model explains this decrease mainly based on the balance of abiotic iron sources and sinks such as dissolution and particle adsorption, considering sinking and ag- gregation of different-sized particles. A concept of a critical concentration of dissolved iron, above which dust deposition acts as a net sink of dissolved iron, rather than a source, has been developed from the study. Taking into account the role of excess iron-binding ligands, this concept might be applied to explain the short-term variability of dissolved iron after natural dust deposition events. Iron can impact the marine productivity not only by directly limiting the growth of primary producers, in particular in the high-nutrient, low-chlorophyll regions, but also by limiting N2 fixation which is characterised by a high Fe requirement, and thus limiting the availability of reactive nitrogen for other primary producers. To study the impact of iron supplied by dust deposition on marine productivity, an ecosystem model including diazotrophs is coupled with a complex Fe speciation model for the tropical eastern North Atlantic. The seasonality and the limitation pattern of N2 fixation is investigated in this model study. Diazotrophs and other phytoplankton have been found in competitive as well as mutually beneficial interactions in regard to the availability of nitrogen, phosphorus, iron and light to grow. In this ocean region, dust deposition is necessary to support diazotrophy and also impacts the growth of other phytoplankton significantly. A simple relationship between dust fluxes and the magnitude of N2 fixation is however not found.