Food web modelling for investigating ecosystem behaviour in large reservoirs of the south-eastern United States: lessons from Lake Lanier, Georgia

Abstract

This paper presents the development and evaluation of a macroscopic food web model, for the primary purpose of identifying the key and redundant uncertainties associated with the constituent processes of the reservoir ecosystem of Lake Lanier (Georgia, USA). The ultimate goal is to provide assistance in directing the focus of future ecological research on Lake Lanier. The model interweaves the conventional grazing food chain with a microbial loop, and incorporates sediment–water–nutrient interactions. Model evaluation is performed by hypothesis screening, using the regionalized sensitivity analysis (RSA) procedure. RSA is conditioned on two target endpoints defined by nutrient concentration and fish biomass, one {nutrient-rich; fish-depleted} and the other {nutrient-depleted; fish-rich}. The results suggest that: (i) watershed inputs are most critical to attaining both target endpoints, particularly the sorptive association between phosphorus and the dominant iron-rich clay sediments; (ii) the phytoplankton-based grazing food chain is critical to reaching the {nutrient-rich; fish-depleted} endpoint, whereas the bacteria-based microbial loop determines the attainability of the {nutrient-depleted; fish-rich} endpoint; (iii) fisheries management practices could significantly impact ecosystem behaviour; and (iv) the availability and distribution of dissolved phosphorus in the water column is critical to productivity for both endpoints. Uncertainties associated with sediment loading, secondary and microbial production, and phosphorus dynamics indicate significant gaps in current scientific knowledge about the food web ecology of Lake Lanier. It is therefore suggested that future research be directed towards reducing these uncertainties, thereby providing a more reliable knowledge base for managing and protecting the integrity of Lake Lanier’s ecosystem.