Baltic Sea food web dynamics and response to environmental change
2011 (English)Licentiate thesis, comprehensive summary (Other academic)
The Baltic Sea ecosystem is subject to a combination of external pressures such as fishery, changing climate and land-based nutrient inputs. In combination with internal food web mechanisms these external forces can lead to changes in the food web structure and function. In this licentiate thesis a new Baltic Sea Ecopath with Ecosim food web model for the open Baltic Proper (BaltProWeb) is introduced as a tool to study the past (1974-2006) food web dynamics (paper 1) and the sensitivity of this model to data uncertainties is analyzed (paper 2).
BaltProWeb includes 22 functional groups from phytoplankton to top-predators, is calibrated with biomass data across trophic levels and can accommodate external forcing such as fishery and environmental change. The model can reproduce the main trends observed in the biomass development of most Baltic Proper functional groups as well as the late 1980s ecosystem reorganization, or regime shift, from the cod (Gadus morhua) dominated state into the sprat (Sprattus sprattus) dominated one. In addition to changes in the functional group biomasses, changes in ecosystem flows were observed. For example the energy flow from the benthic system into the pelagic one (via cod diet) was large during the pre-regime shift period, but decreased in importance after the sharp post-regime shift increase in the sprat biomass. The proportions of both clupeid and zooplankton production required to support the Baltic Sea fisheries were large throughout the model run suggesting that fisheries have potential to be an ecosystem structuring actor in the Baltic Proper. Successful reproduction of the past biomass dynamics required the inclusion of fishery and environmental forcing together with appropriate trophic control during the calibration process. This finding motivates to develop ecosystem models that can incorporate both external pressures as well as internal ecosystem dynamics.
The BaltProWeb model turned out to be sensitive to uncertainty in both model input data (1974 biomasses) as well as environmental forcing functions used. The functional groups the parameterization of which the model was most sensitive to were, in decreasing order of importance, (i) 2-3 year old cod, (ii) adult sprat, (iii) macrozoobenthos, (iv) Pseudocalanus sp. and (v) other mesozooplankton. Changing the initial biomass values of these groups within the boundaries of input data uncertainty (measured as coefficients of variation) together with model recalibration resulted in a set of new Ecosim models with a rather large range in model fit, biomass projections and trophic control. When future (2007-2100) fishery and climate change scenarios were run using the different models, and by this way taking the uncertainties in the input data into account, the range of model outcomes was yet larger than observed during the calibration period (1974-2006). This indicates that model uncertainties cannot be ignored when modeling results are used as basis for ecosystem management. Further, the results highlight that evaluating the model only based on the model fit with data is not sufficient, but also studying the realized trophic control of different models is crucial and that sensitivity analysis can actually be one tool to study possible trophic control mechanisms with. This sensitivity study gave valuable information about both the model sensitivity and the need to communicate the uncertainties when the BaltProWeb model is used in the project ECOSUPPORT to project the future climate change effects on the Baltic Sea ecosystem.
Place, publisher, year, edition, pages
Stockholm: Department of Systems Ecology, Stockholm university , 2011.
food web, modelling, Ecopath with Ecosim, model uncertainty
Research subject Marine Ecology
IdentifiersURN: urn:nbn:se:su:diva-61941OAI: oai:DiVA.org:su-61941DiVA: diva2:438880
2011-09-26, FB127, Frescati Backe, Stockholms Universitet, Stockholm, 13:00 (English)
Huss, Magnus, Dr.
Blenckner, Thorsten, docentHjerne, Olle, Dr.