The credibility of the scientific methodology of mathematical models and their adequacy to form the basis of public policy decisions has frequently been challenged. Skeptical views of the scientific value of modelling argue that there is no true model of an ecological system, but rather several adequate descriptions of different conceptual basis and structure. The purpose of this work was to first advance the Bayesian calibration of process-based models for guiding the water quality criteria setting process in Hamilton Harbour, Ontario, Canada. The analysis suggests that the water quality targets for total phosphorus and chlorophyll a concentrations will likely be met, if the recommendation for phosphorus loading at the level of 142 kg day-1 is achieved. My dissertation also examines how the Bayesian approach can effectively support the decision making process by synthesizing the predictions of different models developed for the same system. The model averaging approach consolidates the finding that the existing total phosphorus goal is most likely unattainable. The discrepancy between the chlorophyll a predictions of the two models pinpoints the need to delve into the dynamics of phosphorus in the sediment-water column interface. This work also aims to examine statistical formulations that explicitly accommodate the covariance among the process error terms for various model endpoints. The analysis suggests that the statistical characterization of the model error can be influential to the inference drawn by a modelling exercise. Finally, my dissertation challenges the capacity of the ecological foundation of eutrophication models to predict the role of nutrient regeneration. It shows that the recycled nutrients can be significant drivers in low as well as in high-productivity ecosystems depending on the period of the year examined. My dissertation also discusses several prescriptive guidelines that should be helpful towards a structural and methodological improvement of eutrophication modelling.

The credibility of the scientific methodology of mathematical models and their adequacy to form the basis of public policy decisions has frequently been challenged. Skeptical views of the scientific value of modelling argue that there is no true model of an ecological system, but rather several adequate descriptions of different conceptual basis and structure. The purpose of this work was to first advance the Bayesian calibration of process-based models for guiding the water quality criteria setting process in Hamilton Harbour, Ontario, Canada. The analysis suggests that the water quality targets for total phosphorus and chlorophyll a concentrations will likely be met, if the recommendation for phosphorus loading at the level of 142 kg day-1 is achieved. My dissertation also examines how the Bayesian approach can effectively support the decision making process by synthesizing the predictions of different models developed for the same system. The model averaging approach consolidates the finding that the existing total phosphorus goal is most likely unattainable. The discrepancy between the chlorophyll a predictions of the two models pinpoints the need to delve into the dynamics of phosphorus in the sediment-water column interface. This work also aims to examine statistical formulations that explicitly accommodate the covariance among the process error terms for various model endpoints. The analysis suggests that the statistical characterization of the model error can be influential to the inference drawn by a modelling exercise. Finally, my dissertation challenges the capacity of the ecological foundation of eutrophication models to predict the role of nutrient regeneration. It shows that the recycled nutrients can be significant drivers in low as well as in high-productivity ecosystems depending on the period of the year examined. My dissertation also discusses several prescriptive guidelines that should be helpful towards a structural and methodological improvement of eutrophication modelling.

The Gulf of Finland is the most eutrophied water body in the Baltic Sea, which is mainly caused by nutrient loads produced by human activities in its surrounding cities. In order to solve this environmental problem, a computational model based on the understanding the relations between eutrophication, water quality and sediments is needed to forecast the water quality variance in response to the natural and anthropogenic influences. A precise water quality model can be useful to assist the policy making in the Gulf of Finland, and even for the whole Baltic Sea. Kiirikki model, as one of these models describing the water quality of Baltic Sea in response of water quality variance, is a sediment and ecosystem based model, treating different sub-basins and layers as boxes. This study aims to assess the parameters’ sensitivity level on the scale of the Gulf of Finland. Firstly, the Morris sampling strategy is applied to generate economic OAT (One factor At a Time) samples before screening 50 out of 100 trajectories with distance as large as possible. In order to assess their sensitivity, index and indicator are needed. EE (elementary effect) is adopted to be the assessment index and four core eutrophication indicators from HELCOM 2009a are to be analyzed. By comparing the (σ,μ) and (σ,μ*) plots of each parameters’ EE values (σ is standard deviation, μ is mean value and μ* is the absolute mean value), some parameters are identified as potential sensitive parameter, such as the minimum biomass of cyanobacteria (Cmin), critical point of CO2 flux (CCr), the optimal temperature for detritus phosphorous mineralization (Toptgamma), maximum loss rate of algae (RAmax), optimal temperature for the growth of other algae (ToptmuA), Coefficient for temperature limiting factor for the growth of cyanobacteria (aTmuC), half-saturation coefficient of radiation for cyanobacteria (KIC) and so on. In contrast, the other parameters are ruled out as having very low values in terms of σ, μ and μ*. This is because the feature of Morris sampling strategy makes it easier to achieve high variance of the outputs, resulting into generally higher σ. Therefore, further investigation with different strategies is needed after the initial screening of the non-sensitive parameters in this study.