Drinking water reservoirs are an important water source in many regions around the globe. As man-made structures, they are actively managed to provide raw water of adequate quality. Global change, and in particular global warming, is influencing the available water quantity and quality and is putting pressure on the management of drinking water reservoirs. To adapt management to the changing conditions, tools to evaluate different measurements, as well as predictions of potential future conditions, are required. Mechanistic models are the tool of choice to test adaptation measures and estimate potential future conditions. In this thesis, the impact of climate warming on the water quality of drinking water reservoirs, as well as potential mitigation strategies are investigated using one-dimensional mechanistic lake models. Therefore, three reservoirs located in Saxony, Germany are chosen as study sites. Specifically, the thesis evaluates the potential of adapting the withdrawal depth to mitigate the impact of global warming onto the water quality of reservoirs. To estimate the uncertainty associated with the used models, a framework for ensemble modeling of lakes and reservoirs is developed and applied to simulate climate impact predictions for a drinking water reservoir.
The main findings of this thesis are: Adapting the withdrawal depth has the potential to mitigate some impact of global warming. Especially, the simulated impact of withdrawal depth on deep water temperature was about the same magnitude as the observed impact of global warming in the last 30 years. By using an ensemble approach, different sources of uncertainty were quantified and compared. For the climate impact simulation the largest uncertainty was found to be the epistemic uncertainty, which is related to the model structure. Nevertheless, the estimated trends for the climate scenario from all five applied models were coherent.
The thesis furthers our knowledge of drinking water reservoir management in a warming climate. Using mechanistic models and ensemble techniques were shown to be an effective tool to compare different management strategies and evaluate uncertainties related to the modeling process. Further mitigation measures need to be developed to safeguard drinking water production under global warming, and methods presented in this thesis help to better evaluate strategies for managing drinking water reservoirs.:1 Introduction
1.1 Background
1.1.1 Down by the river: Drinking water reservoirs
1.1.2 Take care of business: Impact of management
1.1.3 All the world is green: Phytoplankton and lake physics
1.1.4 The times they are a-changing: Climate warming
1.1.5 OK computer: Process-based models
1.2 Study area
1.2.1 Investigated reservoirs
1.2.2 Local impact of climate warming
1.3 Aims and structure of the thesis
1.3.1 Research questions
1.3.2 Structure of the thesis
2 Managing climate change in drinking water reservoirs: potentials and limitations of dynamic withdrawal strategies
2.1 Background
2.2 Methods
2.2.1 Study area
2.2.2 Model input
2.2.3 Model setup
2.2.4 Management strategies
2.2.5 Statistical evaluation
2.3 Results
2.3.1 Observed trend
2.3.2 Validation of simulated temperature profiles
2.3.3 Modeled impact of management
2.3.4 External and internal forcing
2.4 Discussion
2.4.1 Observed trends in hydrophysical features
2.4.2 Modeled impact of management
2.4.3 Statistical model
2.4.4 Climate change and implications for management
2.5 Conclusion
3 LakeEnsemblR: An R package that facilitates ensemble modelling of lakes
3.1 Introduction
3.2 Methods
3.2.1 Model description
3.2.2 R package description
3.2.3 Getting started
3.2.4 Calibration algorithms
3.2.5 Combining multiple ensemble runs
3.3 Example application of LakeEnsemblR
3.3.1 Lough Feeagh: water temperature dynamics
3.3.2 Langtjern: lake ice dynamics
3.3.3 Uncertainty partitioning
3.3.4 Multi-parameter ensemble
3.3.5 Discussion
3.4 Summary
3.4.1 Framework
3.4.2 Recommendations for use
3.4.3 Outlook
4 Ensemble of models shows coherent response of a reservoir’s stratification and ice cover to climate warming
4.1 Introduction
4.2 Methods
4.2.1 Study site
4.2.2 Climate scenarios and data
4.2.3 Lake model ensemble
4.2.4 Calibration
4.2.5 Data evaluation
4.2.6 Uncertainty partitioning
4.3 Results
4.3.1 Observed trends
4.3.2 Model performance
4.3.3 Climate predictions
4.3.4 Uncertainty partitioning
4.4 Discussion
4.5 Conclusions
5 From T to P: Impact of withdrawal on water quality of a drinking water reservoir
5.1 Introduction
5.2 Methods
5.2.1 Study site and data
5.2.2 Applied models
5.2.3 Model calibration and sensitivity analysis
5.2.4 Withdrawal strategies
5.3 Results
5.3.1 Model calibration
5.3.2 Withdrawal strategies
5.4 Discussion
5.5 Conclusion
6 Synthesis
6.1 Lessons learned
6.1.1 Impact of global warming on reservoir water quality
6.1.2 Mitigation through adaptive withdrawal strategies
6.1.3 Using models for scenario analysis
6.2 Next steps
6.2.1 Further management and climate simulations
6.2.2 Model based real time decision support
6.2.3 Improve biogeochemical models
6.3 Concluding remarks
A Appendix Chapter 2
A.1 Additional plots
A.2 Principal component analysis
A.3 Full linear model
B Appendix Chapter 3
B.1 Format input files
B.2 Additional figures
B.3 Additional tables
C Appendix Chapter 4
C.1 Additional Tables
C.2 Additional Figures
D Appendix Chapter 5
D.1 rodeoFABM
D.2 Model description
D.2.1 State variables
D.2.2 Processes
D.2.3 Parameter
D.2.4 Stoichiometry
D.3 Sensitivity analysis
D.4 Additional Tables
D.5 Additional Figures
| Identifer | oai:union.ndltd.org:DRESDEN/oai:qucosa:de:qucosa:94346 |
| Date | 18 November 2024 |
| Creators | Feldbauer, Johannes |
| Contributors | Petzoldt, Thomas, Berendonk, Thomas U., Soetaert, Karline, Schuwirth, Nele, Technische Universität Dresden |
| Source Sets | Hochschulschriftenserver (HSSS) der SLUB Dresden |
| Language | English |
| Detected Language | English |
| Type | info:eu-repo/semantics/publishedVersion, doc-type:doctoralThesis, info:eu-repo/semantics/doctoralThesis, doc-type:Text |
| Rights | info:eu-repo/semantics/openAccess |
Page generated in 0.0016 seconds