Title: Climate Change and Urban Expansion Impact on High and Low Flows and the Overall Water Availability (Impact van klimaatverandering en stedelijke groei op hoog- en laagwaterafvoeren en de algemene waterbeschikbaarheid)
Other Titles: Climate Change and Urban Expansion Impact on High and Low Flows and the Overall Water Availability
Authors: Vansteenkiste, Thomas
Issue Date: 14-Dec-2012
Abstract: Observational records and projections on the future climate and land use provide abundant evidence that water resources are vulnerable and have the potential to be strongly impacted by climatic and land use changes, with wide-ranging consequences for human societies and ecosystems. This research focuses on the determination of the changes of a hydrological system to a number of climate and land use scenarios, which describe plausible future changes for Belgium. The Grote Nete-Grote Laak catchment, a medium-scale catchment (365 km²) with a groundwater-based hydrological system in the northeast of the country was selected for the analysis. This research did not only assess the changes of the hydrological processes, but also the influence of the applied hydrological model and the complexity of its process descriptions on the simulated hydrological variables. Five hydrological models with different spatial resolutions and process descriptions were considered for the impact simulations. The models varied from lumped conceptual models (NAM, PDM and VHM) over the intermediate conceptual-physically based, distributed WetSpa model, to the highly detailed physically based and fully distributed MIKE SHE model. The latter model accounts for the 3D groundwater processes and interacts bi-directionally with a full hydrodynamic MIKE 11 river model, which allowed simulation results to be obtained for internal discharges and groundwater heads within the catchment. A consistent protocol to model calibration was applied to all models. This protocol uses information on the response behaviour of the catchment, extracted from the river flow and input time series. It explicitly focuses on reproducing the extreme high and low flows next to testing the conventional model performance statistics. Also the model predictive capacity under high rainfall intensities, which might become more extreme under future climate change was explicitly verified for the models. The tail behaviour of the extreme flow distributions was graphically evaluated as well as the changes in runoff coefficients in relation to the changing rainfall intensities. It was concluded that after such calibration, all models succeed to produce high performance for the total runoff and quick and slow runoff sub-flow dynamics and volumes, peak and low flow extremes and their frequency distributions. Calibration of the lumped parameter models is however much less time consuming and produced higher overall model performance in comparison to the more complex distributed models. The five models respond in a similar way to the future climate scenarios for Belgium. Future projections on peak flows are highly uncertain with expected increases as well as decreases depending on the climate scenario. The projections on future low flows are more uniform: low flow decreases (up to 60% by 2100) are simulated by all models and for all climate scenarios. However, the uncertainties in the impact projections are high, mainly in the dry season. The PDM model simulates significantly higher runoffpeak flows under future wet scenarios (up to 20% by 2100), which is explained by its specific model structure. For the low flow extremes, the MIKE SHE model estimated the expected low flow decreases to be less severe under these scenarios with on average a lower decrease of about 30%. This is probably due to its more detailed physically-based process descriptions for the groundwater and the groundwater-river interactions of the MIKE SHE model. Results of all models moreover suggest that the model structure is more critical in low flow impact results of climate scenarios than in high flow conditions. Similar impact trends in response to climate change were simulated at the internal gauging stations within the catchment by the MIKE SHE model. The impact magnitude, however, strongly differed over the catchment in function of the local characteristics. High responses were found in the sub-basins with high fraction of urban land and to a lesser extent in cultivated areas. Soil characteristics were found to have only a small influence on these impact results. Within the catchment, the Grote Laak sub-basin was found to exert the highest impact, given that it is more urbanized and cultivated than other sub-basins. The expected changes in rainfall and evapotranspiration also affect the flow patterns in the groundwater system. Low groundwater heads could decrease during future summer and autumn periods by 2100; whereas the impact on high groundwater levels during winter is highly uncertain. This drying trend might have adverse consequences towards ecosystems, water supply, nature conversation, and other types of water use in dry summer periods. However, high spatial variability in the responses is present and corresponds largely to the topography. Winter heads might vary by 2100 between a few centimeters in the lower, river valleys up to 1 m in the interfluvial and elevated zones. Summer heads will generally stagnate in the valleys to decrease up to 1 m in the higher zones.With respect to the urban growth projections in the catchment, both the distributed MIKE SHE and WetSpa models simulated an increase in high flow rates and volumes and a corresponding reduction in baseflow rates and volumes, particularly during summer periods. However there is a large uncertainty in the rate and magnitude of the simulated impact. The WetSpa model simulated an increase of the peak flows almost 4 times higher than the MIKE SHE model, which is particularly attributed to the consideration of an imperviousness factor for the urban land in the WetSpa model structure. The changes at the internal flow gauging stations, estimated with the MIKE SHE model, largely vary in function of the local characteristics. The changes were positively correlated to the percentage increase in urban land, but also to the total proportion of urban land. The highest changes were again expected along the Grote Laak. The study has confirmed that future climate change and urban development are likely to significantly affect future water resources, but the precise magnitude of their impacts is hard to assess, especially during summer periods. These uncertainties in runoff change projections are largely induced by the hydrological model structure and constrain the ability to make clear conclusions about the impact of the climate change and the urban development pressures. The runoff predictions by hydrological models should therefore be interpreted with care and they should only be considered as indicative. The model-based uncertainties have to be accounted for in the decision making process on the adaptation and mitigation against future climate change and urban growth, and future research needs to focus on the rigorous quantification and reduction of these uncertainties.
Table of Contents: Chapter 1 - Introduction 1
1.1 Research introduction 1
1.2 Research objectives and questions 4
1.3 Research outline 6

Chapter 2 - Review on the impact research of climate change and urban expansion in Belgium 9
2.1 Introduction 9
2.2 Scenarios for Belgium 9
2.2.1 Climate change scenarios for Belgium 9
2.2.2 Urban expansion scenarios for Belgium 14
2.3 Classification and comparison of hydrological models 17
2.3.1 System representation: deterministic and stochastic 17
2.3.2 Spatial representation: lumped, distributed and semi-distributed models 18
2.3.3 Process representation: empirical, conceptual and physically based models 19
2.3.4 Comparison of the main hydrological model types 21
2.4 Impact methodology 22
2.4.1 Assessing the impact on surface water 22
2.4.2 Assessing the impact on groundwater conditions 24
2.5 Impact studies for Belgium 25
2.5.1 Impact studies of climate change 25
2.5.2 Impact studies of land use change and urban expansion 31
2.6 Conclusion 34

Chapter 3 - Hydrological impact modeling: overview of data and modeling procedures 37
3.1 Introduction 37
3.2 Study area 37
3.3 Hydrological models 40
3.3.1 NAM 41
3.3.2 PDM 44
3.3.3 VHM 47
3.3.4 MIKE SHE 49
3.3.5 WetSpa 51
3.4 Data sources and pre-processing 53
3.4.1 Climate data 54
3.4.2 Geographical data 59
3.4.3 Calibration and validation data 67
3.5 Parameterisation and calibration of the hydrological models 75
3.5.1 Parameterisation and calibration of lumped models 76
3.5.2 Parameterisation and calibration of distributed models 77
3.6 Model performance evaluation 79
3.6.1 Classical model performance testing 80
3.6.2 Model performance of peak and low flows 82
3.6.3 Predictive power on extreme flow changes under changing rainfall conditions 82
3.6.4 Groundwater model performance 84
3.7 Conclusion 85

Chapter 4 - Development of lumped models 89
4.1 Introduction 89
4.2 Model set-up 89
4.3 Model calibration and validation 90
4.3.1 Calibration pre-processing steps 90
4.3.2 Calibration of NAM 94
4.3.3 Calibration of PDM 96
4.3.4 Calibration of VHM 97
4.4 Model performance of hydrological models 99
4.4.1 Classical model performance testing 100
4.4.2 Model performance of peak and low flows 111
4.4.3 Predictive power on extreme flow changes under changing rainfall conditions 118
4.5 Conclusions 124

Chapter 5 - Development of distributed models 127
5.1 Introduction 127
5.2 Model set-up 127
5.3 Model calibration & validation 128
5.3.1 Calibration of MIKE SHE 128
5.3.2 Calibration of WetSpa 133
5.4 Model performance 136
5.4.1 Classical model performance testing 136
5.4.2 Model performance of peak and low flows 149
5.4.3 Predictive power on extreme flow changes under changing rainfall conditions 155
5.4.4 Groundwater results 160
5.5 Conclusion 166

Chapter 6 – Climate change and its hydrological impact 171
6.1 Introduction 171
6.2 Impact on surface water 172
6.2.1 Impact on peak flows 172
6.2.2 Impact on low flows 188
6.3 Impact on groundwater 203
6.3.1 Temporal changes of groundwater levels 203
6.3.2 Spatial variation of the groundwater impact 206
6.4 Conclusion 211

Chapter 7 – Urban development and its hydrological impact 215
7.1 Introduction 215
7.2 Urban expansion scenarios for the Grote Nete catchment 216
7.3 Impact of urban expansion scenarios on peak flows 219
7.4 Climate change versus urban expansion 228
7.5 Conclusion 233

Chapter 8 – Conclusion 235
8.1 Recapitulation 235
8.2 Recommendations 243

References 249

Appendix 273
Appendix 1 - Parameters of geographical data 273
Appendix 2 - WetSpa default land use parameters 277
ISBN: 978-94-6018-602-8
Publication status: published
KU Leuven publication type: TH
Appears in Collections:Hydraulics Section

Files in This Item:
File Status SizeFormat
PhD_text-Thomas Vansteenkiste.pdf Published 7475KbAdobe PDFView/Open Request a copy

These files are only available to some KU Leuven Association staff members


All items in Lirias are protected by copyright, with all rights reserved.