Institut für Erd‐ und Umweltwissenschaften Arbeitsgruppe Hydrologie / Klimatologie INTEGRATED WATER QUALITY MODELLING IN MESO‐ TO LARGE‐SCALE CATCHMENTS OF THE ELBE RIVER BASIN UNDER CLIMATE AND LAND USE CHANGE Kumulative Dissertation zur Erlangung des akademischen Grades "doctor rerum naturalium" (Dr. rer. nat.) in der Wissenschaftsdisziplin "Geoökologie" eingereicht an der Mathematisch‐Naturwissenschaftlichen Fakultät der Universität Potsdam von C ORNELIA H ESSE Forschungsfeld II Potsdam‐Institut für Klimafolgenforschung Klimawirkung und Vulnerabilität Tag der Einreichung: 12. Juni 2018 Tag der Verteidigung: 08. November 2018 Veröffentlichung: Januar 2019 Gutachter/innen: Prof. Dr. Axel Bronstert Universität Potsdam, Fachgebiet Hydrologie / Klimatologie Prof. Dr. Dörthe Tetzlaff Leibniz‐Institut für Gewässerökologie und Binnenfischerei, Abteilung Ökohydrologie & Humboldt‐Universität zu Berlin Prof. Dr. Britta Schmalz Technische Universität Darmstadt, Fachgebiet Ingenieur‐ hydrologie und Wasserbewirtschaftung Dieses Werk ist unter einem Creative Commons Lizenzvertrag lizenziert: Namensnennung 4.0 International Um die Bedingungen der Lizenz einzusehen, folgen Sie bitte dem Hyperlink: http://creativecommons.org/licenses/by/4.0/ Online veröffentlicht auf dem Publikationsserver der Universität Potsdam: https://nbn-resolving.org/urn:nbn:de:kobv:517-opus4-422957 https://doi.org/10.25932/publishup-42295 Acknowledgements “Wer nichts wagt, der darf nichts hoffen.” Friedrich Schiller Acknowledgements Many people – including myself – were not sure whether this dissertation would ever come to an end. A lot of ups and downs in motivation and scientific enthusiasm had to be overcome while several changes in personal circumstances and working conditions took place. But with the increasing number of family members the number of scientific publications increased, too, providing a base to compile a cumulative dissertation and to finalize this period of life now. During these years as a doctoral candidate several people supported me in different ways. Sincere thanks are given to them all, although only some will be individually mentioned here. My deepest thanks go to Dr. Valentina Krysanova, who was my day‐to‐day supervisor at the Potsdam‐Institute for Climate Impact Research (PIK). She benevolently supported my scientific career and family development and never lost her confidence in a final good end of the dissertation project. I admire her for being involved in so many working processes without losing the overview and always knowing how to improve the research approaches and scientific publications of several people. Thank you for the fruitful discussions, detailed corrections at all times of day (and night) and your wholehearted friendliness. Many thanks go to Prof. Dr. Axel Bronstert, who gave me the opportunity to do the doctorate at the University of Potsdam by being my supervisor there. He was always available for professional and organisational questions as well as confidently motivating. Thank you for giving me the time it needed to complete this dissertation. I thank Prof. Dr. Dörthe Tetzlaff and Prof. Dr. Britta Schmalz for their acceptance to be the external evaluators within the examination process and to write a review. The Potsdam‐Institute for Climate Impact Research (PIK) provided an inspiring working environment, which was characterized by a large number of enthusiastic colleagues, kind administrative staff and a wonderful location at the Telegrafenberg campus. I always enjoyed being a part of the PIK team and thank all of you for helping in scientific, organisational or personal matters. During my employment I saw a lot of colleagues come and go and wish you all the best for your future at PIK or somewhere else in the world. My new colleagues at the Brandenburg State Office of Environment (LfU) supported me with the possibility to take some free days, showing interest and sending good wishes during the last finalizing steps of the dissertation. I also want to thank my parents and siblings for their steadily help and support in all matters of life and especially for taking care of the children during several holiday weeks, allowing me to concentrate on writing the text of the Introduction and Conclusions. Thank you for being interested but not asking too often about the actual status of the dissertation. My own family – Andreas, Mira, Jonte and Enno – had to travel alone several weekends to give me time to continue with the text. I am very grateful that you all are at my site and I am looking forward to join you the next times. Thanks for your patience and all your love. “It always seems impossible, until it is done.” Nelson Mandela Cornelia Hesse i Summaries Summary In a changing world facing several direct or indirect anthropogenic challenges the freshwater resources are endangered in quantity and quality, and need protection as well as actions for improvement in order to ensure an adequate status for human and ecosystem wellbeing in the future. Various substances can influence the water quality of river ecosystems, including those naturally occurring in the watersheds. An excessive supply of nutrients, for example, can cause disproportional phytoplankton development and oxygen deficits in large rivers, leading to failure of the aims requested by the Water Framework Directive (WFD). Such problems can be observed in many European river catchments including the Elbe basin, where effective measures for improving water quality status are highly appreciated. During the last decades computer‐based modelling tools became more and more common in water resources management and protection. Models can help to understand the dominant nutrient processes in a watershed and to identify the main sources of nutrient pollution in the river network. Furthermore, they can be effective tools for impact assessments investigating the effects of changing climate or socio‐economic conditions on the status of surface water bodies, and for testing the usefulness of possible protection measures. An important prerequisite for successful model applications is a sufficiently detailed process description in a model including all relevant ecosystem compartments, as well as a careful model setup and calibration/ validation. Due to the high number of interrelated processes, ecohydrological model approaches containing water quality components are much more complex than the pure hydrological ones, their setup and calibration/validation require more efforts, and thus they are less applied in the scientific community for impact studies. Such models, including the process‐based semi‐ distributed Soil and Water Integrated Model (SWIM), still need some further development and improvement for a more realistic water quality modelling, especially in large catchments. Therefore, this cumulative dissertation focuses on two main objectives: 1) the approach‐related objectives aiming in the SWIM model improvement and further development regarding nutrient (nitrogen and phosphorus) process description, and 2) the application‐related objectives in meso‐ to large‐scale Elbe river basins to support adaptive river basin management in view of possible future changes. The dissertation is based on five scientific papers published in international journals and dealing with these research questions. Several adaptations were implemented in the model code to improve the representation of nutrient processes in the catchments under investigation. Firstly, the model application in the meso‐scale Rhin catchment was improved by including a simple approach for a better simu‐ lation of specific water and nutrient processes in wetland soils. In the Saale case study, the ammonium pool was added to the model for a more comprehensive description of the nitrogen cycle. While calibrating the SWIM model for the Rhin and Saale catchments it was found, that the simple routing of nutrients through the river network is not sufficient to represent their seasonally observed concentrations, especially of those nutrients coming mainly from point sources. Therefore, a detailed in‐stream module was added to the SWIM model structure, simulating algal growth, nutrient transformation processes and oxygen conditions in the river reaches, mainly driven by water temperature and light. This new approach created a highly complex ecohydrological model with a large number of additional parameters, which could be calibrated. However, testing less complex methods to represent retention processes in the ii Integrated water quality modelling Summaries landscape and/or river network did not result in comparably good model performances for all nutrients under investigation. The calibration and validation of the SWIM model enhanced by the new approaches in the selected subcatchment and the entire Elbe river basin delivered satisfactory to good model results in terms of criteria of fit. Thus, the calibrated and validated model provided a sound base for the assessment of possible future changes and impacts. Simple climate sensitivity experiments were applied in the smaller catchments, and finally a detailed climate change impact assessment for the entire transboundary Elbe river basin was conducted, driven by a set of 19 regionally downscaled climate scenarios for the reference and two future periods. The ensemble of climate scenarios projects rising average temperature and precipitation for the Elbe watershed with increasing trends, leading to higher river discharge, and spatially variable changes in nutrient loads in