Climate change and watershed planning : understanding the related impacts and risks

This dissertation consists in studying the possible effects of climate change in watershed management. Specifically, the dissertation explores the possibility of using a Artificial Neural Networks – ANN models coupled with physically based hydrological models as method for improving the description of the regional hydrologic cycle of a watershed by considering the relationships among geophysical and human systems when stressed by both climatic and anthropogenic factors and under long-term time periods. The selected physically-based model is named Soil and Water Assessment Tool -- SWAT, As watershed management is a broad theme involving the consideration of several interconnected systems, such as biogeophysical and socio-economic systems, the dissertation first reviews basic concepts regarding the management of watersheds, the utilisation of hydrological models as tool in supporting the development of integrated watershed management plans, and the presentation of two case studies: i. The first one, named the Venice Lagoon Watershed – VLW, in Italy, consisting in the study of the possible impacts of climate and land-use changes on the availability of water for irrigation water, both in terms of quantity and quality, and; The second case study, focusing in the study of the possible effects of climate and land-use changes on river flood events the Itajaí River Watershed – IRW, in Brazil. In terms of structure the thesis is divided into five chapters: i. The first chapter reviews concepts of watershed management, while introducing hydrological modelling techniques and proposed innovations to be used throughout the dissertation, such as the consideration of the carbon dioxide fertilisation effect due to an increase concentration of carbon dioxide in the atmosphere; ii. The second chapter presents a literature review and a case study describing the uncertainty in employing ANN models in hydrologic systems; iii. The third chapter describes the complexity of the VLW while presenting two case studies evaluating the feasibility of coupling the two considered models (i.e. ANN and SWAT models); iv. The fourth chapter presents the Italian case study, covering the topic of climate change impacts on irrigation water, and; v. The fifth chapter presents the Brazilian case study, covering the topic of climate change effects on river flood risk. The source code of the proposed ANN model and the modifications in the SWAT model are presented as Annex information, as well as a detailed description of the ANN model.