Water contamination during urban flood events can have a negative impact on human health and the environment. Prior flood studies lack investigation into how GIS can map and analyze this at a large scale (cadastral) level. This  thesis  focused on how GIS can  help map and analyze water contamination risk in urban  areas  using  LiDAR  elevation  data,  at  a  large-scale  (cadastral)  level,  and  symbology  and  flood classification  intervals  specifically  selected  for  contamination  risk.  This  was  done  by  first  completing a literature review about past research and studies  of similar scope. Based on  the findings, a method to map and analyze water contamination risk during sea-based flood scenarios was tested in the Näringen district of Gävle, Sweden. This study area was investigated and flood contamination risk maps were produced  for two different  flood scenarios which illustrated  which properties are vulnerable to flooding and at what depth, what their contamination risk is, and if they are hydrologically connected to the ocean.  The findings from this investigation  are that this method of examining water contamination risk could be useful to planning officials who are in charge of policies relating to land-use. These findings could help guide landuse  or  hazardous  material  storage  regulations  or  restrictions.  To  further  research  in  this  topic,  it  is recommended  that  similar  studies  are  performed  that  use  a  more  detailed  land-use  map  which  has information  on  what  type  and  quantity  of  possible  contaminants  are  stored  on  individual  properties. Furthermore, flood modeling should be employed in place of the flood mapping which was conducted in this thesis.

The escalating impact of climate change has become a significant global concern, particularly in urban environments through the risk of flooding, due to intensified precipitation patterns. Nature Based Solu-tions (NBS), offer effective strategies for mitigating flood risks by enhancing stormwater management and promoting urban resilience. Multicriteria Analysis (MCA) has shown to be useful for identifying suitable areas for NBS, however, there is limited research on its application specifically for urban flood resilience in Sweden. Therefore, this study aims to develop a GIS-based suitability mapping framework within MCA method for allocating suitable areas for two NBS measures: Retention Pond (RP) and Detention Basin (DB), applying it to a case study in Sweden. The study employs a mixed-method approach and consists of (1) framework develop-ment through a literature review, geospatial data assessment, and key-informant interviews, and (2) application of the framework to a case study area in Sweden. In the case study area, the resulting suitability map indicates that 7.5 % of DBs and 7% of RPs met all criteria. Key-informant interviews with local experts provided valuable insights, particularly the exclusion of hazardous zones as well as emphasizing the importance of considering not only biophysical characteristics, but also socio-cultural factors. In conclusion, this study contributes to the body of knowledge on NBS suitability mapping. The findings offer guidance to climate strategists and urban planners on a municipal level, selecting optimal locations for NBS strategies for urban flood resilience and stormwater management.

Artificial Neural Networks (ANNs) have been comprehensively researched, both from a computer scientific perspective and with regard to their use for predictive modelling in a wide variety of applications including hydrology and the environment. Yet their adoption for live, real-time systems remains on the whole sporadic and experimental. A plausible hypothesis is that this may be at least in part due to their treatment heretofore as “black boxes” that implicitly contain something that is unknown, or even unknowable. It is understandable that many of those responsible for delivering Early Warning Systems (EWS) might not wish to take the risk of implementing solutions perceived as containing unknown elements, despite the computational advantages that ANNs offer. This thesis therefore builds on existing efforts to open the box and develop tools and techniques that visualise, analyse and use ANN weights and biases especially from the viewpoint of neural pathways from inputs to outputs of feedforward networks. In so doing, it aims to demonstrate novel approaches to self-improving predictive model construction for both regression and classification problems. This includes Neural Pathway Strength Feature Selection (NPSFS), which uses ensembles of ANNs trained on differing subsets of data and analysis of the learnt weights to infer degrees of relevance of the input features and so build simplified models with reduced input feature sets. Case studies are carried out for prediction of flooding at multiple nodes in urban drainage networks located in three urban catchments in the UK, which demonstrate rapid, accurate prediction of flooding both for regression and classification. Predictive skill is shown to reduce beyond the time of concentration of each sewer node, when actual rainfall is used as input to the models. Further case studies model and predict statutory bacteria count exceedances for bathing water quality compliance at 5 beaches in Southwest England. An illustrative case study using a forest fires dataset from the UCI machine learning repository is also included. Results from these model ensembles generally exhibit improved performance, when compared with single ANN models. Also ensembles with reduced input feature sets, using NPSFS, demonstrate as good or improved performance when compared with the full feature set models. Conclusions are drawn about a new set of tools and techniques, including NPSFS and visualisation techniques for inspection of ANN weights, the adoption of which it is hoped may lead to improved confidence in the use of ANN for live real-time EWS applications.