Political boundaries make watershed planning difficult despite the influence of many state and federal programs. Broad, top-down, watershed initiatives fail to reach many municipalities due to human resources, time and legalities. Thus, a watershed ecosystem based approach to city planning should be utilized in order to integrate a holistic and scientific foundation for land use decisions. However, there is a need for research for developing and applying a watershed approach to urbanizing watersheds. The goal of this study is to provide a series of science based transferable recommendations upon which municipalities can make land use planning decisions. These recommendations are informed by a watershed modeling and prioritization study conducted with the community of Northampton, Massachusetts. Analyses of water resource planning options were made concerning future development scenarios using an approach which links water quality and quantity, land use and government. A required component of the ecosystem approach, stakeholder participation, applied the Deliberative Attribute Prioritization Procedure (DAPP) for the first time in this context to assess the relative of different environmental concerns. The results of these stakeholder focus groups showed the importance of several key attributes including land use, water quality, water quantity, and impacts to neighborhing communities that were utilized in the watershed models. This thesis provides an integrated tool for water resource planning at the municipal level. However, without the effective transfer of these recommendations into existing policies like zoning, the results of the study have limited use. Therefore implementation of recommendations within municipal planning documents is an important component. This information will be utilized to evaluate priority water resource protection overlays by providing quantitative information and decision making within a community. A citywide watershed model and analysis used to guide policy-making and decision-making will assist in fulfilling the community of Northampton’s continuing commitment to work toward economic, environmental, and equitable sustainability, as well as provide a model for other communities.

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.