A carregar...
Publicação
Combining Artificial Neural Networks and GIS fundamentals for Coastal erosion prediction modeling
| Nome: | Descrição: | Tamanho: | Formato: | |
|---|---|---|---|---|
| 2.42 MB | Adobe PDF |
Orientador(es)
Resumo(s)
The complexities of coupled environmental and human systems across the space and time of fragile systems challenge new data-driven methodologies. Combining geographic information systems (GIS) and artificial neural networks (ANN) allows us to design a model that forecasts the erosion changes in Costa da Caparica, Lisbon, Portugal, for 2021, with a high accuracy level. The GIS–ANN model proves to be a powerful tool, as it analyzes and provides the “where” and the “why” dynamics that have happened or will happen in the future. According to the literature, ANNs present noteworthy advantages compared to the other methods that are used for prediction and decision making in urban coastal areas. In order to conduct a sensitivity analysis on natural and social forces, as well as dynamic relations in the dune–beach system of the study area, two types of ANNs were tested on a GIS environment: radial basis function (RBF) and multilayer perceptron (MLP). The GIS–ANN model helps to understand the factors that impact coastal erosion changes, and the importance of having an intelligent environmental decision support system to address these risks. This quantitative knowledge of the erosion changes and the analytical map-based frame are essential for an integrated management of the area and the establishment of pro-sustainability policies.
Descrição
Palavras-chave
Geographic information systems Artificial neural networks Backpropagation Coastal urban zones Erosion changes prediction
Contexto Educativo
Citação
Peponi, A., Morgado, P., & Trindade, J. (2019). Combining Artificial Neural Networks and GIS fundamentals for Coastal erosion prediction modeling. Sustainability, 11, 975. https://doi.org/10.3390/su11040975
Editora
MDPI