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.
Este estudo trata da caracterização multitemporal de barras e ilhas fluviais no baixo curso do rio Jaguaribe, particularmente no segmento a jusante da barragem do açude Castanhão, em anos secos e chuvosos. Metodologicamente, recorreu-se à determinação de anos-padrão secos e chuvosos, à utilização das ferramentas de sensoriamento remoto, através do uso de fotografias aéreas, imagens de satélite e softwares de geoprocessamento, para o mapeamento das diferentes tipologias de barras e ilhas fluviais. Os resultados indicaram que, embora haja especificidades nos dados morfológicos de quantidade e área para as diferentes tipologias de barras fluviais e ilhas, em geral, a formação destas feições reflete as variações de vazão no baixo curso do rio Jaguaribe, como comprovam os dados do coeficiente de correlação linear de Pearson e do coeficiente de determinação. Ou seja, nos anos considerados secos, observou-se uma redução das feições devido à ausência de fluxos significativos para promover o transporte de sedimentos, ao passo que, em anos chuvosos, verificou-se um incremento de barras fluviais, sobretudo pela ação cíclica dos processos erosivos e sedimentares, que promoveram a distribuição e redistribuição dos sedimentos no canal. Em adição, verificou-se que a correlação é mais expressiva quando se considera a posição da barra no canal (lateral ou central), em detrimento da forma/geometria (pontal), cuja formação está diretamente associada a outros fatores geomorfológicos.
Coastal areas are among the most dynamic earth systems as they are exposed to powerful agents. Nearshore wave energy is one of the most important triggering factors for erosion and flooding and is often neglected for severe infrastructure damaging, property losses and loss of life. These consequences are amplified with high population density and heavy infrastructure implantation as it happens in Lisbon (Portugal). In this context, it is of great importance for coastal stakeholders, decision-makers and civil protection entities to estimate precisely the spatial distribution of storm hazard for prevention and mitigation purposes, as well as to design adjusted answers for calamity responses. We apply a coastal storm hazard index (CSHI) considering triggering and conditioning variables involved in the effects of an extreme storm, namely: 100-year return period of SWAN modelled H s , and its spatial distribution across the study area, land use, number of buildings, height, slope, geology, geomorphology, erosion/ accretion rates, width of the systems, exposure of the coastline, bathymetry and legally protected areas. The variables were weighted according to a hierarchical analysis process and classified into five classes of exposure. A validation process was then implemented by comparing the occurrences identified in the last two decades newspapers and the storm hazard classification, showing a satisfactory validation results. The results show a classified storm hazard map that identifies the most and the less exposed areas. High values of CSHI occur in areas with excessive human pressure, low heights sandy systems with significant costal erosion rates. The main type of consequences identified are associated with inland flooding and erosion, resulting in the destruction of coastal protection infrastructures, and population displacement leading to great economic and social impacts and loss of life.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.