Keywords
knowledge-based modelling, water resource, neural networks, forecasting, drought, karst
Location
Session D4: Water Resources Management and Planning - Modelling and Software for Improving Decisions and Engaging Stakeholders
Start Date
12-7-2016 8:30 AM
End Date
12-7-2016 8:50 AM
Abstract
Karst aquifers provide water resource for a large part of the Mediterranean population and water resource becomes a strategic problem during summer when population increases due to tourism. To help managers to optimize the exploitation of water, this work studies the ability of a neural network model to efficiently simulate water levels in the Cèze River, connected to a karst aquifer, few months ahead during the dry season. The neural model is based on recurrent multilayer perceptrons that learn the relations between inputs (mainly rainfall and ETP) and output (water level). After a training step using 17 years of data, the model is assessed on a never seen year to be validated. A particular attention is devoted to the dry season (from May to September). The model achieved good forecast of the maximal observed drawdown. Several architectures were run, each related to a supposed physical processes or to a strategy of modelling. Recurrent multilayer perceptrons were used to achieve these hypotheses, and thanks to a rigorous process of variables and model complexity selection, the resulting performances were very satisfying. Assessment of the model was done on one of the drier summer of the database composed of 19 years of daily water-levels. For the five tested architectures, Nash criteria evolve between 0.84 to 0.9 on the whole year, and between 0 to 0.6 on the drier period (June to August). As this modelling strategy can be fed by rainfall scenario, it could help managers to optimize anthropogenic impacts on the river and preserve natural ecosystems. The methodology is generic and thus can be used with profit by managers on other hydro systems.
Included in
Civil Engineering Commons, Data Storage Systems Commons, Environmental Engineering Commons, Hydraulic Engineering Commons, Other Civil and Environmental Engineering Commons
Towards a neural networks-based prediction tool devoted to low water-levels forecasting: relevant architecture selection based on main physical processes
Session D4: Water Resources Management and Planning - Modelling and Software for Improving Decisions and Engaging Stakeholders
Karst aquifers provide water resource for a large part of the Mediterranean population and water resource becomes a strategic problem during summer when population increases due to tourism. To help managers to optimize the exploitation of water, this work studies the ability of a neural network model to efficiently simulate water levels in the Cèze River, connected to a karst aquifer, few months ahead during the dry season. The neural model is based on recurrent multilayer perceptrons that learn the relations between inputs (mainly rainfall and ETP) and output (water level). After a training step using 17 years of data, the model is assessed on a never seen year to be validated. A particular attention is devoted to the dry season (from May to September). The model achieved good forecast of the maximal observed drawdown. Several architectures were run, each related to a supposed physical processes or to a strategy of modelling. Recurrent multilayer perceptrons were used to achieve these hypotheses, and thanks to a rigorous process of variables and model complexity selection, the resulting performances were very satisfying. Assessment of the model was done on one of the drier summer of the database composed of 19 years of daily water-levels. For the five tested architectures, Nash criteria evolve between 0.84 to 0.9 on the whole year, and between 0 to 0.6 on the drier period (June to August). As this modelling strategy can be fed by rainfall scenario, it could help managers to optimize anthropogenic impacts on the river and preserve natural ecosystems. The methodology is generic and thus can be used with profit by managers on other hydro systems.
