Guilherme Garcia de Oliveira

Simplifying artificial neural network models of river basin behaviour by an automated procedure for input variable selection

The objective of the present work is to present a simplified and automated method for identifying and excluding unnecessary input variables, with a consequent reduction in dimensionality of ANN-based hydrological models. The proposed method is iterative and computationally efficient: it consists of perturbing the input variables, recording the change in model performance, establishing an index showing the contribution of each variable to the ANN (the relative contribution index, RCI) and excluding the least-influential variables that fall below a threshold. The method was used to simulate mean daily flow for a 20-year period 1989-2009 from four drainage basins nested at different scales ranging from 19.4km? to 9426km?, in the Southern Brazil. The main result of this method of simplifying ANN-based hydrological models was to increase the Nash-Sutcliffe (NS) coefficient and to reduce RMSE in all the simulations undertaken. The potential of ANN models was therefore improved by eliminating unnecessary and/or redundant variables. Simulating the intermediate basin with area 5414km? (Santo i?ngelo), for example, the initial performance (12 inputs; NS=0.894) improved when a simpler and more parsimonious model was used (4 inputs; NS=0.944). To validate the simplification procedure, a comparison was made between the proposed method (RCI) and the well-known methods of Overall Connection Weights (OCW) and Forward Stepwise Addition (FSA). For the comparison between RCI and OCW methods, in most cases, the ordering of selected variables was similar, confirming that the two procedures satisfactorily identify the more important variables, although the RCI is computationally more efficient giving a small advantage in the resulting model performance. In the FSA method, although the performance of the obtained models has also been satisfactory, the computational effort was much greater than with the other two methods because of the excessive number of the neural network training performed (117 training procedures in Combination 2, against only six for the RCI method, for example). An excessive number of input variables can reduce the efficiency of ANN simulation.A criterion was established by which input variables could be selected for exclusion.The proposed method satisfactorily identifies the most important variables.Simplification of the ANN improved performance of hydrological simulations.

AS INCERTEZAS ASSOCIADAS ÀS CONDIÇÕES CLIMÁTICAS OBTIDAS PELO MODELO ETA CPTEC/HADCM3: AVALIAÇÃO COMPARATIVA ENTRE OS DADOS SIMULADOS E OBSERVADOS DE PRECIPITAÇÃO, EVAPOTRANSPIRAÇÃO E VAZÃO NA BACIA HIDROGRÁFICA DO RIO IJUÍ, BRASIL.

This study aims to evaluate the climate scenarios simulated by the Eta CPTEC/HadCM3 model, conducted by four members of the HadCM3 global climate model (CNTRL, LOW, MID and HIGH), at the Ijui River Basin, Brazil. The used control period was from 1961 to 1975, looking for the assessment of climate scenarios and river flow during the 1976 and 1990 period. The task was divided into five stages: spatial interpolation of climate data, correction of simulated climatic variables series (bias correction), calculation of reference evapotranspiration, hydrological simulation of monthly river flow, comparison between simulated and observed conditions of precipitation, evapotranspiration and river flow. Although the correction methods used to eliminate the simulated climate series biases have originated very different scenarios, neither methods presented a better performance in all examined criteria. Sometimes, the differences between the simulated values based on the Eta Model and the observed values were higher than 20%, both in both rainfall and river flow resulting from the hydrologic modeling processes. Therefore, one must consider that these uncertainties will propagate to future scenarios, when analyzing the effects of climate change on water availability.