Monthly runoff prediction using phase space reconstruction

A nonlinear prediction method, developed based on the ideas gained from det erministic chaos theory, is employed: (a) to predict monthly runoff; and (b ) to detect the possible presence of chaos in runoff dynamics. The method f irst reconstructs the single-dimensional (or variable) runoff series in a m ulti-dimensional phase space to represent its dynamics, and then uses a loc al polynomial approach to make predictions. Monthly runoff series observed at the Coaracy Nunes/Araguari River basin in northern Brazil is studied. Th e predictions are found to be in close agreement with the observed runoff, with high correlation coefficient and coefficient of efficiency values, ind icating the suitability of the nonlinear prediction method for predicting t he runoff dynamics. The results also reveal the presence of low-dimensional chaos in the runoff dynamics, when an inverse approach is adopted for iden tification, as: (a) an optimal embedding dimension exists, and (b) the pred iction accuracy decreases with an increase in prediction lead lime.