A technique for simulating future climate change variable using improved K-Nearest neighbors algorithm (k-NN)

A technique for simulating future rainfall events using improved K-NN (K-nearest neighbors) algorithm is presented in this study. The K-NN calculation don't deliver new informational collections however simply reshuffles the chronicled information to create new informational indexes. A simulation day was selected in the month of August. The algorithm steps and resampling with ancient documents was applied to simulate rainfall events in Kaduna River catchment as a basis for future understanding of the characteristics of the basin. Simulated data sets for the months of April, May, June, July, August, September and October yielded nearly exact replica of the ancient documents. In performing the model, statistical characteristics such as mean, standard deviation, variance, cumulative probability, covariance, skewness, cross correlation are all preserved by the K-NN model. The technique clearly demonstrates that it can be used to generate future rainfall events that can be applied for hydrological investigation of the characteristics of a basin for future developments.