Abstract

Some types of land development can be associated with increased impervious area that causes increase in surface runoff and decrease in ground water recharge. Both of these processes can have large-scale ramifications through time. Increased runoff results in higher flows during rainfall events. On the other hand, groundwater recharge decreases due to increase impervious surfaces and decrease rate. Hence, there is a need to quantify the impacts of landuse changes from the point of minimizing potential environmental degradation. The objective of this study is to develop a model for assessing the impacts on the watershed runoff due to changes in landscape patterns. While conceptual or physical based models are of importance in the understanding of hydrologic processes, there are many practical situations where the main concern is with making accurate predictions at specific locations. For this purpose, artificial neural network (ANN) model was developed. Landsat data was used in this study in view of its ability to provide useful information on landuse dynamics. The model’s performance in both training and testing phases were evaluated  using mean absolute error (MAE), mean square error (MSE), U Theil’s coefficient and regression analysis. The correlation coefficients between simulated and real data were found to be 0.94 and 0.89 for the training and testing phases respectively. Most of the data points were within the confidence level of 95%. The model can be used as a decision making tool when formulating landuse policies. It can be a practical tool for hydrologists, engineers, and town and country planners.

Key words: Artificial neural network, river flow, landsat data, land use changes.