Modelagem e análise temporal da dinâmica do uso e ocupação do solo na bacia do rio Cuiá

Abstract

The two decades of population growth accelerated the process of human design on areas, various environmental problems. The objective of this work was to analyze the changes in soil use and occupation in the Cuia River Basin and to estimate the use and occupation of the soil for the Cuia basin for the year 2030 Artificial neural network conditioning: Neural Network Perceptron Multi- Layer. In this way, the satellite image file of the years 1998, 2001 and 2005 was carried out, a prediction of the use of the ground was made for the year 2001 in Neural Networks for the validation, then the result of the prediction of the land use was published in accordance with the Kappa and after unloading the Kappa with a result according to a classification predicted by Landis & Koch (1977) a prediction of land use and occupation was made for 2030. the years. The dynamic modeling of soil use is based on the Neural Network satisfactory results for the Cuia River basin with accuracy of 98.64%, after 10,000 iterations and Kappa comparisons equal to 0.94, classification as excellent. The land use forecast for 2030 increased the area of the area of occupation and an area of expansion. This does not mean that the LCM may be used as the tool of the water resources, as well as the tool for planning and ordering of use and occupation of future soil.