Estimation of Vegetation and Land Use Changes Using Remote Sensing Techniques and Geographical Information System (Case Study: Roodab Plain, Sabzevar City)

Abstract

Land use may be regarded as one of the most important factors affecting the<br />environment with respect to human activities. So far, destroying the rangelands and<br />changing them into the waste lands and poor rangelands has been proposed as the most<br />significant variations of land use done by human beings. This paper has been conducted to<br />evaluate the variations of vegetation percentage and land uses in Barabad-Darook village<br />with the area of 1522.99 km2 in Sabzevar city during 1987-2007. Thus, using satellitebased<br />images of TM and ETM+, the most appropriate band composition has been selected<br />and a mapping of vegetation cover and land use was provided through maximum<br />likelihood algorithms to correct the errors of geometer and radiometer highlights. At last,<br />the accuracy of extracted maps was to be determined by the means of overall accuracy test<br />and Kappa coefficient in order to achieve the validation of research process. Results<br />indicate that waste lands have been increased from 84.75 to 89.49 and third-rated<br />rangelands have been reduced from 6.85 to 4.14 percent. On the other hand, first-rated<br />rangelands were reduced from 0.03 to 0.01 percent which covers 5170791.45 m2 of total<br />area in the district. Also, the results show that irrigated agricultural lands are to be<br />decreased from 6.53 to 0.07 percent. In total, due to improper exploitations of regional<br />water resources and vegetation cover, land uses have been changed into fallow and waste<br />lands leading to the decrease in the percentage cover of high quality rangelands. Research<br />findings demonstrate that considering the accepted accuracy, new remote sensing<br />technology can be applied to exactly estimate the area changes of land use and vegetation.