Monday, February 1, 2010

Study of Land Cover Types In Nablus Area using Satellite Remote Sensing

Study of Land Cover Types In Nablus Area using Satellite Remote Sensing PDF

Manar Mohammad Ahmed Sholi

Supervisor(s)
Dr. Ahmed Ra'fat Ghodieh -
Discussion Commity

160 صفحة
Abstract :

Abstract

This study aimed at providing updated data about the land cover use patterns in the study area using remote sensing technique. It also aimed at analyzing these patterns to produce accurate maps for them using GIS technique to know variations in land use distribution in the study area that comprised Nablus District and some surrounding areas. The total area of the study area is (1142 km2)

The study relied on a digital multi spectral satellite image acquired from Spot satellite with a spatial resolution of (20 m) with three optical wavebands namely: green, red, and near red.

The image of the study area was classified using the (Envi_4) Remote Sensing software. Image Classification relied on the Gausian Maximum Likelihood Classifier method as one of the supervised classification methods. This required collecting field data from different places in the study area in the form of samples called (Training Areas). In this stage, different small areas were selected to represent different land cover patterns of the area. The radiometric characteristics for these samples were then studied. Thereafter, the 14 samples of training areas that represent 14 patterns of land cover types were projected on the image and were consequently used in classifying the image.

Image Classification was carried out by comparing the classification results against selected samples from the field during the fieldwork. The (Overall Accuracy) of the classified image was (75.3 %). This accuracy is moderate. It is recommended to use sufficient number of spectral band to improve classification accuracy.

This study came up with several results and conclusions such as the emphasis on the capability of remote sensing technique in producing accurate land use maps. In addition, the study emphasized the important rule of this technique in environments of complicated topographic structures such as mountainous areas، since they are difficult to be accessed. The study also showed the olive tree as one of the main land cover patterns occupies the highest percentage of the area studied since it is planted in different environments..

The study recommended the use of remote sensing in studying land use change since this technique has the capability of providing updated data on continuous basis. Furthermore, this technique is considered to be as one of the less costly ones in studying large areas. The study also recommends using Spot data to study agricultural areas characterized by small to moderate field size. this reduces the distortion elements that appears when classifying the image. The study finally recommends increasing the training groups to facilitate analysis process of the satellite image.

Full Article

No comments:

Post a Comment