Evaluation of plant communities by Remote Sensing Methods (Case study: preserved area of Goloul va Sarani -Northern Khorasan province)

The nature of plant communities is one of the issues that have been always debated by ecologists and plant scientists. In fact, theories such as individualism and community units that included two main poles of these discussions are partly developed by mentality or problems such as sampling, calculation, analysis and conclusions. It seems that the use of new techniques such as remote sensing (RS) and geographical information system (GIS) can largely help to solve the above problems. In any case, it seems that a regional view of the above and survey that how the plants have been distributed by more advanced tools such as satellites sensors and computer besides of field observations can be more effective in comparison with traditional methods. In this study we used Landsat ETM+ and IRS LISS III satellites data to verify that how plant groups distributed in Goloul va Sarani protected area. Field data included 143 training samples that were polled in homogeneous areas in each plant type. Then the samples were analyzed on the satellite images using the index of resolution and error matrix. The results showed that the highest resolution is related to the fields of pure alfalfa. Evaluation of training samples resolution about of the other vegetation classes also presented good results. Meanwhile, the plant communities maps that produced by classification of each satellite images showed clear differences with the other. These differences are likely related to differences of satellite sensors to detect of transition zones (ecoton). However, in this study it was confirmed homogeneous areas of vegetation that exist among of them the vague areas created by Tangled mixed communities.