تعيين نقشه كاربري اراضي با استفاده از سنجندهETM+ (مطالعه موردي حوزه آبخيز هندودر)

Preparation of Land Use Map Using ETM+ Landsat (A Case Study in Hendodar Watershed)

زﻣﯿﻨﻪ و ﻫﺪف: رﺷﺪ ﺳﺮﯾﻊ ﺟﻤﻌﯿﺖ ﺳﺒﺐ اﺳﺘﻔﺎده ﺑﯿﺶ از ﺣﺪ ﻇﺮﻓﯿﺖ و ﻓﺸﺎر ﻣﻀﺎﻋﻒ ﺑﺮ ﻣﻨﺎﺑﻊ ﻃﺒﯿﻌﯽ ﺷﺪه ﮐﻪ در ﻧﺘﯿﺠﻪ آن ﺗﻐﯿﯿﺮات ﺳﺮﯾﻊ ﮐﺎرﺑﺮي اراﺿﯽ اﺗﻔﺎق اﻓﺘﺎده اﺳﺖ. ﺑﻨﺎﺑﺮاﯾﻦ ﺷﻨﺎﺧﺖ ﺳﺮﯾﻊ و دﻗﯿﻖ اﻧﻮاع ﭘﻮﺷﺶ زﻣﯿﻦ ﻣﯽﺗﻮاﻧﺪ ﻧﻘﺶ ﻣﺆﺛﺮي در ﺑﺮﻧﺎﻣﻪرﯾﺰي و ﻣﺪﯾﺮﯾﺖ داﺷﺘﻪ ﺑﺎﺷﺪ. دادهﻫﺎي ﻣﺎﻫﻮارهاي ﺑﻪدﻟﯿﻞ دﯾﺪ وﺳﯿﻊ و ﯾﮑﭙﺎرﭼﻪ، در ﺑﺮﮔﺮﻓﺘﻦ ﺑﺨﺶ ﻋﻤﺪهاي از ﻃﯿﻒ اﻟﮑﺘﺮوﻣﻐﻨﺎﻃﯿﺴﯽ و ﺑﻪروز ﺑﻮدن ﺗﺼﺎوﯾﺮ ﺑﺮاي ﺗﻬﯿﻪ ﻧﻘﺸﻪﻫﺎي ﮐﺎرﺑﺮي اراﺿﯽ ﻣﻨﺎﺳﺐ ﻣﯽﺑﺎﺷﻨﺪ. ﻣﻄﺎﻟﻌﻪ ﺣﺎﺿﺮ ﺑﺎ ﻫﺪف ﺗﻌﯿﯿﻦ ﻧﻘﺸﻪ ﮐﺎرﺑﺮي اراﺿﯽ ﺑﺎ اﺳﺘﻔﺎده از ﺳﻨﺠﻨﺪه+ETM ﻣﻄﺎﻟﻌﻪ ﻣﻮردي ﺣﻮزه آﺑﺨﯿﺰ ﻫﻨﺪودر( اﻧﺠﺎم ﮔﺮﻓﺘﻪ اﺳﺖ. روش ﺑﺮرﺳﯽ: از ﺗﺼﺎوﯾﺮ ﻣﺎﻫﻮارهاي ﻟﻨﺪﺳﺖ 7 ﺑﺮاي ﺗﻌﯿﯿﻦ ﭘﻮﺷﺶ و ﮐﺎرﺑﺮي اراﺿﯽ ﺣﻮزه آﺑﺨﯿﺰ ﻫﻨﺪودر اﺳﺘﺎن ﻣﺮﮐﺰي اﺳﺘﻔﺎده ﺷﺪ. ﺑﺮداﺷﺖ ﻣﻮﻗﻌﯿﺖ اﻧﻮاع ﮐﺎرﺑﺮي و ﭘﻮﺷﺶ اراﺿﯽ از ﺳﻄﺢ ﺣﻮزه ﻣﻮرد ﻧﻈﺮ ﺑﺎ ﻧﻘﺎط ﺗﺴﺖ و ﮐﻨﺘﺮل ﺑﺎ اﺳﺘﻔﺎده از GPS اﻧﺠﺎم ﮔﺮﻓﺖ. ﻃﺒﻘﻪﺑﻨﺪي ﺑﻪ روش ﻧﻈﺎرت ﺷﺪه و ﺑﺎ ﭼﻬﺎر اﻟﮕﻮرﯾﺘﻢ ﻣﺨﺘﻠﻒ ﻃﺒﻘﻪﺑﻨﺪي ﮐﻪ ﺷﺎﻣﻞ ﺣﺪاﮐﺜﺮ اﺣﺘﻤﺎل، ﺣﺪاﻗﻞ ﻓﺎﺻﻠﻪ ﺗﺎ ﻣﯿﺎﻧﮕﯿﻦ،Minimum Mahalanobis Distance و ﻣﺘﻮازي اﻟﺴﻄﻮح ﺑﺮ روي ﮔﺮوهﻫﺎي ﺳﻪ ﺑﺎﻧﺪي ﺻﻮرت ﮔﺮﻓﺖ. ﯾﺎﻓﺘﻪﻫﺎ: ﺟﻬﺖ اﻧﺘﺨﺎب ﺑﻬﺘﺮﯾﻦ ﮔﺮوه ﺳﻪ ﺑﺎﻧﺪي از روش ﻣﺤﺎﺳﺒﻪ ﺣﺪ ﺷﺎﺧﺺ ﻣﻄﻠﻮﺑﯿﺖ )OIF( ﺑﺮاي ﺑﺎﻧﺪﻫﺎي اﺻﻠﯽ، ﺷﺎﺧﺺﻫﺎ و ﺗﺠﺰﯾﻪ ﻣﺆﻟﻔﻪﻫﺎ اﺳﺘﻔﺎده ﺷﺪ. ﺗﺼﺎوﯾﺮ ﻣﺎﻫﻮارهاي ﺑﺎ اﺳﺘﻔﺎده از اﻟﮕﻮرﯾﺘﻢﻫﺎي ﻃﺒﻘﻪﺑﻨﺪي ﻧﻈﺎرت ﺷﺪه و اﻋﻤﺎل ﻧﻤﻮﻧﻪﻫﺎي ﺑﺮداﺷﺘﯽ از ﺳﻄﺢ ﻣﻨﻄﻘﻪ، ﻃﺒﻘﻪﺑﻨﺪي ﮔﺮدﯾﺪ. از ﺑﯿﻦ اﻟﮕﻮرﯾﺘﻢﻫﺎ، اﻟﮕﻮرﯾﺘﻢ ﻃﺒﻘﻪﺑﻨﺪي ﺣﺪاﮐﺜﺮ اﺣﺘﻤﺎل داراي ﻧﺘﺎﯾﺞ ﺑﻬﺘﺮي از اﻧﻮاع ﭘﻮﺷﺶ و ﮐﺎرﺑﺮيﻫﺎي اراﺿﯽ ﺑﺮ روي ﺗﺼﺎوﯾﺮ ﺑﻮد. ﺑﺤﺚ و ﻧﺘﯿﺠﻪﮔﯿﺮي: ﺑﻬﺘﺮﯾﻦ ﻧﺘﯿﺠﻪ ﻃﺒﻘﻪﺑﻨﺪي در ﻣﻘﺎﯾﺴﻪ ﺑﺎ ﻧﻘﺸﻪ واﻗﻌﯿﺖ زﻣﯿﻨﯽ، ﺑﺎ اﺳﺘﻔﺎده از ﻣﺠﻤﻮﻋﻪ ﺑﺎﻧﺪﻫﺎي b4 ، b1 و b7 ﺑﺎ دﻗﺖ ﮐﻠﯽ81/25 درﺻﺪ در اﻟﮕﻮرﯾﺘﻢ ﻃﺒﻘﻪﺑﻨﺪي ﺣﺪاﮐﺜﺮ اﺣﺘﻤﺎل ﺑﻪدﺳﺖ آﻣﺪ ﮐﻪ ﺑﯿﺎﻧﮕﺮ ﻗﺎﺑﻠﯿﺖ و ﮐﺎراﯾﯽ ﻧﺴﺒﺘﺎً ﺧﻮب دادهﻫﺎي +ETM در تهيه ﻧﻘﺸﻪ ﻧﻬﺎﯾﯽ ﮐﺎرﺑﺮي اراﺿﯽ ﻣﯽﺑﺎﺷﺪ.

Background and Objective: High population growth rate has led to excessive use of capacity and double pressure on natural resources, resulting in rapid land use changes. Therefore, quick and accurate identification of types of land cover can play an effective role in planning and management. Satellite data because of vast and integrated sight covering with different electromagnetic spectrums and updated images are very suitable for making applicable Land use maps. The aim of this study is preparation of land use map using ETM+ landsat (a Case Study in Hendodar Watershed) Method: The Landsat 7 satellite images were used to determine the land use changes of Hendodar watershed in Markazi province. The GPS was used to determine the position of land use and land cover types on the basis of taking test and ground control points on field investigation. Obtained samples were used for supervised classification with four different algorithms including maximum Likelihood, minimum distance, Minimum Mahalanobis Distance and Box Classification. Findings: The optimum index factor (OIF) for the main bands and PCA (principal coordinate analysis) were used to select the optimum combination of three bands in a satellite image to create a color composite, sample set and other operation and classification. Among the algorithms, the maximum likelihood classification algorithm had better results from the types of coverage and lands use on the images. Discussion and Conclusion: The maximum likelihood classification algorithm with combination of b7, b4, b1 bands with 81.25% accuracy is the best algorithms of land use determination and classification comparing with real ground map of the area.