ﺑﺮونﯾﺎﺑﯽ ﻣﮑﺎﻧﯽ ﮔﺮوه ﺑﺰرگ ﺧﺎك ﺑﺎ اﺳﺘﻔﺎده از روش ﺟﻨﮕﻞ ﺗﺼﺎدﻓﯽ در ﻣﻨﻄﻘﮥ ﺧﺸﮏ اﯾﺮان ﻣﺮﮐﺰي (ﻓﺎرﯾﺎب-ﮐﻬﻨﻮج)

The spatial extrapolation of soil great group by application of Random Forest in arid region of central Iran (Faryab-Kahnooj)

در ﻧﻘ ﺸﻪﺑﺮداري رﻗﻮﻣﯽ، ﻗﻮاﻧﯿﻦ ﺣﺎﮐﻢ در ﺑﯿﺎن وﯾﮋﮔﯽﻫﺎ و اﻧﻮاع ﺧﺎكﻫﺎ را ﻣﯽﺗﻮان ﺑﺎ ا ﺳﺘﻔﺎده از ﺗﻌﯿﯿﻦ رواﺑﻂ ﺧ ﺼﻮ ﺻﯿﺖ ﺗﺤﺖ ﻣﻄﺎﻟﻌﻪ-ﻣﺘﻐﯿﺮﻫﺎي ﭘﯿﺶﺑﯿﻨﯽ ﮐﻨﻨﺪة ﻣﺤﯿﻄﯽ و روشﻫﺎي ﻣﺪﻟﺴـﺎزي ﮐﻤّﯽ و ﻋﺪدي، اﺳـﺘﺨﺮاج ﻧﻤﻮد. در ﻧﺘﯿﺠﻪ، ﻗﻮاﻧﯿﻦ اﯾﺠﺎد ﺷـﺪه از اﯾﻦ رواﺑﻂ، ﻗﺎﺑﻞ ﺑﺴﻂ و ﺑﺮازش ﺑﻪ ﻣﻮﻗﻌﯿﺖﻫﺎي ﺑﺎ ﺷﺮاﯾﻂ ﻣﺸﺎﺑﻪ ﻣﯽﺑﺎﺷﻨﺪ ﮐﻪ از آن ﺗﺤﺖ ﻋﻨﻮان ﺑﺮونﯾﺎﺑﯽ ﯾﺎد ﻣﯽﺷﻮد. در ﻣﻄﺎﻟﻌﮥ ﺣﺎﺿﺮ، دﺳﺘﯿﺎﺑﯽ ﺑﻪ ﻧﻘﺸﮥ رﻗﻮﻣﯽ ﺧﺎك در ﻣﻨﻄﻘﻪاي ﮐﻪ ﻓﺎﻗﺪ ﻫﺮﮔﻮﻧﻪ اﻃﻼﻋﺎت ﺧﺎك اﺳﺖ )ﻣﻨﻄﻘﮥ ﭘﺬﯾﺮﻧﺪه(، ﺑﺮونﯾﺎﺑﯽ ﺑﺎ ﺑﻪﮐﺎرﮔﯿﺮي ﻣﺪل ﺟﻨﮕﻞ ﺗﺼﺎدﻓﯽ از ﻣﻨﻄﻘﮥ داراي اﻃﻼﻋﺎت ﺧﺎك )ﻣﻨﻄﻘﮥ ﻣﺮﺟﻊ ﯾﺎ دﻫﻨﺪه( ﻣﻮرد ﺑﺮرﺳــﯽ ﻗﺮار ﮔﺮﻓﺖ. ﻣﻨﻄﻘﮥ ﻓﺎرﯾﺎب ﺷــﻬﺮﺳــﺘﺎن ﮐﻬﻨﻮج ﺑﻪ ﻋﻨﻮان ﻣﻨﻄﻘﮥ ﻣﺮﺟﻊ و ﻣﻨﻄﻘﮥ ﮐﻬﻨﻮج ﺑﻪ ﻋﻨﻮان ﻣﻨﻄﻘﮥ ﭘﺬﯾﺮﻧﺪه اﻧﺘﺨﺎب ﮔﺮدﯾﺪ. ﺗﺸـﺎﺑﻪ ﻓﺎﮐﺘﻮرﻫﺎي ﺧﺎﮐﺴـﺎزي دو ﻣﻨﻄﻘﮥ ﻣﺮﺟﻊ و ﭘﺬﯾﺮﻧﺪه ﺑﺎ اﺳـﺘﻔﺎده از ﺷﺎﺧﺺﻫﺎي ﺗﺸﺎﺑﻪ اﻗﻠﯿﻤﯽ، ﭘﺴﺘﯽ و ﺑﻠﻨﺪي، زﻣﯿﻦ ﺷﻨﺎﺳﯽ و ﺷﮑﻞ زﻣﯿﻦ ﺑﺮرﺳﯽ ﮔﺮدﯾﺪ. در ﮔﺎم ﻧﺨﺴﺖ، ﻣﺪل ﺟﻨﮕﻞ ﺗﺼﺎدﻓﯽ ﺑﻪ ﻣﻨﻈﻮر ﺗﻌﯿﯿﻦ ارﺗﺒﺎط ﮐﻼسﻫﺎي ﺧﺎك ﺑﺎ ﻣﺘﻐﯿﺮﻫﺎي ﻣﺤﯿﻄﯽ در ﻣﻨﻄﻘﮥ ﻣﺮﺟﻊ اﺳـــﺘﻔﺎده ﮔﺮدﯾﺪ. ﻧﺘﺎﯾﺞ ﺣﺎﺻـــﻞ از ﺑﺮازش اﯾﻦ ﻣﺪل در ﻣﻨﻄﻘﮥ ﻣﺮﺟﻊ دﻗﺖ ﮐﻠﯽ 88% و ﮐﺎﭘﺎ 0/77 را ﻧﺸــﺎن ﻣﯽدﻫﺪ. ﺑﺎ ﺗﻮﺟﻪ ﺑﻪ ﻋﻤﻠﮑﺮد ﻣﻨﺎﺳــﺐ ﻣﺪل در ﻣﻨﻄﻘﮥ ﻣﺮﺟﻊ و ﺷــﺒﺎﻫﺖ دو ﻣﻨﻄﻘﻪ، اﻣﮑﺎن اﺳـﺘﻔﺎده از راﺑﻄﮥ ﮐﻼسﻫﺎي ﺧﺎك و ﻣﺘﻐﯿﺮﻫﺎي ﻣﺤﯿﻄﯽ اﺳـﺘﺨﺮاج ﺷـﺪه در ﻣﻨﻄﻘﮥ ﭘﺬﯾﺮﻧﺪه وﺟﻮد دارد. ﻧﺘﺎﯾﺞ ﺣﺎﺻـﻞ از ﺑﺮونﯾﺎﺑﯽ دﻗﺖ ﮐﻠﯽ 81% و ﮐﺎﭘﺎ 0/61 را در ﺗﺨﻤﯿﻦ ﮐﻼسﻫﺎي ﺧﺎك ﻣﻨﻄﻘﮥ ﭘﺬﯾﺮﻧﺪه ﻧﺸـﺎن ﻣﯽدﻫﺪ. در ﻧﺘﯿﺠﻪ دﺳـﺘﯿﺎﺑﯽ ﺑﻪ ﻧﻘﺸـﮥ ﮐﻼس ﺧﺎك ﻣﻨﻄﻘﮥ ﭘﺬﯾﺮﻧﺪه ﺑﺎ ﺻــﺮف ﻫﺰﯾﻨﻪ و زﻣﺎن ﺑﻪ ﻣﺮاﺗﺐ ﮐﻢﺗﺮي ﺑﻮدهاﺳــﺖ، ﻟﺬا ﻣﯽﺗﻮان ﺑﺮونﯾﺎﺑﯽ را ﺑﻪ ﻋﻨﻮان روﺷــﯽ ﮐﺎرآﻣﺪ در ﺗﺨﻤﯿﻦ و ﭘﯿﺶﺑﯿﻨﯽ ﮐﻼس ﺧﺎك در ﻣﻨﺎﻃﻖ ﻓﺎﻗﺪ اﻃﻼﻋﺎت ﻣﻌﺮﻓﯽ ﮐﺮد.

In digital soil mapping, soil characteristic and classes could be extracted truly by numerical and quantitative modelling. Hence, derived rules could be fitted to similar regions for achieving ruled relations on areas without soil information which is called as extrapolation. In the present study, achieving digital soil class map of an area without adequate soil information by Random forest was tested by extrapolation at great group level. The results show overall accuracy 88% and kappa 0.77 of donor area which is able to fit over its similar region. Results of extrapolation show overall accuracy 81% and kappa 0.61 of recipient area which could show logical concordance to produce soil class map of recipient area without applying related soil samples of this area in short time and low cost. Since, extrapolation could be as an efficient way to predict soil classes of unknown areas.