طبقه‌بندي ناهمواري‌هاي كارستي با استفاده از شاخص‌هاي ژئومورفومتريك و شبكۀ عصبي مصنوعي (مطالعۀ موردي: بخشي از حوضه‌هاي خرم‌آباد، بيرانشهر و الشتر)

Landform classification of karstic area by Goemorphometric Index and Artificial Neural Network (Case study: A part of Korram Abad, Biran Shahr and Alashtar Watersheds)

اﺳﺘﻔﺎده از ﺷﺎﺧﺺﻫﺎي ژﺋﻮﻣﻮرﻓﻮﻣﺘﺮي در ﺗﻔﮑﯿﮏ ﻧﺎﻫﻤﻮاريﻫﺎي ﺳﻄﺢ زﻣﯿﻦ ﮐﺎرﺑﺮد ﮔﺴﺘﺮدهاي را ﻃﯽ دﻫﮥ ﮔﺬﺷﺘﻪ در ﻋﻠﻢ ژﺋﻮﻣﻮرﻓﻮﻟﻮژي داﺷﺘﻪ اﺳﺖ. در اﯾﻦ ﺗﺤﻘﯿﻖ از روش ﭘﺮﺳﭙﺘﺮون ﭼﻨﺪ ﻻﯾﮥ ﺷﺒﮑﮥ ﻋﺼﺒﯽ ﻣﺼﻨﻮﻋﯽ ﺑﺮاي ﻃﺒﻘﻪﺑﻨﺪي ﻧﺎﻫﻤﻮاريﻫﺎي ﮐﺎرﺳﺘﯽ اﺳﺘﻔﺎده ﺷﺪ. اﺑﺘﺪا ﺑﺎ اﺳﺘﻔﺎده از ﻧﻘﺸﮥ ﻣﺪل رﻗﻮﻣﯽ ارﺗﻔﺎع، ﺷﺎﺧﺺﻫﺎي ژﺋﻮﻣﻮرﻓﻮﻣﺘﺮي ﺗﻬﯿﻪ ﺷﺪ و ﺳﭙﺲ اﯾﻦ ﺷﺎﺧﺺﻫﺎ ﺑﻪﻋﻨﻮان ﻧﺮونﻫﺎي ﻻﯾﮥ ورودي در ﺷﺒﮑﮥ ﻋﺼﺒﯽ ﻣﺼﻨﻮﻋﯽ اﺳﺘﻔﺎده ﺷﺪ. ﻋﻼوه ﺑﺮ اﯾﻦ از ﻧﻤﻮدارﻫﺎي ﺟﻌﺒﻪاي ﺑﺮاي ﺗﺤﻠﯿﻞ ارﺗﺒﺎط ﻧﺎﻫﻤﻮاريﻫﺎي ﮐﺎرﺳﺘﯽ ﻫﻤﭽﻮن دوﻟﯿﻦ، ﺗﭙﻪ، د ﺷﺖ ﮐﺎر ﺳﺘﯽ، درة ﮐﺎر ﺳﺘﯽ و ﭘﺮﺗﮕﺎه ﺑﺎ ﺷﺎﺧﺺﻫﺎي ژﺋﻮﻣﻮرﻓﻮﻣﺘﺮي ا ﺳﺘﻔﺎده ﺷﺪ. ﻧﺘﺎﯾﺞ ﻃﺒﻘﻪﺑﻨﺪي ﻧ ﺸﺎن داد ﮐﻪ ﻧﺎﻫﻤﻮاريﻫﺎي ﻣﻨﻄﻘﮥ ﻣﻮرد ﻣﻄﺎﻟﻌﻪ ﺑﻪﺗﺮﺗﯿﺐ ﺷﺎﻣﻞ 48/5 ،1/07 ،6/9 ،34 و 9/51 درﺻﺪ دره، دﺷﺖ، دوﻟﯿﻦ، ﭘﺮﺗﮕﺎه و ﺗﭙﻪ ﻣﯽﺑﺎﺷﺪ. ﻋﻼوه ﺑﺮ اﯾﻦ، ﻧﺘﺎﯾﺞ ﻧ ﺸﺎن داد ﮐﻪ ﻣﺪل ﺑﻬﯿﻨﮥ ﺷﺒﮑﮥ ﻋ ﺼﺒﯽ ﻣ ﺼﻨﻮﻋﯽ ﺑﺮاي ﻃﺒﻘﻪﺑﻨﺪي ﻧﺎﻫﻤﻮاريﻫﺎ، ﻣﺪل 12-9-1 ﺑﺎ ﺿﺮﯾﺐ ﯾﺎدﮔﯿﺮي 0/1 و ﺿﺮﯾﺐ ﺗﺒﯿﯿﻦ 87/18 درﺻﺪ ﺑﻮد و دﻗﺖ روش اﺑﺪاﻋﯽ ﺑﺮاي ﻃﺒﻘﻪﺑﻨﺪي ﻧﺎﻫﻤﻮاريﻫﺎي ﮐﺎرﺳﺘﯽ 90/58 درﺻﺪ ﻣﯽﺑﺎﺷﺪ. ﻫﻤﭽﻨﯿﻦ ﺗﺤﻠﯿﻞﻫﺎ ﻧﻤﺎﯾﻨﺪه اﯾﻦ اﺳﺖ ﮐﻪ ﺗﻐﯿﯿﺮات ﺷﺎﺧﺺﻫﺎي ژﺋﻮﻣﻮرﻓﻮﻣﺘﺮي در ﻧﺎﻫﻤﻮاريﻫﺎي ﺗﭙﻪ، ﭘﺮﺗﮕﺎه و درهﮐﺎرﺳﺘﯽ ﺑﺴﯿﺎر ﻧﻤﺎﯾﺎن ﺑﻮده وﻟﯽ در دﺷﺖ و دوﻟﯿﻦ ﮐﻤﯽ داراي ﻫﻤﭙﻮﺷﺎﻧﯽ ﻫﺴﺘﻨﺪ.

The geomorphometric indexes have been widely used for separation of surface landform features in the geomorphology science over the past decades. In this study, Multilayer Perceptron Neural Network (MPNN) was used to provide karstic landform classification. To that regard, initially, geomorphometric indicators were extracted from Digital Elevation Model (DEM), and then these indexes were used as neurons of input layer in artificial neural network. Furthermore, the box plots were applied to analyze the relationship between karstic landforms (such as dolines, hills, karstic plains, karstic valley and headland) and geomorphometric indexes. The results showed that 34, 6.9, 1.07, 48.5, 9.51 percent of the studying area are spatially covered by valleys, plains, dolines, highlands and hills respectively. It has also been found that the optimal structure of artificial neural networks for classification of landform is model No. 12-9-1 by having the learning rate 0.1 and 87.18 percent of determination coefficient. Also, it should be noted that the accuracy of the innovative method for classification of karstic landform is 90.58 percent. The analysis revealed that variations in geomorphometric indexes are very visible in the landform of hills, highlands and karstic valleys, whereas there are slightly overlapping in the plains and dolines.