كليدواژه :
اﻋﺘﺒﺎرﺳﻨﺠﯽ , MaxEnt , ﻣﻨﺎﺑﻊ آب , AHP , ﻣﻨﺤﻨﯽ ﺗﺸﺨﯿﺺ ﻋﻤﻠﮑﺮد
چكيده فارسي :
ﺑﺎ اﻓﺰاﯾﺶ ﺟﻤﻌﯿﺖ، اﻫﻤﯿﺖ ﻣﻨﺎﺑﻊ آب زﯾﺮزﻣﯿﻨﯽ ﺑﻪﻋﻨﻮان ﯾﮑﯽ از ﻣﻬﻤﺘﺮﯾﻦ ﻣﻨﺎﺑﻊ ﺗﺄﻣﯿﻦ ﮐﻨﻨﺪه آب ﺷﺮب در ﻣﻨﺎﻃﻖ ﺧﺸﮏ ﺑﯿﺶ از ﭘﯿﺶ آﺷﮑﺎر ﻣﯽ ﺷﻮد. در اﯾﻦ ﭘﮋوﻫﺶ، ﺑﻪﻣﻨﻈﻮر ﺗﻌﯿﯿﻦ ﻣﻨﺎﻃﻖ داراي ﭘﺘﺎﻧﺴﯿﻞ آب زﯾﺮزﻣﯿﻨﯽ ﺷﻬﺮﺳﺘﺎن ﺗﺮﺑﺖ ﺟﺎم و اوﻟﻮﯾﺖﺑﻨﺪي ﻋﻮاﻣﻞ ﻣﻮﺛﺮ از روشﻫﺎي ﺗﺤﻠﯿﻞ ﺳﻠﺴﻠﻪ ﻣﺮاﺗﺒﯽ و روش ﺑﯿﺸﯿﻨﻪ آﻧﺘﺮوﭘﯽ ﺑﺎ اﺳﺘﻔﺎده از ﻣﺪل MaxEnt و ﻋﻮاﻣﻞ ﻓﺎﺻﻠﻪ و ﺗﺮاﮐﻢ ﮔﺴﻞ، ﺳﻨﮓﺷﻨﺎﺳﯽ، ﺷﯿﺐ، ﺟﻬﺖ ﺷﯿﺐ، ﻓﺎﺻﻠﻪ از آﺑﺮاﻫﻪ و ﺗﺮاﮐﻢ زﻫﮑﺸﯽ، ﻃﺒﻘﺎت ارﺗﻔﺎﻋﯽ، ﮐﺎرﺑﺮي اراﺿﯽ، اﻧﺤﻨﺎي داﻣﻨﻪ، ﺷﺎﺧﺺ رﻃﻮﺑﺖ ﺗﻮﭘﻮﮔﺮاﻓﯿﮏ و ﺷﺎﺧﺺ ﻣﻮﻗﻌﯿﺖ ﺗﻮﭘﻮﮔﺮاﻓﯿﮏ اﺳﺘﻔﺎده ﺷﺪه اﺳﺖ. ﻫﻤﭽﻨﯿﻦ، ﺑﺮاي ارزﯾﺎﺑﯽ اﯾﻦ دو روش، از ﻣﻨﺤﻨﯽ ﺗﺸﺨﯿﺺ ﻋﻤﻠﮑﺮد ﻧﺴﺒﯽ )ROC( اﺳﺘﻔﺎده ﺷﺪ. از ﻣﺠﻤﻮع 220 ﭼﺸﻤﻪ ﻣﻮﺟﻮد، ﺑﻪ ﺻﻮرت ﺗﺼﺎدﻓﯽ 30 درﺻﺪ ﺑﻪﻋﻨﻮان دادهﻫﺎي اﻋﺘﺒﺎرﺳﻨﺠﯽ و 70 درﺻﺪ ﺑﻪﻋﻨﻮان دادهﻫﺎي آزﻣﻮن )روش ﺑﯿﺸﯿﻨﻪ آﻧﺘﺮوﭘﯽ( ﻃﺒﻘﻪﺑﻨﺪي ﺷﺪﻧﺪ. ﺑﺮ اﺳﺎس روش ﺑﯿﺸﯿﻨﻪ آﻧﺘﺮوﭘﯽ، ﻧﺘﺎﯾﺞ ﻧﺸﺎن داد ﮐﻪ 29/6 درﺻﺪ ﺣﻮزه آﺑﺨﯿﺰ داراي ﭘﺘﺎﻧﺴﯿﻞ آب زﯾﺮزﻣﯿﻨﯽ ﺑﺎﻻﯾﯽ اﺳﺖ. ﺑﺮ اﺳﺎس ﻧﻤﻮدار ﺟﮑﻨﺎﯾﻒ، ﻻﯾﻪﻫﺎي ارﺗﻔﺎع )DEM(، ﺷﯿﺐ، ﻓﺎﺻﻠﻪ از ﮔﺴﻞ و ﺳﻨﮓﺷﻨﺎﺳﯽ ﺑﻪ ﺗﺮﺗﯿﺐ ﻣﻬﻤﺘﺮﯾﻦ ﻋﻮاﻣﻞ ﺗﺎﺛﯿﺮﮔﺬار ﺑﺮ ﭘﺘﺎﻧﺴﯿﻞ آب زﯾﺮزﻣﯿﻨﯽ ﺑﻮدﻧﺪ. ﺳﻄﺢ زﯾﺮ ﻣﻨﺤﻨﯽ )AUC( در روش ﺑﯿﺸﯿﻨﻪ آﻧﺘﺮوﭘﯽ، ﻧﺸﺎندﻫﻨﺪه دﻗﺖ 91 درﺻﺪ )ﻋﺎﻟﯽ( در ﻣﺮﺣﻠﻪ آﻣﻮزش و 80 درﺻﺪ )ﺧﯿﻠﯽ ﺧﻮب( در ﻣﺮﺣﻠﻪ اﻋﺘﺒﺎرﺳﻨﺠﯽ ﺑﺮاي ﺗﻌﯿﯿﻦ ﻣﻨﺎﻃﻖ داراي ﭘﺘﺎﻧﺴﯿﻞ آب زﯾﺮزﻣﯿﻨﯽ ﺑﻮد. ﺑﺮ اﺳﺎس روش ﺗﺤﻠﯿﻞ ﺳﻠﺴﻠﻪ ﻣﺮاﺗﺒﯽ، 34/4 درﺻﺪ ﺣﻮﺿﻪ داراي ﭘﺘﺎﻧﺴﯿﻞ آب زﯾﺮزﻣﯿﻨﯽ ﻣﯽﺑﺎﺷﺪ و ﻻﯾﻪﻫﺎي ﺷﯿﺐ، ﺳﻨﮓﺷﻨﺎﺳﯽ، ارﺗﻔﺎع و ﻓﺎﺻﻠﻪ از ﮔﺴﻞ ﺑﻪ ﺗﺮﺗﯿﺐ ﻣﻬﻤﺘﺮﯾﻦ ﻋﻮاﻣﻞ ﺗﺎﺛﯿﺮﮔﺬار ﺑﻮدﻧﺪ و دﻗﺖ اﯾﻦ روش 73 درﺻﺪ ﺑﺮآورد ﺷﺪ. ﻧﺘﺎﯾﺞ ﻧﺸﺎن داد ﮐﻪ ﺑﻪ ﮐﺎرﮔﯿﺮي روشﻫﺎي ﺗﺤﻠﯿﻞ ﺳﻠﺴﻠﻪ ﻣﺮاﺗﺒﯽ و روش ﺑﯿﺸﯿﻨﻪ آﻧﺘﺮوﭘﯽ، ﺿﻤﻦ ﺻﺮﻓﻪﺟﻮﯾﯽ در زﻣﺎن و ﻫﺰﯾﻨﻪ، ﻗﺎﺑﻠﯿﺖ ﻣﻨﺎﺳﺒﯽ در ﭘﯿﺶ ﺑﯿﻨﯽ ﭘﺘﺎﻧﺴﯿﻞ آب زﯾﺮزﻣﯿﻨﯽ دارﻧﺪ و روش ﺑﯿﺸﯿﻨﻪ آﻧﺘﺮوﭘﯽ ﺑﺮﺗﺮي ﺑﯿﺸﺘﺮي ﻧﺴﺒﺖ ﺑﻪ روش ﺗﺤﻠﯿﻞ ﺳﻠﺴﻠﻪ ﻣﺮاﺗﺒﯽ دارد.
چكيده لاتين :
With the increase in population, the importance of groundwater resources as one of the most important sources of drinking water in the arid regions becomes apparent. In this research, in order to determine the areas with groundwater potential in the city of Torbate Jam and prioritizing the effective factors, hierarchical analysis methods and maximum entropy method using MaxEnt model and the factors of distance from fault and fault density, lithology, slope, slope direction, distance from the waterway and drainage density, elevation, land use, slope curvature, topographic humidity index and topographic position indicator was used. Also, for assessing these two methods, the Receiver Operating Characteristic (ROC) was used. From 220 sources, 30% were randomly assigned as validation data and 70% were categorized as test data in maximum entropy method. Results showed that 29.6% of the watershed had high groundwater potential according to the maximum entropy method. Based on Jack-Knife Diagram, DEM, slope, distance from fault and lithology were the most important factors affecting groundwater potential, respectively. The Area Under the Curve (AUC) in the maximum entropy method indicated a precision of 91% (excellent) at the training period and 80% (very good) in the validation period to determine areas with potential for groundwater. Based on AHP method, 34.4% of the area has groundwater potential, and the slope layers, lithology, elevation and distance from the faults were the most important factors, respectively and accuracy of this method was 73%. The results showed that applying AHP and maximum entropy methods, while saving time and cost, have a good ability to predict the potential of groundwater and the maximum entropy method has more superiority than the hierarchical analysis method.
