همکاری با انجمن علمی مدیریت و کنترل مناطق بیابانی ایران

نوع مقاله : مقاله پژوهشی

نویسندگان

1 استادیار گروه مهندسی طبیعت و گیاهان دارویی، دانشکده کشاورزی و منابع طبیعی، دانشگاه تربت‌حیدریه، خراسان رضوی، ایران

2 کارشناس ارشد آبخیزداری، شرکت سامانه آب کومش، خراسان رضوی، ایران

3 کارشناس گروه آموزشی مدیریت مناطق بیابانی، مرکز تحقیقات بین‌المللی بیابان، دانشگاه تهران، ایران

چکیده

گسل از فرآیندهای اصلی زمین­ریختی است که با پدیده‌هایی هم­چون زمین‌لرزه ارتباط دارد. لذا آگاهی از اثر عوامل محیطی در بروز گسل و شناخت مناطق پرخطر اهمیت زیادی دارد که این اهداف در پژوهش حاضر بررسی شده­اند. عرصه پژوهش حوضه قره­قوم است که پس از تعیین گسل­های آن، متغیرهای ارتفاع، شیب، جهت، اقلیم، کاربری، ژئومورفولوژی، سنگ­شناسی، فرسایش، بارش، پهنه­های سیلابی، خاک­شناسی، شاخص­های قدرت جریان، ناهمواری زمین و فاصله از جاده، آبراهه، تاقدیس و ناودیس آن استخراج گردید. سپس الگوریتم­های داده­کاوی شامل درخت­تصمیم، جنگل­تصادفی، تجمیعی­بوستینگ، ماشین­بردارپشتیبان، رگرسیون­لوجیستیک و شبکه­عصبی در نرم‌افزار R جهت شناسایی ارزش متغیرها و روش­های آماری دومتغیره شامل ارزش اطلاعات و تراکم سطح برای شناسایی ارزش کلاس­های هر متغیر در وقوع گسل برازش شد. ارزیابی دقت طبقه­بندی الگوریتم­ها با منحنی ROC نشان داد الگوریتم­های جنگل­تصادفی و ماشین­بردارپشتیبان به ترتیب با سطح زیر منحنی  88 و 86 درصد بهترین عملکرد را در طبقه­بندی وقوع گسل بر مبنای متغیرهای ورودی دارند. در نهایت با ضرایب جینی در الگوریتم جنگل تصادفی، نقشه­های پهنه­بندی به­دست آمده از ترکیب این الگوریتم با روش­های آماری دومتغیره،تهیه و اعتبارسنجی شد. طبق این الگوریتم، به­ترتیب متغیرهای ارتفاع، خاک­شناسی و شاخص ناهمواری زمین مهم‌ترین پارامترها در وقوع گسل شناخته شدند. طبق ارزیابی نقشه­های پهنه­بندی­، در روش ارزش اطلاعات و تراکم سطح به­ترتیب حدود 52 و 35 درصد گسل­ها در کلاس خطر خیلی­زیاد واقع شدند. لذا روش ارزش اطلاعات در تعیین پهنه­های حساس به وقوع گسل دقت بیش­تری داشت. با توجه به نتایج، روش­های داده­کاوی به عنوان ابزاری مفید در مدیریت ریسک گسل معرفی گردید. همچنین لزوم توجه به متغیرهای محیطی به­ویژه توپوگرافی در مراحل مدیریت و تغییر کاربری حوضه ضرورت می­یابد.

کلیدواژه‌ها

عنوان مقاله [English]

Evaluation of data mining and bivariate statistical methods in risk zoning of fault occurrence (Case study: Qara-Qum watershed)

نویسندگان [English]

  • Mehdi Bashiri 1
  • Seyedeh Maedeh Kavosi Davodi 2
  • Ali Afzali 3

1 Assistant Professor, Department of Nature Engineering and Medicinal Plants, Faculty of Agriculture and Natural Resources, University of Torbat Heydarieh, Razavi-Khorasan, Iran

2 M.Sc. in Watershed Management, Samane-Ab-Koomesh Company, Razavi-Khorasan, Iran

3 M.Sc. in Watershed Management, Department of Management of Desert Regions, International Desert Research Center, Tehran University, Iran

چکیده [English]

Fault is one of the main processes in tectonics which has relationship with phenomena such as earthquake. Therefore, awareness of the effect of environmental factors on fault occurrence and recognition of high risk areas is very important, that these goals in the present study have been investigated. The research area is Qara-Qum where after determining its faults, altitude, slope, direction, climate, land use, geomorphology, geology, erosion, precipitation, flood zones, pedology, stream power, topographic ruggedness index, distance from road, waterway, anticline and syncline were extracted. Then, data mining algorithms including of decision tree, random forest, cumulative, backing machine, logistic regression and neural network in R software are used to identify the value of variables and bivariate statistical methods including of information value and area density for identification of the values for each variable class fitted in fault occurrence. The accuracy of classification algorithms with ROC curve showed that based on input variables, random forest and support vector machine algorithms with 88% and 86% area under a curve had the best performance in classifying fault occurrence, respectively. Finally, according to the Gini coefficients in random forest algorithm, the zoning maps obtained by combining this algorithm were prepared and validated by bivariate statistical methods. According to this algorithm, the height, pedology and topographic ruggedness index variables, identified as the most important parameters in the fault occurrence respectively. Based on the zoning maps evaluation, information value and area density methods, around 52 and 35 percent of faults placed in very high risk class respectively. Therefore, the information value method was more accurate in identifying fault-sensitive zones. According to the results, data mining methods were introduced as a useful tool in fault risk management. It is also necessary to pay attention to environmental variables, especially topography, during the basin management and land use change stages.

کلیدواژه‌ها [English]

  • Density Area
  • Information Value
  • ROC Curve
  • Classification Algorithm
-  Anbalagan, R., 1992. Landslide hazard development and zonation mapping in mountainous Terrain. Journal of Engineering Geology, 32: 269-277.
-  Andre, S. and Norman, K., 2010. Combining random forests and object-oriented analysis for landslide mapping from very high resolution imagery. Procedia Environmental Sciences, 3: 123-129.
-  Ayalew, L., Ymagishi, H., Marui, H. and Kanno, T., 2005. GIS-based susceptibility mapping with comparisons of result from methods and verifications. Journal of Engineering Geology, 81: 432-445.
-  Ebrahimi, P., Eslah, M. and Azarakhshi, M., 2017. Landslide hazard zonation using SMCE method and AHP technic (case study: Hafshejan watershed, Chaharmahal-O-Bakhtiari). Journal of Range and Watershed Management, 70(1): 1-17.
-  Esfandiyari-Darabadi, F. and Beheshti Javid, E., 2016. Landslides susceptibility zoning using bayes' theorem-ANP hybrid model (case study: Heyran defile). Journal of Hydrogeomorphology, 2(8): 93-111.
-  Fathizad, H., Safari, A., Bazgir, M. and Khosravi, G. H., 2017. Evaluation of SVM with Kernel method (linear, polynomial, and radial basis) and neural network for land use classification. Iranian Journal of Range and Desert Research, 23 (4): 729-743.
-  Fawcett, T., 2006. An introduction to ROC analysis, Pattern Recognition Letters, 27: 861–874
-  Fayyad, U., Piatetsky-Shapiro, G. and Smyth, P., 1996. From data mining to knowledge discovery in databases. AI magazine, 17(3): 37-54.
-  Ghazanfari, M., Alizadeh, S. and Teimourpour, B., 2008. Data Mining and knowledge discovery. University of Science & Technology Publication, Tehran, 403p.
-  Gohardoust, A., Sadoddin, A., Ownegh, M. and Najafinejad, A., 2017. Identification of hazard areas by using land use planning (Case Study: Chehelchai Minodasht Watershed-Golestan Province). Iranian Journal of Range and Desert Research, 24 (3): 524-536.
-  Hur, J. H., Ihm, S. Y. and Park, Y. H., 2017. A Variable Impacts Measurement in random forest for mobile cloud computing. Wireless Communications and Mobile Computing, 2017: 1-13.
-  Ilanlou, M., Moghimi, E. and Servati, M. R., 2009. Mass movement hazard zonation with analyzing hierarchy process (AHP) method (case study: Karaj dam basin). Journal of Physical Geography, 2(5): 85-95.
-  Jamshidi, M., Eftekhari, K., Navidi, M. N. and Momeni, A., 2015. 40 years of pedological studies in soil and water research institute. Agricultural Research, Education and Extension Organization, Tehran, 60p.
-  Jedari-Eyvazi, J. and Mahmoudi, F., 2001. Dynamic Geomorphology. Payam-Noor University Publication, Tehran, 326p.
-  Komac, M., 2006. A landslide susceptibility model using the analytical hierarchy process method and multivariate statistics in perialpine Slovenia. Geomorphology, 74: 17-28.
-  Lee, S., 2004, Application of likelihood ratio and logistic regression models to landslide susceptibility mapping using GIS. Journal of Environmental Management, 34: 223-232.
-  Lee, S. and Sambath, T., 2006. Landslide susceptibility mapping in the Damrei Romel area, Cambodia using frequency ratio and logistic regression models. Journal of Environmental Geology, 50(6): 847-855.
-  Luca, F., Conforti, M. and Robustelli, G., 2011. Comparison of GIS-based gullying susceptibility mapping using bivariate and multivariate statistics: Northern Calabria, South Italy. Journal of Geomorphology, 134(3): 297-308.
-  Mahdavi, R., Alievazi, A., Gholami, H. and Kamali, A., 2017. Identifying the sediment source zones using maximum likelihood, minimum distance and parallelepiped algorithms (Case Study: South Roudbar, Kerman). Iranian Journal of Range and Desert Research, 24 (3): 610-622.
-  Mininstry of Energy., 2015. Iran Water Statistical Yearbook 2011-2012. Water and Waste Water Macro Planning Bureau, Tehran, 283p.
-  Mohammadzadeh, K., Bahmani, S. and Fathi, M. H., 2017. Logistic regression assessment in the investigation of the landslide potential (case study: from Nasirabad to Sattar khan dam). Hydrogeomorphology, 3(11): 127-148.
-  Naderi, F. and Karimi, H., 2011. Efficiency assessment of two Information Value and Gopta-Joshi methods in landslide hazard mapping in the Talkhab watershed of Ilam. Journal of Watershed Management Research (Pajouhesh & Sazandegi), 92: 95-103.
-  Naderi, F., Naseri, B., Karimi, H. and Habibi Bibalani, G. H., 2010. Efficiency evaluation of different landslide susceptibility mapping methods (Case study: Zangvan watershed, Ilam province). First international conference of soil and roots engineering relationship (LANDCON1005). Iran, Ardebil Province, 24-26 May: 1-8.
-  -Papadakis, M. and Karimalis, A., 2017. Producing a landslide susceptibility map through the use of analytic hierarchical process in Finikas watershed, North Peloponnese, Greece. American Journal of Geographic Information System, 6(1): 14-22.
-  Parmar, M. K., Malik, A. and Godiyal. M., 2012. Landslide hazard zonation using remote sensing and GIS: a case study of Giri valley. District Sirmaur Himachal Pradesh. International Journal of Environmental Sciences, 1: 26-39.
-  Pourghasemi, H. R., Moradi, H. R., Fatemi Aghda, S. M., Gokceoglu, C. and Pradhan, B., 2014. GIS-based landslide susceptibility mapping with probabilistic likelihood ratio and spatial multi-criteria evaluation models (North of Tehran, Iran). Arabian Journal of Geosciences, 7(5): 1857-1878.
-  Pourhashemi, S., Amirahmadi, A. and Akbari, E., 2014. Among bivariate method selection for landslide hazard zonation in GIS (case study: Baghi basin). Arid Regions Geographic Studies, 4(15): 71-89.
 
-  Riley, S. J., De-Gloria, S. D. and Elliot, R., 1999. A terrain ruggedness that quantifies topographic heterogeneity. Intermountain Journal of Science, 5(1-4): 23-27.
-  Sarp, G., 2014. Evolution of neotectonic activity of East Anatolian Fault System (EAFS) in bingol Pull-aport basin based on fractal dimension and morphometric. Journal of Asian Earth Sciences, 88: 168-177.
-  Shadfar, S., Yamani, M. and Namaki, M., 2011. Landslide hazard zonation using information value, area density and LNRF models in Chalkrood catchment. Journal of Watershed Engineering and Management, 3(1): 40-47.
Wilson, J. P. and Gallant, J. C., 2000. Digital terrain analysis. Terrain analysis: Principles and applications, 6(12): 1-27.