مقایسه قابلیت کاربرد دو روش رگرسیون لجستیک و شبکه عصبی در پهنه بندی حساسیت آتش سوزی عرصه های جنگلی و مرتعی استان مازندران

نوع مقاله : مقاله پژوهشی

نویسندگان

1 گروه مهندسی نقشه برداری، دانشگاه آزاد اسلامی، واحد ممقان

2 استادیار دانشگاه حکیم سبزواری

3 گروه محیط زیست، دانشگاه حکیم سبزواری

چکیده

آتش‌سوزی در عرصه‌های طبیعی یکی از عوامل کاهش سطح جنگل ها و مراتع ایران است. در این پژوهش، حساسیت آتش‌سوزی بر روی عرصه‌های جنگلی و مرتعی استان مازندران با استفاده از روش‌های داده مبنا مورد تجزیه‌وتحلیل قرار گرفت. از اینرو، 14 متغیر مستقل محیطی جهت تعیین پاسخ به نقاط فعال آتش‌سوزی سنجنده مادیس استفاده گردید. این متغیرها در روش‌های داده مبنای رگرسیون لجستیک و شبکه‌های عصبی که روش مؤثری برای استخراج خودکار اهمیت عوامل بر روی حساسیت آتش‌سوزی است، استفاده شد. نتایج نمودار منحنی تشخیص عملکرد نسبی برای داده های اعتبارسنجی نشان داد که هر دو روش از دقت بالایی (بیش‌تر از 89 درصد) در تشخیص نقاط فعال حریق سنجنده مادیس برخوردار هستند که روش شبکه عصبی از تشخیص بالاتری با 88 درصد برای نمایش مناطق با حساسیت بالا نسبت به روش رگرسیون لجستیک با حدود 85 درصد برخوردار است. ضریب همبستگی بین دو روش نشان داد که 97/0 پهنه‌های حساسیت در دو روش نسبت به هم یکسان هستند. 6/21 درصد از مساحت کل استان مازندران در پهنه های با حساسیت بالا و بسیار بالای آتش سوزی جنگل و مراتع قرار دارد. نقشه‌ی حساسیت آتش‌سوزی جنگل‌ها و مراتع ارائه‌شده در این تحقیق می‌تواند به‌عنوان نقشه اساسی برنامه راهبردی در استان مازندران مورداستفاده قرار گیرد تا در ارزیابی برنامه‌های آسیب‌پذیری و برنامه‌ریزی برای تقلیل این آسیب‌ها مورداستفاده قرار بگیرد.

کلیدواژه‌ها

عنوان مقاله [English]

Comparison of logistic regression and neural network methods in fire susceptibility of forest and rangelands, Mazandaran province

نویسندگان [English]

  • Kianoush Vidamanesh 1
  • Azadeh Atabati 3

1 Islamic Azad University، Mamaghan Branch

2

3 Environmental Science, Hakim Sabzevari University

چکیده [English]

Fires in natural areas are one of the factors decreasing forested area of northern Iran. In this study, forest and rangelands susceptibility to fire were analyzed using data-driven methods over Mazandaran Province. Fourteen important environmental and anthropogenic parameters influencing forest and rangelands susceptibility to fire were used to model probability of fire susceptibility. Binary logistic regression and artificial neural network, as two well-known data driven methods was then used to evaluate environmental and anthropogenic performance on landfire and map of forest fire susceptibility estimates were prepared in GIS environment. The area under the successive rate curve (AUSC) showed that ANN method modeled forest fire susceptibility with an accuracy of around 88% and BLR with 85%. 21.6% of the total area of Mazandaran province is located in areas with high and very high susceptibility levels of forest and rangeland fire. Overall, ANN method showed promising results to estimate landfire susceptibility. The forestry and rangelands fire susceptibility map presented in this study can be used as a basic map of the strategic planing in Mazandaran province to reduce probability fire damages.

کلیدواژه‌ها [English]

  • Fire susceptibility
  • Binary logistic regression
  • Artificial Neural Network
  • Geographic Information System
  • Mazandaran province
Adab, H., Amirahmadi, A., & Atabati, A. (2014). Relating Vegetation Cover with Land Surface Temperature and Surface Albedo in Warm Period of Year Using MODIS Imagery in North of Iran. Physical Geography Research Quarterly, 46(4), 419-434.  (In Farsi)
Ahmadian, S.H. (2002). Using Remote Sensing (RS) and Geographic Information System (GIS) in Land Use Rectification in Mazandaran Province, National Conference on Land Management - Soil Eradication and Sustainable Development, Arak, Iran Agricultural Research, Education and Extension Organization (ARREO).COI: FMSE01_058
Ali, A. A., O. Blarquez, M. P. Girardin, C. Hély, F. Tinquaut, A. El Guellab, V. Valsecchi, A. Terrier, L. Bremond and A. Genries (2012). "Control of the multimillennial wildfire size in boreal North America by spring climatic conditions." Proceedings of the National Academy of Sciences 109(51): 20966-20970.
Altman، D. G. (1990). Practical statistics for medical research: CRC press.
، M. G.، Aranha، J.، & Amraoui، M. (2014). Land cover fire proneness in Europe. 2014، 23(3)، 13.
Ardakani, A. S., Zoej, M. J. V., Mohammadzadeh, A., & Mansourian, A. (2011). Spatial and temporal analysis of fires detected by MODIS data in Northern Iran from 2001 to 2008. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 4(1), 216-225.
Baker، L.، & Ellison، D. (2008). Optimisation of pedotransfer functions using an artificial neural network ensemble method. Geoderma، 144(1–2)، 212-224.
Banejad, H., Mohebzadeh, H., & Olyaie, E. (2013). Applying ANN and GIS for estimation of effective parameters in determination of plant pattern (Case Study: Nahavand City). Journal of Environmental Science and Technology, 15(1), 23-35. (In Farsi)
Barros, A. M. and J. M. Pereira (2014). "Wildfire selectivity for land cover type: does size matter?" PloS one 9(1): e84760.
Bartsch، A.، Balzter، H.، & George، C. (2009). The influence of regional surface soil moisture anomalies on forest fires in Siberia observed from satellites. Environmental Research Letters، 4(4)، 045021.
Bhople، A. D.، & Tijare، P. (2012). Fast Fourier Transform Based Classification of Epileptic Seizure Using Artificial Neural Network. International Journal of Advanced Research in Computer Science and Software Engineering، 2(4).
Boger, Z. and H. Guterman (1997). Knowledge extraction from artificial neural network models. Systems, Man, and Cybernetics, 1997. Computational Cybernetics and Simulation., 1997 IEEE International Conference on, IEEE. Orlando, FL, USA.
Bond, W. J., Woodward, F. I., & Midgley, G. F. (2005). The global distribution of ecosystems in a world without fire. New Phytologist, 165(2), 525-538.
Buchanan، B.، Fleming، M.، Schneider، R.، Richards، B.، Archibald، J.، Qiu، Z.، & Walter، M. (2014). Evaluating topographic wetness indices across central New York agricultural landscapes. Hydrology and Earth System Sciences، 18(8)، 3279.
Bunruamkaew، K.، & Murayam، Y. (2011). Site Suitability Evaluation for Ecotourism Using GIS & AHP: A Case Study of Surat Thani Province، Thailand. Procedia - Social and Behavioral Sciences، 21، 269-278.
Chuvieco، E.، & Congalton، R. G. (1989). Application of remote sensing and geographic information systems to forest fire hazard mapping. Remote Sensing of Environment، 29(2)، 147-159. 
Chuvieco، E.، & Salas، J. (1996). Mapping the spatial distribution of forest fire danger using GIS. International Journal of Geographical Information Science، 10(3)، 333-345.
Conrad، O.، Bechtel، B.، Bock، M.، Dietrich، H.، Fischer، E.، Gerlitz، L.، . . . Böhner، J. (2015). System for Automated Geoscientific Analyses (SAGA) v. 2.1. 4. Geoscientific Model Development Discussions، 8(2)، 2271-2312.
Davis, E. J., Abrams, J., Wollstein, K., & Meacham, J. E. (2017). Rangeland fire protection associations: an alternative model for wildfire response. Ecosystem Workforce Program. Oregon, Institute for a Sustainable Environment: 20.
Debano، L. F.، Savage، S. M.، & Hamilton، D. A. (1976). The Transfer of Heat and Hydrophobic Substances During Burning1. Soil Science Society of America Journal، 40، 779-782.
Dong، X.، Li-min، D.، Guo-fan، S.، Lei، T.، & Hui، W. (2005). Forest fire risk zone mapping from satellite images and GIS for Baihe Forestry Bureau، Jilin، China. Journal of Forestry Research، 16(3)، 169-174.
Duguy, B., J. A. Alloza, M. J. Baeza, J. De la Riva, M. Echeverría, P. Ibarra, J. Llovet, F. P. Cabello, P. Rovira and R. V. Vallejo (2012). "Modelling the ecological vulnerability to forest fires in Mediterranean ecosystems using geographic information technologies." Environmental management 50(6): 1012-1026.
Eskandari, S. (2015). Evaluation of forest fire risk potential using Dong model, case study: District Three of Neka-Zalemroud forests. Geographical Planning of Space, 5(15), 195-210. (In Farsi)
Feinstein، A. R.، & Cicchetti، D. V. (1990). High agreement but low kappa: I. The problems of two paradoxes. Journal of clinical epidemiology، 43(6)، 543-549.
Gelman, A., Jakulin, A., Pittau, M. G., & Su, Y.-S. (2008). A weakly informative default prior distribution for logistic and other regression models. The Annals of Applied Statistics, 1360-1383.
Giglio, L., Descloitres, J., Justice, C. O., & Kaufman, Y. J. (2003). An enhanced contextual fire detection algorithm for MODIS. Remote sensing of environment, 87(2), 273-282.
Giglio، L.، Schroeder، W.، & Justice، C. O. (2016). The collection 6 MODIS active fire detection algorithm and fire products. Remote Sensing of Environment، 178، 31-41.
Gonzalez-Olabarria، J. R.، Brotons، L.، Gritten، D.، Tudela، A.، & Teres، J. A. (2012). Identifying location and causality of fire ignition hotspots in a Mediterranean region. International Journal of Wildland Fire، 21(7)، 905-914.
Güngöroğlu، C. (2017). Determination of forest fire risk with fuzzy analytic hierarchy process and its mapping with the application of GIS: The case of Turkey/Çakırlar. Human and Ecological Risk Assessment: An International Journal، 23(2)، 388-406.
He, C., S.-y. ZHANG, F. CHEN and Y. SUN (2013). "Forest fire division by using MODIS data based on the temporal-spatial variation law." Spectroscopy and Spectral Analysis 33(9): 2472-2477.
Heydari, M., Poorbabaei, H., & Esmaaelzade, O. (2015). Indicator plant species in monitoring forest soil conditions using logistic regression model in Zagros Oak (Quercus brantii var.persica) forest ecosystems, Ilam city. Journal of Plant Researches, 27(5), 811-828. (In Farsi)
Hijmans، R. J.، Cameron، S. E.، Parra، J. L.، Jones، P. G.، & Jarvis، A. (2005). Very high resolution interpolated climate surfaces for global land areas. International journal of climatology، 25(15)، 1965-1978.
Jaiswal، R. K.، Mukherjee، S.، Raju، K. D.، & Saxena، R. (2002). Forest fire risk zone mapping from satellite imagery and GIS. International Journal of Applied Earth Observation and Geoinformation، 4(1)، 1-10.
Jamshidi Bakhtar, A., Marvie Mohadjer, M. R., Sagheb Talebi, K., Namiranian, M., & Maroufi, H. (2013). Alteration of plant diversity after fire in Zagros forest stands, case study: Marivan forests. Iranian Journal of Forest and Poplar Research, 21(3), 529-541. (In Farsi)
Keane, R. E., J. K. Agee, P. Fulé, J. E. Keeley, C. Key, S. G. Kitchen, R. Miller and L. A. Schulte (2009). "Ecological effects of large fires on US landscapes: benefit or catastrophe? A." International Journal of Wildland Fire 17(6): 696-712.
Le, L., Lin, Q., & Wang, Y. (2017). Landslide susceptibility mapping on a global scale using the method of logistic regression. Natural Hazards and Earth System Sciences, 17(8), 1411.
Lee، B.، Kim، S. Y.، Chung، J.، & Park، P. S. (2008). Estimation of fire severity by use of Landsat TM images and its relevance to vegetation and topography in the 2000 Samcheok forest fire. Journal of Forest Research، 13(4)، 197-204.
MacMillan، R.، Shary، P.، Hengl، T.، & Reuter، H. (2008). Geomorphometry: concepts، software، applications (Vol. 33، pp. 772). Amsterdam، The Netherlands: Elsevier Science. Chap. Landforms and Landforms elements in geomorphometry.
Maeda, E. E., Formaggio, A. R., Shimabukuro, Y. E., Arcoverde, G. F. B., & Hansen, M. C. (2009). Predicting forest fire in the Brazilian Amazon using MODIS imagery and artificial neural networks. International Journal of Applied Earth Observation and Geoinformation, 11(4), 265-272.
Maroco، J.، Silva، D.، Rodrigues، A.، Guerreiro، M.، Santana، I.، & de Mendonça، A. (2011). Data mining methods in the prediction of Dementia: A real-data comparison of the accuracy، sensitivity and specificity of linear discriminant analysis، logistic regression، neural networks، support vector machines، classification trees and random forests. BMC Research Notes، 4(1)، 299.
Mirzayi M, Riyahi Bakhtiyari A, Salman Mahini A, Gholamalifard M. Investigating the Land Cover Changes in Mazandaran Province Using Landscape Ecology’s Metrics Between 1984 - 2010. ijae. 2013; 2 (4) :37-55
Moretti، M.، Obrist، M. K.، & Duelli، P. (2004). Arthropod biodiversity after forest fires: winners and losers in the winter fire regime of the southern Alps. Ecography، 27(2)، 173-186.
Moritz، M. A.، Parisien، M.-A.، Batllori، E.، Krawchuk، M. A.، Van Dorn، J.، Ganz، D. J.، & Hayhoe، K. (2012). Climate change and disruptions to global fire activity. Ecosphere، 3(6)، 1-22.
Mukti، A.، Prasetyo، L. B.، & Rushayati، S. B. (2016). Mapping of Fire Vulnerability in Alas Purwo National Park. Procedia Environmental Sciences، 33، 290-304.
Nakas، C. T.، Dalrymple‐Alford، J. C.، Anderson، T. J.، & Alonzo، T. A. (2013). Generalization of Youden index for multiple‐class classification problems applied to the assessment of externally validated cognition in Parkinson disease screening. Statistics in Medicine، 32(6)، 995-1003.
Neshat, M., Yaghubi, M., Naghibi, M. B., & Esmailzadeh, A. (2008). Designing an expert system of liver disorders by using neural network Paper presented at the Eleventh Student Conference on Electrical Engineering, Zanjan. 1-8. (In Farsi)
Papanastasis, V. P., Bautista, S., Chouvardas, D., Mantzanas, K., Papadimitriou, M., Mayor, A. G., . . . Vallejo, R. V. (2017). Comparative assessment of goods and services provided by grazing regulation and reforestation in degraded Mediterranean rangelands. Land Degradation & Development, 28(4), 1178-1187.
Pereira, M. G., Aranha, J., & Amraoui, M. (2014). Land cover fire proneness in Europe. 2014, 23(3), 13.
Pezzatti، G. B.، Bajocco، S.، Torriani، D.، & Conedera، M. (2009). Selective burning of forest vegetation in Canton Ticino (southern Switzerland). Plant Biosystems - An International Journal Dealing with all Aspects of Plant Biology، 143(3)، 609-620.
Qin، X.، Yan، H.، Zhan، Z.، & Li، Z. (2014). Characterising vegetative biomass burning in China using MODIS data. International Journal of Wildland Fire، 23(1)، 69-77.
Rafiee, F., Ejtahadi, H., & Jankju, M. (2015). Plant diversity at different interval times after burnings in a semiarid rangeland. Journal of Plant Researches, 27(5), 854-864. (In Farsi)
Ray، R. L. (2016). Moisture Stress Indicators in Giant Sequoia Groves in the Southern Sierra Nevada of California، USA. Vadose Zone Journal، 15(10).
Renard، Q.، Pélissier، R.، Ramesh، B. R.، & Kodandapani، N. (2012). Environmental susceptibility model for predicting forest fire occurrence in the Western Ghats of India. International Journal of Wildland Fire، 21(4)، 368-379.
Sall، J.، Lehman، A.، Stephens، M. L.، & Creighton، L. (2012). JMP start statistics: a guide to statistics and data analysis using JMP: SAS Institute.
Sarvaleh, F., Bavaghar, M., & Shabanian, N. (2014). Application of artificial neural network for forest fire risk mapping based on physiographic, human and climate factors in Sarvabad, Kurdistan province. Iranian Journal of Forest and Range Protection Research, 11(2). , 97-107. (In Farsi)
Schroeder، W.، Ruminski، M.، Csiszar، I.، Giglio، L.، Prins، E.، Schmidt، C.، & Morisette، J. (2008). Validation analyses of an operational fire monitoring product: The Hazard Mapping System. International Journal of Remote Sensing، 29(20)، 6059-6066.
Setiawan، I.، Mahmud، A. R.، Mansor، S.، Mohamed Shariff، A. R.، & Nuruddin، A. A. (2004). GIS‐grid‐based and multi‐criteria analysis for identifying and mapping peat swamp forest fire hazard in Pahang، Malaysia. Disaster Prevention and Management: An International Journal، 13(5)، 379-386.
Shafiei, A., Akbarinia, M., Azizi, P., & Rad, J. (2010). Impacts of fire on some chemical properties of forest soil in north of Iran (case study: Kheyroudkenar forest). Iranian Journal of Forest and Poplar Research, 18(3), 365-379. (In Farsi)
Shen, J. and S. Gao (2008). "A solution to separation and multicollinearity in multiple logistic regression." Journal of data science: JDS 6(4): 515.
Srivastava، P.، & Garg، A. (2013). Forest Fires in India: Regional and Temporal Analyses. Journal of Tropical Forest Science، 228-239.
Vadrevu، K. P.، Eaturu، A.، & Badarinath، K. V. (2010). Fire risk evaluation using multicriteria analysis—a case study. Environmental monitoring and assessment، 166(1)، 223-239.
Vasilakos، C.، Kalabokidis، K.، Hatzopoulos، J.، & Matsinos، I. (2009). Identifying wildland fire ignition factors through sensitivity analysis of a neural network. Natural hazards، 50(1)، 125-143.
Wang، S.، Zhou، Y.، Wang، L.، & Zhang، P. (2003). A research on fire automatic recognition using MODIS data. Paper presented at the Geoscience and Remote Sensing Symposium، 2003. IGARSS'03. Proceedings. 2003 IEEE International.
Yin، H.-w.، Kong، F.-h.، & Li، X.-z. (2004). RS and GIS-based forest fire risk zone mapping in Da Hinggan Mountains. Chinese geographical science، 14(3)، 251-257.
Zarekar, A., Zamani, B. K., Ghorbani, S., Moalla, M. A., & Jafari, H. (2013). Mapping spatial distribution of forest fire using MCDM and GIS (case study: three forest zones in Guilan Province). Iranian Journal of Forest and Poplar Research, 21(2), 218-230. (In Farsi)
Zevenbergen، L. W.، & Thorne، C. R. (1987). Quantitative analysis of land surface topography. Earth surface processes and landforms، 12(1)، 47-56.
Zhang، Q.-f.، & Chen، W.-j. (2007). Fire cycle of the Canada’s boreal region and its potential response to global change. Journal of Forestry Research، 18(1)، 55.
Zhu, W., N. Zeng and N. Wang (2010). "Sensitivity, specificity, accuracy, associated confidence interval and ROC analysis with practical SAS implementations." NESUG proceedings: health care and life sciences, Baltimore, Maryland 19.
Zumbrunnen، T.، Menéndez، P.، Bugmann، H.، Conedera، M.، Gimmi، U.، & Bürgi، M. (2012). Human impacts on fire occurrence: a case study of hundred years of forest fires in a dry alpine valley in Switzerland. Regional Environmental Change، 12(4)، 935-949.