Journals / Technological Applied Sciences / 2026 / Cilt: 9 Sayı: 2

FOREST FIRE BEHAVIOR PREDICTION WITH MULTIVARIATE REGRESSION

ORMAN YANGIN DAVRANIŞININ ÇOK DEĞİŞKENLİ REGRESYONLA TAHMİN EDİLMESİ

Pages
13–21
Publication Date
2014-03-01
🏛️Bibliographic metadata and full text access for this article are provided via DergiPark.
DergiPark Record ↗
✨

AI Summary & Key Takeaways

AI Insight
⚡ TL;DR(Too Long; Didn't Read)

One of these regression models that contains coefficients of independent variables multiplied together and its square was R2 with the value of 0.854 (p

📌 3 Key Takeaways

🎯
Objective & Hypothesis:A forest fire starts on the surface of forest and is quite difficult to extinguish when it spreads the hill.
📊
Key Findings & Data:In this study, 151 data set was obtained from 21 controlled grassland fire, and then four different regression models were tested with this data and forest fire behavior was tried to model.
💡
Conclusion & Impact:One of these regression models that contains coefficients of independent variables multiplied together and its square was R2 with the value of 0.854 (p
🔬Design:Modeling & Simulation
📍Scope:Comprehensive Sample
⚙️Variables:Çok Değişkenli · Yangın Davranış Modelleme · Yangın Davranış Tahmini

Abstract

A forest fire starts on the surface of forest and is quite difficult to extinguish when it spreads the hill. Therefore, it is very important to extinguish a forest fire and estimate its behavior before it jumps to hill of trees. In this study, 151 data set was obtained from 21 controlled grassland fire, and then four different regression models were tested with this data and forest fire behavior was tried to model. Hence, weed, litter and humus layer samples were taken from forest floor, and then they were dried in the oven. Additionally, mobile weather station was established in the region, and every stage of fire was measured and recorded on camera. Later in the MATLAB environment, four different regression models were developed with fire data. In all of the models developed, it was used the burned area as the dependent variable and weed, litter, humus, temperature, dew point, relative humidity, wind speed and time as independent variable. One of these regression models that contains coefficients of independent variables multiplied together and its square was R2 with the value of 0.854 (p

Özet

Bir orman yangını örtüde başlar ve tepeye sıçradığında ise söndürmek oldukça zorlaşır. Bu yüzden çıkan bir orman yangınının tepeye sıçramadan söndürülmesi ve yangın davranışının tahmin edilmesi oldukça önemlidir. Bu çalışmada ise çıkartılan 21 kontrollü örtü yangınından elde edilen 151 veri seti, dört farklı regresyon modeliyle denenip, orman yangın davranışı modellenmeye çalışılmıştır. Bunun için orman yüzeyindeki bulunan diri örtü, ölü örtü ve humus katmanlarından örnekler alınmış ve bu örnekler fırında kurutulmuştur. Bunun dışında yangın çıkartılacak bölgede mobil meteoroloji istasyonu kurulmuş ve yangının her aşaması ölçülüp, kamerayla kaydedilmiştir. Daha sonra matlab ortamında veriler üzerinde dört farklı regresyon modeli geliştirilmiştir. Geliştirilen modellerin tamamında bağımlı değişken olarak yanan alan, bağımsız değişken olarak ise diri örtü, ölü örtü, humus, sıcaklık, çiğ noktası, bağıl nem, rüzgar hızı ve zaman kullanılmıştır. Bu regresyon modellerinden bağımsız değişkenlerin katsayılarının birbiriyle çarpımını ve karelerini içeren eğri R2 0.854 (p

🔖

Cite This Article

Tuna, H., Erdem, A. (2014). FOREST FIRE BEHAVIOR PREDICTION WITH MULTIVARIATE REGRESSION. Technological Applied Sciences, Cilt: 9 Sayı: 2, 13–21. https://doi.org/10.12739/NWSA.2014.9.2.2A0085

📄 Full Text PDF Viewer

📄

PDF görüntüleyici hazırlanıyor...

References (9)

  1. Dickinson, M.B., Johnson, E.A., and Artiaga, R., (2013). Fire spread probabilities for experimental beds composed of mixedwood boreal forest fuels. Canadian Journal of Forest Research-Revue Canadienne De Recherche Forestiere, Volume: 43, Number: 4, pp: 321330, DOI: 10.1139/cjfr-2012-0291.
  2. Zhang, H., Han, X., and Dai, Sha., (2013). Fire Occurrence Probability Mapping of Northeast China With Binary Logistic Regression Model. Ieee Journal of Selected Topıcs in Applıed Earth Observatıons and Remote Sensıng, Volume: 6, Number: 1, pp: 127-127, DOI: 1109/JSTARS.2012.2236680.
  3. Bisquert, M., Caselles, E., Sanchez, J.M., and Caselles, V., (2013). Application of artificial neural networks and logistic regression to the prediction of forest fire danger in Galicia using MODIS data. International Journal of Wildland Fire, Volume: 21, Number: 8, pp: 1025-1029, DOI: 10.1071/WF11105.
  4. Bisquert, M.M., Sanchez, J.M., and Caselles, V., (2011). Fire danger estimation from MODIS Enhanced Vegetation Index data: application to Galicia region (north-west Spain). International Journal of Wildland Fire, Volume: 20, Number: 3, pp: 465-473, DOI: 1071/WF10002.
  5. Martinez-Fernandez, J., Chuvieco, E., and Koutsias, N., (2013). Modelling long-term fire occurrence factors in Spain by accounting for local variations with geographically weighted regression. Natural Hazards and Earth System Sciences, Volume: 13, Number: 2, pp: 311-327, DOI: 10.5194/nhess-13-311-2013. del Hoyo, L.V. and Isabel, M.P.M, (2011). Logistic regression models for human-caused wildfire risk estimation: analysing the effect of the spatial accuracy in fire occurrence data. European Journal of Forest Research, Volume: 130, Number: 6, pp: 983-996, DOI: 1007/s10342-011-0488-2.
  6. Cruz, M.G., McCaw, W.L., Anderson, W.R., and Gould, J.S., (2013). Fire behaviour modelling in semi-arid mallee-heath shrublands of southern Australia. Environmental Modelling and Software, Volume: 40, pp: 21-34, DOI:10.1016/j.envsoft.2012.07.003. Kucuk, O., Bilgili, E., Bulut, S., and Fernandes, P.M., (2012). Rates of Surface Fire Spread in A Young Calabrian Pine (Pinus Brutia Ten.) Plantation. Environmental Engineering and Management Journal, Volume: 11, Number:8, pp: 1475-1480.
  7. Cao, X., Cui, X.H.L., Yue, M., Chen, J., Tanikawa, H., and Ye, Y., (2013). Evaluation of wildfire propagation susceptibility in grasslands using burned areas and multivariate logistic regression. International Journal of Remote Sensing, Volume: 34, Number:19, pp: 6679-6700, DOI: 10.1080/01431161.2013.805280. Xue, Y., Liu, S.L., Zhang, L., and Hu, Y.M., (2013). Integrating fuzzy logic with piecewise linear regression for detecting vegetation greenness change in the Yukon River Basin,Alaska. International Journal of Remote Sensing, Volume: 34, Number:12, pp: 4242-4263, DOI: 10.1080/01431161.2013.775532. Woolley, T., Shaw, D.C.L., Ganio, L.M., and Fitzgerald, S., (2012). A review of logistic regression models used to predict post-fire tree mortality of western North American conifers. International Journal of Wildland Fire, Volume: 21, Number:1, pp: 1-35, DOI: 1071/WF09039.
  8. Weinberg, S.L. and Abramowitz, S.K., (2008). Statistics Using SPSS: An Integrative Approach. United Kingdom: Cambridge University Press.
  9. Gordon R.A., (2012). Applied Statistics for the Social and Health Sciences. Routledge.