Improvement of fire danger modelling with geographically weighted logistic model
Haijun Zhang A , Pengcheng Qi A B and Guangmeng Guo A
+ Author Affiliations
- Author Affiliations
A Remote Sensing Center, Department of Geography, Nanyang Normal University, Nanyang, 473061, China.
B Corresponding author. Email: pengchengqi_ny@126.com
International Journal of Wildland Fire 23(8) 1130-1146 https://doi.org/10.1071/WF13195
Submitted: 17 November 2013 Accepted: 19 June 2014 Published: 3 November 2014
Abstract
Global models dominate historical documents on fire danger modelling. However, local variations may exist in the relationships between fire presence and fire-influencing factors. In this study, 50 fire danger models (10 global logistic models and 40 geographically weighted logistic models, i.e. local models), were developed to model daily fire danger in Heilongjiang province in north-east China and cross-validation was performed to evaluate the predictive performance of the various developed models. In modelling, multi-temporal spatial sampling and repeated random sub-sampling were applied to obtain 10 groups of training sub-samples and inner testing sub-samples. For each of the 10 groups of training sub-samples, principal component analysis, in which muticollinearity among variables can be removed, was used to create nine principal components that were then employed as covariates to develop one global logistic model and four geographically weighted logistic models. Compared to global models, all local models showed better model fitting, less spatial autocorrelation of residuals and more desirable modelling of fire presence. In particular, not only was local spatial variation in fire–environment relationships accounted for in the adaptive Gaussian geographically weighted logistic models, but spatial autocorrelation of residuals was significantly reduced to acceptable levels, indicating strong inferential performance.
Additional keywords: fire danger, geographically weighted regression, Heilongjiang, local model, north-east China, spatial autocorrelation of residuals, spatial non-stationarity.
References
Akaike H (1973) Information theory and an extension of the maximum likelihood principle. In ‘Second International Symposium on Information Theory’, Akademiai Kiado, Budapest (Eds BN Petrov, F Csaki) pp. 267–281.Anselin L (1995) Local indicators of spatial association – LISA. Geographical Analysis 27, 93–115.
| Local indicators of spatial association – LISA.Crossref | GoogleScholarGoogle Scholar |
Anselin L, Griffith DA (1988) Do spatial effects really matter in regression analysis? Papers/Regional Science Association. Regional Science Association. Meeting 65, 11–34.
| Do spatial effects really matter in regression analysis?Crossref | GoogleScholarGoogle Scholar |
Atkinson PK, German SE, Sear DA, Clark MJ (2003) Exploring the relations between riverbank erosion and geomorphological controls using Geographically Weighted Logistic Regression. Geographical Analysis 35, 58–82.
| Exploring the relations between riverbank erosion and geomorphological controls using Geographically Weighted Logistic Regression.Crossref | GoogleScholarGoogle Scholar |
Augustin NH, Kublin E, Metzler B, Meierjohann E, von Wühlisch G (2005) Analyzing the spread of beech canker. Forest Science 51, 438–448.
Bisquert M, Caselles E, Sánchez JM, Caselles V (2012) 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 21, 1025–1029.
| Application of artificial neural networks and logistic regression to the prediction of forest fire danger in Galicia using MODIS data.Crossref | GoogleScholarGoogle Scholar |
Boschetti L, Roy DP, Justice CO, Giglio L (2010) Global assessment of the temporal reporting accuracy and precision of the MODIS burned area product. International Journal of Wildland Fire 19, 705–709.
| Global assessment of the temporal reporting accuracy and precision of the MODIS burned area product.Crossref | GoogleScholarGoogle Scholar | 1:CAS:528:DC%2BC3cXhtFOkur%2FO&md5=5574debd0f355dc43d18ac8cb4aed9bcCAS |
Brillinger DR, Preisler HK, Benoit JW (2003) Risk assessment: a forest fire example. Statistical Science: A Festschrift for Terry Speed 40, 177–196.
Brillinger DR, Preisler HK, Benoit JW (2006) Probabilistic risk assessment for wildfires. Environmetrics 17, 623–633.
| Probabilistic risk assessment for wildfires.Crossref | GoogleScholarGoogle Scholar |
Brooks ML (1999) Alien annual grasses and fire in the Mojave Desert. Madrono 46, 13–19.
Catry FX, Rego FC, Bação FL, Moreira F (2009) Modeling and mapping wildfire ignition risk in Portugal. International Journal of Wildland Fire 18, 921–931.
| Modeling and mapping wildfire ignition risk in Portugal.Crossref | GoogleScholarGoogle Scholar |
Ceccato P, Gobron N, Flasse S, Pinty B, Tarantola S (2002) Designing a spectral index to estimate vegetation water content from remote sensing data: part 1– theoretical approach. Remote Sensing of Environment 82, 188–197.
| Designing a spectral index to estimate vegetation water content from remote sensing data: part 1– theoretical approach.Crossref | GoogleScholarGoogle Scholar |
Chuvieco E, Aguado I, Yebra M, Nieto H, Salas J, Martín MP, Vilar L, Martínez J, Martín S, Ibarra P, de la Riva J, Baeza J, Rodríguez F, Molina JR, Herrera MA, Zamora R (2010) Development of a framework for fire risk assessment using remote sensing and geographic information system technologies. Ecological Modelling 221, 46–58.
| Development of a framework for fire risk assessment using remote sensing and geographic information system technologies.Crossref | GoogleScholarGoogle Scholar |
Crase B, Liedloff AC, Wintle BA (2012) A new method for dealing with residual spatial autocorrelation in species distribution models. Ecography 35, 879–888.
| A new method for dealing with residual spatial autocorrelation in species distribution models.Crossref | GoogleScholarGoogle Scholar |
Deng O, Li YQ, Feng ZK, Zhang DY (2012) Model and zoning of forest fire risk in Heilongjiang province based on spatial Logistic. Transactions of the Chinese Society of Agricultural Engineering 28, 200–205. . [In Chinese]
Di XY, Li YF, Sun J, Yang G (2011) Dynamics of forest fire weather indices in Tahe county of Great Xing’ an Mountains region, Heilongjiang province. Chinese Journal of Applied Ecology 22, 1240–1246. . [In Chinese]
Dickson BG, Prather JW, Xu YG, Hampton HM, Aumack EN, Sisk TD (2006) Mapping the probability of large fire occurrence in northern Arizona, USA. Landscape Ecology 21, 747–761.
| Mapping the probability of large fire occurrence in northern Arizona, USA.Crossref | GoogleScholarGoogle Scholar |
Dilts TE, Sibold JS, Biondi F (2009) A Weights-of-Evidence model for mapping the probability of fire occurrence in Lincoln county, Nevada. Annals of the Association of American Geographers. Association of American Geographers 99, 712–727.
| A Weights-of-Evidence model for mapping the probability of fire occurrence in Lincoln county, Nevada.Crossref | GoogleScholarGoogle Scholar |
Dlamini WM (2011) Application of Bayesian networks for fire risk mapping using GIS and remote sensing data. GeoJournal 76, 283–296.
| Application of Bayesian networks for fire risk mapping using GIS and remote sensing data.Crossref | GoogleScholarGoogle Scholar |
Dormann CF, McPherson JM, Araújo MB, Bivand R, Bolliger J, Carl G, Davies RG, Hirzel A, Jetz W, Kissling WD, Kühn I, Ohlemüller R, Peres-Neto PR, Reineking B, Schröder B, Schurr FM, Wilson R (2007) Methods to account for spatial autocorrelation in the analysis of species distributional data: a review. Ecography 30, 609–628.
| Methods to account for spatial autocorrelation in the analysis of species distributional data: a review.Crossref | GoogleScholarGoogle Scholar |
Foody GM (2003) Geographical weighting as a further refinement to regression modelling: an example focused on the NDVI–rainfall relationship. Remote Sensing of Environment 88, 283–293.
| Geographical weighting as a further refinement to regression modelling: an example focused on the NDVI–rainfall relationship.Crossref | GoogleScholarGoogle Scholar |
Fotheringham AS (2009) ‘The problem of spatial autocorrelation’ and local spatial statistics. Geographical Analysis 41, 398–403.
| ‘The problem of spatial autocorrelation’ and local spatial statistics.Crossref | GoogleScholarGoogle Scholar |
Fotheringham AS, Brunsdon C, Charlton M (2002) ‘Geographically Weighted Regression: the Analysis of Spatially Varying Relationships.’ (Wiley: New York)
García CV, Woodard PM, Titus SJ, Adamowicz WL, Lee BS (1995) A logit model for predicting the daily occurrence of human caused forest fires. International Journal of Wildland Fire 5, 101–111.
| A logit model for predicting the daily occurrence of human caused forest fires.Crossref | GoogleScholarGoogle Scholar |
Garrigues S, Allard D, Baret F, Weiss M (2006) Quantifying spatial heterogeneity at the landscape scale using variogram models. Remote Sensing of Environment 103, 81–96.
| Quantifying spatial heterogeneity at the landscape scale using variogram models.Crossref | GoogleScholarGoogle Scholar |
Giglio L, Descloitres J, Justice CO, Kaufman YJ (2003) An enhanced contextual fire detection algorithm for MODIS. Remote Sensing of Environment 87, 273–282.
| An enhanced contextual fire detection algorithm for MODIS.Crossref | GoogleScholarGoogle Scholar |
Guangmeng G, Mei Z (2004) Using MODIS land surface temperature to evaluate forest fire risk of northeast China. IEEE Geoscience and Remote Sensing Letters 1, 98–100.
| Using MODIS land surface temperature to evaluate forest fire risk of northeast China.Crossref | GoogleScholarGoogle Scholar |
Guo L, Ma ZH, Zhang LJ (2008) Comparison of bandwidth selection in application of geographically weighted regression: a case study. Canadian Journal of Forest Research 38, 2526–2534.
| Comparison of bandwidth selection in application of geographically weighted regression: a case study.Crossref | GoogleScholarGoogle Scholar |
Hernandez-Leal PA, Gonzalez-Calvo A, Arbelo M, Barreto A, Alonso-Benito A (2008) Synergy of GIS and remote sensing data in forest fire danger modeling. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing 1, 240–247.
| Synergy of GIS and remote sensing data in forest fire danger modeling.Crossref | GoogleScholarGoogle Scholar |
Hu HQ, Liu YC, Jiao Y (2007) Estimation of the carbon storage of forest vegetation and carbon emission from forest fires in Heilongjiang Province, China. Journal of Forest Research 18, 17–22.
| Estimation of the carbon storage of forest vegetation and carbon emission from forest fires in Heilongjiang Province, China.Crossref | GoogleScholarGoogle Scholar | 1:CAS:528:DC%2BD2sXkt12gu7Y%3D&md5=5cbbf5d3e8d05b33217b06ec24c48769CAS |
Huete A, Didan K, Miura T, Rodriguez EP, Gao X, Ferreira LG (2002) Overview of the radiometric and biophysical performance of the MODIS vegetation indices. Remote Sensing of Environment 83, 195–213.
| Overview of the radiometric and biophysical performance of the MODIS vegetation indices.Crossref | GoogleScholarGoogle Scholar |
Huh D, Flaherty BP, Simoni JM (2012) Optimizing the analysis of adherence interventions using logistic generalized estimating equations. AIDS and Behavior 16, 422–431.
| Optimizing the analysis of adherence interventions using logistic generalized estimating equations.Crossref | GoogleScholarGoogle Scholar | 21553253PubMed |
Hurvich CM, Simonoff JS, Tsai C (1998) Smoothing parameter selection in nonparametric regression using an improved Akaike information criterion. Journal of the Royal Statistical Society. Series B, Statistical Methodology 60, 271–293.
| Smoothing parameter selection in nonparametric regression using an improved Akaike information criterion.Crossref | GoogleScholarGoogle Scholar |
Jung J, Kim C, Jayakumar S, Kim S, Han S, Kim DH, Heo J (2013) Forest fire risk mapping of Kolli Hills, India, considering subjectivity and inconsistency issues. Natural Hazards 65, 2129–2146.
| Forest fire risk mapping of Kolli Hills, India, considering subjectivity and inconsistency issues.Crossref | GoogleScholarGoogle Scholar |
Karegowda AG, Jayaram MA, Manjunath AS (2010) Combining Akaike’s information criterion (AIC) and the golden-section search technique to find optimal numbers of k-nearest neighbors. International Journal of Computers and Applications 2, 80–87.
| Combining Akaike’s information criterion (AIC) and the golden-section search technique to find optimal numbers of k-nearest neighbors.Crossref | GoogleScholarGoogle Scholar |
Kühn I (2007) Incorporating spatial autocorrelation may invert observed patterns. Diversity & Distributions 13, 66–69.
Lasaponara R (2006) On the use of principal component analysis (PCA) for evaluating interannual anomalies from SPOT/VEGETATION NDVI temporal series. Ecological Modelling 194, 429–434.
| On the use of principal component analysis (PCA) for evaluating interannual anomalies from SPOT/VEGETATION NDVI temporal series.Crossref | GoogleScholarGoogle Scholar |
Legendre P (1993) Spatial autocorrelation: trouble or new paradigm? Ecology 74, 1659–1673.
| Spatial autocorrelation: trouble or new paradigm?Crossref | GoogleScholarGoogle Scholar |
Lei XL, Zhou GS, Jia BR, Li S (2012) Relationships of forest fire with lightning in Daxing’anling Mountains, Northeast China. Chinese Journal of Applied Ecology 23, 1743–1750. . [In Chinese]
Li LM, Song WG, Ma J, Satoh K (2009) Artificial neural network approach for modeling the impact of population density and weather parameters on forest fire risk. International Journal of Wildland Fire 18, 640–647.
| Artificial neural network approach for modeling the impact of population density and weather parameters on forest fire risk.Crossref | GoogleScholarGoogle Scholar |
Liu ZH, Yang J, Chang Y, Weisberg PJ, He HS (2012) Spatial patterns and drivers of fire occurrence and its future trend under climate change in a boreal forest of Northeast China. Global Change Biology 18, 2041–2056.
| Spatial patterns and drivers of fire occurrence and its future trend under climate change in a boreal forest of Northeast China.Crossref | GoogleScholarGoogle Scholar |
Liu ZH, Yang J, He HS (2013) Identifying the threshold of dominant controls on fire spread in a boreal forest landscape of Northeast China. PLoS ONE 8, 1–10.
Lloyd CD (2010) Nonstationary models for exploring and mapping monthly precipitation in the United Kingdom. International Journal of Climatology 30, 390–405.
Loboda TV (2009) Modeling fire danger in data-poor regions: a case study from the Russian Far East. International Journal of Wildland Fire 18, 19–35.
| Modeling fire danger in data-poor regions: a case study from the Russian Far East.Crossref | GoogleScholarGoogle Scholar |
Lozano FJ, Suárez-Seoane S, de Luis E (2007) Assessment of several spectral indices derived from multi-temporal Landsat data for fire occurrence probability modeling. Remote Sensing of Environment 107, 533–544.
| Assessment of several spectral indices derived from multi-temporal Landsat data for fire occurrence probability modeling.Crossref | GoogleScholarGoogle Scholar |
Lozano FJ, Suárez-Seoane S, Kelly M, Luis E (2008) A multi-scale approach for modeling fire occurrence probability using satellite data and classification trees: a case study in a mountainous Mediterranean region. Remote Sensing of Environment 112, 708–719.
| A multi-scale approach for modeling fire occurrence probability using satellite data and classification trees: a case study in a mountainous Mediterranean region.Crossref | GoogleScholarGoogle Scholar |
Martínez-Fernández J, Chuvieco E, 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 13, 311–327.
| Modelling long-term fire occurrence factors in Spain by accounting for local variations with geographically weighted regression.Crossref | GoogleScholarGoogle Scholar |
Mathew J, Jha VK, Rawat GS (2009) Landslide susceptibility zonation mapping and its validation in part of Garhwal Lesser Himalaya, India, using binary logistic regression analysis and receiver operating characteristic curve method. Landslides 6, 17–26.
| Landslide susceptibility zonation mapping and its validation in part of Garhwal Lesser Himalaya, India, using binary logistic regression analysis and receiver operating characteristic curve method.Crossref | GoogleScholarGoogle Scholar |
Miller JA (2012) Species distribution models: spatial autocorrelation and non-stationarity. Progress in Physical Geography 36, 681–692.
| Species distribution models: spatial autocorrelation and non-stationarity.Crossref | GoogleScholarGoogle Scholar |
Nieto H, Aguado I, García M, Chuvieco E (2012) Lightning-caused fires in Central Spain: development of a probability model of occurrence for two Spanish regions. Agricultural and Forest Meteorology 162–163, 35–43.
| Lightning-caused fires in Central Spain: development of a probability model of occurrence for two Spanish regions.Crossref | GoogleScholarGoogle Scholar |
Páez A, Scott DM (2004) Spatial statistics for urban analysis: a review of techniques with examples. GeoJournal 61, 53–67.
| Spatial statistics for urban analysis: a review of techniques with examples.Crossref | GoogleScholarGoogle Scholar |
Parisien MA, Snetsinger S, Greenberg JA, Nelson CR, Schoennagel T, Dobrowski SZ, Moritz MA (2012) Spatial variability in wildfire probability across the western United States. International Journal of Wildland Fire 21, 313–327.
| Spatial variability in wildfire probability across the western United States.Crossref | GoogleScholarGoogle Scholar |
Pearce J, Ferrier S (2000) Evaluating the predictive performance of habitat models developed using logistic regression. Ecological Modelling 133, 225–245.
| Evaluating the predictive performance of habitat models developed using logistic regression.Crossref | GoogleScholarGoogle Scholar |
Preisler HK, Brillinger DR, Burgan RE, Benoit JW (2004) Probability based models for estimation of wildfire risk. International Journal of Wildland Fire 13, 133–142.
| Probability based models for estimation of wildfire risk.Crossref | GoogleScholarGoogle Scholar |
Preisler HK, Westerling AL, Gebert KM, Munoz-Arriola F, Holmes TP (2011) Spatially explicit forecasts of large wildland fire probability and suppression costs for California. International Journal of Wildland Fire 20, 508–517.
| Spatially explicit forecasts of large wildland fire probability and suppression costs for California.Crossref | GoogleScholarGoogle Scholar |
Prestemon JP, Chas-Amil ML, Touza JM, Goodrick SL (2012) Forecasting intentional wildfires using temporal and spatiotemporal autocorrelations. International Journal of Wildland Fire 21, 743–754.
| Forecasting intentional wildfires using temporal and spatiotemporal autocorrelations.Crossref | GoogleScholarGoogle Scholar |
Puri K, Areendran G, Raj K, Mazumdar S, Joshi PK (2011) Forest fire risk assessment in parts of Northeast India using geospatial tools. Journal of Forest Research 22, 641–647.
| Forest fire risk assessment in parts of Northeast India using geospatial tools.Crossref | GoogleScholarGoogle Scholar |
Reineking B, Weibel P, Conedera M, Bugmann H (2010) Environmental determinants of lightning- v. human-induced forest fire ignitions differ in a temperate mountain region of Switzerland. International Journal of Wildland Fire 19, 541–557.
| Environmental determinants of lightning- v. human-induced forest fire ignitions differ in a temperate mountain region of Switzerland.Crossref | GoogleScholarGoogle Scholar |
Renard Q, Pélissier R, Ramesh BR, Kodandapani N (2012) Environmental susceptibility model for predicting forest fire occurrence in the Western Ghats of India. International Journal of Wildland Fire 21, 368–379.
| Environmental susceptibility model for predicting forest fire occurrence in the Western Ghats of India.Crossref | GoogleScholarGoogle Scholar |
Su YF, Foody GM, Cheng KS (2012) Spatial non-stationarity in the relationships between land cover and surface temperature in an urban heat island and its impacts on thermally sensitive populations. Landscape and Urban Planning 107, 172–180.
| Spatial non-stationarity in the relationships between land cover and surface temperature in an urban heat island and its impacts on thermally sensitive populations.Crossref | GoogleScholarGoogle Scholar |
Swets JA (1986) Indices of discrimination or diagnostic accuracy: their ROCs and implied models. Psychological Bulletin 99, 100–117.
| Indices of discrimination or diagnostic accuracy: their ROCs and implied models.Crossref | GoogleScholarGoogle Scholar | 1:STN:280:DyaL283gs1GgtA%3D%3D&md5=0ac51ab802364db3fda5c9bd4c234fbaCAS | 3704032PubMed |
Vilar L, Woolford DG, Martell DL, Martn MP (2010) A model for predicting human-caused wildfire occurrence in the region of Madrid, Spain. International Journal of Wildland Fire 19, 325–337.
| A model for predicting human-caused wildfire occurrence in the region of Madrid, Spain.Crossref | GoogleScholarGoogle Scholar |
Wang FH (2006) Principal components, factor, and cluster analyses, and application in social area analysis. In ‘Quantitative Methods and Applications in GIS’. (Ed. FH Wang) pp. 127–131. (CRC Press: Boca Raton, FL)
Wang Q, Ni J, Tenhunen J (2005) Application of a geographically weighted regression analysis to estimate net primary production of Chinese forest ecosystems. Global Ecology and Biogeography 14, 379–393.
| Application of a geographically weighted regression analysis to estimate net primary production of Chinese forest ecosystems.Crossref | GoogleScholarGoogle Scholar |
Wang LT, Zhou Y, Zhou WQ, Wang SX (2013) Fire danger assessment with remote sensing: a case study in Northern China. Natural Hazards 65, 819–834.
| Fire danger assessment with remote sensing: a case study in Northern China.Crossref | GoogleScholarGoogle Scholar |
Woolford DG, Bellhouse DR, Braun WJ, Dean CB, Martell DL, Sun J (2011) A spatio–temporal model for people-caused forest fire occurrence in the Romeo Malette Forest. Journal of Environmental Statistics 2, 2–16. Available at http://www.jenvstat.org/v02/i01/paper [Verified 25 September 2014]
Wu W, Zhang LJ (2013) Comparison of spatial and non-spatial logistic regression models for modeling the occurrence of cloud cover in north-eastern Puerto Rico. Applied Geography 37, 52–62.
| Comparison of spatial and non-spatial logistic regression models for modeling the occurrence of cloud cover in north-eastern Puerto Rico.Crossref | GoogleScholarGoogle Scholar |
Zhang HJ, Han XY, Dai S (2013) Fire occurrence probability mapping of Northeast China with binary logistic regression model. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing 6, 121–127.
| Fire occurrence probability mapping of Northeast China with binary logistic regression model.Crossref | GoogleScholarGoogle Scholar |
Zhang HJ, Qi PC (2012) Mapping fire occurrence susceptibility in Northeast China: comparison of frequency ratio and binary logistic regression. Geography and Geo-information Science 28, 35–38. [In Chinese] Available at http://en.cnki.com.cn/Article_en/CJFDTOTAL-DLGT201205009.htm [Verified 25 September 2014]
Zhou W, Zhou Y, Wang S, Zhao Q (2003) Early warning for grassland fire danger in North China using remote sensing. In ‘International Geoscience and Remote Sensing Symposium (IGARSS03)’, 21–25 July 2003, Toulouse, France. IEEE Conference Publications, pp. 2505–2507. https://doi.org/10.1109/IGARSS.2003.1294490
