Wildfire ignition-distribution modelling: a comparative study in the Huron–Manistee National Forest, Michigan, USA
Avi Bar Massada A D , Alexandra D. Syphard B , Susan I. Stewart C and Volker C. Radeloff A
+ Author Affiliations
- Author Affiliations
A Department of Forest and Wildlife Ecology, University of Wisconsin – Madison, 1630 Linden Drive, Madison, WI 53706, USA.
B Conservation Biology Institute, 10423 Sierra Vista Avenue, La Mesa, CA 91941, USA.
C Northern Research Station, USDA Forest Service, 1033 University Avenue, Suite 360, Evanston, IL 60201, USA.
D Corresponding author. Email address: barmassada@wisc.edu
International Journal of Wildland Fire 22(2) 174-183 https://doi.org/10.1071/WF11178
Submitted: 18 December 2011 Accepted: 16 July 2012 Published: 18 September 2012
Abstract
Wildfire ignition distribution models are powerful tools for predicting the probability of ignitions across broad areas, and identifying their drivers. Several approaches have been used for ignition-distribution modelling, yet the performance of different model types has not been compared. This is unfortunate, given that conceptually similar species-distribution models exhibit pronounced differences among model types. Therefore, our goal was to compare the predictive performance, variable importance and the spatial patterns of predicted ignition-probabilities of three ignition-distribution model types: one parametric, statistical model (Generalised Linear Models, GLM) and two machine-learning algorithms (Random Forests and Maximum Entropy, Maxent). We parameterised the models using 16 years of ignitions data and environmental data for the Huron–Manistee National Forest in Michigan, USA. Random Forests and Maxent had slightly better prediction accuracies than did GLM, but model fit was similar for all three. Variables related to human population and development were the best predictors of wildfire ignition locations in all models (although variable rankings differed slightly), along with elevation. However, despite similar model performance and variables, the map of ignition probabilities generated by Maxent was markedly different from those of the two other models. We thus suggest that when accurate predictions are desired, the outcomes of different model types should be compared, or alternatively combined, to produce ensemble predictions.
Additional keywords: GLM, Maxent, Random Forests.
References
Bar Massada A, Radeloff VC, Stewart SI, Hawbaker TJ (2009) Wildfire risk in the wildland–urban interface: a simulation study in northwestern Wisconsin. Forest Ecology and Management 258, 1990–1999.| Wildfire risk in the wildland–urban interface: a simulation study in northwestern Wisconsin.Crossref | GoogleScholarGoogle Scholar |
Bar Massada A, Syphard AD, Hawbaker TJ, Stewart SI, Radeloff VC (2011) Effects of ignition location models on the burn patterns of simulated wildfires. Environmental Modelling & Software 26, 583–592.
| Effects of ignition location models on the burn patterns of simulated wildfires.Crossref | GoogleScholarGoogle Scholar |
Breiman L (2001) Random forests. Machine Learning 45, 15–32.
Cleland DT, Crow TR, Saunders SC, Dickmann DI, Maclean AL, Jordan JK, Watson RL, Sloan AM, Brosofske KD (2004) Characterizing historical and modern fire regimes in Michigan (USA): a landscape ecosystem approach. Landscape Ecology 19, 311–325.
| Characterizing historical and modern fire regimes in Michigan (USA): a landscape ecosystem approach.Crossref | GoogleScholarGoogle Scholar |
Cutler DR, Edwards TC, Beard KH, Cutler A, Hess KT, Gibson J, Lawler JJ (2007) Random forests for classification in ecology. Ecology 88, 2783–2792.
| Random forests for classification in ecology.Crossref | GoogleScholarGoogle Scholar |
Diaz-Avalos C, Peterson D, Alvarado E, Ferguson S, Besag J (2001) Space-time modelling of lightning-caused ignitions in the Blue Mountains, Oregon. Canadian Journal of Forest Research 31, 1579–1593.
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 |
Elith J, Graham CH (2009) Do they? How do they? WHY do they differ? On finding reasons for differing performances of species distribution models. Ecography 32, 66–77.
| Do they? How do they? WHY do they differ? On finding reasons for differing performances of species distribution models.Crossref | GoogleScholarGoogle Scholar |
Elith J, Graham CH, Anderson RP, Dudík M, Ferrier S, Guisan A, Hijmans RJ, Huettmann F, Leathwick JR, Lehmann A, Li J, Lohmann LG, Loiselle BA, Manion G, Moritz C, Nakamura M, Nakazawa Y, Overton JMM, Townsend Peterson A, Phillips SJ, Richardson K, Scachetti-Pereira R, Schapire RE, Soberón J, Williams S, Wisz MS, Zimmermann NE (2006) Novel methods improve prediction of species’ distributions from occurrence data. Ecography 29, 129–151.
| Novel methods improve prediction of species’ distributions from occurrence data.Crossref | GoogleScholarGoogle Scholar |
Elith J, Phillips SJ, Hastie T, Dudík M, Chee YE, Yates CJ (2011) A statistical explanation of MaxEnt for ecologists. Diversity & Distributions 17, 43–57.
| A statistical explanation of MaxEnt for ecologists.Crossref | GoogleScholarGoogle Scholar |
Finney MA (2005) The challenge of quantitative risk analysis for wildland fire. Forest Ecology and Management 211, 97–108.
| The challenge of quantitative risk analysis for wildland fire.Crossref | GoogleScholarGoogle Scholar |
Franklin J (2010) ‘Mapping Species Distributions.’ (Cambridge University Press: New York)
Freeman EA, Moisen GG (2008) A comparison of the performance of threshold criteria for binary classification in terms of predicted prevalence and kappa. Ecological Modelling 217, 48–58.
| A comparison of the performance of threshold criteria for binary classification in terms of predicted prevalence and kappa.Crossref | GoogleScholarGoogle Scholar |
Gavier-Pizarro GI, Radeloff VC, Stewart SI, Huebner CD, Keuler NS (2010) Housing is positively associated with invasive exotic plant species richness in New England, USA. Ecological Applications 20, 1913–1925.
| Housing is positively associated with invasive exotic plant species richness in New England, USA.Crossref | GoogleScholarGoogle Scholar |
Guisan A, Edwards TC, Hastie T (2002) Generalized linear and generalized additive models in studies of species distributions: setting the scene. Ecological Modelling 157, 89–100.
| Generalized linear and generalized additive models in studies of species distributions: setting the scene.Crossref | GoogleScholarGoogle Scholar |
Guisan A, Zimmermann NE, Elith J, Graham CH, Phillips S, Peterson AT (2007) What matters for predicting the occurrences of trees: techniques, data, or species’ characteristics? Ecological Monographs 77, 615–630.
| What matters for predicting the occurrences of trees: techniques, data, or species’ characteristics?Crossref | GoogleScholarGoogle Scholar |
Hanley JA, McNeil BJ (1982) The meaning and use of the area under a receiver operating characteristic (ROC) curve. Radiology 143, 29–36.
Hastie T, Tibshirani R (1990) ‘Generalized Additive Models.’ (Chapman and Hall: New York)
Homer C, Dewitz J, Fry J, Coan M, Hossain N, Larson C, Herold N, McKerrow A, VanDriel JN, Wickham J (2007) Completion of the 2001 National Land Cover Database for the conterminous United States. Photogrammetric Engineering and Remote Sensing 73, 337–341.
Hosmer DW, Jovanovic B, Lemeshow S (1989) Best subsets logistic regression. Biometrics 45, 1265–1270.
| Best subsets logistic regression.Crossref | GoogleScholarGoogle Scholar |
Krawchuk M, Cumming S, Flannigan M, Wein R (2006) Biotic and abiotic regulation of lightning fire initiation in the mixedwood boreal forest. Ecology 87, 458–468.
| Biotic and abiotic regulation of lightning fire initiation in the mixedwood boreal forest.Crossref | GoogleScholarGoogle Scholar | 1:STN:280:DC%2BD283jtVWrtA%3D%3D&md5=2e2ac8ae90c27b7264feb12c5e420fb1CAS |
Latham D, Williams E (2001) Lightning and forest fires. In ‘Forest Fires’. (Eds E Johnson, K Miyanishi) pp. 375–418. (Academic Press: San Diego, CA)
Li H, Wu J (2004) Use and misuse of landscape indices. Landscape Ecology 19, 389–399.
| Use and misuse of landscape indices.Crossref | GoogleScholarGoogle Scholar |
Liaw A, Wiener M (2002) Classification and regression by random forest. R News 2, 18–22.
Lobo JM, Jiménez-Valverde A, Real R (2008) AUC: a misleading measure of the performance of predictive distribution models. Global Ecology and Biogeography 17, 145–151.
| AUC: a misleading measure of the performance of predictive distribution models.Crossref | GoogleScholarGoogle Scholar |
Marmion M, Parviainen M, Luoto M, Heikkinen RK, Thuiller W (2009) Evaluation of consensus methods in predictive species distribution modelling. Diversity & Distributions 15, 59–69.
| Evaluation of consensus methods in predictive species distribution modelling.Crossref | GoogleScholarGoogle Scholar |
McCune B, Grace J (2002) ‘Analysis of ecological communities.’ (MJM Software Design: Glenden Beach, OR)
McLeod AI, Xu C (2010) bestglm: Best subsets GLM. In ‘R package version 0.31’. Available at http://cran.r-project.org/web/packages/bestglm/vignettes/bestglm.pdf [Verified 22 August 2012]
Moritz MA, Hessburg PF, Povak NA (2011). Native fire regimes and landscape resilience. In ‘The Landscape Ecology of Fire’. (Eds D McKenzie, C Miller, DA Falk) pp. 51–86. (Springer: Dordrecht, the Netherlands)
Narayanaraj G, Wimberly M (2011) Influences of forest roads on the spatial pattern of wildfire boundaries. International Journal of Wildland Fire 20, 792–803.
| Influences of forest roads on the spatial pattern of wildfire boundaries.Crossref | GoogleScholarGoogle Scholar |
Nelder JA, Wedderburn RWM (1972) Generalized linear models. Journal of the Royal Statistical Society. Series A (General) 135, 370–384.
| Generalized linear models.Crossref | GoogleScholarGoogle Scholar |
Parisien M-A, Moritz MA (2009) Environmental controls on the distribution of wildfire at multiple spatial scales. Ecological Monographs 79, 127–154.
| Environmental controls on the distribution of wildfire at multiple spatial scales.Crossref | GoogleScholarGoogle Scholar |
Pearson RG, Raxworthy CJ, Nakamura M, Townsend Peterson A (2007) Predicting species distributions from small numbers of occurrence records: a test case using cryptic geckos in Madagascar. Journal of Biogeography 34, 102–117.
| Predicting species distributions from small numbers of occurrence records: a test case using cryptic geckos in Madagascar.Crossref | GoogleScholarGoogle Scholar |
Phillips SJ, Dudík M (2008) Modeling of species distributions with Maxent: new extensions and a comprehensive evaluation. Ecography 31, 161–175.
| Modeling of species distributions with Maxent: new extensions and a comprehensive evaluation.Crossref | GoogleScholarGoogle Scholar |
Phillips SJ, Anderson RP, Schapire RE (2006) Maximum entropy modeling of species geographic distributions. Ecological Modelling 190, 231–259.
| Maximum entropy modeling of species geographic distributions.Crossref | GoogleScholarGoogle Scholar |
Pontius RG, Millones M (2011) Death to kappa: birth of quantity disagreement and allocation disagreement for accuracy assessment. International Journal of Remote Sensing 32, 4407–4429.
| Death to kappa: birth of quantity disagreement and allocation disagreement for accuracy assessment.Crossref | GoogleScholarGoogle Scholar |
Prasad A, Iverson L, Liaw A (2006) Newer classification and regression tree techniques: bagging and random forests for ecological prediction. Ecosystems 9, 181–199.
| Newer classification and regression tree techniques: bagging and random forests for ecological prediction.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 |
Radeloff VC, Hammer RB, Stewart SI, Fried JS, Holcomb SS, McKeefry AJ (2005) The wildland–urban interface in the United States. Ecological Applications 15, 799–805.
| The wildland–urban interface in the United States.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 |
Sturtevant BR, Cleland DT (2007) Human and biophysical factors influencing modern fire disturbance in northern Wisconsin. International Journal of Wildland Fire 16, 398–413.
| Human and biophysical factors influencing modern fire disturbance in northern Wisconsin.Crossref | GoogleScholarGoogle Scholar |
Syphard AD, Franklin J (2009) Differences in spatial predictions among species distribution modeling methods vary with species traits and environmental predictors. Ecography 32, 907–918.
| Differences in spatial predictions among species distribution modeling methods vary with species traits and environmental predictors.Crossref | GoogleScholarGoogle Scholar |
Syphard AD, Franklin J (2010) Species traits affect the performance of species distribution models for plants in southern California. Journal of Vegetation Science 21, 177–189.
| Species traits affect the performance of species distribution models for plants in southern California.Crossref | GoogleScholarGoogle Scholar |
Syphard AD, Radeloff VC, Keely JE, Hawbaker TJ, Clayton MK, Stewart SI, Hammer RB (2007) Human influence on California fire regimes. Ecological Applications 17, 1388–1402.
| Human influence on California fire regimes.Crossref | GoogleScholarGoogle Scholar |
Syphard AD, Radeloff VC, Keuler NS, Taylor RS, Hawbaker TJ, Stewart SI, Clayton MK (2008) Predicting spatial patterns of fire on a southern California landscape. International Journal of Wildland Fire 17, 602–613.
| Predicting spatial patterns of fire on a southern California landscape.Crossref | GoogleScholarGoogle Scholar |
Syphard AD, Radeloff VC, Hawbaker TJ, Stewart SI (2009) Conservation threats due to human-caused increases in fire frequency in Mediterranean-climate ecosystems. Conservation Biology 23, 758–769.
| Conservation threats due to human-caused increases in fire frequency in Mediterranean-climate ecosystems.Crossref | GoogleScholarGoogle Scholar |
Ward G, Hastie T, Barry SC, Elith J, Leathwick JR (2009) Presence-only data and the EM algorithm. Biometrics 65, 554–563.
| Presence-only data and the EM algorithm.Crossref | GoogleScholarGoogle Scholar |
Yang J, He H, Shifley S, Gustafson E (2007) Spatial patterns of modern period human-caused fire occurrence in the Missouri Ozark Highlands. Forest Science 53, 1–15.
