Risk assessment for wildland fire aerial detection patrol route planning in Ontario, Canada

McFayden, Colin B.; Woolford, Douglas G.; Stacey, Aaron; Boychuk, Den; Johnston, Joshua M.; Wheatley, Melanie J.; Martell, David L.

doi:10.1071/wf19084

Cited by 16 publications

(14 citation statements)

References 48 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Well-calibrated fire occurrence predictions are much easier for a fire management agency to interpret, whereas miscalibrated probabilities and a scale from low to high danger (without a probabilistic interpretation or a connection to the expected number of fires) are more difficult to interpret and thus are less useful as a decision support tool. In addition, the output of FOP models may be used as input in other decision support tools such as the aerial detection planning tool described in McFayden et al (2020) or in wildland fire risk modelling (see Xi et al 2019;Johnston et al 2020 for summaries). The risk to a resource or asset (e.g.…”

Section: Pitfalls With Current Approachesmentioning

confidence: 99%

See 1 more Smart Citation

Guidelines for effective evaluation and comparison of wildland fire occurrence prediction models

Phelps

Woolford

2021

Int. J. Wildland Fire

Self Cite

View full text Add to dashboard Cite

Daily, fine-scale spatially explicit wildland fire occurrence prediction (FOP) models can inform fire management decisions. Many different data-driven modelling methods have been used for FOP. Several studies use multiple modelling methods to develop a set of candidate models for the same region, which are then compared against one another to choose a final model. We demonstrate that the methodologies often used for evaluating and comparing FOP models may lead to selecting a model that is ineffective for operational use. With an emphasis on spatially and temporally explicit FOP modelling for daily fire management operations, we outline and discuss several guidelines for evaluating and comparing data-driven FOP models, including choosing a testing dataset, choosing metrics for model evaluation, using temporal and spatial visualisations to assess model performance, recognising the variability in performance metrics, and collaborating with end users to ensure models meet their operational needs. A case study for human-caused FOP in a provincial fire control zone in the Lac La Biche region of Alberta, Canada, using data from 1996 to 2016 demonstrates the importance of following the suggested guidelines. Our findings indicate that many machine learning FOP models in the historical literature are not well suited for fire management operations.

show abstract

Section: Pitfalls With Current Approachesmentioning

confidence: 99%

“…FOP models also feed into other decision support tools, such as those that aid aerial detection planning (e.g. McFayden et al 2020).…”

Section: Introductionmentioning

confidence: 99%

Guidelines for effective evaluation and comparison of wildland fire occurrence prediction models

Phelps

Woolford

2021

Int. J. Wildland Fire

Self Cite

View full text Add to dashboard Cite

show abstract

“…Aerial patrol may also be used to improve the chance of detecting wildland fires at earlier stages. [71] presented a three-step process to spatially quantify the risk of not patrolling a specified area for the detection of wildland fires in Canada. The proposed process produced a daily updated fine-scale risk index map that can be used to design optimal aerial patrol routes.…”

Section: Early Detection Of Wildfiresmentioning

confidence: 99%

Three Lines of Defense for Wildfire Risk Management in Electric Power Grids: A Review

et al. 2021

View full text Add to dashboard Cite

Wildfires pose a significant challenge to the natural and the built environments, as well as the safety and economic wellbeing of the communities residing in wildfire-prone areas. The electric power grid is specifically among the built environments most affected by, and contributing to, wildfires. In this paper, we propose a three lines of defense (3LD) framework for wildfire risk management in electric power infrastructure and review the literature from this lens. An overview of the physics and phenomenology of the wildfires as it relates to power grids is presented, and the logic for the proposed 3LD framework is discussed. The reviewed literature based on the 3LD theme includes the most relevant and emerging research work on wildfire prevention as the first line of defense, wildfire mitigation and proactive response as the second line of defense, and wildfire recovery preparedness as the third line of defense. This study reveals that while the state of the art, to a large extent, stands comprehensive in various aspects of power system resilience and wildfire risk management, there is a gap in the literature in addressing this emerging risk in a holistic, interdisciplinary approach.

show abstract

“…FFMC is used by Canadian wildland fire management agencies as a measure of the receptivity of surface fuels to ignition (Wotton 2009) and has been found to be a significant predictor for human-caused FOP in other regions in Canada (e.g. Cunningham and Martell 1973;Woolford et al 2011;Magnussen and Taylor 2012;McFayden et al 2020). Weather variables that are used to calculate fuel moisture, namely relative humidity, temperature and precipitation, were excluded to avoid possible multicollinearity effects with FFMC, which is calculated as a function of those variables (Van Wagner 1987).…”

Section: Model Building and Variable Selectionmentioning

confidence: 99%

Comparing calibrated statistical and machine learning methods for wildland fire occurrence prediction: a case study of human-caused fires in Lac La Biche, Alberta, Canada

Phelps

Woolford

2021

Int. J. Wildland Fire

Self Cite

View full text Add to dashboard Cite

Wildland fire occurrence prediction (FOP) modelling supports fire management decisions, such as suppression resource pre-positioning and the routeing of detection patrols. Common empirical modelling methods for FOP include both model-based (statistical modelling) and algorithmic-based (machine learning) approaches. However, it was recently shown that many machine learning models in FOP literature are not suitable for fire management operations because of overprediction if not properly calibrated to output true probabilities. We present methods for properly calibrating statistical and machine learning models for fine-scale, spatially explicit daily FOP followed by a case-study comparison of human-caused FOP modelling in the Lac La Biche region of Alberta, Canada, using data from 1996 to 2016. Calibrated bagged classification trees, random forests, neural networks, logistic regression models and logistic generalised additive models (GAMs) are compared in order to assess the pros and cons of these approaches when properly calibrated. Results suggest that logistic GAMs can have similar performance to machine learning models for FOP. Hence, we advocate that the pros and cons of different modelling approaches should be discussed with fire management practitioners when determining which models to use operationally because statistical methods are commonly viewed as more interpretable than machine learning methods.

show abstract

Risk assessment for wildland fire aerial detection patrol route planning in Ontario, Canada

Cited by 16 publications

References 48 publications

Guidelines for effective evaluation and comparison of wildland fire occurrence prediction models

Guidelines for effective evaluation and comparison of wildland fire occurrence prediction models

Three Lines of Defense for Wildfire Risk Management in Electric Power Grids: A Review

Comparing calibrated statistical and machine learning methods for wildland fire occurrence prediction: a case study of human-caused fires in Lac La Biche, Alberta, Canada

Contact Info

Product

Resources

About