Assessment of post-wildfire debris flow occurrence using classifier tree

Abstract

Besides the dangers of an actively burning wildfire, a plethora of other hazardous consequences can occur afterwards. Debris flows are among the most hazardous of these, being known to cause fatalities and extensive damage to infrastructure. Although debris flows are not exclusive to fire affected areas, a wildfire can increase a location's susceptibility by stripping its protective covers like vegetation and introducing destabilizing factors such as ash filling soil pores to increase runoff potential. Due to the associated dangers, researchers are developing statistical models to isolate susceptible locations. Existing models predominantly employ the logistic regression algorithm; however, previous studies have shown that the relationship between the predictors and response is likely better predicted using nonlinear modeling. We therefore propose the use of nonlinear C5.0 decision tree algorithm, a simple yet robust algorithm that uses inductive inference for categorical data modelling. It employs a tree-like decision making system that makes conditional statements to split data into homogeneous classes. Our results showed the C5.0 approach to produce stable and higher validation metrics in comparison to the logistic regression. A sensitivity of 81% and specificity of 78% depicts improved predictive capability and gives credence to the hypothesis that data relationships are likely nonlinear. ARTICLE HISTORY

A Method for Improving Controlling Factors Based on Information Fusion for Debris Flow Susceptibility Mapping: A Case Study in Jilin Province, China

“…In addition, the logistic regression model [ 9 ], evidence belief function [ 10 ], weight of evidence [ 11 ], frequency ratio [ 10 , 12 ], and statistical index (SI) [ 13 ] have been used extensively. Regarding heuristic models, Addison et al [ 14 ] used the classifier tree to analyze debris flow. Other models, such as multistandard analysis [ 15 ] and the heuristic fuzzy model [ 16 ], have also been used extensively.…”

Application of remote sensing and machine learning modeling to post-wildfire debris flow risks

“…The two models considered in this study are the linear-based logistic regression model from Staley et al, (2017) and the nonlinear-based C5.0 decision tree model from Addison et al, (2018). It is important to note that these models predict the origination points of DFs and not their inundation paths downstream.…”

“…The model was then tested on the validation set, which yielded a sensitivity of 81% and a specificity of 58% (Staley et al, 2017;Addison et al, 2018). A sensitivity of 81% means that the model has the capability of correctly predicting ~8 out of 10 DF producing locations, whereas a specificity of 58% means that ~6 out of 10 "DF safe" locations within a burned area will be correctly identified.…”

“…linear (Rupert et al, 2008;Cannon et al, 2010;Staley et al, 2017), and nonlinear 51 machine learning modeling (Kern et al, 2017, Addison et al, 2018, both of which have inherent advantages and disadvantages.…”

“…The objective of this paper is to apply representative models from these two approaches to the DF event that occurred after the Thomas Fire and assess their predictive capabilities. The models considered are the linear-based logistic regression model from Staley et al, (2017) and the nonlinear-based C5.0 decision tree model from Addison et al, (2018). To do this, DF scars were delineated by running a change detection protocol with the Normalized Difference Vegetation Index (NDVI) algorithm, using highresolution data from Planet Labs Inc.…”