Journal cover Journal topic
Natural Hazards and Earth System Sciences An interactive open-access journal of the European Geosciences Union
Journal topic

Journal metrics

Journal metrics

  • IF value: 2.281 IF 2.281
  • IF 5-year value: 2.693 IF 5-year 2.693
  • CiteScore value: 2.43 CiteScore 2.43
  • SNIP value: 1.193 SNIP 1.193
  • SJR value: 0.965 SJR 0.965
  • IPP value: 2.31 IPP 2.31
  • h5-index value: 40 h5-index 40
  • Scimago H index value: 73 Scimago H index 73
Volume 18, issue 9 | Copyright
Nat. Hazards Earth Syst. Sci., 18, 2331-2343, 2018
© Author(s) 2018. This work is distributed under
the Creative Commons Attribution 4.0 License.

Research article 04 Sep 2018

Research article | 04 Sep 2018

Evaluation of predictive models for post-fire debris flow occurrence in the western United States

Efthymios I. Nikolopoulos1, Elisa Destro2, Md Abul Ehsan Bhuiyan1, Marco Borga2, and Emmanouil N. Anagnostou1 Efthymios I. Nikolopoulos et al.
  • 1Department of Civil and Environmental Engineering, University of Connecticut, Storrs, CT, USA
  • 2Department of Leaf, Environment, Agriculture and Forestry, University of Padova, Legnaro, PD, Italy

Abstract. Rainfall-induced debris flows in recently burned mountainous areas cause significant economic losses and human casualties. Currently, prediction of post-fire debris flows is widely based on the use of power-law thresholds and logistic regression models. While these procedures have served with certain success in existing operational warning systems, in this study we investigate the potential to improve the efficiency of current predictive models with machine-learning approaches. Specifically, the performance of a predictive model based on the random forest algorithm is compared with current techniques for the prediction of post-fire debris flow occurrence in the western United States. The analysis is based on a database of post-fire debris flows recently published by the United States Geological Survey. Results show that predictive models based on random forest exhibit systematic and considerably improved performance with respect to the other models examined. In addition, the random-forest-based models demonstrated improvement in performance with increasing training sample size, indicating a clear advantage regarding their ability to successfully assimilate new information. Complexity, in terms of variables required for developing the predictive models, is deemed important but the choice of model used is shown to have a greater impact on the overall performance.

Publications Copernicus
Short summary
Debris flows, following wildfires, constitute a significant threat to downstream populations and infrastructure. Therefore, developing measures to reduce the vulnerability of local communities to debris flows is of paramount importance. This work proposes a new model for predicting post-fire debris flow occurrence on a regional scale and demonstrates that the proposed model has notably higher skill than the currently used approaches.
Debris flows, following wildfires, constitute a significant threat to downstream populations and...