Natural Hazard Uncertainty Assessment

Tierz, Pablo; Sandri, Laura; Costa, Alessandro Martins da; Sulpizio, Roberto; Zaccarelli, Lucia; Vito, M. A. Di; Marzocchi, Warner

doi:10.1002/9781119028116

Cited by 9 publications

References 66 publications

(93 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

Modeling wildfire activity in the western United States with machine learning

Buch

Williams

Juang

et al. 2022

Preprint

View full text Add to dashboard Cite

The annual area burned due to wildfires in the western United States (WUS) increased by more than 300% between 1984 and 2020. However, accounting for the nonlinear, spatially heterogeneous interactions between climate, vegetation, and human predictors driving the trends in fire frequency and sizes at different spatial scales remains a challenging problem for statistical fire models. Here we introduce a novel stochastic machine learning (ML) framework to model observed fire frequencies and sizes in 12 km x 12 km grid cells across the WUS. This framework is implemented using Mixture Density Networks trained on a wide suite of input predictors. The modeled WUS fire frequency corresponds well with observations at both monthly (r= 0.94) and annual (r= 0.85) timescales, as do the monthly (r= 0.90) and annual (r= 0.88) area burned. Moreover, the annual time series of both fire variables exhibit strong correlations (r >= 0.6) in 16 out of 18 ecoregions. Our ML model captures the interannual variability and the distinct multidecade increases in annual area burned for both forested and non-forested ecoregions. Evaluating predictor importance with Shapley additive explanations, we find that fire month vapor pressure deficit (VPD) is the dominant driver of fire frequencies and sizes across the WUS, followed by 1000-hour dead fuel moisture (FM1000), total monthly precipitation (Prec), mean daily maximum temperature (Tmax), and fraction of grassland cover in a grid cell. Our findings serve as a promising use case of ML techniques for wildfire prediction in particular and extreme event modeling more broadly.

show abstract

Modeling wildfire activity in the western United States with machine learning

Buch

Williams

Juang

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

Towards Quantitative Volcanic Risk of Pyroclastic Density Currents: Probabilistic Hazard Curves and Maps Around Somma‐Vesuvius (Italy)

et al. 2018

View full text Add to dashboard Cite

Pyroclastic density currents (PDCs) are hot flowing mixtures of gas and pyroclasts that can cause widespread loss of life and structural damage around the erupting volcano. Hazard assessments that include quantification of aleatory and epistemic uncertainty are a necessary step toward calculating volcanic risk of PDCs in an accurate and complete manner. We develop a three‐stage procedure to quantify such uncertainties for dense PDCs. First, the TITAN2D model is parameterized to simulate the PDC phenomenology at the target volcano. Second, TITAN2D is coupled with Polynomial Chaos Quadrature to propagate aleatory uncertainty from model parameters to hazard intensity measures (flow depth and speed). Third, the TITAN2D‐PCQ analysis is merged with the Bayesian Event Tree for Volcanic Hazard to include other volcano‐specific aleatory uncertainty and estimates of epistemic uncertainty. A comprehensive collection of probabilistic hazard curves and maps for flow depth and speed around the volcano is obtained through this methodology and its application is illustrated at Somma‐Vesuvius (Italy). Our results indicate that, given an eruption from the current central crater, exceedance probabilities are around 30% (aleatory uncertainty only) and between 10% and 60% (aleatory and epistemic uncertainty), for flow depth = 1 m and flow speed = 2 m/s, over the first 2–3 km around the vent. Dense PDCs faster than 30 m/s may cover areas about 50 km2 around the vent, on average, 1 every 10 eruptions. This type of probabilistic hazard assessment represents a crucial step toward quantitative volcanic risk of dense PDCs at Somma‐Vesuvius and worldwide.

show abstract