Quantifying location error to define uncertainty in volcanic mass flow hazard simulations

Abstract. The use of mass flow simulations in volcanic hazard zonation and mapping is often limited by model complexity (i.e. uncertainty correct values parameters), a lack quantification, approaches to incorporate this into maps. When quantified, simulation errors are typically evaluated on pixel-pair basis, using the difference between simulated observed (“actual”) map-cell evaluate performance model. However, these comparisons conflate location quantification errors, neglecting possible spatial autocorrelation errors. As result, assessments yield moderate accuracy values. In paper, similarly were found assessment three depth-averaged numerical models 2012 debris avalanche from Upper Te Maari crater, Tongariro Volcano, as benchmark. To provide fairer covariance we fuzzy set approach indicate proximity valued map cells. This “fuzzification” results yields improvements targeted metrics relative length scale parameter at expense decreases opposing (e.g. fewer false negatives result more positives) reduction resolution. generate zones incorporating identified associated trade-offs demonstrated indicates potential for informed stakeholders reducing estimation supporting decision-making data.