How much does the rainfall temporal resolution affect rainfall thresholds for landslide triggering?

• DOI: 10.1007/s11069-019-03830-x
• Publication Date: 2019-12-31T10:02:25Z
• Authors: Stefano Luigi Gar...; Massimo Melillo; Silvia Peruccacci; Maria Teresa Brun...
• Keywords: Natural hazard; Temporal resolution

Abstract In many areas of the world, prediction rainfall-induced landslides is usually carried out by means empirical rainfall thresholds. Their definition complicated several issues, among which are evaluation and quantification diverse uncertainties resulting from data methods. Threshold effectiveness reliability strongly depend on quality quantity measurements landslide information used as input. this work, influence temporal resolution calculation landslide-triggering thresholds evaluated discussed. For purpose, hourly collected 172 rain gauges geographical occurrence 561 in Liguria region (northern Italy) period 2004–2014 used. To assess impact different resolutions thresholds, clustered increasing bins 1, 3, 6, 12 24 h. A comprehensive tool applied to each dataset automatically reconstruct conditions responsible for failures calculate frequentist cumulated event rainfall–rainfall duration ( ED ) Then, using a quantitative procedure, calculated validated. The main finding work that use with results considerable variations shape validity range Decreasing resolution, smaller intercepts, higher slopes, shorter ranges obtained. On other hand, it seems does not validation procedure threshold performance indicators. Overall, coarse causes systematic underestimation at short durations, relevant drawbacks (e.g. false alarms) if implemented operational systems prediction.