Data limitations and potential of hourly and daily rainfall thresholds for shallow landslides

Abstract

Abstract. Rainfall thresholds are a simple and widely used method to predict landslide occurrence. In this paper we provide a comprehensive data-driven assessment of the effects of rainfall temporal resolution (hourly versus daily) on landslide prediction performance in Switzerland, with sensitivity to two other important aspects which appear in many landslide studies – the normalisation of rainfall, which accounts for local climatology, and the inclusion of antecedent rainfall as a proxy of soil water state prior to landsliding. We use an extensive landslide inventory with over 3800 events and several daily and hourly, station and gridded rainfall datasets, to explore different scenarios of rainfall threshold estimation. Our results show that although hourly rainfall did show best predictive performance for landslides, daily data were not far behind, and the benefits of hourly resolutions can be masked by the higher uncertainties in threshold estimation connected to using short records. We tested the impact of several typical actions of users, like assigning the nearest raingauge to a landslide location and filling in unknown timing, and report their effects on predictive performance. We find that localisation of rainfall thresholds through normalisation compensates for the spatial heterogeneity in rainfall regimes and landslide erosion process rates and is a good alternative to regionalisation. On top of normalisation by mean annual precipitation or a high rainfall quantile, we recommend that non-triggering rainfall be included in rainfall threshold estimation if possible. Finally, we demonstrate that there is predictive skill in antecedent rain as a proxy of soil wetness state, despite the large heterogeneity of the study domain, although it may not be straightforward to build this into rainfall threshold curves.

Deciphering meteorological influencing factors for Alpine rockfalls: a case study in Aosta Valley

“…The lower dispersion obtained with normalized thresholds in comparison to those derived from the simple intensity-duration formulation confirms the utility of this approach in areas characterized by the juxtaposition of different climate types. This result agrees with findings of numerous previous studies (Guzzetti et al 2007(Guzzetti et al , 2008Peruccacci et al 2017;Leonarduzzi and Molnar 2020).…”

“…For the precipitationrelated indices (STR, EWI), two types of normalization were carried out. The first threshold normalization was performed based on the mean annual precipitation (MAP), while the second normalization was based on the rainy day normal (RDN), which is the ratio of the MAP to the mean annual number of rainy days (Guzzetti et al 2007(Guzzetti et al , 2008Peruccacci et al 2017;Leonarduzzi and Molnar 2020). A normalized threshold was built also for the freeze-thaw cycles, taking inspiration particularly from the RDN normalization.…”

Rainfall-induced shallow landslides and soil wetness: comparison of physically-based and probabilistic predictions

“…To this end, we estimate the a and b parameters of the power law curve by maximising the True Skill Statistic (TSS=True Positive Ratio -False Positive Ratio), as in Leonarduzzi et al (2017). This allows us to classify the rainfall events into the following groups by the calibrated ED threshold (see also Leonarduzzi and Molnar, 2020): observed and correctly predicted triggering events above the ED curve (True Positives), observed triggering events which fall below the ED curve (Misses), observed non-triggering events which fall above the ED curve (False Alarms), and observed non-triggering events which fall below the ED curve (True Negatives).…”

The role of geomorphology, rainfall and soil moisture in the occurrence of landslides triggered by 2018 Typhoon Mangkhut in the Philippines

“…Aspects such as the geographical distribution, the intended use of the thresholds, or simply the resources available determine the source of the rainfall data used to construct thresholds. Uncertainties on the source and resolution of the data as well as on the methods used to define the rainfall events or the rainfall thresholds will be key to their accuracy (Abancó et al, 2016;Leonarduzzi and Molnar, 2020;Nikolopoulos et al, 2015). Rainfall thresholds are used with early warning purposes in several countries and regions all over the world, although not in the Philippines so far (Guzzetti et al, 2020).…”