What controlled the occurrence of more than 116,000 human-mapped landslides triggered by Cyclone Gabrielle, New Zealand?
Chris Massey, Kerry Leith, Tom Robinson, Biljana Luković, Sam McColl, Trevor Carey‐Smith, Brenda Rosser, Liam Wotherspoon, Hugh G. Smith, Harley Betts, Robert Buxton, Janine Bidmead
Abstract
Abstract This paper uses an extract of 116,655 landslide source areas from the >160,000 human-mapped landslides contained in the Version One (V1.0) landslide inventory triggered by Cyclone Gabrielle (11–17 February 2023). It describes the analysis of the human-mapped landslide distribution and the relationships between landslide occurrence, their susceptibility factors, and their hydrological forcing factors. The results show that rain is one of the main controlling factors on landslide probability, but the relationship is complex. Our results show the amount of 24-h rain becomes less important, relative to other landslide susceptibility factors, at high (>220 mm) to very high (>330 mm) 24-h rainfalls, suggesting that such extreme rain amounts may not drive the highest probability of landslide occurrence. Results also suggest that the 24-h rainfalls experienced by some areas during Cyclone Gabrielle were likely to have been sufficiently high enough to “overprint” antecedent soil water conditions. In these overprint areas, soil-water conditions prior to Cyclone Gabrielle rainfall may not have materially influenced the spatial landslide distributions. The results from this study suggest that during a storm, more intense rainfall may not directly translate to increasing probability of landslide occurrence; however, it may diminish the effect of pre-event soil water content.