Debris flow modelling and early warning system for disaster risk reduction in Kundasang area, Sabah

Abstract

Landslides are recurring geological disasters resulted in many human- and economic losses, even so with the rapid urbanization and extreme climate. Approximately, 21,000 landslides – debris flows were recorded in Malaysia. There are many uncertainties in the underlying root causes, understanding triggering factors, and ways to reduce local risk in a changing environment. Even more challenging to prepare a cost-effective Early Warning System (EWS) and to enhance resilient communities in a tectonically active region. This study aims to develop an integrated framework for debris flow EWS with the case study in Mesilau watershed, Kundasang, Sabah. These are the three objectives; (i) to map and characterize the debris flow induced by the 2015 Ranau earthquake, (ii) to model and simulate the debris flow runout, and (iii) to develop an integrated framework for debris flow EWS, in supporting local disaster risk reduction and resilience strategy. The study started by characterizing the watershed and landslide areas using the Geographic Information System. The results showed that the earthquake stripped at least 1.44 km2 of vegetation cover within the upstream of Mesilau watershed, and increased the Stripped Earth Material (SEM) by 1.32 km2. Thus, the increased SEM contributed to the temporary landslide damming formation. Next, critical rainfall, discharge, and hydrographs were extracted using the empirical method, and Hydrological Modelling System to understand the triggering factor for debris flow event. The extractions suggested the breaching of temporary landslide dam was due to the rainfall intensity of 14.2 mm/h, and 7-days critical rainfall that exceeded 66.3 mm.Hence, remobilized the temporary landslide dam, and initiated the debris flow that travelled for 18.6 km to Liwagu Dam, Ranau town. The extracted parameters were imported into the HyperKANAKO software to model and simulate the best-fit debris flow run out. The obtained best-fit debris flow run out was utilized to estimate the debris flow velocity and the lead time to evacuate. The best-fit simulation results suggested that the debris flow velocity as 12.5 m/s, with the suitable discharge at 550m3/s. The result indicates that the required lead time for the community to evacuate is4.5 min before the debris flow arrives at Mesilau village. These simulation results were validated through the field evidence, image correlations, expert and local judgments. Subsequently, the debris flow EWS was designed by referring to the TAKUWA’s guideline. All the obtained scientific results were then used to gain societal inputs by understanding demands and needs for people-centered EWSvia structured EWS surveys and open-ended interviews. The societal inputs highlight that the EWS is critically needed for Mesilau watershed, as the area is proneto the earthquake and cascading geohazards (i.e., debris flows). In conclusion, the proposed integrated framework for debris flow EWS is aligned to Malaysia’s commitment to increase the access to multi-hazard EWS and disaster risk information (Target G) of the Sendai Framework for Disaster Risk Reduction 2015 –2030.