Improving debris flow monitoring in Taiwan by using high-resolution rainfall products from QPESUMS

Abstract

Debris flow is commonly initiated by torrential rain and its triggering is correlated to the hydrological, geological, and geomorphic conditions on site. In spite of the important effects of geology and topography, rainfall characteristic is the main external triggering factor to debris flow and is a predominant parameter for real-time monitoring. Due to the scarcity of sufficient spatial ground-based rainfall data in hill areas, quantitative precipitation estimation using remote-sensing techniques such as radar and satellite is needed for debris flow pre-warning. The QPESUMS (Quantitative Precipitation Estimation and Segregation Using Multiple Sensors) system was acquired to retrieve spatial rainfall data during the rainfall period from June 30 to July 6 in 2004 when Typhoon Mindulle and southwesterly flow struck Taiwan. The retrieved data were used for setting up the debris flow monitoring algorithm. With the aid of multiple platforms of meteorological observations, a rainfall threshold isohyet in a pilot area was mapped for debris flow monitoring. The rainfall monitoring algorithm based on QPESUMS provides more detailed information than the limited number of ground-based rainfall stations for interpreting the spatial distributions of rainfall events, and therefore is more suitable for debris-flow monitoring

Multi-objective optimization of typhoon inundation forecast models with cross-site structures for a water-level gauging network by integrating ARMAX with a genetic algorithm

“…QPESUMS forecasts future rainfall patterns by predicting the movement paths of cloud cells. Data are provided for a wide range of applications, including typhoon rainfall forecasts (Lee et al, 2006), river flooding forecasts (Vieux et al, 2003), and landslide forecasts (Chen et al, 2007).…”

Analysis of a major storm over the Dongjiang reservoir basin associated with Typhoon Bilis (2006)

“…Weather radar has significant advantage over rain gauge since it can obtain the spatial-temporal variations of the precipitation field (Chang et al 2008;Chen et al 2007;Morrissey et al 2004;Yu and Cheng 2008). In order to understand the major storm contribution to the precipitation distribution over the Dongjiang reservoir basin, our investigations tried to examine the radar reflectivity data of 1200-1800 UTC on July 14, 2006, at 1-km 2 spatial resolution and about 6-min temporal resolution to analyze the evolution of the major storm (Figs.…”

Rainfall‐induced landslides and debris flows

“…An effort is needed to seek for a unifying theory that can embrace the different behaviours described by experimentalists within a common framework. Regarding the spatial description of rainfall at large-scale, some promising contributions can arise from the growing availability of radar data Jakob et al, 2006;Chen et al, 2007;Yu et al, 2007), and local meteorological models. Notwithstanding, future research needs to account more explicitly for the orogenic effects on precipitation and for the recognition of small scale rainfall patterns.…”