Meteorological and earth observation remote sensing data for mass movement preparedness

Abstract

One of the major problems of the efficient synergetic use of remote sensing data for natural disaster mitigation is the fusion of various meteo- and geodata sets of significantly different spatial resolution. On the other hand, different morphological types of mass movements are based on alternate concepts i.e. of generation which, again has to be initially reflected in differing methodological approaches. The unifying idea and stronghold, however, of the presented approach is the precipitation parameters which trigger the debris flows. Albeit, frequently there are no relevant precipitation climatic data available. Despite significant drawbacks in the integration of Landsat and/or SPOT data sound hazard zonation maps can be generated. For a test area in the French Alps it has been shown that remote sensing data can be used to predict potential debris flow events. High temporal frequency remote sensing data from the Meteosat series of satellites allow the identification of cloud clusters most likely to result in intense rainfall which are, in turn, likely to initiate debris flow activity. Video evidence, field observations and an empirical debris flow model linked to an instantaneous rain gauge were used to ascertain the exact times of debris flow initiation. Due to the high spatial and temporal variability of rainfall in mountainous regions and the large areas vulnerable to debris flows compared to the coverages of these observations, there are, however, restrictions on the use of these data for large regions to provide early warning of debris flow events operationally. Additionally, the possible timeliness of warnings using such observations is restricted to the relatively short interval between the onset of the triggering phenomena and the hazard event. The remote sensing techniques developed in this study, allow warning of potential debris flow events to be derived before the triggering phenomenon occurs, by attempting to recognise the evolution if intense rain-bearing clouds. In this role the meteorological remote sensing data are not used to retrieve rainfall amounts but, instead, to derive rain cloud properties that produce debris flow triggering conditions.