Presentation on theme: "指導教授：李錫堤 教授 學生：邱奕勛 報告日期："— Presentation transcript:
1 指導教授：李錫堤 教授 學生：邱奕勛 報告日期：100.6.16 The rainfall intensity-duration control of shallow landslides and debris flows : an updateGuzzetti, F., Peruccacci, S., Rossi, M., and Stark, C. P., The rainfall intensity-duration control of shallow landslides and debris flows: an update. Landslides, 5(1), 3-17.指導教授：李錫堤 教授學生：邱奕勛報告日期：
2 Outline Introduction Rainfall and landslide information Rainfall intensity–duration dataObjective identification of rainfall thresholdsGlobal rainfall ID thresholdsConclusions
3 Introduction (1/3)In 1980, Nel Caine published a paper entitled “The rainfall intensity–duration control of shallow landslides and debris flows”.I：rainfall intensity in millimeter per hour (mm/h)D：rainfall duration in hours (h)
4 Introduction (2/3)Rainfall ID thresholds are used to forecast the occurrence of shallow landslides and debris flows based on continuous rainfall measurements.To account for climatic effects, the rainfall intensity is normalized by the mean annual precipitation (MAP) and the rainy-day normal (RDN ,Wilson and Jayko 1997).Rainfall thresholds can be defined physically or empirically.Regional thresholds and Local thresholds.
5 Introduction (3/3)Rainfall and climate variables used in the literature for the definition of rainfall thresholds for the initiation of landslides
6 Rainfall and landslide information (1/6) The collected information was organized in a catalogue listing 2,626 landslide events covering the 89-year period between 1917 and 2005.areas for which rainfall characteristics resulting in shallow landslides
7 Rainfall and landslide information (2/6) Geographical precision for thelocation of the rainfall andlandslide events.Proportion of landslide eventsfor which antecedent rainfallinformation is available.
8 Rainfall and landslide information (3/6) Proportion of different shallowLandslide types.Number of reported landslidesfor each rainfall event.
9 Rainfall and landslide information (4/6) Accuracy of the temporalinformation.Information on the exactor approximate time, date, or period of failure.Information on lithology.
10 Rainfall and landslide information (5/6) Climate classification, based onthe Köppen climate classificationsystem.
11 Rainfall and landslide information (6/6) Frequency of rainfall events for classesof mean annual precipitation (MAP).Frequency of rainfall events for classesof average number of rain days (RDs)
12 Rainfall intensity–duration data (1/3) moving-average filtering48 hrPercentile estimates of rainfall IDconditions.Raw ID data.
14 Rainfall intensity–duration data (3/3) Climate information for the 15 climate regions considered in this work.
15 Objective identification of rainfall thresholds (1/1) In general, a rainfall threshold is drawn visually as the (apparent) lower boundary for the rainfall conditions.New methodA. threshold curve of the form was selected to describethe threshold.B. Bernoulli probability approach was used to find the scale (intercept) αand the shape (slope) β of the power law curve.
16 Objective identification of rainfall thresholds (2/2) C. Bayesian inferenceDie1/3Car accidentDo not die1/102/3DeadDie1/5Non-car accident9/10Do not die4/5WinBUGS package (http://www.mrcbsu.cam.ac.uk/bugs/)
17 Global rainfall ID thresholds (1/4) MAPGlobal rainfall ID thresholds for the initiation of shallow landslides and debris flows.Raw dataRDN0.25 mm/hr
18 Global rainfall ID thresholds (2/4) Validation of obtained ID and normalized-ID rainfall thresholds.RawdataMAPRDN
19 Global rainfall ID thresholds (3/4) Comparison with existing global thresholds1. Caine (1980)2 .Innes (1983)3 .Clarizia et al. (1996)4. Crosta and Frattini (2001)5. Cannon and Gartner (2005)6.Threshold inferred from the entire set of ID rainfall data (this work)7. Thresholds for two different rainfall periods(D < 48 h, and D ≥ 48 h)(this work)
20 Global rainfall ID thresholds (4/4) Climatically subdivided thresholds
21 Conclusions (1/1)New statistical treated outliers in an unbiased way and reduced interpretation errors.Collection of a large number of ID rainfall data through a literature search.Results of the normalization indicate that MAP and RDN did not reduce significantly.A global landslide warning system could contribute to mitigatelandslide risk.