TOPOGRAPHIC FEATURES
AND THE INITIATION OF DEBRIS FLOWS Chien-Yuan Chen1*, Wen-Jeng Lee2
ABSTRACT
A minimum topographic criterion to initiate debris flows originated from 11 creeks was studied. The topographic features were derived from 10 m resolution digital terrain data using spatial and 3-D analysis tools and terrain analysis in GIS. The topographic indices for sediment transport capacity index, hypsometric integral, elevation-relief ratio, stream power index, form factor, topographic wetness index, effective basin area, slope gradient, plan and profile curvatures were used for the analysis. The elevation-relief ratio is linear and positively correlated to the index of the hypsometric integral. The debris flows were initiated from steep slopes or landslide source areas with higher topographic wetness indices. The stream power index clearly changed after a debris flow. There is a minimum topographic features existed to initiate debris flow. The combination of the stream power index and the topographic wetness index provided a critical state line for before and after debris flows. The indices values depend upon the resolution of digital terrain data. A index combination both of the stream power index (SPI) and topographic wetness index (TWI) is proposed herein to identify the minimum topographic features necessary to initiate debris flow.
Key Words: Debris flow, Landslide, Geomorphic analysis, Topography, Digital terrain model
INTRODUCTION
There are 1,420 debris flow prone creeks in Taiwan (COA, 2005), categorized into three groups for high, medium, and low potential for debris flow by their topographic and geologic conditions (Lin et al., 2002; Lin et al., 2006b). Among these debris flow prone creeks, there were 685 landslide-induced debris flows. Climate change has caused an increase in the frequency of massive debris flows and landslides in recent years, especially in the mountainous areas in Taiwan (Chen et al., 2008). These disasters are attributed to the steep topographic characteristics and fragile geologic conditions of these areas, in addition to climate change.
Numerous methodologies may be used to model the downstream debris flow affected area. In spatial analysis by field topographic characteristics (Yu et al., 2006), the topographic characteristics after the debris flow are divided into three areas, the source, transportation, and deposition areas. Wichmann et al. (2007) modeled the debris flow initiation locations by
1 Associate Professor, Department of Civil and Water Resources Engineering, National Chiayi University, Chiayi City 60004, Taiwan, R.O.C. (*Corresponding Author; Tel.:+886-5-2717686; Fax:+886-5-2717693;
Email:chienyuc@mail.ncyu.edu.tw)
2 Associate Researcher, National Science and Technology Center for Disaster Reduction, Sindian City, Taipei County 231, Taiwan, R.O.C.
determining which cells have a high enough channel gradient in relation to discharge and exhibit a minimum sediment contributing area, using GIS. In the model, the locations of source sites are crucial and must be identified first. The topographic form controls the initiation sites of debris channel (Montgometry and Dietrich, 1994a; Vandaele et al., 1996) and a threshold relation exists between the topographic slope angle and the contributing area (Dietrich et al., 1992; Montgometry and Dietrich, 1994b). The topographic variables derived from 10-m digital elevation data show that debris flows slope angles > 32° and upslope contributing areas < 3,000 m2 were preferentially susceptible to debris flow (Godt and Coe, 2007).
The initiation of debris flow is related to the slope of the source areas, a typical value is between 27o and 38o (Rickenmann and Zimmermann, 1993; Hungr et al., 1984; Takahashi, 1981). In addition, there is a specified curvature in topography in debris flow source area (Heinimann et al., 1998; Bachmann, 2001). The geomorphic characteristics of debris flows show that their sediment transportation is related to the topographic features. One topography attribute used for identification of the topographic features is the sediment transport capacity index (LS) (Moore and Burch, 1986). The calculation of the LS value is given in the equation below:
LS = (m + 1)(As/22.13)m(sinβ/0.0896)n (1) where As is the specific catchment area (m2/m), β is the slope gradient (in degrees), and the values of m and n are 0.4 and 1.3 respectively (Moore and Wilson, 1992). The hypsometric integral (Hi) is identified as the integral value of the hypsometric curve for the elevation versus area in a basin and is expressed as a percentage (Willgoose and Hancock, 1998; Awasthi et al., 2002; Bishop et al., 2002). It is an important indicator of watershed conditions (Ritter et al., 2002). The hypsometric curve divides watersheds into three types:
un-equilibrium for Hi > 0.6 (young stage), equilibrium for Hi between 0.4～0.6 (mature stage), and Monadonock phase for Hi < 0.4 (Strahler, 1952). It has been shown mathematically that the elevation-relief ratio is identical to the hypsometric integral, but the latter has the advantage of being much easier to calculate within the GIS environment (Abebe1 and Foerch, 2006; Singh et al., 2008). The elevation-relief ratio (E) is defined as (Pike and Wilson, 1971):
E = (mean elevation – min elevation)/(max elevation – min elevation) (2) The stream power index could be used to identify the erosive effects of concentrated surface runoff (Wilson and Gallant, 2000). The stream power index (SPI) is defined as:
SPI = ln(As × tanβ) (3) Form factor (or shape factor) is the ratio of the minor axis to the major axis of the basin (Pareschia et al., 2002). The debris flow is highly correlated to the shape of the basin (Wan et al., 2008). A larger form factor has a rounder basin shape and larger peak flow rate (Wohl and Pearthree, 1991). The form factor is defined as:
F = W/Lo = A/Lo2 (4) where Lo is the length of the river, W is the average width of the basin (W = A/Lo), and A is the area of the basin. The topographic wetness index (TWI) has been used to describe the spatial soil moisture patterns (Kirkby, 1975; Beven and Kirkby, 1979; Wilson and Gallant, 2000). The locations of higher TWI host more favorable conditions for landslide formation (Conoscenti et al., 2008). It is a thus useful conditioning indice in landslide susceptibility studies (Gorum et al., 2008, Nefeslioglu et al., 2008). The TWI is defined as:
TWI = ln(As/tanβ) (5) A topographic characteristics analysis using Digital Terrain Model (DTM) for 11 creeks is studied for identification the geomorphic features of the basin and threshold conditions
necessary to initiate debris flows. The results may help to define debris flow potential and the sources of debris masses for disaster prevention.
SITE LOCATIONS OF THE STUDY CREEKS AND BACKGROUNDS
Eleven historical debris flow basins for the precision of the digital elevation model available in 10 × 10 m were selected in the study. Eight creeks in the sample, Nanpingkeng, Junkeng, Erbu, Shanbu, Chushui, Fengqiu, Heshe No. 1, and Heshe No. 3 are located in the Chenyoulan Basin in Nantou County in central Taiwan (Fig. 1). The Chenyouland Basin has been well documented (Lin et al. 2002; Chen et al. 2005; Lin et al. 2005; Lin et al.
2006a, 2006b; Yu et al. 2006; Wan et al., 2008). Two additional basins located in Taipei County in northern Taiwan, those of the Houtong and Chonghe creeks, were also used. In Taichung County in central Taiwan, the Songhe creek was selected.
Fig. 1 Site locations of the 11 study creeks in the Chenyoulan Basin (Nantou County), Taipei County, and Taichang County
Table 1 Database of the study debris flows (source COA, 2005)
Debris No. Length (m) Basin area (104 m2) A15_area (104 m2)a Stratum X (m) Y (m) Damageb
Chonghe 1950 124 68 sedimentary309839 2790517 H
Chushui 5755 863 524 sedimentary234517 2602913 H
Erbu 1871 158 96 metamorphic234685 2626755 H
Fengqiu 1929 162 162 metamorphic236644 2618531 H
Heshe No. 1 3770 311 161 sedimentary235394 2606789 H
Heshe No. 3 1509 163 163 sedimentary236571 2608149 M
Houtong 1541 191 173 sedimentary333158 2776496 H
Junkeng 1572 51 51 metamorphic235097 2627860 M
Nanpingkeng 1684 148 108 metamorphic235622 2629355 M
Shanbu 2442 334 147 metamorphic234408 2625719 H
Songhe No. 1 1339 371 339 metamorphic247514 2674952 H
a) area of slope over 15o in the basin b) H: high, M: middle
These creeks are stream flow type and debris flows were initiated from landslides. The slides were mainly triggered during typhoons Herb (1996), Xiangsane (2000), Toraji (2001), and
Mindulle (2004). The basic data of the 11 creeks are listed in Table 1.
METHODS AND APPROACH
The major input data for topographic feature analysis are the digital terrain models (DTMs) before and after debris flow. The DTMs were made by the Council of Agriculture (http://www.afasi.gov.tw/) in 10 × 10 m format from aerial photos taken in 1993 and 1996 before and after Typhoon Herb induced debris flows for the creeks in the Chenyoulan Basin.
The photos of Houtong and Chonghe creeks were made in 1994 and 2002, while the aerial photos of Songhe creek were taken in 2003 and 2004. Debris flows initiated from headwall landslide are categorized as “headwall,” “sidewall” when originating from sidewall landslides, and “headwall-sidewall” when sourced by both mechanisms. The DTMs were converted into raster grid images using spatial analysis in ArcGIS 9. Flow enforcement for the depression and flats for pre-processing of the terrain analysis was carried out first. Flow direction, accumulation, and specific contribution area were analyzed in the terrain analysis. The extractions of watershed and stream network were based on user-defined thresholds. Slope angle, upslope contributing area, effective basin area, plan and profile curvatures above the debris flow initiation points were compiled from the 10-m DTMs. Slope angles were computed using the SLOPE commands in the GRID module of ArcGIS 9. Upslope contributing areas were computed using the D-infinity method (Tarboton, 1997). The flowchart for the analysis is shown in Fig. 2.
The major terrain indices used and their relations to the debris flow are listed in Table 2. The calculations of the TWI and SPI indices were carried out in the raster calculator function of Geta GIS (SINICA, 2003). The hypsometric integral Hi is calculated by the trapezoidal numerical integration method using an avenue extension of ArcView 3.2 GIS from the ESRI support center (http://arcscripts.esri.com/). The sedimentary transport capacity index LS was calculated by SATEEC (Lim et al., 2005).
Fig. 2 Flow chart to identify the topographic features of the basins
The horizontal runout distance for each debris flow was measured in ArcGIS by manually digitizing a curvilinear line from the debris flow initiation point along the debris flow path to the deposition front with the aids of field Investigation and aerial images verification. The initiation point was identical to the fan apex by aerial photo interpretation and modified by the
changes of DTMs before and after debris flow. Fig. 3 shows the interpretation of debris flow source area, flow path, and deposit area by the difference of DTMs before and after debris flow in the Junkeng creek. The mapping of landslide head scarps is performed using the elevation change in the DTMs. The types of landslide curvature in plan and profile were decided by the elevation in the raster calculation function in GIS. Landslide volume is estimated by the erosion area multiplied by the digital elevation change in the basin, while debris volume delivered to the channel is the landslide volume minus the deposit volume in the landslide area.
Table 2 Major terrain indices and their relation to debris flow
Terrain indices Relation to debris flow
Relief ratio △H/L, debris flow mobility from landslide head scarp to debris deposition front
Elevation-relief ratio E, surface runoff and sediment yield prediction Stream power index SPI, the erosive effects of concentrated surface runoff Sediment transport capacity index LS, the sediment transport capacity
Effective basin area A15, the steep topographic characteristics in the basin Curvature (plan and profile) κ, debris flow source zone topographic feature
Topographic wetness index TWI, the movement of water in terrain slope or local drainage by downslope topography
Form factor F, the topographic feature of basin and peak flow rate
Fig. 3 the interpretations of landslide head scarp, debris flow source area, flow path, deposit area, and initiation point by the difference of digital elevation data before and after debris flow after Typhoon Herb in Junkeng creek (elevation unit: m)
TOPOGRAPHIC FEATURES ANALYSIS
The terrain indices are used for the debris flow source area, deposit area, and for the whole basin. Terrain attributes for plan and profile curvatures and TWI in the head landslide area were used. The basin area, slope, A15, △H/L, Hi, E, F, LS, SPI, and TWI indices were identified for the whole basin. Slope, runout distance, affected area, and magnitude were studied in the downstream deposit area. Table 3 lists the results of the topographic features for the creeks.
CONCLUSIONS AND DISCUSSION
The debris flow topographic features were discussed separately by landslide source area, the whole basin, and deposit area. The elevation changes of the topographic features were
analyzed and their relationship obtained by statistical analysis.
Topography features of source area
Debris flows were divided into three categories by the type of initiation: shallow landsliding, rilling, and the “firehose effect” in the alpine landscape (Godt and Coe, 2007). Most of the head source zones of the study creeks were initiated under planer plan and profile curvatures (Table 3). Field investigation verified that most of the head source landslides were shallow landslides. The field topographic curvature may not be the main feature in initiating landslides and subsequent debris flows. The debris flows were not attributed to either purely headwall or sidewall landslides. Instead, the interaction of head and sidewall slides created more abundant debris masses.
Table 3 Topographic features of the study basins Debris flow curvature
(plan/
profile)a
LS TWI (source /basin) (%)
SPI E F Hi (%) △H/L Landslide volume (m3)
Debris volume delivered to the channel (m3) Chonghe v/p 14.7 5.29/5.40 3.86 0.37 0.33 35.8 0.21 57,481 57,481
Chushui p/p 19.0 5.4/5.36 4.17 0.52 0.26 52.6 0.29 1,505,612 1 351,741 Erbu p/c 18.0 4.9/5.17 4.08 0.40 0.45 39.7 0.27 121,400 108,400 Fengqiu p/p 20.4 4.75/4.86 4.40 0.58 0.44 56.6 0.44 73,711 65,494 Heshe No. 1 p/p 18.5 4.9/5.08 4.18 0.46 0.22 45.0 0.33 47,703 39,590 Heshe No. 3 p/p 19.7 4.8/4.80 4.46 0.47 0.72 47.2 0.26 606,354 464,258
Houtong p/p 17.2 5.12/5.18 4.08 0.54 0.80 53.3 0.26 218,000 201,700 Junkeng p/p 17.5 5.21/5.26 4.01 0.48 0.21 49.1 0.35 6,947 4,526 Nanpingkeng c/p 17.1 5.1/5.29 3.97 0.50 0.52 47.4 0.32 104,100 102,900
Shanbu v/p 15.6 5.0/5.35 3.91 0.41 0.56 40.8 0.20 311,165 292,706 Songhe No. 1 p/p 17.6 4.75/4.80 4.46 0.45 2.07 45.1 0.25 522,000 408,000
a) c: concave, p: planar, v: convex
The slope of the source landslide areas is between 30o and 42o, slightly higher than the typical value of 27o to 38o (Rickenmann and Zimmermann, 1993; Hungr et al., 1984; Takahashi, 1981). A large TWI indicates a lower slope gradient and/or high catchment area in the valley floor area. Further, a large TWI features high slope gradient and very small catchment areas in the drainage divides (Conoscenti et al., 2008). The debris flows were initiated from a steeper slope or large topographic wetness index for the landslide area, especially for the Chushui creek (Fig. 4). The potential slope instability area, by its regional mean slope (SL) and the mean topographic wetness index (TWIL) of debris flow source area, can be estimated by:
TWIL = -0.04SL + 6.34, r2 = 0.95 (6) In the Chushui creek, the slope in the debris flow head landslide area had a steeper slope and a higher TWI than the other creeks. This shows that a steep slope and a higher TWI may feasibly generate a landslide.
Topographic features of the basin
The topographic features of the basins for their slope, basin area, A15, △H/L, Hi, E, F, LS, TWI, and SPI before debris flow were identified. These indices were not found to display obvious changes before and after the debris flows. The indices A15, △H/L, Hi, E, LS, and SPI all provided a minimum criterion to estimate the debris flow magnitude. The maximum LS in the creeks is about 45.4 with minor changes for different creeks, and the LS is proportional to
the SPI. The form factor F is not proportional to the other indices. The effective basin area A15
is proportional to the basin area. The elevation relief ratio E was found to be proportional to the hypsometric integral Hi (Fig. 5), and can be expressed as:
Hi (%) = 98E + 0.4, r2 = 0.97 (7) The linear positive correlations of Hi and E are identical to those of Abebe1 and Foerch (2006), and Singh et al. (2008).
28 32 36 40 44 48
Landslide source area mean slope (%) 4.6
4.8 5 5.2 5.4
Landslide source area mean TWI
Y = -0.04X + 6.34, r2 = 0.95 Sanpu
Junkeng
Chonghe Chushui
Songhe No.1 Heshe No.1 Erpu
Nanpingkeng Houtong
FengchouHeshe No.3
0.35 0.4 0.45 0.5 0.55 0.6
Elevation-relief ratio, E 35
40 45 50 55 60
Hypsometric integral, Hi (%)
Y = 98X + 0.4, r2 = 0.97 Sanpu
Junkeng
Chonghe
Chushui
Songhe No.1
Erpu
Heshe No.1 Nanpingkeng
Houtong Fengchou
Heshe No.3
Fig. 4 Mean landslide slope and mean topographic
wetness index TWIL in debris flow head landslide area Fig. 5 Elevation relief ration E versus hypsometric integral Hi
The basin mean slope (Sav in %) is proportional to the indices of the stream power index SPI, and the sediment transport capacity index LS, and is in inverse proportion to the topographic wetness index TWI (Fig. 6). The steeper slope is attributed to higher sediment transport capacity and higher SPI and LS indices. Lower SPI and larger TWI were found in the Chushui creek basin, compared to the same mean slope of basins. The relationships can be expressed as:
SPI = 0.04Sav + 2.73, r2 = 1.0 (8) LS = 0.33Sav + 6.83, r2 = 0.86 (9) TWI = -0.04Sav + 6.53, r2 = 1.0 (10)
24 28 32 36 40 44
Basin mean slope (%) 3.8
4 4.2 4.4 4.6
Stream power index, SPI
Y = 0.04X + 2.73, r2 = 1.0
Sanpu Junkeng
Chonghe
Chushui Songhe No.1
Erpu Heshe No.1
Nanpingkeng Houtong
Fengchou Heshe No.3 (a)
24 28 32 36 40 44
Basin mean slope (%) 14
16 18 20 22
Sedimentary transport capacity index, LS
Y = 0.33X + 6.83, r2 = 0.86 Sanpu
Junkeng
Chonghe
Chushui
Songhe No.1 Erpu
Heshe No.1
Nanpingkeng Houtong
Fengchou Heshe No.3 (b)
24 28 32 36 40 44
Basin mean slope (%) 4.7
4.9 5.1 5.3 5.5
Topographic wetness index, TWI
Y = -0.04X + 6.53, r2 = 1.0 Sanpu
Junkeng Chonghe
Chushui
Songhe No.1 Erpu
Heshe No.1 Nanpingkeng
Houtong
Fengchou Heshe No.3
(c)
Fig. 6 Basin mean slope and (a) stream power index SPI, (b) topographic wetness index TWI, (c) sediment transport capacity index LS
Both of the SPI and TWI are well correlated to the basin mean slope before debris flows (Fig.
6). The SPI versus TWI also shows a good fit relationship before and after the debris flow (Fig. 7) and can be expressed by:
TWI = -SPI + 9.3, r2 = 1.0 (before debris flow) (11) TWI = -0.39SPI + 7.0, r2 = 0.97 (after debris flow) (12)
One exception for Eq. (11) is in the Chushui creek, where the debris flow was initiated under
he SPI is an indicator of sedimentary transport capacity, while TWI is an alternative indicator a higher combination of TWI and SPI than the equation suggests. The steep streambed slope and underground water flow could explain the different topographic features that initiate debris flow (Chen et al., 2007b). Songhe No. 1 experienced visible topographic changes after the debris flow and its SPI value was far beyond the analysis range. The Junkeng creek showed fewer topographic indices changed after the debris flow. The TWI was marginally reduced and the SPI increased markedly after debris flow in the basins. The mean basin slope gradient (%) is 37.2 and 37.4 before and after debris flow, respectively, in the Chushui creek, and 40.4 and 40.7, respectively, in the Songhe creek. The reduced TWI may be due to the increased mean basin slope after debris flow. The resolution of DTM also affects the indices value. The TWI derived from 40 m DTM in the Houtong creek is changed from 5.2 to 8.1 and the SPI changed from 4.1 to 6.7. The indices values reduced markedly with the finer of DTM resolution.
T
for potential to landslide. The combination of both indices is meaningful in predicting landslide-induced debris flow. A new Debris Flow Topographic Index (DFTI) combining both of the SPI and TWI indices is thus proposed to identify the minimum topographic features to initiate landslide-induced debris flow in a basin. The critical line can be expressed by changing Eq. (11) to:
DFTI = SPI + TWI ≧ 9.3 (13) This equation represents a critical state line that reveals the minimum topographic features of a basin necessary to initiate debris flows.
3.6 4 4.4 4.8 5.2 5.6
Stream power index, SPI 4.8
5 5.2 5.4
4.7 4.9 5.1 5.3 5.5
Topographic wetness index, TWI
before debris flow after debris flow
Y = -1.0X + 9.3, r2 = 1.0
Y = -0.39X + 7.0, r2 = 0.97
initiation of debris flow before debris flow
after debris flow
Songhe No.1 Heshe No.3 Fengchou
Erpu
Heshe No.1 Houtong Nanpingkeng
Junkeng Sanpu Chonghe
Chus
Fengc .3
Erpu Nanpingke hui Chonghe
hou Chushui Heshe No
Junkeng Sanpu
ng
Houtong
Heshe No.1
Fig. 7 Stream power index SPI and topographic wetness index TWI before and afte bris flow
CKNOWLEDGEMENTS
inancial supports from the National Science Council of Taiwan under contract No. NSC
EFERENCES
r de
A F
95-2221-E-492-002 and National Chiayi University under contract No. NCYU 97T001-05-04-002 are appreciated.
R
Abebe, A. and Foerch, G. (2006). “Catchment characteristics as predictors of base flow index
Awas 02). “Land use change in
Bachm d wildland fire risk analysis. PhD thesis, Department of Beven, K.J. and Kirkby, M.J. (1979). “A physically based, variable contributing area model
Bisho (2002). “Geomorphic change in high
Chen, C.Y., Yu, F.C., Lin, S.C. and Cheung, K.W. (2007b). “Discussion of landslide
Chen, C.Y., Chen, L.K., Yu, F.C., Lin, S.C., Lin, Y.C., Lee, C.L., Wang, Y.T. and Cheung,
Chen (2005). “Study of Judging Debris Flow Occurrence by
COA (2005). Council of Agriculture, Republic of China. “The Map of Hazardous
Cono sis to assess landslide
Dietri ion
Godt, by a 28 July 1999
Gorum tion of
Heini
Hungr, O., Morgan, G.C. and Kellerhals, P. (1984). “Quantitative analysis of debris hazards
Kirkb . and
Lim, sed
Lin, G ment of risk due to debris flow events: A
Lin, ow run off simulation and
verification – case study of Chen-You-Lan watershed, Taiwan,” Natural Hazards and Earth System Sciences, 5: 439-445.
(BFI) in Wabi-shebele river basin, East Africa,” Prosperity and poverty in a globalised world-challenges for agricultural research. Conference on International Agricultural Research for Development. Tropentag, October 11-13, Bonn.
thi, K.D., Sitaula, B.K., Singh, B.R. and Bajacharaya, R.M. (20
the Nepalese watersheds: GIS and geomorphometric analysis,” Land Degradation and Development, 13: 495-513.
ann, A. (2001). GIS-base
Geography. University of Zurich, Zurich.
of basin hydrology,” Hydrol. Sci. Bull., 24: 43-69.
p, M.P., Shroder, J.F., Bonk, R. and Olsenholler, J.
mountains: a western Himalayan perspective,” Global and Planetary Change, 32:
311-329.
self-organized criticality and the initiation of debris flow,” Earth Surface Processes and Landform, 32: 197-209.
K.W. (2008). “Characteristics analysis for the flash flood-induced debris flows,”
Natural Hazards, 47: 245-261.
, W.F., Li, Y.H. and Wu, H.L.
combing the Physiographical Factors and Rainfall Condition Chen-Yu-Lan Stream Watershed as an Example,” Journal of Taiwan Geographic Information Science, 2:
27-44. (in Chinese)
Debris-Flow Streams in Taiwan,” Scale: 1/250,000. (in Chinese) scenti, C., Maggio, C.D. and Rotigliano, E. (2008). “GIS analy
susceptibility in a fluvial basin of NW Sicily (Italy),” Geomorphology, 94: 325-339.
ch, W.E.,Wilson, C.J., Montgomery, D.R., McKean, J. and Bauer, R. (1992). “Eros thresholds and land surface morphology,” Geology, 20: 675-679.
J.W. and Coe, J.A. (2007). “Alpine debris flows triggered
thunderstorm in the central Front Range, Colorado,” Geomorphology, 84: 80-97.
, T., Gonencgil, B., Gokceoglu, C. and Nefeslioglu, H.A. (2008). “Implementa
reconstructed geomorphologic units in landslide susceptibility mapping: the Melen Gorge (NW Turkey),” Natural Hazards, 46: 323-351.
mann, H.R., Hollenstein, K., Kienholz, H., Krummenacher, B., and Mani, P. (1998).
Methoden zur Analyse und Bewertung von Naturgefahren. Bundesamt fu¨r Umwelt, Wald und Landschaft (BUWAL), Bern.
for design of remedial measures,” Canadian Geotechnical Journal, 21: 663-677.
y, M.J. (1975). “Hydrograph modelling strategies,” In Peel, R., Chisholm, M Haggett, P. (Eds.). Process in physical and human geography, Heinemann, London.
K.J., Sagong, M., Engel, B.A., Tang, Z., Choi, J. and Kim, K.S. (2005). “GIS-ba sediment assessment tool,” Catena, 64: 61-80.
.F., Chen, L.H., and Lai, J.N. (2006a). “Assess
case study in central Taiwan,” Natural Hazards, 39: 1-14.
M.L., Wang, K.L. and Huang, J.J. (2005). “Debris fl
Lin, P.S., Lin, J.Y., Hung, J.C. and Yang, M.D. (2002). “Assessing debris-flow hazard in a watershed in Taiwan,” Engineering Geology, 66: 295-313.
Lin, P.S., Lin, J.Y., Lin, S.Y. and Lai, J. (2006b). “Hazard Assessment of Debris Flows by
Moore, I.D. and Burch, G.J. (1986). “Modelling erosion and deposition: topographic effects,”
Moor vised Universal Soil
New York, pp. 221-246.
Nefes andslide susceptibility mapping
gy, 43: 219-231.
Ritter, D.F., Kochel. R.C. and Miller, J.R. (2002). Process geomorphology. McGraw Hill,
Singh integral estimation methods
sources Management, 22: 1545-1560.
Statistical Analysis and GIS in Central Taiwan,” International Journal of Applied Science and Engineering, 4: 165-187.
Millard, T. (1999). “Debris Flow Initiation in Coastal British Columbia Gullies,” Forest Research Technical Report, TR-002 Geomorphology, September.
Trans. Am. Soc: Agr. Enqrx., 29: 1624-1640.
e, I.D. and Wilson, J.P. (1992). “Length-slope factors for the Re
Loss Equation: simplified method of estimation,” Journal of soil and water conservation, 47: 423-428.
Montgomery, D.R. and Dietrich, W.E. (1994a). “Landscape dissection and drainage area–slope thresholds,” In Kirkby, M.J. (Ed.), Process Models and Theoretical Geomorphology. John Wiley,
Montgomery, D.R. and Dietrich, W.E. (1994b). “A physically-based model for the topographic control of shallow landsliding,” Water Resources Research, 30: 1153-1171.
lioglu, H.A., Duman, T.Y. and Durmaz, S. (2008). “L
for a part of tectonic Kelkit Valley (Eastern Black Sea region of Turkey),”
Geomorphology, 94: 401-418.
Pareschia, M.T., Santacroceb, R., Sulpiziob, R., and Zanchetta, G. (2002). “Volcaniclastic debris flows in the Clanio Valley (Campania, Italy): insights for the assessment of hazard potential,” Geomorpholo
Pike, R.J. and Wilson, S.E. (1971). “Elevation-relief ratio, hypsometric integral, and geomorphic area-altitude analysis,” Bulletin of the Geological Society of America, 82:
1079-1084.
Rickenmann, D., and Zimmermann, M. (1993). “The 1987 debris flows in Switzerland:
documentation and analysis,” Geomorphology, 8: 175-189.
Boston.
, O., Sarangi, A. and Sharma, M.C. (2008). “Hypsometric
and its relevance on erosion status of North-Western Lesser Himalayan Watersheds,”
Water Re
SINICA (2003). “The introduction and applications of digital elevation models, GETA: A grid-based program for estimating terrain attributes V1.0”, operation manual, http://webgis.sinica.edu.tw/ies/. (in Chinese)
Takah ris flows and their hazardous zones; soft
Tarbo of flow directions and
Wan, S., Lei, T.C., Huang, P.C. and Chou, T.Y. (2008). “The knowledge rules of debris flow Strahler, A.N. (1952). “Hypsometric (area-altitude) analysis of erosional topography,”
Bulletin of the Geological Society of America, 63: 1117-1142.
ashi, T. (1981). “Estimation of potential deb
countermeasures for a disaster,” Journal of Natural Disaster Science, 3: 57–89.
ton, D.G. (1997). “A new method for the determination
contributing areas in grid digital elevation models,” Water Resources Research, 33:
309-313.
Vandaele, K., Poesen, J., Govers, G. and van Wesemael, B. (1996). “Geomorphic threshold conditions for ephemeral gully incision,” Geomorphology, 16: 161-173.
event: A case study for investigation Chen Yu Lan River, Taiwan,” Engineering Geology, 98: 102-114.
Wichmann, V., Heckmann, T., Haas, F. and Becht, M. (2007). „Modelling alpine sediment cascades: Process interaction and landscape connectivity,” Geophysical Research Abstracts, 9: 06140.
Willgoose, G. and Hancock, G. (1998). “Revisiting the hypsometric curve as an indicator of form and process in transport-limited catchment,” Earth Surface Processes and landforms, 23: 611-623.
Wohl, E.E. and Pearthree, P.P. (1991). “Debris flows as geomorphic agents in the Huachuca Yu, F
Wilson, J.P. and Gallant, J.C. (2000). Digital terrain analysis. In Wilson JP, Gallant JC (eds) Terrain analysis. John Wiley & Sons, New York, 1-27.
Mountains of Southeastern Arizona,” Journal of Geomorphology, 4: 273-292.
.C., Chen, C.Y., Chen, T.C., Hong, F.Y. and Lin, S.C. (2006). “A GIS process for debris flow potentially endangered areas delineating,” Natural Hazards, 37: 169-189.
