Content uploaded by Alfredo Mahar Lagmay
Author content
All content in this area was uploaded by Alfredo Mahar Lagmay on Mar 26, 2016
Content may be subject to copyright.
131© Springer International Publishing Switzerland 2016
J.L. Drake et al. (eds.), Communicating Climate-Change and Natural Hazard
Risk and Cultivating Resilience, Advances in Natural and Technological
Hazards Research 45, DOI 10.1007/978-3-319-20161-0_9
Chapter 9
Shallow Landslide Hazard Mapping for Davao
Oriental, Philippines, Using a Deterministic
GIS Model
Ian Kaye Alejandrino , Alfredo Mahar Lagmay , and Rodrigo Narod Eco
Abstract Davao Oriental located at 7°30′N and 126°50′E is one of the many
landslide- prone provinces in the Philippines experiencing severe rainfall throughout
the year. With the increase in population and other infrastructural developments, it
is necessary to map the landslide potential of the area, to assure the safety of the
people and delineate suitable land for development. In order to produce rainfall-
induced shallow landslide hazard maps, Stability Index Mapping (SINMAP) was
used over a 5-m interferometric synthetic aperture radar (IFSAR)-derived digital
terrain model (DTM). SINMAP is based on the infi nite slope stability model.
Topographic, soil geotechnical, and hydrologic parameters (cohesion, angle of fric-
tion, bulk density, infi ltration rate, and hydraulic transmissivity) were assigned to
each pixel of the DTM with the total area of 5,164.5 km
2 to compute for the corre-
sponding factor of safety. The landslide hazard maps generated using SINMAP are
found to be accurate when compared to the landslide inventory from 2003 to 2013.
The landslide susceptibility classifi cation was translated to zoning maps indicating
areas that are safe from shallow landslides, areas that can be built upon with slope
intervention and monitoring, and the no-build areas. These maps complement the
structurally controlled landslide, debris fl ow, and other natural hazard maps that are
being prepared to aid proper zoning for residential and infrastructural
developments.
I. K. Alejandrino , B.S. (*)
Institute of Civil Engineering , University of the Philippines ,
Diliman , Quezon City , Philippines
Project NOAH, Deparment of Science and Technology , University of the Philippines ,
Diliman , Quezon City , Philippines
e-mail: ian_931@yahoo.com
A. M. Lagmay , Ph.D. • R. N. Eco , B.S.
National Institute of Geological Sciences , University of the Philippines, Diliman ,
Quezon City , Philippines
Project NOAH, Deparment of Science and Technology , University of the Philippines ,
Diliman , Quezon City , Philippines
132
Keywords Landslide • Hazard mapping • Deterministic model • Philippines •
Davao Oriental • Natural hazards
9.1 Introduction
Landslides triggered by rainfall pose signifi cant threat to human lives and property
in the Philippines . As the country’s population grows, landslides have become a
major concern to the safety of citizens. Unlike fl ooding which causes damage to
structures that more often can be fi xed, landslides may leave irreparable damage.
The province of Davao Oriental is classifi ed as one of the areas with high suscepti-
bility to landslides according to the Mines and Geosciences Bureau (MGB), the
government agency tasked to produce landslide hazard maps for the country. While
it is not advisable to locate communities or build infrastructure in areas identifi ed by
MGB as susceptible to landslides, displacement of people from their communities
and consequent loss of their source of livelihood have great social implications.
This calls for detailed landslide hazard maps that identify specifi c areas within an
existing community (i.e., municipality or village) to be used as part basis for proper
zoning of residential, industrial, and agricultural areas. By identifying safe zones
from detailed hazard maps, unnecessary relocation of communities and their conse-
quent ill effects may be avoided. Empirical and descriptive land models, because of
their easy implementation, have been developed for hazard analysis and landslide
monitoring and are applicable in a regional scale (Caine 1980 ; Cannon and Gartner
2005 ). Previously, MGB produced descriptive landslide hazard maps using a check-
list of factors to be assessed in the fi eld (see Table 9.1 ), which may contribute to the
mapping of landslide susceptibility. Figure 9.1 shows the 1:50,000 landslide hazard
map done by MGB for a part of Davao Oriental.
In recent years, physically based landslide models have been developed to assess
landslide hazard using a range of topographic, geologic, and hydrologic parameters
(Baum et al. 2008 ; Dietrich et al. 1995 ; Godt et al. 2009 ; Hong et al. 2007 ; Iverson
2000 ; Lu and Godt 2008 ; Ren et al. 2009 ; Wu and Sidle 1995 ). “Apart from the
process of understanding, the quality of data and data availability, the degree of suc-
cess of any particular research depends on the wise selection of statistical methods,
guided by the knowledge of the limits and strength of each method” (Felicisimo
et al. 2013 , p. 175).
Statistical models require landslide inventory as an input. In creating regional
maps for landslide susceptibility, the results of these models will incline toward the
variables assigned to with more records of landslides which can lead to biased
results, and the success of the model may vary for different locations due to the
quality of data available per location.
Deterministic models, on the other hand, do not require landslide inventories as
inputs but rather as a way to check and validate the acceptability of the model. The mod-
els rely on the spatial resolution of the terrain and the precision of the soil and geological
parameters used to simulate processes that contribute to the change in the stability.
I.K. Alejandrino et al.
133
Table 9.1 Sample fact sheet, methodology, and rating system in the determination of landslide
hazard susceptibility as used in the MGB geohazard mapping program (Open-source image from
MGB-UNDP 2004 )
Geohazard Levels of susceptibility
Low to absent Use infl ection point for separation of levels
Medium
High
Weights of evidence method, expert driven
1. Landslide inventory, by previous reports, aerial photo interpretation/ remote sensing images,
topographic map interpretation, with or without actual fi eld survey on budget and time
2. GIS
Minimum requirements for thematic map inputs
Geologic maps on (use lithology rather than formations) TMU, slope, faults, roads, gully
heads
Optional requirements (only if available) – landslide inventory, land use/land cover map, soil
map, vegetation map
For 1:50,000, expert driven, subject to fi eld verifi cation
For 1:10,000, data driven, fi eld mapping
Map calculations
Simple addition for thematic maps, uniform weights for all themes
After fi eld checking, the weights can be exchanged depending on the acquired fi eld data
Make histogram of the rated pixels and identify infl ection points to get the different
susceptibility levels
Buffer Distance Rating
1. Faults 0–50 km 3
51–100 2
>100 1
2. Roads 0–25 2
>25 1
3. Gully heads 0–25 2
>25 1
4. Slope 0–2 % 1
3–7 % 2
8–13 % 3
14–20 % 4
21–55 % 5
56–140 % 6
>140 % 7
5. Landslides Old 5
Active 7
6. Land cover Classify accordingly Rate accordingly
7. Lithology Classify accordingly Rate accordingly
8. TMU Classify accordingly Rate accordingly
9. Vegetation Classify accordingly Rate accordingly
9 Shallow Landslide Hazard Mapping for Davao Oriental, Philippines, Using…
135
A deterministic model can only address one type of landslide process. It is advis-
able to use a different model for each type (i.e., Debris-2D or Flow-R for debris
fl ows and COLTOP-3D and Matterocking for deep-seated or structural failure). For
the case of the Philippines , without a landslide inventory and a good coverage of
satellite imageries to construct one, the option to create a landslide hazard map for
the entire country using a statistical model cannot be completed. Having to resort on
deterministic models , studies were initially conducted on provinces that have good
quality of landslide inventories. Our study area Davao Oriental was one of those.
This study aims to generate a rainfall-induced shallow landslide hazard map for
Davao Oriental using a deterministic GIS model called Stability Index Mapping
(SINMAP). It is part of the fl agship program of the Philippine government for disas-
ter prevention and mitigation called Project Nationwide Operational Assessment of
Hazards (NOAH), which seeks to use the best available tools and scientifi c methods
for mitigating the impacts of natural hazard s (Lagmay 2012 ). The shallow landslide
hazard maps generated from SINMAP complement the deep-seated, structurally
controlled landslide and debris fl ow hazard maps that are also being simulated for
the province. These computer-simulated maps also complement the fi eld data gath-
ering effort by MGB using the empirical and descriptive methods of landslide map-
ping. By doing so, the landslide simulation output is verifi ed and calibrated by
MGB data for best results. It at the same time maximized utilization of newly avail-
able and high-resolution topographic data. Together with the storm surge and fl ood
hazard maps, these will be freely accessible through the Project NOAH website to
help in the disaster risk reduction and management efforts of the country.
9.2 Methodology
9.2.1 Stability Index Mapping (SINMAP)
SINMAP is an ArcView GIS extension and an objective terrain stability mapping
tool that complements other types of terrain stability mapping methods. The analy-
sis done by SINMAP is based upon the infi nite slope stability model which balances
the resisting components of friction and destabilizing components of gravity on a
failure plane parallel to the ground surface. It is implemented for shallow landslide
phenomena controlled by groundwater fl ow and convergence. It does not apply to
deep-seated instability including rotational slumps and deep earthfl ows (Montgomery
and Dietrich 1994 ).
Slope failures occur frequently during or following a continuous period of heavy
rainfall. The mechanism for rainfall-induced landslides is mainly by the infi ltration of
rainwater from an initial unsaturated state, which then causes a decrease in negative
pore pressure. This also leads to a decrease in effective normal stress acting through
the failure plane and reduces the available shear strength to a point where equilibrium
can no longer be sustained in the slope (Orense
2004 ). Thus, slope stability models
that consider hydrologic wetness and shear strength-related properties of soil and
topography of the region could possibly predict zones of mass movement in slopes.
9 Shallow Landslide Hazard Mapping for Davao Oriental, Philippines, Using…
136
The input data comprises the topographic slope, specifi c catchment, geotechnical
and hydrologic soil parameters, and climate . Topographic variables such as catch-
ment area and slope are computed from the digital terrain model (DTM). The DTM
used was generated from an airborne interferometric synthetic aperture radar
(IFSAR) survey of the Philippines and has 5-m horizontal resolution and 0.5-m
vertical accuracy.
SINMAP does not require numerically precise input and accepts other input
parameters in terms of upper and lower bounds on their range of values. The meth-
ods implemented rely on a grid-based data structure where geotechnical parameters
are assigned to each cell in the grid over the study area. The accuracy of the output
is therefore heavily reliant on the accuracy of the DTM used (Pack et al. 2005 ).
The primary output of the program is a stability index (SI) representing the pos-
sibility of landslide occurrence per cell. “Stability indices output by the analysis
should not be interpreted as numerically precise and are most appropriately inter-
preted in terms of relative hazard” (Pack et al. 2005 , p. 2). The SI is defi ned as the
probability of the location as being stable assuming uniform distribution of param-
eters over the uncertainty ranges. It is not capable of predicting when landslides will
occur but gives the location where it is most likely to take place. The most conserva-
tive of the SI values is defi ned as the factor of safety (ratio between stabilizing and
destabilizing forces) in a given location. If the factor of safety is less than 1, it is
then defi ned as the probability that the location is stable given the range of param-
eters used (Pack et al. 1998 ).
Table 9.2 represents the relative interpretation (upper threshold to stable) of the
range of SI values based on the breakpoints given by the program. The program also
suggests the relative degree of destabilizing factors required for stability and insta-
bility for each range of values. The term defended slope is used to classify regions
where the slope is held in place by forces not considered by the model. These may
Table 9.2 Stability class defi nitions (Open-source image by Pack et al. 2005 )
Condition Class
Predicted
state Parameter range
Possible infl uence of
factors not modeled
SI > 1.5 1 Stable slope
zone
Range cannot
model instability
Signifi cant destabilizing
factors are required for
instability
1.5 > SI > 1.25 2 Moderately
stable zone
Range cannot
model instability
Moderate destabilizing
factors are required for
instability
1.25 > SI > 1.0 3 Quasi-stable
slope zone
Range cannot
model instability
Minor destabilizing factors
could lead to instability
1.0 > SI > 0.5 4 Lower
threshold
slope zone
Pessimistic half of
range required for
instability
Destabilizing factors are
not required for instability
0.5 > SI > 0.0 5 Upper
threshold
slope zone
Optimistic half of
range required
for stability
Stabilizing factors may be
responsible for stability
0.0 > SI 6 Defended
slope zone
Range cannot
model stability
Stabilizing factors are
required for stability
I.K. Alejandrino et al.
137
be areas with existing slope protection, exposed bedrock layer, and other factors that
contribute to the stability of the slope.
For mapping purposes in the Philippines , the six classes are reduced to three
major hazard ratings with corresponding interpretations. Class 3 refers to areas of
low susceptibility where slope intervention is recommended, while Class 4 refers to
areas of moderate susceptibility where slope intervention and engineering interven-
tion are recommended. Lastly, Class 5 refers to areas of high susceptibility that are
recommended as no-build zones . The defended region was not included in the
reclassifi cation due to its nature, and the possible trigger for these regions is most
likely not governed by the analysis of the program.
9.2.2 Stability Index
SINMAP uses the following formula illustrated in Fig. 9.2 to calculate the stability
index based on the infi nite slope equation proposed by Hammond et al. ( 1992 ):
where:
FS = factor of safety
a = topographic catchment area
C = dimensionless cohesion
= (C
r + C s )/hpg
h = soil thickness
p = soil density
g = gravity constant
r = water density to soil density ratio
T = soil transmissivity
Φ= soil internal angle of friction
Fig. 9.2 Infi nite slope
model (Adapted from
open-source image
(Hammond et al.
1992 ) )
9 Shallow Landslide Hazard Mapping for Davao Oriental, Philippines, Using…
138
Variables a and θ are obtained from the DTM, while the rest of the geotechnical
and hydrologic parameters are manual inputs to the model. The combination of the
smallest C and angle of friction together with the largest R/T defi nes the most con-
servative analysis or worst-case scenario within the assumed variability in the input
parameters (Pack et al. 1998 ).
9.2.3 Climate Type and Soil Cover
The province of Davao Oriental falls under type II of Philippine Atmospheric,
Geophysical and Astronomical Services Administration (PAGASA) modifi ed
Coronas Climate Classifi cation Scheme. The rainy season prevails the whole year-
round with pronounced heavy rainfall during December.
Davao Oriental has soil cover which is mainly loam and sandy clay loam (see
Fig. 9.3 ) with a section of rough mountainous land that has unidentifi ed soil type.
The other soil types cover the relatively fl at sections and the coastline of the
province.
The area of Davao Oriental was divided into two sets of input parameters: the
upper portion’s (Camasan sandy clay loam, mountain undifferentiated soil, Bolinao
clay, and San Manuel silty clay loam) cells were assigned with parameter values
from Sandy Clay Loam Classifi cation, while the other part (Malalag loam, San
Manuel silty clay loam, and a small part of Bolinao clay) was assigned with param-
eter values from loam. Other soil covers, which were smaller in area, were not
refl ected in the simulations due to their small contribution in the total land cover
(Table 9.3 ).
9.2.4 Landslide Inventory
The landslide inventory was identifi ed from high-resolution satellite imagery from
2003 to 2013. However, the type of landslide, trigger, or threshold was not available
due to the lack of information on the exact dates when the landslides occurred.
Figure 9.4 shows the DTM and the location of the landslide inventory for the
province.
9.2.5 Process Flow
The outline for creating the landslide susceptibility map using SINMAP is shown as
a schematic diagram in Fig. 9.5 . The program extracts the topographic data from the
DTM, and the soil map is used as calibration regions for the input parameters.
I.K. Alejandrino et al.
139
Fig. 9.3 Soil map of Davao Oriental , Philippines , from Bureau of Soils
9 Shallow Landslide Hazard Mapping for Davao Oriental, Philippines, Using…
140
Table 9.3 Model calibration parameters
Soil density
Internal angle of
friction Cohesion
Approximate
T/R
kg/m 3 Min Max Min Max Min Max
Loam 2114 28 32 0 0.642 0.97 40.29
Sandy clay loam 1681 31 34 0 0.579 3.75 119.25
Fig. 9.4 IFSAR DTM and landslide inventory of Davao Oriental from various satellite imageries
I.K. Alejandrino et al.
141
9.3 Results
The SINMAP simulations reveal that the mountainous regions of Davao Oriental
are areas susceptible to medium to high hazards (see Fig. 9.6 ). Medium susceptibil-
ity is depicted as color orange in the hazard maps, while the high landslide hazard
susceptibility is shown as red. Most of the high landslide-susceptible areas are con-
fi ned to the very steep slopes and ravines. The fl at areas especially near the coast-
lines are in general devoid of any indication of landslide susceptibility. Table 9.4
shows the area percentage with high, moderate, and low susceptibility compared to
the total area of the DTM used in simulation. Landslide inventories were also used
to evaluate the accuracy and acceptability of the generated hazard map.
Compared to the existing 1:50,000 scale landslide hazard maps generated by
MGB, the maps produced from SINMAP simulations are more detailed and appear
to delineate the degree of landslide hazard better (see Fig. 9.7 ). Areas depicted in
the MGB map as high and moderate hazard are shown as very broad regions, while
Fig. 9.5 Schematic diagram of the methodology
9 Shallow Landslide Hazard Mapping for Davao Oriental, Philippines, Using…
142
those shown as high and moderate susceptibility in the SINMAP-generated hazard
maps are site specifi c and localized. More areas are found to be stable and safe from
landslide with the computer-simulated (SINMAP) hazard maps.
There are many deterministic model s that were available to use such as
SHALSTAB and TRIGRS, but we selected SINMAP because of its ease of use, and
it accepts a range of input parameters, which is desirable for regional scale mapping
because it allows room for uncertainty and nonuniform soil types. Though the
inventory may be used as an input to improve the program’s output, it was avoided
Fig. 9.6 Rainfall-induced shallow landslide hazard susceptibility map of Davao Oriental
I.K. Alejandrino et al.
143
to see how the program would perform for areas with low number of available land-
slide inventory. Also there were parameter adjustments or back calculations that
were done in other studies prior to ours to assign areas of landslide inventory to high
susceptibility areas but were not applied to our study since they would defeat the
purpose of the method being reproducible for other provinces of the country. Model
results assigned moderate and high hazard areas to cover around 65 % of the prov-
ince. In these areas fall 97.05 % of the total number of landslides observed from
high-resolution imagery from 2003 to 2013 (see Fig. 9.8 ). The remaining 35 % of
the province has little chance (around 3 %) of experiencing rainfall-induced shallow
landslides. From this we can say that the program has very good accuracy (98.31 %)
to determine possible unstable slopes.
Compared to the previous hazard map from MGB, the computer-simulated land-
slide hazard map was observed to be more detailed in terms of hazard delineation
and characterization of the degree of hazard present in various areas (see Fig. 9.7 ).
The classifi cation in the MGB, due to its descriptive nature, is very subjective to the
surveyor, and the assignment of the extent is greatly dependent on the coverage of
the investigation conducted in the fi eld. In comparison, SINMAP’s accuracy and
precision in identifying and delineating landslide hazards are greatly affected by the
resolution of the DTM used in the program (Pack et. al 2005 ). Higher-resolution
DTMs will likely yield more accurate results provided that the range of other input
parameters are realistic. Furthermore, the effects of land cover to rainfall infi ltration
and root cohesion were not considered in this simulation. Thus, the results may be
interpreted to be the worst-case scenario.
The applicability of the results as well as the classifi cation of the relative degree
of susceptibility depends on the intended use. In terms of identifying and detailed
delineation of rainfall-induced shallow landslide hazards over a wide area, the pro-
gram has produced very good results. The landslide inventory was found to be
highly consistent with the areas that were assigned with high and moderate suscep-
tibility (see Fig. 9.8 ). Even though almost all landslides in the inventory fall under
areas classifi ed as susceptible to landslides, the generated hazard maps are incom-
plete which require further analysis of landslides triggered by factors other than the
effects of rainfall on shallow soil surfaces. Hence, the SINMAP results are best used
Table 9.4 Summary of rainfall-induced shallow landslide susceptibility and landslide inventory
Stability classifi cation
Stable
Low
susceptibility
Moderate
susceptibility
High
susceptibility Total area
Area (km 2 ) 1324.40 431.34 2512.26 863.88 5131.88
Percentage of
the area
25.81 8.41 48.95 16.83 100.00
Number of
landslides
4 3 77 153 237
Percentage of
landslides
1.69 1.27 32.49 64.56 100
9 Shallow Landslide Hazard Mapping for Davao Oriental, Philippines, Using…
144
Fig. 9.7 Comparison between MGB landslide hazard map and SINMAP-simulated susceptibility map
I.K. Alejandrino et al.
145
with other landslide susceptibility maps such as those that depict debris fl ow and
deep-seated or structurally controlled landslide hazards.
9.4 Conclusions and Recommendations
The consideration of land cover can give a signifi cant effect on the classifi cation of
shallow landslide hazards. Depending on the land cover, it will have varying infl u-
ence on degrees of root cohesion and rainfall infi ltration. A site investigation to
verify the range of values of the input parameters will help improve the assessment
for critical areas in a smaller scale such as inhabited areas and proposed areas for
development. It was noted by Witt ( 2005 ): “Neither model [SinMap nor ShalStab]
takes into account antecedent moisture nor the effect that geologic structure can
have on concentrating groundwater fl ow” (p. 120).
Since we are mapping landslide hazard for regional scale, we opt not to address
the effect of geologic structure on concentrating groundwater fl ow. A geological
survey or validation may be done to incorporate its effects for critical areas (i.e.,
populated areas). Antecedent moisture or rainfall intensity and duration may be
more important to consider with the goal of creating an early warning system that
assesses stability in actual events. Though SINMAP has proven to be a good tool for
Fig. 9.8 Verifi cation using landslide inventory
9 Shallow Landslide Hazard Mapping for Davao Oriental, Philippines, Using…
146
detailed planning over a wide region, programs that consider transient dynamical
response of the subsurface moisture to spatiotemporal variability of rainfall in com-
plex terrains (e.g., TRIGRS - Transient Rainfall Infi ltration and Grid-Based Regional
Slope-Stability Model by Baum et al. 2008 and SEGMENT-landslide by Ren et al.
2008 ) are of great signifi cance especially to communities currently residing on the
identifi ed hazard areas and for motorists traveling across roads that are constructed
on steep slopes.
The MGB maps, if followed strictly, necessitate the relocation of communities or
development of areas as not susceptible to landslides. Given the broad areas identi-
fi ed as landslide susceptible in the MGB maps, this can mean large-scale displace-
ment of people from their communities and consequent loss of their source of
livelihood which have great social implications. With the simulated landslide haz-
ard maps that detail specifi c areas within an existing community (i.e., municipality
or village), safe zones are identifi ed and can be used as settlements, thereby avoid-
ing unnecessary relocation of communities and their consequent ill effects. The
computer-simulated maps complement the MGB fi eld data gathering efforts, which
use the empirical and descriptive methods of landslide mapping. The MGB data can
be used to verify and calibrate the landslide simulations to produce more detailed
and accurate maps for safe development planning of communities.
Acknowledgments We would like to thank the creators of SINMAP (Pack et al.) for making this
program available to the research and development community and communicating with us in the
early stages. We also thank the National Mapping and Resource Information Authority (NAMRIA)
for the IFSAR DTM used in this simulation. Funding for the project titled Enhancing Landslide
Hazard Maps Through LIDAR and Other High Resolution Imageries is from the Department of
Science and Technology (DOST), government of the Philippines .
Other Notes DOST Project NOAH is a program implemented by the Philippine government to
assess the different hazards present in the Philippines . Assessment of fl ood, landslide, and storm
surge hazards is part of the program. Completed maps are to be added to the NOAH website
(
www.noah.dost.gov.ph ) for free access to the general public to aid in the information dissemina-
tion to reduce effects of meteorological hazards in the country. The website, in partnership with
PAGASA , also displays various weather sensors and visualizations to aid in the understanding of
weather data.
References
Baum RL, Savage WZ, Godt JW (2008) TRIGRS—A Fortran program for transient rainfall infi l-
tration and grid-based regional slope-stability analysis, version 2.0: U.S. Geological Survey
Open-File Report, 2008–1159, p 75
Caine N (1980) The rainfall intensity: duration control of shallow landslides and debris fl ows.
Geogr Ann Ser A Phys Geogr 62:23–27. doi:
10.2307/520449
Cannon SH, Gartner J (2005) Wildfi re related debris fl ow from a hazards perspective. In: Debris-
fl ow hazards and related phenomena, Springer-Praxis books in geophysical sciences. Springer,
Berlin, pp 321–344. doi:
10.1007/3-540-27129-5_15
I.K. Alejandrino et al.
147
Dietrich WE, Reiss R, Hsu ML, Montgomery DR (1995) A process-based model for colluvial soil
depth and shallow landsliding using digital elevation data. Hydrol Process 9:383–400.
doi:
10.1002/hyp.3360090311
Felicisimo M, Cuartero A, Remondo J, Quirós E (2013) Mapping landslide susceptibility with
logistic regression, multiple adaptive regression splines, classifi cation and regression trees, and
maximum entropy methods: a comparative study. Landslides 10(2):175–189. doi:
10.1007/
s10346-012-0320-1
Godt JW, Baum RL, Lu N (2009) Landsliding in partially saturated materials. Geophys Res Lett
36:L02403. doi:
10.1029/2008GL035996
Hammond C, Hall D, Miller S, Swetik P (1992) Level I Stability Analysis (LISA) Documentation
for Version 2.0 (General technical Report INT-285).
http://forest.moscowfsl.wsu.edu/cgi-bin/
engr/library/searchpub.pl?pub=1992a . Accessed 16 Jun 2014
Hong Y, Adler RF, Huffman GJ (2007) Use of satellite remote sensing data in the mapping of
global landslide susceptibility. Nat Hazards 43(2):245–256. doi:
10.1007/s11069-006-9104-z
Iverson RM (2000) Landslide triggering by rain infi ltration. Water Resour Res 36(7):1897–1910.
doi:
10.1029/2000WR900090
Lagmay AMF (2012) Disseminating near-real time hazards information and fl ood maps in the
Philippines through web-GIS. Project NOAH Open File Reports. V1:21–36: ISSN-23627409
Lu N, Godt JW (2008) Infi nite slope stability under steady unsaturated seepage conditions. Water
Resour Res 44:W11404. doi:
10.1029/2008WR006976
MGB Geoportal (2010) Landslide hazard map of Davao Oriental, Mines and Geosciences Bureau
Geoportal.
http://gdis.denr.gov.ph/mgbviewer/ . Accessed 4 Jun 2014
MGB-UNDP (2004, March) Manual for standardization, geohazard mapping program. Mines and
Geosciences Bureau project, funded by the United Nations Development Program, Montgomery
DR, Dietrich WE (1994) A physically based model for the topographic control on shallow
landsliding. Water Resour Res 30(4):1153–1171. doi:
10.1029/93WR02979
Orense R (2004) Slope failures triggered by heavy rainfall. Philipp Eng J 25(2):73–90.
http://
journals.upd.edu.ph/index.php/pej/article/view/2343 . Accessed 5 Apr 2014
Pack RT, Tarboton DG, Goodwin CN (1998) Terrain stability mapping with SINMAP, technical
description and users guide for version 1.00, Report number 4114-0, Terratech Consulting,
Salmon Arm, BC, Canada
Pack RT, Tarboton DG, Goodwin CN, Prasad A (2005) SINMAP 2- A stability index approach to
terrain stability mapping. Utah State University.
http://hydrology.usu.edu/sinmap2/ . Accessed
5 May 2013
Ren D, Leslie LM, Karoly D (2008) Mudslide risk analysis using a new constitutive relationship
for granular fl ow. Earth Interact 12:1–16. doi:
http://dx.doi.org/10.1175/2007EI237.1
Ren D, Wang J, Fu R, Karoly D, Hong Y, Leslie LM, Fu C, Huang G (2009) Mudslide-caused
ecosystem degradation following Wenchuan earthquake 2008. Geophys Res Lett 36:L05401.
doi:
10.1029/2008GL036702
Witt AC (2005) Using a GIS (Geographic Information System) to model slope instability and
debris fl ow hazards in the French Board river watershed. Thesis, North Carolina State
University.
http://repository.lib.ncsu.edu/ir/bitstream/1840.16/552/1/etd.pdf . Accessed 6 Oct
2014
Wu W, Sidle RC (1995) A distributed slope stability model for steep forested basins. Water Resour
Res 31(8):2097–2110. doi:
10.1029/95WR01136
9 Shallow Landslide Hazard Mapping for Davao Oriental, Philippines, Using…
