Natural Hazards (2024) 120:2179–2231

https://doi.org/10.1007/s11069-023-06277-3

REVIEW ARTICLE

Risk assessment and management of rainfall‑induced

landslides in tropical regions: a review

M. P. Amarasinghe1,2 · S. A. S. Kulathilaka2 · D. J. Robert1 · A. Zhou1 ·

H. A. G. Jayathissa3

Received: 29 May 2023 / Accepted: 18 October 2023 / Published online: 26 November 2023
© The Author(s), under exclusive licence to Springer Nature B.V. 2023

Abstract
Excessive rainfall is considered the major landslide triggering mechanism, especially in
tropical climate regions. During rainfall, water infiltrates into the subsurface; reducing the
matric suction, increasing pore water pressure, and decreasing the shear strength of the
soil. The prevailing unfavourable ground and geomorphological conditions can further
exacerbate the vulnerability and severity of catastrophic landslides. Hence, it is vital to
identify different landslide mechanisms, key drivers for rainfall-induced landslides, and
risk assessment methods for adopting appropriate failure mitigation strategies. This study
captures a comprehensive review and in-depth analysis based on 200 articles published in
literature including authors own case studies to describe the risk management strategies of
rain-induced landslides in tropical countries. First, a clear relationship between the rainfall
patterns and the landslide events has been proposed through the comprehensive data sets
reviewed. Then key influencing factors for landslides in the tropical region have been iden-
tified with in-depth discussion from past reported studies. Moreover, landslide risk assess-
ment and management framework are discussed with the key steps involved. The frame-
work provides a better-structured approach to discuss on identifying, analysing, evaluating,
and managing risk associated with landslides. The complex geological conditions, lack of
rainfall and impact data, and rapid change in land use make quantitative risk assessment
challenging in the tropical region. The review finally recommends effective risk mitigation
strategies from the authors’ experience on past projects and reported literature case studies.
The outcomes from the review are beneficial for engineers and authorities for adopting risk
mitigation approaches in tropical regions.

Keywords Landslides · Rainfall-induced · Tropical · Risk assessment · Risk management ·

Mitigation

1 Introduction

Landslides are mass movements of rocks, soils or debris and one of the most frequently
occurring natural hazards that affect human lives, properties, and the economy (Carrión-
Mero et al. 2021). Landslides are responsible for about 17% of the annual average natural

Extended author information available on the last page of the article

13
Vol.:(0123456789)
2180 Natural Hazards (2024) 120:2179–2231

disaster losses from 1980 to 2013, which is about US $20 billion annually (Haque et al.
2016; Ozturk et al. 2022). It is expected that with climate changes and urbanisation around
the globe, landslide incidents will further increase (Gariano and Guzzetti 2016; Ozturk
et al. 2022). Rainfall can be considered as the major triggering factor for landslides. About
90% of the casualties due to landslides in the world result from rainfall-induced landslides
(Haque et al. 2016). In the recent past, an increase in rainfall-induced landslides has been
observed due to climate changes, increase in population, urbanisation, infrastructure devel-
opment, and land use changes (Petley 2010). However, the effect of climate change on
landslide occurrence has been debated in the recent past (Crozier 2010; Manchado et al.
2021; Picarelli et al. 2021). Although the frequency of landslides is increasing, the lack of
long-term trends of landslide activity has been a critical issue in distinguishing between
the effect of climate change and other anthropogenic drivers. Nevertheless, the studies of
Cendrero et al. 2020; Crozier 2010; Huggel et al. 2012; Manchado et al. 2021; Ozturk et al.
2022; Puente-Sotomayor et al. 2021; Quesada-Román et al. 2021; Scheidl et al. 2020 have
shown that the effect of human activities has a more significant influence on the increase in
landslide frequency than climate change.
Tropical climate regions usually consist of high average annual rainfall, high average
annual temperature and low seasonal thermal variations (Dewitte et al. 2022). These char-
acteristics contribute to the wet and dry climatic conditions that exist in these parts of the
world. Rather than limiting the study confined to the tropical region, which is defined as
the area between the Tropic of Cancer (23.5° North) and Tropic of Capricorn (23.5° South)
in this study, the region between 30° North and 30° South latitudes covering some part of
the subtropical region also considered allowing to examine the larger geographical area
with shared climate characteristics. This provides a more comprehensive understanding
and practical application of the outcomes for the areas with consistent climate attributes
in the extended tropical zone. More than 80% of rainfall-induced landslides are taking
place in tropical countries. From the review of Maes et al. (2017), 99 landslide-prone coun-
tries have been identified in the tropical region. Rapid urbanisation in the tropical region
which is characterised by the growth of cities and an increase in population, can increase
the strain on infrastructure and resources (García-Soriano et al. 2020; Quesada-Román and
Campos-Durán, 2023). This can lead to the expansion of cities towards steeper sloping
areas, increasing the risk of landslide hazards (Ozturk et al. 2022). The impact of land-
slides has been high in the Global South, predominantly in the developing tropical coun-
tries considering the high economic, social, political and cultural vulnerability (Alcántara-
Ayala 2002; Quesada-Román and Campos-Durán 2023).
Although it is clear that the stability of the landslide is compromised by the triggering
event, the underlying causes or the preparatory factors acting simultaneously with the trig-
gering factors result in the instability (Casagli et al. 2017; McColl 2015). Studying these
preparatory and triggering factors of landslides is important to understand the key mecha-
nisms of slope failure and to evaluate the hazard it can generate (Dille et al. 2019). The
discontinuities in the rock or soil mass have been identified as one of the major indicators
of a zone of weakness where the material strength is reduced and facilitate other factors
favourable for landslides such as weathering of material and slope hydrology modifica-
tion (Casagli et al. 2017; Dille et al. 2019; McColl 2015; Scott and Wohl 2019). Tropi-
cal regions exhibit increased susceptibility to landslides considering the unique geological
characteristics associated with the climate. Weathering is predominant in tropical climates
compared to other parts of the world due to the high temperature, high humidity and the
presence of organic acids derived from soil and vegetation (Duzgoren-Aydin and Aydin
2006; Oilier 2010), leading towards deeper weathered layers (Migoń 2010; Thomas 1994).

13
Natural Hazards (2024) 120:2179–2231 2181

Nevertheless, the influence of weathering of rock on slope instability in the tropics is rarely
investigated. Furthermore, the weakening of the slope due to seismic activities and subse-
quent slope failures due to intense hydro-meteorological events have also been identified
in the tropical region (Chen et al. 2019; Quesada-Román et al. 2019). Less attention has
been given within the scientific literature to the landslide processes in the tropics compared
to the other regions where the significance of tropical systems for landslides is underes-
timated and has led to undervaluing the possible hazard in the region (Dille et al. 2019;
Gariano and Guzzetti 2016; Maes et al. 2017; Thomas 1994). Understanding the factors
influencing landslides in tropical regions is essential for effective susceptibility assessment.
However, many studies in the area of landslide susceptibility in the region is limited to
basic approaches due to lack of data, poor quality or non-availability of the data in the
appropriate format (Puente-Sotomayor et al. 2021).
Similar to identifying causes, it is important to evaluate the hazard, risk and potential
risk management strategies, which can be crucial in reducing the negative consequences of
landslide events. For this, understanding the potential impact and consequences are essen-
tial. The limited availability of data on rainfall-induced landslides and quality remains a
significant limitation hindering the accurate assessment and decision-making process in
the tropical region (Quesada-Román, 2021). Significant research has been conducted on
landslide susceptibility, hazard and risk assessment all over the world (Aleotti and Chowd-
hury 1999; Dahal and Dahal 2017; Dai et al. 2002; Hungr 2016; Lacasse and Nadim 2009;
Luna et al. 2014). Both quantitative and qualitative methods can be used in landslide haz-
ard assessment while quantitative methods are much more detailed and accurate compared
to qualitative methods. A large amount of historical data and resources are required for
quantitative analysis, which can be challenging to obtain in many areas (Corominas et al.
2014; van Westen et al. 2008). Additionally, it requires significant expertise to implement
as these quantitative methods are complex. In industrialised countries, quantitative risk
assessment is gradually becoming a trend, but considering the fact that most of the tropical
countries are still developing, with limited resources and economic constraints, qualitative
methods are adopted, and quantitative methods are rarely used (Arroyo-Solórzano et al.
2022; Turner 2018). Moreover, these are focused on landslide susceptibility rather than
hazard evaluation. Furthermore, the effectiveness of the implemented risk management
measures is significantly lacking in the scientific literature (Maes et al. 2017).
This article aims to consolidate the existing knowledge on landslide risk assessment and
management in the tropical region. As a preliminary to risk assessment and management
of landslides in the tropical area, the relationship between rainfall patterns and landslides
in the region was discussed. This was followed by an extensive review of the causative fac-
tors considering complex geological conditions and hydrology for tropical regions. These
are broad topics studied in the field of landslides and extensive literature available over the
years. However, the authors have critically evaluated and summarised the seminal contri-
butions in these areas. Next, each component of risk assessment and management accord-
ing to the framework by Fell et al. (2005) was evaluated from the literature and from the
authors’ past experience on projects and case studies. Here the susceptibility and hazard
assessment, consequence analysis, risk estimation, risk evaluation, and risk management
are considered in detail. Although there is no significant difference in the risk assessment
techniques between tropical regions and other regions, as the fundamental principles and
methodologies remain consistent regardless of the location, a critical review of the tech-
niques that can be used is summarised in this article. Special attention was given to land-
slide remediation measures and landslide monitoring and early warning under risk man-
agement. A review of several case studies from Sri Lanka for landslide risk reduction was

13
2182 Natural Hazards (2024) 120:2179–2231

also presented which are good examples of the cost-effective risk mitigation strategies used
in developing countries in the tropical region, considering site-specific conditions such as
geology, geomorphology, and hydrology.

2 Significance of the review

Substantial efforts have been made over the past five decades to classify landslides, identify
the mechanism of landslides and the causative factors. In addition, techniques have been
developed to identify and evaluate landslide susceptibility, hazard, and risk. Furthermore,
landslide risk management has also been a key area of study. Although reviews have been
conducted on risk assessment and risk management studies, comprehensive reviews on the
rainfall-induced landslide risk assessment and management, along with the key driving
factors prevailing in tropical countries are limited. Firstly, this review focussed on identify-
ing the relationship between landslides and rainfall seasonality. Then, the causative factors
for landslides are summarised by analysing data from the reported literature. Special atten-
tion was given to the unfavourable geological conditions and hydrology associated with
landslides in the tropical region. Limited studies have dealt with the complex subsurface
conditions prevailing in tropical countries and the effect of heterogeneous behaviour due
to weathering. Additionally, there appears to be a limited number of studies on the specific
effect of rainfall on landslides in colluvium soils, which makes this aspect of the research
important and relevant.
In this article, efforts have been made to explain these different aspects with examples.
Mainly examples from Sri Lanka have been explained along with some other tropical
countries in the study. Sri Lanka was chosen as a representative tropical country consid-
ering its high susceptibility to landslides and highly complex subsurface conditions. Sri
Lanka is among the top five countries with the highest number of landslide records per
1000 ­km2 and among top ten countries in terms of the number of landslide fatalities per
million inhabitants. Almost all the landslides in Sri Lanka are triggered by rainfall. The
country is far away from the Indo-Australian plate boundaries and have only witnessed
minor tremors during the history. Therefore, landslides in Sri Lanka are well suited for the
study of rainfall-induced landslides in the tropics. This makes it an ideal location to study
the effect of ground conditions and effect of hydrology on the landslide failure mechanism
in tropical regions.
Next, the review is extended for the risk assessment and risk management of rainfall-
induced landslides along with the engineering solutions for risk treatment, monitoring and
early warning measures. Because the fundamental principles and methodologies do not
vary with the geographical location, there is no significant variation in the risk assessment
techniques in the tropical region from other areas. However, the authors have summarised
the current methods in landslide risk assessment to ensure the comprehensiveness of the
article. Based on the landslide risk treatment measures conducted in the tropical region,
the engineering solutions for landslide mitigation, monitoring and early warning measures
adopted in developing countries have been summarised in the article. These case stud-
ies provide a clear understanding of the requirement of different site-specific rectification
measures to control the risk at acceptable or tolerable levels. The examples and case studies
are carefully tailored mainly focussing Sri Lanka considering the availability and access to
data for the authors’ as opposed to other similar tropical countries. This allows the authors
to evaluate the effectiveness of mitigation measures based on the monitoring data available

13
Natural Hazards (2024) 120:2179–2231 2183

Fig. 1  An overview of the review (numbers in parentheses refer to the corresponding section in the paper)

and in-depth analysis as the background of the landslide case study is known. An overview
of this review is illustrated in Fig. 1.

3 Relationship between rainfall and landslide occurrence

in the tropical region

According to Froude and Petley (2018), rainfall-induced landslides are scattered across the
world with the majority within tropical and subtropical regions of high relief as observed
from the data (Fig. 2) of the Global Fatal Landslide Database (GFLD). Asia recorded the
majority of rainfall-triggered landslides (75% of the total landslides) between 2004 and
2016. Out of the 1534 recorded rainfall-triggered fatal landslides between 2003 and 2009,
44% occurred in South Asia, 30% in East Asia and 26% in Southeast Asia (Petley 2010).
In Asia, the fatal incidents are clustered around the southern Himalayan arc, along the
Southwest coast of India and Sri Lanka, along the southern and eastern coastal regions of
China, along the western edge of the Philippines sea plate, in the Indonesian islands, coun-
tries such as Bangladesh, Laos, Myanmar and Malaysia (Petley 2012).The variation of the
monthly percentage of fatal landslide events with the rainfall distribution for Central Amer-
ica, South America, the Caribbean, South Asia, Southeast Asia and Sri Lanka is shown in
Fig. 3. The majority of rainfall-triggered landslides in Asia occur between May to Octo-
ber, peaking in July. In South Asia, the monsoon climate controls the landslide occurrence,
especially the Southwest monsoon. Although there is a seasonality of landslide occurrence
in East Asia, tropical cyclones affect landslides significantly. Petley (2010) states that there
is no seasonal variation dominant in Southeast Asia.
In contrast, Lee et al. (2014) have reported that most of the landslide events in
Malaysia have occurred in the months of May and November, during which the high-
est rainfalls are recorded in the country. The two monsoons, Southwest and Northeast

13
2184 Natural Hazards (2024) 120:2179–2231

Europe, Turkey, and Iran cluster

0
(35 N)

(300 N)

0
(30 S)
0
(35 S) Asian cluster

Fig. 2  Rain-induced landslide distribution from the GFLD (modified after Froude and Petley 2018)

also begin in these two months for this region, confirming the seasonality of land-
slides. The fatal landslide events in South America, Central America and the Carib-
bean also follow seasonality with the annual rainfall pattern, as most of the landslide
events are climate-oriented (Fig. 3a–c). In Central America and the Caribbean islands,
the hurricane season affects landslide occurrence, and in South America, it is during
storms that most of the landslides occur (Sepúlveda and Petley 2015). In the multi-
annual time scale, climatic oscillations altering the precipitation in tropical climate
regions can occur due to the coupled atmosphere–ocean phenomena El Niño-South-
ern Oscillations. These changes to rainfall patterns can affect tropical monsoons and
cyclones in the Pacific Ocean and Indian Ocean. There is a strong correlation between
the likelihood of landslides and the El Niño/La Niña cycles in certain regions, such as
Latin America and some parts of Southeast Asia and India, due to the extreme rainfalls
received (Emberson et al. 2021; Sepúlveda and Petley 2015).
Understanding the relationship between landslides and rainfall patterns is important
for risk management and early warning. Identifying the seasonality of landslide occur-
rence with rainfall for a region allows for identifying periods of increased landslides
and prioritising vulnerable areas and resource allocation. Based on the authors’ experi-
ence with landslides in Sri Lanka and the data available, it is evident that most of the
landslides occur in the months of May and June due to heavy rainfalls received during
the Southwest monsoons (Fig. 3g). However, when the average monthly rainfall for the
country is considered, the rainfall is higher for the months of October and November
than in May as these months witness heavy rainfall not only in hilly regions but also in
most parts of the country. This highlights two important points. Firstly, it is crucial to
establish rainfall thresholds for landslide early warnings specifically for the local area
rather than solely relying on national-level rainfall thresholds. Secondly, continuous
records of landslide occurrence along with the triggering rainfalls must be collected
systematically to better predict landslides, which is lacking in many countries in the
tropical region.

13
Natural Hazards (2024) 120:2179–2231 2185

Fig. 3  Variation of the monthly percentage of fatal landslide distribution and mean monthly rainfall for a
Central America, b South America, c Caribbean, d South Asia, e East Asia, f Southeast Asia and g Sri
Lanka. Data are sourced from Froude and Petley (2018); Petley (2010); Sepúlveda and Petley (2015);
UNISDR (2022)

4 Landslide causative factors

Landslide events are activated by the combined effect of intrinsic and extrinsic factors
and it is important to identify those causative factors in landslide susceptibility, hazard
and risk identification (Terzaghi 1950; Dahal and Dahal 2017). In addition, for landslide
risk management using mitigation measures, monitoring and early warning, which should
be designed and implemented based on site-specific conditions, identifying the causative

13
2186 Natural Hazards (2024) 120:2179–2231

factors is vital. Intrinsic or internal factors which affect landslide occurrence include geol-
ogy, geomorphology, land use, land cover, soil type and depth, drainage, proximity to water
bodies and road cuts (Bera et al. 2019; Shano et al. 2020). Intense rainfall, earthquakes,
volcanic eruptions, rapid snowmelt, rapid change in water levels and anthropogenic activi-
ties are some of the external triggering factors which stimulate landslides (Wieczorek
1996). A combination of time-dependent and time-independent factors can lead to the fail-
ure of slopes when a favourable combination of these factors is present (McColl 2015;
Sidle and Bogaard 2016).
A more systematic classification of contributory factors is explained in WP/WLI (1994).
According to Crozier (1986), the stability of a slope can exist in one of the following three
stages: stable, marginally stable and actively unstable. Based on the function of causal fac-
tors for transforming a slope into these three stages, causal factors can be classified as pre-
paratory causal factors and triggering causal factors (WP/WLI 1994). Preparatory causal
factors can place the slope in a marginally stable state without actually initiating the move-
ment. The causal factors which initiate the movement shifting the stability of the slope
from a marginal stage to an actively unstable stage are known as triggering causative fac-
tors. These causal factors can be further classified as ground conditions, geomorphological
processes, physical processes and man-made processes. A checklist for the landslide causes
under these four categories has been provided in Cruden and Varnes (1996).
Except for ground conditions, the other three factors can perform as both preparatory
and triggering causal factors (Cruden and Lan 2015; McColl 2015) and one or more of
these factors can contribute to landslide events. For rainfall-induced landslides, human
intervention also has a significant contribution to placing the slope into a marginally sta-
ble state. Actions such as deforestation, excavation of the slope or toe, loading of slope or
crest, mining, artificial vibration, irrigation for cultivation and water leakage from defec-
tive drainage and pipes can contribute to failure. Nevertheless, this article is focussed on
the impact of rainfall and ground conditions on landslides in tropical regions considering
the significant influence of these two factors for landslides in the region. High heteroge-
neity and significant weathering of the subsurface are characteristic features in the tropi-
cal region which make the landslide mechanism more complex and difficult to evaluate
(Dewitte et al. 2022; Dille et al. 2019; Thomas 1994). Special emphasis has been given to
examples from Sri Lanka to demonstrate how these factors contribute to the likelihood of
landslides occurring.

4.1 Effect of ground conditions for landslides

4.1.1 The effect of geological structures on landslides

Petrology, geological structures, topography, in situ stress conditions and degree of

weathering affect the rock slope instability in natural and engineered slopes (Borrelli
et al. 2007; McColl 2015; Regmi et al. 2013). Rocks can be broadly classified as sedi-
mentary, igneous, and metamorphic rocks based on the method of formation. The stabil-
ity or instability of rock slopes is significantly affected by the structures present in the
rock mass such as bedding, folds, discontinuities, and faults (McColl 2015; Stead and
Wolter 2015). Joint spacing and orientation significantly influence the failure in rock
slopes. Figure 4a illustrates the joints closely spaced in rock. In addition, these joints
assist the weathering of the rocks. The almost impermeable rocks become permeable

13
Natural Hazards (2024) 120:2179–2231 2187

(a) (b)
Joint set 1
Corestones
Joint set 2
Joint set 1
Feldspar rich gneisses weathered
to form whitish clay rich soil
highly susceptible to landslides
Joint set 2

(c) (d)

Thinly banded foliations

Bedding plane direction

(e) Ridge feature (f)

Basinal feature
Wrapping layers of the boudin

Boudin (competent rock)

Fig. 4  Different structural features in metamorphic rock formations (a). rock joints, (b). the more competent
rocks remain unweathered within the weathered less competent layers resulting in the heterogeneous soil
profile, (c). thinly banded rock foliations, (d). bedding planes arranged in parallel orientation, (e). compe-
tent rock remains unweathered & (f). tension cracks appeared due to the weakening of the subsurface due to
heterogeneity and pore water pressure

and pore water pressure can build-up in these joints. The infills of intrusions or the
weathering products inside these joint cracks such as clay can lubricate the planes or
blocks of parting (Cooray 1994).
Due to the closely spaced bedding planes in sedimentary rock formations, they often
behave anisotropically and slippage may result along the weak planes. The horizontal bed-
ding planes of sedimentary rocks are subjected to uplifting, tilting, and folding frequently
and this results in variations in the stability of the slopes (Hart 2000). Due to foliation
orientation, translational sliding or toppling-type failures in horizontal folds and com-
plex transitional and rotation-type failures in plunging slopes can be observed (Fleuty
1964; Ramsay 1967). In open folds, longitudinal or transverse discontinuities may appear
and when the folds are tight and isoclinal, foliations and cleavages may appear which are
signs of degradation of the quality of the rock mass. Faults can form sliding surfaces that
would steepen the rock slope (Stead and Wolter 2015).
In igneous rock formations, the weakening of the high-strength rock mass is mainly due
to the structures within the rock and instability varies depending on the topography, degree

13
2188 Natural Hazards (2024) 120:2179–2231

of weathering and in situ stress conditions. The discontinuities within the rock mass, ori-
entation and frequency of discontinuity control the instability and weathering of the rock.
Sheeting joints are a common feature in igneous rocks. They run parallel to the ground sur-
face in both flat and steep slopes at shallow depths (< 30 m) due to high differential com-
pressive stresses (Hencher et al. 2011; Ziegler et al. 2014). Meanwhile, the faults within
the rock control the failure location and behaviour. Igneous rocks are composed of different
minerals with different decomposition rates, and this affects the weathering rate of the rock
mass. In addition, grains and lattice distortions also contribute to weathering and strength
(Stead and Wolter 2015).
Foliations, cleavage, joints, and other structural features affect the instability in met-
amorphic rocks. Large catastrophic landslides have occurred in both high-grade foliated
metamorphic rocks such as gneiss and low and medium-grade foliated slate, phyllite and
schist (Zádorová and Penížek 2018). Grains and mineral composition also affect the weath-
ering process in metamorphic rock similar to igneous rocks. However, the weathering pro-
files in metamorphic rocks such as gneiss are more complex than in igneous rocks, affect-
ing the stability significantly due to the change in landslide hydrology upon weathering.
Hence, it is essential to discuss these complexities further.
Chemical weathering is prominent in tropical and subtropical regions, which signifi-
cantly changes the mineralogical, chemical and textural composition of rock, reducing the
strength (Che et al. 2012; Regmi et al. 2014). The chemical and mineralogical composition
and heterogeneities of the parent rock significantly contributes to the weathering and the
formation of the residual soil and residual minerals, which is associated with landslides in
the tropical and subtropical region (Duzgoren-Aydin and Aydin 2006). Quartz, clay miner-
als and sometimes iron oxides are typically left as residual minerals after complete weath-
ering. The implications for landslides with the chemical and mineralogical contributions in
tropical and subtropical igneous (Che et al. 2012; Duzgoren-Aydin and Aydin 2006) sedi-
mentary (Fontoura et al. 2023) and metamorphic rock (Regmi et al. 2014; Sajinkumar et al.
2011) soils have been studied. In igneous rocks such as Granitic rocks quartz and feldspar
minerals are in abundance. Quartz and mica usually remain resistant to weathering and
feldspar converts to clay. Relatively thick weathering prevails in these types of rocks. How-
ever, in mafic rock formations, only a thin weathering profile remains with quartz, feldspar
and ferromagnesian minerals. In clastic sedimentary rock formations, the clay or quartz
will be the residual mineral and composition will vary depending on the type. Weathered
non-clastic rocks release carbonates and may get dissolved upon weathering. The weather-
ing profile is usually thin in sedimentary rock formations.
Gneiss, granulites and quartzite which are rich in quartz, alter to quartz residual miner-
als, while slates, schists, and phyllite disintegrate and form clay minerals in abundance.
Thick weathering profiles can be seen in metamorphic rock formations. Gneiss consists of
altering coarse-grained feldspar or quartz with fine-grained minerals and schist consists of
micaceous layers. Due to the presence of feldspar, these rocks weather into clay (Fig. 4b)
and hence it is highly susceptible to landslides. The metamorphic rocks consist of folia-
tions or schistosity. Planar structures are formed due to foliation/schistosity present in the
rock (Fig. 4c) due to flattish minerals such as mica and can fail along these planes. The
banding of rock is visible due to the light colour of feldspar and quartz and the dark colour
of hornblende and mica (Cooray 1994).
Metamorphic rocks are arranged in parallel or almost parallel bands which are closely
associated with each other but act differently (Fig. 4d). Usually, the more competent bands
deform less compared to the less competent layers in the rocks (Dahanayake et al. 2011).
During extensions, tensile stresses are applied to the bands and the less ductile competent

13
Natural Hazards (2024) 120:2179–2231 2189

bands start necking and thinning which eventually break apart into several pieces with
pinch and swell structures. The less competent ductile material moves into these spacings
(Ramberg 1955). In geology, these structures are known as boudinage structures and are
often referred to as cornerstones in the residual soil formations (Fig. 4b and e). Further-
more, large-scale boudin structures may be stable under gravity, but mini-boudin structures
are vulnerable to landslides due to the heterogeneity in materials and associated pore water
pressure increase. Selective erosion and piping within the soil can reduce the shear stress
and increase the probability of slope instability (Dahanayake et al. 2011). Specifically, in
cut slopes, the cornerstones or mini boudins can fall owing to this reason. A failure due to
the heterogeneity in material and pore water pressure increase is illustrated in Fig. 4f.

4.1.2 Effects of soil formations and their characteristics for landslides in the tropics

Soil is generally referred to as loose, unconsolidated material at the surface of the earth
which is variable in thickness (Terzaghi and Peck 1967) and categorised based on the
mode of formation as residual soil or transported soil. In essence, residual soils are in situ
soil formations above the parent rock due to weathering of parent rock material through
physical and chemical processes which have completely altered its rock texture (Brand and
Phillipson 1985). Factors such as climate, parent material, topography, time and biological
activities affect the characteristics of formed residual soils (Sing and Huat 2012). Trans-
ported or sedimentary soils are materials formed elsewhere but transported to the present
location by physical processes involving transporting agents. Based on the transporting
agent water, wind, ice and gravity, sedimentary soils can be categorised as alluvial, aeolian,
glacial and colluvial soils, respectively.
The spatial variation of geology and overburden is substantial throughout the world,
and it significantly affects slope stability. The stability of slopes and landslides in tropical
residual soils has been investigated in a number of studies (Lacerda 2007; Wesley 1990,
2011). The prevailing deeply weathered residual soil formations, deep groundwater levels
and thick unsaturated zones are significant factors affecting these failures (Lee et al. 2014).
Moreover, rainfall-induced slope instability in the colluvial formations has been substantial
in the tropical region (Wesley 2009). Therefore, understanding the mechanism of residual
and colluvial soil formation is required to explain the mechanism of rainfall-induced land-
slide failure in tropical regions. Hence, the effect of soil formation and its characteristics
affecting the landslides are discussed in the subsequent paragraphs.
Although the landslides are prominent in residual soils, the in situ weathered mantle
can be divided into different subdivisions based on weathering grade. Geological Society
Engineering Group Working Party (1990) classified the weathered mantle in granitic rocks
into six grades which were derived from the five principal classes defined by Ruxton and
Berry (1957), as illustrated in Fig. 5. This six-fold classification is widely used worldwide
(Migoń 2010). However, weathering profiles are not always required to be continuous as
in the weathering sequence. In granitoid rocks, clear vertical differentiation of different
weathering grades is visible. Migoń (2010) showed that depending on the weathering
grade, the type of movement can be different (Fig. 5a) and the reason for failure in rainfall-
induced landslides can vary with the weathering grade as shown in Table 1. Nevertheless,
it is important to note that the landslides in tropical regions may show complex behaviour
involving different types of movements.
The weathering of rocks can extend to great depths depending on the temperature and
humidity (Thomas 1994). Although deep weathering is not restricted to tropical regions,

13
2190 Natural Hazards (2024) 120:2179–2231

(a) (b)

Residual soil

Completely weathered rock

Highly weathered rock

Moderately weathered rock

Slightly weathered rock

Fresh rock

Fig. 5  a Weathering grade and movement type relationship (modified after Migoń, (2010) and b a typical
weathering profile

the favourable high temperature and high humidity factors result in deep weathering in
these regions. In Hong Kong and Southeast Brazil, weathered layer thicknesses of 70–90 m
have been encountered (Shaw 1997; Thomas 1994). In Sri Lanka, residual soil layers of
around 30–40 m have been encountered in the wet zone of the country (Cooray 1994) and
it is about 100 m in Malaysia (Chigira et al. 2011).
Most of these slope failures are initiated by the existence of relict discontinuities in
these layers and have been reported in tropical countries such as Hong Kong, Brazil, Sri
Lanka and Singapore (Irfan 1998; Wesley 2011). The structural discontinuities present in
the rock may be preserved even after weathering as relict discontinuities. These disconti-
nuities are more prominent in the less weathered saprolite layer compared to the residual
layer as more fine particles may be present in the residual layer. According to Irfan (1998),
a failure due to relict discontinuities can be a transitional failure where the surface of fail-
ure is fully or partially specified by the discontinuities. Alternatively, it may result in mass
toppling or circular or progressive failures. When the discontinuities are not closely ori-
ented or with the presence of major joints, bedding planes and faults, planar or wedge-
type failure takes place as shown in Fig. 6. In contrast, when the discontinuities are closely
spaced, it increases the infiltration and reduces the shear strength of the soil which results
in circular or nonlinear failure surfaces. When vertical or almost vertical relict discontinui-
ties are present, toppling failures take place along the weakened discontinuities. Creep or
progressive type slope movements take place in the fine-grained saprolite initiated due to
internal deformation and stress redistribution owing to mass structures such as relict joints
and bedding planes. The effect of discontinuities on the groundwater is noteworthy under
tropical conditions which is discussed briefly in Sect. 4.2
According to Wesley (2011), the slope stability in residual soils and sedimentary
soils are somewhat different from one another. Alluvial, aeolian and glacial soil forma-
tions are transported to the lowland areas and get deposited where the risk of stability
is not significant in most cases. The transported and deposited sedimentary soils are
subjected to consolidation under their own weight and develop interparticle bonding
with time, leading to increased soil strength. In contrast, strength decreases with time
in residual soils (Wesley 1990). Tropical residual soils are more heterogeneous due to
variability in parent rock and in situ weathering processes (Wesley 2009) and can act as

13
 Table 1  Different landslide movement types based on the weathering grade of granitic rock (Durgin 1977; Migoń 2010)
Migoń (2010) Durgin (1977)
Grade Formal name Type of slope failure Main reasons for failure Formal name Type of slope failure Main reasons for failure

VI Residual soil (rock is Rotational slides Pore water pressure Saprolite (completely Slump or rotational slides Decrease in shear strength
completely converted to increases due to converted to soil) due to the presence of rel-
soil and mass structure variations of physical ict joints and increase in
destroyed) properties infiltration through them
V Completely weathered Piping within corestone Decomposed granitoid Debris flows or debris Pore water pressure
(Decomposed and areas (> 85% weathered rock) avalanches build-up in the boundary
disintegrated to the soil The disappearance of between heterogeneous
Natural Hazards (2024) 120:2179–2231

but mass structure may discontinuity planes layers

be intact) and homogeneity within
saprolite
IV Highly weathered (>50% Corestone (15–85% Rockfall avalanches or Undermining of corestones
of rock material is weathered rock) rolling rocks due to erosion
decomposed. Rock is
present as corestones or
discontinuous frame-
work)
III Moderately weathered Shallow debris slides Selective erosion and
(<50% of rock material and corestone falls reduced friction Fresh rock (< 15% weath- Rockfall, rockslides, bock The angle of the joints,
decomposed. Rock is between sliding sur- ered rock) glides shearing resistance,
present as corestones or faces of slickensides or Debris avalanches and decrease in effective
continuous framework) rock joints debris slides over sheet- cohesion and seepage
II Slightly weathered Rockslides and rockfalls Opening of major ing surface pressure within rock
(Rock material may be fractures and clay accu- joints.
discoloured indicating mulation within rock Reduced shear strength in
weathering)) discontinuities the surficial layers above
I Fresh rock (Weathering the near-surface fresh
not visible) rock due to pore water
pressure and reduced
friction angle between
contacts
2191

13
2192 Natural Hazards (2024) 120:2179–2231

Fig. 6  Block failure due to relict discontinuities

weak zones considering the mobilisation and redistribution of elements and accumula-
tion of clay minerals (Che et al. 2012; Duzgoren-Aydin and Aydin 2006). Sedimentary
soil is more homogeneous due to the sorting process that takes place during erosion,
transportation, and deposition. As such, the behaviour of residual soils is more complex
(Sing and Huat 2012).
The effect of clay minerals in landslides in humid tropical regions has been studied pre-
viously (Duzgoren-Aydin 2002; Duzgoren-Aydin et al. 2002). The majority of the weath-
ered clay minerals are kaolinite in humid tropical climates (Che et al. 2012; Duzgoren-Aydin
2002; Regmi et al. 2014). Studies have shown that the existence of a possible landslide plane
can be identified from the clay mineralogy and soil chemistry (Wen et al. 2004). The pres-
ence of clay horizons in soils can decrease the permeability, increase the pore water pres-
sure, and affect the stability of residual soil slopes. The effect of clay accumulation in relict
joints for landslides has been described in Irfan (1998); Prior and Ho (1972).
In contrast, colluvium is heterogeneous and complex due to mixing with different mate-
rials during the down-slope movements (Zádorová and Penížek 2018). Colluvial soils are
defined as heterogeneous, loose soil or rock fragments of different particle sizes accumu-
lated on the lower part of the slope due to transportation by gravity during former land-
slides (Eggleton 2001; Schulz et al. 2008). The thickness of the colluvium layer depends on
factors such as the process of formation, time and number of cycles of deposition, topog-
raphy, land cover, human impact and the material source (Zádorová and Penížek, 2018).
Instability or reactivation of landslides in colluvial soil can occur due to many reasons.
Cuts made at the toe of the slopes where colluvium layers stretch along the slope, loads
applied on the layer due to constructions or impact from rock slides or falls, and the ele-
vation of groundwater level are a few reasons for the reactivation of the colluvium layer
(Lacerda 2007). Many major landslides have occurred in the colluvium layers due to the
heterogeneity in the colluvium itself and the heterogeneity between colluvium-weathered
rock contact. When debris or rock materials are present in the colluvium, water infiltration
gets hindered and piping or selective erosion can occur, resulting in instability. Similarly,
pore water pressure can build-up along the colluvium-weathered rock contact due to the
contrasting permeability making it a potential failure surface.

13
Natural Hazards (2024) 120:2179–2231 2193

4.2 Landslide hydrology

Rainfall infiltration can be considered as the main triggering factor for landslide occur-
rence in tropical and subtropical regions which was concluded by several studies (Fou-
rie 1996; Rahardjo et al. 1995, 2007; Zhang et al. 2011). This section focuses on the
effect of hydrology on landslides pertaining to ground conditions in the tropical region.
The rainfall-induced slope failures are often shallow failures with a depth of less than
2–3 m and parallel to the surface of the slope (Rahardjo et al. 1995; Sidle and Ochiai
2013). The failure plane of these shallow landslides is often located at hydrologic dis-
continuities such as soil–bedrock contact or colluvium-residual soil contact (Che et al.
2012; Dhakal and Sidle 2004; Vieira and Fernandes 2004). Shallow rapid landslides
are often initiated due to heavy short-duration rainfalls and antecedent rainfall condi-
tions may influence slopes with the unsaturated vadose zone (Sidle and Bogaard 2016;
Zhang et al. 2016). Further, Rahardjo et al. (2007) concluded that the antecedent rainfall
on slope stability depends on soil permeability, where antecedent rainfall is critical for
soils with a low saturated coefficient of permeability. Soils with a high coefficient of
permeability are affected by short-duration high-intensity rainfalls. Deep-seated slow-
moving landslides are mostly due to prolonged rainfall which increases the pore water
pressure within the failed mass and along the slip surface (Iverson and Major 1987;
Zhang et al. 2016). Deep-seated rapid landslides can occur as a result of prolonged rain-
fall or as a response to individual storms (Sidle and Bogaard 2016; Sidle and Chigira
2004).
Often rainfall-induced shallow landslides in the soil can be classified as slides, flows
or slide-to-flow movements (Cascini et al. 2010). In slide-to-flow type transformations,
the landslide is initiated as a slide and transformed into a flow with the mixing of dif-
ferent materials in the presence of water from excessive rainfall and different water
sources such as streams, canals, and surface water flows. These flow-type landslides can
be catastrophic as they have high kinetic energy and destructive potential due to their
ability to flow like a fluid and possible entrain material along the path. These can have
long runout distances and reach downhill areas without any prior warnings resulting in
huge loss of lives and property (McDougall 2017). Moreover, deep-seated landslides are
often slow-moving and have a sliding-type movement. These can get reactivated from
time to time during or after heavy rainfalls and cause damage to the properties within
the landslide boundary. The shear zone of the deep-seated landslides is highly sensitive
to the pore water pressures as it can change the stress state in the shear zone (Kang et al.
2022; Wang et al. 2018). Furthermore, movements can occur as a single block or as sev-
eral blocks with different movement directions or velocities. However, if transformed to
a flow-type movement, it can be catastrophic due to the large mass of landslide materi-
als involved and rapid to extreme rapid movements.
Changes in pore water pressure and seepage forces with rainfall are shown in Fig. 7.
The failure is identified under two key mechanisms. Firstly, rainfall-induced slope insta-
bility can be attributed to the advancement of the pore pressure front and associated
shear strength reduction due to the loss of matric suction (Ng and Shi 1998; Rahardjo
et al. 1995). This condition is critical for residual soils as the groundwater table is often
deep and steep residual soil slopes can remain stable due to shear strength increase from
the matric suction above the groundwater table. Hence, with the rainfall and infiltration
of water into the soil, negative pore water pressure decreases towards zero, reducing the
shear strength of the soil and would result in slope instability. Secondly, perched water

13
2194 Natural Hazards (2024) 120:2179–2231

table conditions can occur due to the heterogeneity in the regolith resulting in positive
pore water pressures with the wetting front development (Collins and Znidarcic 2004;
Huat et al. 2006). In the second mechanism, the failure can occur due to pore water
pressure build-up at the bottom of the slopes or along the contact surface between rock
and the overlaying residual soil or colluvial deposits (above hydraulic restrictive layers)
(Collins and Znidarcic 2004; Wang and Sassa 2001). In addition, the surcharge due to
the increase in soil weight during rainfall can also contribute to the instability (Chigira
and Yokoyama 2005; Galeandro et al. 2013). The depth to the groundwater table and the
slope geometry has a secondary contribution to the instability of slopes (Rahardjo et al.
2007).
Landslide hydrology is influenced by factors such as lithology, geomorphology, weath-
ering, soil development processes, hydraulic properties of soil, rainfall and vegetation
(Rahardjo et al. 2007; Sidle and Bogaard 2016; Smith et al. 2002). For landslides, the perme-
ability of the soil significantly affects the phenomenon which mainly depends on the hetero-
geneity of the soil (Debieche et al. 2012). In tropical regoliths, groundwater permeability in
residual soil and saprolite is complex due to the presence of discontinuities and geological
structures. The permeability of the residual layer can be low compared to saprolite or par-
tially weathered rock with a high rate of discontinuities. However, when these discontinuities
are occupied by infills of clay, the permeability may be diminished. Similarly, zones of clay
areas, cornerstones, and contact between weathered rock and overlying soils can act as a bar-
rier to the groundwater flow, resulting in perched water tables (Sidle and Bogaard 2016).
Moreover, the presence of macropores and fissures due to biological or mechanical
origin can result in preferential flow affecting the hydrology within the catchment area
resulting in instability of slopes by increasing the capacity to infiltrate into the subsurface

Fig. 7  Schematic illustration of the effect of rainfall on pore pressure distribution in soil slopes

13
Natural Hazards (2024) 120:2179–2231 2195

(Cristiano et al. 2016; Xie et al. 2020; Zhang et al. 2020, 2021). The biological macropo-
res are due to living organism activities, including plant roots (Bogaard and Greco 2016;
Xie et al. 2020). The mechanical origin of macropores can be due to shrinkage or desic-
cation cracks, soil piping due to erosion and mechanical fissures are due to the dynamics
of landslides (Cristiano et al. 2016; Sidle and Bogaard 2016; Zhang et al. 2021). Most of
the unstable slopes have complicated networks of flow paths due to the above fact and sig-
nificantly affect the flow orientation. Vertical flow paths are affected close to the surface
and horizontal paths are affected at deeper levels (Debieche et al. 2012; Sidle and Bogaard
2016). The traditionally used numerical analysis and laboratory experiments for rainfall
infiltration often use homogeneous and isotropic soil layers and the capability of capturing
the preferential flow of water is difficult. In situ experiments are also time-consuming and
laborious. Further, it is difficult to constrain the natural boundaries in in situ experiments
and to induce landslides to identify the preferential flow network. However, recently there
have been studies conducted on dynamic hydrology and the influence on the stability of
slopes with preferential flow using laboratory experiments using model tests (Zhang et al.
2021) and numerical model tests (Krzeminska et al. 2013; Shao et al. 2015). Further geo-
physical and hydrological approaches have also been used to recognise and map the pref-
erential flow path networks (Ellefsen et al. 2002; Kukemilks and Wagner 2021; Revil et al.
2005). Similar to assisting in the accretion of pore water pressure in the slope, preferential
flow paths can contribute to the dissipation of pore water pressure as well (Bogaard and
Greco 2016; Uchida et al. 2001). From the study of Shao et al. (2015) it has been evident
that with preferential flow paths, for long-duration low-intensity rainfalls, there is a posi-
tive effect on stability and for short-duration high-intensity rainfalls a negative effect on
stability.
Another factor that should be considered in landslide hydrology is the bedrock ground-
water flow. Studies have shown that pervious bedrock can influence the stability of a slope
to a great extent by draining out water, whereas perched water table conditions can occur
resulting in failure when more impervious bedrock is present (Onda et al. 2004; Travel-
letti et al. 2012). Sometimes rather than draining water through the bedrock, groundwa-
ter exfiltration from fractured rock into overlain soil can occur which causes shallow and
deep landslides (Brönnimann et al. 2013; Gerscovich et al. 2006; Iverson 2000; Sidle and
Chigira 2004). Meanwhile, confined aquifers in permeable bedrock can also increase the
pore water pressure in the regolith. The artesian outbursts of bedrock due to pore water
pressure build-up inside discontinuities owing to water ingestion during heavy rainfalls can
also trigger landslides due to sudden gushing out of the water. Studies show that even if the
bedrock is overlaid by a low permeable soil, the pore water pressure can increase within
the bedrock which exceeds the soil pressure due to the flow of water through cracks and
fractured outcrop which acts as preferential flow paths (Brönnimann et al. 2013).

5 Landslide risk assessment

Risk has been identified as the probability of an identified hazard occurrence and the mag-
nitude of the damage or consequence on humans, properties or the environment. Landslides
in natural or engineered slopes are always associated with risk and hence risk assessment
and management are required. Risk assessment is a well-established field that has been
widely studied and documented over the years. There is no substantial disparity between
landslide risk assessment methods for tropical regions and other parts of the world as the

13
2196 Natural Hazards (2024) 120:2179–2231

fundamental principles do not vary based on the geographic location. The main difference
in risk assessment for the tropical region compared to other areas is with the identification
of the susceptibility based on the factors which affect the landslide, as discussed in the
previous sections. Tropical regions receive high rainfall compared to other regions and the
seasonality of landslides are affected with the monsoons, tropical cyclones and El Nina/
La Nino cycles which should be given consideration in risk assessment in the region. Land
use can also significantly contribute for the susceptibility in the region. Usually, the tropi-
cal regions are covered with dense vegetation cover compared to other regions. However,
due to the rapid urbanisation and population growth in the tropical region compared to
other regions, cities have been expanded towards steeper sloping areas, increasing the risk
of landslide hazards. Subsequently, with the change in land use and loss of dense vegeta-
tion cover in the hilly areas, the landslide risk may increase compared to other regions.
Furthermore, for landslide risk assessment in the tropical region, the lack of systematically
collected data can be considered a problem as most of the tropical countries are developing
countries and such studies are limited due to economic constraints. Hence, for risk assess-
ments, it is required to consider the unique environmental conditions and the mitigation
and management strategies should be tailored to the conditions prevailing. This analysis
aims to provide a comprehensive overview of the current state of knowledge in risk assess-
ment as a prerequisite for landslide risk management.
A glossary of terms has been developed by the technical committee on risk assessment
and management (TC32) of The International Society of Soil Mechanics and Geotechnical
Engineering (ISSMGE) so that there is consistency in the terminology used in risk assess-
ment and management all over the world (Fell et al. 2005). Different guidelines are avail-
able for risk assessment and risk management although it is not confined to tropical region
(Australian Geomechanics Society 2000, 2002, 2007; Fell et al. 2008; Geotechnical Engi-
neering Office 1998). Furthermore, risk management frameworks have been well-estab-
lished for quantitative risk assessment and management (Dai et al. 2002; Fell et al. 2005;
Hungr et al. 2016). The established framework is shown in Fig. 8 and this can be used irre-
spective of the climate region. The framework consists of 6 stages, starting from the scope
definition and underpinning hazard assessment, consequence analysis, risk estimation, risk
evaluation and risk management. In the following sections, the different techniques used in
these 6 stages were reviewed and facts relevant to the tropical region are also mentioned.
Particular emphasis was given to the risk management stage separately as it is important to
review different risk reduction measures implemented based on the site or local area spe-
cific conditions.

5.1 Scope definition, susceptibility and hazard assessment

In the first step, scope definition or description of intention involves two stages: screening
and scoping (Lee and Jones 2014). At this stage, it is decided whether risk management
is required and the scope of risk assessment and methods are defined. In the second stage
hazard assessment, landslides are characterised based on their size, velocity, mechanism,
location, runout (spatial probability) and probability of occurrence (temporal probabil-
ity). The first step in the landslide hazard assessment or risk assessment is to identify the
landslide susceptibility. Usually, the terms susceptibility and hazard assessment have been
used synonymously in articles. Often susceptibility assessments are interpreted as hazard

13
Natural Hazards (2024) 120:2179–2231 2197

Identify potential hazards, consequences, Scope definition

stakeholders, goals, methods

Hazard Assessment

Hazard definition
Characterize hazard mechanism,
frequency, magnitude, intensity &

RISK ANALYSIS
runout
Analysis of frequency

RISK ASSESSMENT
Consequence Analysis

Characterization of consequence scenarios

RISK MANAGEMENT
Characterize elements at risk,
vulnerability & temporal
probability
Estimation of magnitude and probability
of consequences

Develop risk scenarios, estimate risk Risk estimation

Compare risk estimate to tolerable or

acceptable risk, prioritize risk for
Risk evaluation
mitigation and monitoring

Risk mitigation and

Feedback loop

treatment plan

Identify and compare mitigation options,

implement selected mitigation,
estimate residual risk and review Implementation of plan

Monitoring, review
& feedback

Fig. 8  Framework for landslide risk management (modified after Australian Geomechanics Society 2000;
Fell et al. 2005)

assessments in studies and it is vital to understand the difference between these two terms.
The type of landslide, size and spatial likelihood of the landslides may include in suscepti-
bility maps and susceptibility may be quantitatively indicated using indicators such as the
density of landslides in a specific area or as an affected area. Often, only the initiation area
is identified in these susceptibility maps.
To determine the possible travel distance of landslides and the detrimental area, a
runout assessment should be conducted. Runout assessment can be highly complicated as
the characteristics of the runout area such as slope characteristics, hydraulic properties,
failure mechanism, type of material, downhill paths rheological properties and residual

13
2198 Natural Hazards (2024) 120:2179–2231

shear strength in the failure zone may intensify or restrain the motion (Dai et al. 2002;
Evans et al. 2007). Empirical and process-based methods have been developed for land-
slide runout analysis. However, this area still requires developments to account for varying
complex subsurface conditions along the flow path, the contribution of rainfall and other
water sources to obtain accurate detrimental areas from landslide runout. Many analytical
studies on runout assessment were focused on either validation of the runout model or on
simulating case studies. Forward analysis and estimation of flow-type landslides can be
crucial in susceptibility assessment and risk management, considering most of the cata-
strophic rainfall-induced landslides in the tropical region is flow-type landslides with long
runout distances.
The difficulty with hazard assessment is that it needs to specify the temporal probability
of the landslides with the intensity at a given location (Corominas et al. 2014). Frequency,
return period or exceedance probability can be used to express the probability of occur-
rence in hazard assessment. Out of many landslide susceptibility and hazard assessment
methods from the literature, inventory-based methods are used as a starting point for other
methods as they provide the necessary inputs. Qualitative and quantitative methods are
used in susceptibility assessments. Approaches such as heuristic methods, geomechanical
approaches, empirical probability and indirect approaches can be used to assess temporal
probability. A summary of the susceptibility and hazard assessment techniques is shown in
Table 2
It is challenging to conduct hazard assessment in tropical regions considering the com-
plex ground conditions, highly unpredictable rainfall patterns and difficulty in quantifying,
limited availability of data on past landslide records, rainfall records and geological and
geotechnical information, and due to rapid land use change associated with urbanisation
and population growth in the region. Hence, using physically based methods can be recom-
mended for accurate prediction of landslides in the tropical region. However, it requires a
significant amount of data which can be time-consuming and expensive to obtain. Further
it requires higher computational capabilities, and expertise in many areas such as geotech-
nical engineering, geology and hydrology. As a solution, highly landslide susceptible areas
can be identified from a qualitative assessment technique and it can be coupled with site-
specific physically-based analysis for early warning or remediation.

5.2 Consequence analysis

In the consequence analysis, the elements at risk are identified, and the consequences are
quantified. This is followed by estimating temporal-spatial probability for the elements at
risk and vulnerability of the elements at risk, in the sense of property damage and loss of
life or injury. The impact of landslides has been categorised into two subsets by Alimoham-
madlou et al. (2013) as general impacts and particular impacts. The impact on the socio-
economy from landslides can be categorised as economic impacts and human impacts (Ali-
mohammadlou et al. 2013). The worldwide average annual economic loss from landslides
has been estimated as $20 billion and this is around 17% of the total annual economic
losses due to global disasters (Haque et al. 2016; Klose et al. 2015). The effects of land-
slides on humans can be physical or psychological, where people can sustain injuries or
deaths from landslides and psychological trauma can be associated with the deaths of loved
ones and damage to personal belongings. The particulate impacts are categorised as private
and public losses and direct and indirect losses. Private losses are the physical and men-
tal health of individuals and the loss of their personal belongings. The public losses are

13
 Table 2  Methods for landslide susceptibility and hazard assessment (Corominas et al. 2014; Dai et al. 2002; Pardeshi et al. 2013; Shano et al. 2020)

Susceptibility and Hazard zonation techniques

Hazard assessment Susceptibility assessment Inventory-based methods Landslide distribution maps based on image interpretation
Landslide activity maps based on multi-temporal image interpretation
Generation of inventories based on historical data
Landslide inventory based on radar interferometry
Representation of landslide inventory as density information and spatial density information
Qualitative methods Knowledge-driven methods Direct mapping Geomorphologic hazard maps
Indirect mapping Boolean logic
Natural Hazards (2024) 120:2179–2231

Fuzzy logic
Multiclass overlay
Spatial multi-criteria evaluation
Quantitative methods Data-driven methods Bivariate statistics Weights of evidence
Information value
Frequency ratio
Weighted overlay method
Multivariate statistics Logistic regression
Discriminant analysis
Artificial neural network
Physically based methods GIS-based LEM Static infinite slope modelling
Dynamic infinite slope modelling
Kinematic analysis for rock Stereonet plots, GIS-based analysis
slope of discontinuities
2D and 3D LEM Slope stability with groundwater
flow
Numerical methods Continuum modelling
Discontinuum modelling
2199

13
 Table 2  (continued)
2200

Temporal probability Heuristic methods Event trees

Geomechanical approach Probability of failure First-order second-moment method

13
Point estimate method
Monte Carlo simulations
Empirical probability Binomial distribution
Poisson distribution
Indirect approach Precipitation intensity–duration
Natural Hazards (2024) 120:2179–2231
Natural Hazards (2024) 120:2179–2231 2201

damage to public properties and infrastructure facilities which may affect the governments
and the public (Alimohammadlou et al. 2013). Direct impacts are the immediate effects of
landslides such as the destruction of properties, infrastructure and the natural environment.
In quantifying the direct costs, the expenses for the repair, maintenance and replacement of
the damaged properties from landslides can be considered (Kjekstad and Highland 2009).
The indirect impacts can affect the regions in the long-term and it is difficult to quantify the
influence. The impact on livelihood, reduced land value, productivity and mental health of
people, and the spread of infectious diseases are some of the indirect impacts.
It is important to know about the past impacts from landslides in analysing the con-
sequences. Information regarding impacts from past landslides is important as case study
scenarios for estimating the potential consequences of future landslide events. It also pro-
vides valuable information regarding the level of risk in a particular area. Although many
landslides occur each year worldwide, the damages from these mass movements to proper-
ties and the number of fatalities or injuries are poorly quantified for many reasons. Accord-
ing to Petley et al. (2005) and Petley (2012), there are two reasons for this underestima-
tion of fatalities from landslides. Firstly, due to the lack of systematic data collection and
the difficulty in assessing the damage from landslides as of the poor accessibility in these
terrains. Moreover, small landslide events are systematically under reported and the focus
tends to be on larger events. Secondly, due to the poor categorisation of landslide fatalities
under the triggering events such as earthquakes, storms or floods in multihazard databases.
For instance, the landslide events and the fatalities in the African continent are heavily
under-recorded, although many have been exposed to landslides (Kjekstad and Highland
2009).
There exist few databases with global landslide records such as Emergency Events Data-
base (EM-DAT) (maintained by the Centre for Research on the Epidemiology of Disas-
ters), Global Fatal Landslide Database (GFLD) (maintained by the University of Sheffield),
NatCat (maintained by Munich Reinsurance Company), Sigma (maintained by Swiss Rein-
surance Company), DesInventar Sendai ((maintained by United Nations Office for Disaster
Risk Reduction and the NASA Global Landslide Catalogue (GLC). These databases are
prepared with the intention of providing policymakers and researchers with systematic and
standardised data to support decision-making in disaster mitigation and disaster prepared-
ness (Panwar and Sen 2020). Different characteristics of these global landslide databases
based on coverage, methods, inclusion criteria, data heads, data sources and inherent limi-
tations are demonstrated in Table 3.
There exist different merits and demerits in all these databases. The records may be
biased, uncertain and inaccurate, and reasons are discussed in the literature (Froude and
Petley 2018; Guzzetti 2000; Kirschbaum et al. 2015, 2010; Panwar and Sen 2020). It is
important to maintain the landslide disaster databases on the global and local scale for
landslide susceptibility, hazard, and risk evaluation. These databases should be up to date
and continuous for a longer period to identify the different trends, but it is a tedious task
to collect the data for every landslide event around the world. Hence, it is important to
have a hierarchical methodology for collecting data. Sometimes, this data may be biased
as there is no definition of what scale of landslides or slope failures should be included in
the database. Furthermore, an idea of the scale of the landslide cannot be obtained from
these databases. For instance, large-scale landslides may be missing in the fatal landslide
databases as the landslide may have occurred in a remote area with no fatalities, but a small
cut slope failure with fatalities may have been included in the databases. In representing
the landslide trends over the years, the use of the database may not be reliable as many
of the past event records may be missing. In addition, it is vital to include the landslide

13
 2202

Table 3  Characteristics of landslide disaster databases

13
Database EM-DAT GFLD GLC DesInventar

Coverage 1990-present (consistent data avail- 2004–2016 1915 (oldest entry)-present (con- 1456 (oldest entry)-present (consist-
able from 1995) sistent data from 2007) ent data for some countries from
1970)
Type Multihazard (natural, technological Fatal landslides Landslides Multihazard of natural, technological
and complex disasters) and man-made
Methodology Country level entries Country level entries Country level entries Hierarchical methodology for entries
from town to district level, prov-
inces, county level or country level
Criteria for inclusion 10 or more human deaths and/or Non seismically occurred fatal Rainfall triggered landslide events Events regardless of size, impact or
100 or more people affected and/ landslides regardless of size, regardless of size, impact or location
or declaration of emergency by impact and location location
the state
Data heads Human impacts: deaths + missing, Human deaths Fatalities and injuries Human impacts: deaths, missing,
injuries and affected injured, evacuated, relocated and
Economic impacts: total damage, directly/indirectly affected
reconstruction cost, insured losses Property damage to houses, crops and
animals and infrastructure damages
Data source United Nations agencies, National Mass media, governments and Reported in media, disaster Official emergency management
governments, non-government aid agencies, academic reports, databases, scientific reports and agencies, sectorial institutions,
organisations, insurance compa- personal communications citizen science reports archives of relief or aid organisa-
nies, research institutes and press tions, academic or scientific files,
media releases
Limitations Underestimates the number of Includes only fatal landslide events No inclusion criteria The coverage is only for 89 countries
landslide events Spatial coverage is high but higher Spatial coverage is high but higher No inclusion criteria. Hence, higher
Only major events are included number of missing records as data number of missing records as number of missing records and
Losses may be categorised under are not collected in a hierarchical data are not collected in a hierar- many missing data heads in records
the primary event rather than the method chical method Records can be biased
cause for the loss Records can be biased Records can be biased
Natural Hazards (2024) 120:2179–2231
Natural Hazards (2024) 120:2179–2231 2203

triggering rainfall intensity in these records for future development in early warnings and
rainfall thresholds. However, this can be challenging as landslides may have been triggered
due to cumulative rainfall received over several days or short periods of heavy rainfall.

5.3 Risk estimation

Risk estimation can be performed using quantitative risk estimations, semi-quantitative

or qualitative risk estimations (Australian Geomechanics Society 2002). The decision on
whether to use qualitative or quantitative risk estimations should be based on the problem
of concern, the required accuracy of the outcomes and the available data (Dai et al. 2002).
Quantitative risk estimation can be conducted by calculating the annual risk. In addition,
frequency-consequence (f-N) pairs and cumulative frequency-consequence plots (F-N
plots) are used in quantitative risk analysis. These two methods are widely used to express
the societal life loss risk. They are not only used in landslide hazard events but also for dam
breaches, floods, chemical plant explosions and radiation from nuclear plants (Strouth and
McDougall 2021). The probable life loss (PLL) is also another parameter used in quantify-
ing risk, where it is the expected number of deaths over a period of time. The product of
probability (f) and the number of fatalities (N) generates the probable life loss (Strouth and
McDougall 2021).
Different tools are available for qualitative and semi-quantitative risk analysis such as
risk scoring schemes, risk registers, risk rating systems and risk ranking matrices (Lee and
Jones 2014). To describe the frequency and consequence of landslides, different descrip-
tors and numerical rating scales are used (Fell et al. 2005). These methods are highly sub-
jective and inexact. However, qualitative estimates can be used to compare the risk in dif-
ferent sites to select the most critical sites for risk management. In semi-quantitative risk
estimations, both quantitative and qualitative methods are used in determining the prob-
abilities of frequency and consequences.

5.4 Risk evaluation

Together with risk analysis, the process of risk evaluation can be conducted to systemati-
cally structure the risk assessment in the framework. The results from risk estimations are
compared against risk perceptions of value judgements and tolerance criteria. Values of
the property or financial losses and loss of life from landslides should be compared with
acceptable and tolerable risk levels defined by authorities. It is important to understand
the concept of acceptable risk and tolerable risk. Acceptable risk is the risk that every-
one affected is willing to accept. Taking actions to reduce the risk would not be required
unless there are low cost and reasonably practical measures available. The tolerable risk is
a risk that society could accept to secure certain net benefits by having faith that the risk is
contained. However, the range of risk is not negligible and should be kept under constant
review and lowered as feasible as possible. The principle of As Low As Reasonably Practi-
cal (ALARP) can be applied so that the risk is contained to a marginally and practically tol-
erable level considering the cost (Australian Geomechanics Society 2002; Fell et al. 2005;
Lee and Jones 2014; Sim et al. 2022). Different countries have different risk acceptance
criteria as different levels of risk exist and various factors should be considered in decid-
ing acceptance and tolerance levels (Duzgun and Lacasse 2005). Sim et al. (2022) present a
review of the acceptable and tolerable risk criteria used in different countries such as Hong
Kong, China, Australia, Canada, New Zealand, Norway, Switzerland and Malaysia. F-N

13
2204 Natural Hazards (2024) 120:2179–2231

Fig. 9  Proposed landslide

risk criteria (F-N criteria) for
landslides and boulder falls from
natural terrains in Hong Kong
(after Geotechnical Engineering
Office 1998)

criteria have been used in Hong Kong for the provisional risk criterion in natural slopes as
shown in Fig. 9 (Geotechnical Engineering Office 1998). This figure can also be used to
explain the concepts of acceptable, tolerable and ALARP principles as discussed.
There exists a clear relationship between risk acceptance and the financial willingness
to pay. Considering the fact that most of the tropical countries are middle or developing
countries, the acceptable risk can be high compared to developed countries (Sim et al.
2022). The developed countries have the financial capability to pay and mitigate risk while
they have a low tolerance for risk. In developed countries, the landslide risk assessment is
becoming quantitative. Mitigation includes structural and geotechnical measures and has
implemented a good administrative system that facilitates the endangered society (Turner
2018). From a critical review of articles on landslide disaster risk reduction in the tropi-
cal region by Maes et al. (2017), it was evident that although around 38% of the published
literature recommends the use of risk management and vulnerability reduction as the risk
reduction measure, but the most implemented measure is risk assessment (57%). Further, it
was identified that the major bottlenecks for implementing risk reduction in these tropical
regions as scientific (30%), political (29%), social (14%), economical (12%), disaster risk
management (10%), and geographical (5%) factors.
The possible outcomes of the risk assessment can be catogorized into two posibilities:
the risk is either acceptable or tolerable where risk mitigations are not required or it is
deemed intolerable where mitigation measures are required. Therefore, as the final stage of
the framework, risk management has been included which involves identifying mitigation
strategies accompanied by a treatment plan, implementing the plan and ensuring the effec-
tiveness of the mitigation measures (Australian Geomechanics Society 2002).

6 Landslide risk management

This section is focussed on the landslide risk management strategies that can be applied
once risk analysis and risk assessment have been conducted. Based on identified risk, dif-
ferent treatment options and a treatment plan should be considered to control the risk.
Figure 10 illustrates the different risk treatment strategies that have been identified by
the Australian Geomechanics Society (2000). These strategies are further explained in
detail in this section, especially the strategies that can be used to reduce the likelihood
of landslide occurrence, and monitoring and early warning that can be effectively used in

13
Natural Hazards (2024) 120:2179–2231 2205

Fig. 10  Landslide risk treatment

strategies

rainfall-induced landslides based on the review outcomes and the authors’ past experience
on different slope stability projects. The risk management in developing countries within
the tropical region was mainly focussed in this study because most of the fatalities occur
in this region and the high vulnerability of these nations in terms of economic, social, cul-
tural and political aspects. Moreover, Sri Lanka was considered as a main focus in exam-
ples and case studies considering the in-depth expertise of the authors and easy accessibil-
ity for the documents/data. This would also supplement the review article with the new
insights of effectiveness of different mitigation strategies evident from the monitoring and
performance.
When the risk is within acceptable levels, then there is no need for any mitigation meas-
ures. However, from time to time with the changes, risk should be reviewed and updated.
Furthermore, risk can be avoided for new projects by abandoning the project or relocating
to a place where the risk is acceptable or tolerable. In reducing the consequences of land-
slide risk, governments or local authorities can remove the existing developments in the
identified risk zones or evacuate people in the area and resettle them in areas identified as
having an acceptable or tolerable risk level (Australian Geomechanics Society 2000).
Although rainfall has been identified as the major triggering factor for landslides in
tropical countries, human interferences in the mountainous areas such as improper land
use and cultivation practices, improper developments and constructions, unplanned set-
tlements, obstruction of natural drainage paths and deforestation, have also contributed
to these rainfall-induced landslides. In controlling these anthropogenic activities, proper
land use planning and guidelines for developments should be in place for identified
high-risk zones. As a tool for land use planning and to communicate with the public,
stakeholders and relevant authorities landslide zoning can be used. Here land is divided
into homogenous domains with a similar potential for landslides and can be classified
as landslide inventory maps, susceptibility maps, hazard maps and risk maps (Fell et al.
2008). Within the risk assessment and management framework, landslide zoning maps
such as inventory maps, susceptibility zonation maps, hazard zonation maps and risk
zonation maps may be used for information, design, statutory and advisory purposes.

13
2206 Natural Hazards (2024) 120:2179–2231

Fig. 11  Methods for landslide remediation

Different guidelines are available for landslide zoning (Australian Geomechanics Soci-
ety 2007; Cascini et al. 2005; Fell et al. 2008; Jackson et al. 2012).
Besides the above risk management strategies, postponing the decisions on rectifica-
tion designs can be made if the available data are not adequate to decide. This must be
done with adequate monitoring to identify any catastrophic slope movements. This is a
temporary strategy and work should be carried out to collect the necessary data required
for decision-making. Furthermore, risk can be transferred to a third party to compensate
for any damages or to accept the risk. In risk treatment of rainfall-induced landslides
often measures to reduce the likelihood of landslides, and monitoring and early warning
are implemented. Hence, this study focuses on those two aspects of risk control in detail
in the following sections.

6.1 Reducing the likelihood of landslides

To reduce the likelihood of landslide occurrence, engineering solutions can be used and
these solutions can be categorised under two general approaches. The first is to remedy the
unstable slope (prevent movement in the first place) and the second is to control landslide
movements (control movement after it happens) (Dai et al. 2002). To remedy the unsta-
ble slope, methods such as slope geometry modifications, surface and subsurface drain-
age measures, retaining structures and internal slope reinforcements can be used as shown
in Fig. 11 (Popescu 2001). Landslide movement control measures such as debris barriers,
diversionary structures and installation of rock catch fences for rock falls can be catego-
rised under the control of landslide movements. These structures should be designed to
withstand the kinetic energy from the rapid movement of landslide materials. Based on
a comprehensive review and project experience, the following methods of mitigation of
unstable slopes are recommended for rainfall-induced landslides.

13
Natural Hazards (2024) 120:2179–2231 2207

6.1.1 Slope geometry modification

Slope geometry modification can be used to reduce unstable material such as unfavourably
oriented fractured rocks, boulders, loose earth and debris or remove the landslide driving
mass completely and increase the factor of safety against stability. The slope gradient can be
decreased using this method and the material at the crest of the slope can be reduced, and
additional material such as toe weights, can be provided to increase the stability under this
approach. Nevertheless, this slope geometry modification is more effective when combined
with drainage improvement in the case of rainfall-induced landslides. Geometry modifica-
tion is also used with retention structures to create a safe slope angle at the crest of the struc-
ture. This can be considered one of the cost-effective rectification measures that can be used
for small-scale landslides where enough space is available for slope geometry modifications.
Based on the authors’ experience, such ground modifications have been used in Kahagolla
landslide rectification which is discussed under case studies in the subsequent section.
The minimum factor of safety (FoS) for the slope depends on the acceptable and tolera-
ble risk and the cost. A tool to determine the slope FoS according to the possible risk to life
and economic losses has been developed in Hong Kong, which is also adopted in countries
like Sri Lanka as a guide. According to the Geotechnical Manual for Slopes (1984), for
existing slopes and remedial works where the designer have an idea about the ground con-
dition from rigorous geological and geotechnical studies and where groundwater table and
loading conditions remain almost the same as the existing slope for areas with high-risk of
life loss upon failure, a minimum FoS of 1.2 is recommended. However, when the existing
slope is substantially modified or when the stability is affected by the new work, a mini-
mum FoS of 1.4 is demanded for high-risk areas in terms of economic or life loss. In addi-
tion, the slope should satisfy a minimum FoS of 1.1 for the worst-case groundwater table
condition in a 10 year return period rainfall. Indian Standards (IS14243-2, 1995) stated that
the minimum static FoS for soil, debris or talus slope should be maintained above 1.5 and
for rock slopes above 1.2 for the safety of structures in the slopes.

6.1.2 Surface and subsurface drainage

Surface and subsurface drainage improvement measures are widely used in most slope stabi-
lisation projects due to the intense and prolonged rainfall conditions that trigger the failure.
Improving the surface drainage can reduce the infiltration of rainwater to the ground and sub-
surface drains can facilitate the dissipation of pore water pressures accumulated due to the
already infiltrated rainwater. This would result in an increase in the shear strength of the soil.
Surface drainage is the most cost-effective drainage measure that can be implemented.
Impermeable surface drains such as cut-off drains, berm drains, and kerb drains can be uti-
lised to divert the water flow away from the landslide area and reduce the water ingestion
into the soil (Haugen 2017; Lee et al. 2018). Surface cover can be used to reduce infiltra-
tion and increase surface runoff (Arantes et al. 2021). Shotcrete could be used to minimise
the infiltration but considering the aesthetic appearance and environmental sustainability,
vegetation covers are preferred over artificial covering (Choi and Cheung 2013). Other
than the interception of rainwater infiltration and diversion as runoff, vegetation provides
other stabilising actions. These include root reinforcement, anchorage of surface soil to the
deep soil layers, enhancement of cohesion at shallow depths, influencing soil suction by
water uptake from roots, interception of precipitation from plant canopy and increase in the

13
2208 Natural Hazards (2024) 120:2179–2231

Fig. 12  Different mitigation and movement control measures a Vegetation and surface cover, b Trench
drain construction, c Radial sub horizontal drains, d Inside of the drainage well during drilling for sub-
horizontal drains, e Drainage well, f Surface drainage with retaining structures g Slope reinforcement with
soil nailing and surface drainage with vegetation, h Anchor reinforcement, i Retention of rock, j Steel fence,
and k Hanger nets

normal stress due to biomass (Cohen et al. 2011; Masi et al. 2021; Mohamed et al. 2022;
Pollen-Bankhead and Simon 2010). The surface covers can also minimise soil erosion on
the slope. In most of the rectification works, grass covers are implemented as shown in
Fig. 12a, using techniques like hydroseeding, sodding or sprigging.
Capillary barrier systems are also being used as slope cover in rainfall-induced slope
mitigation work (Lee et al. 2011, 2022; Rahardjo et al. 2012). In capillary barrier sys-
tems, a layer of fine-grained soil overlaying a coarse-grained soil layer is used. This

13
Natural Hazards (2024) 120:2179–2231 2209

method uses the principle of the variations in the hydraulic properties of the soil. Under
unsaturated conditions, the permeability of the fine-grained soil is higher compared to
the permeability of the coarse-grained layer. Hence water flows laterally along the fine
layer and the coarse layer acts as a barrier until a breakthrough occurs. There have been
several studies conducted to evaluate the capillary barrier system performance using lab-
oratory (Lee et al. 2011, 2022; Tami et al. 2004) and field testing (Rahardjo et al. 2012)
as well as using numerical analysis (Morris and Stormont 1998; Ng et al. 2015). Crack
sealing has been used to cover the discontinuities in weathered rock as water can ingest
at a rapid rate through these cracks due to the high permeability and can saturate the soil
through groundwater flow or increase the pore pressure inside the discontinuities.
A detailed description of the subsurface drains for slope stability can be found in For-
rester (2001) and Urciuoli and Pirone (2013). As subsurface drainage measures, shallow
or deep trenches, sub-horizontal drains, small, medium or large diameter vertical wells,
tunnels, galleries or adits can be implemented. Shallow and deep trenches are widely
used above the landslide crest transversely to cut-off the groundwater flowing towards
the landslide area. Shallow trenches can have a depth of up to 5 m and deep trenches
may extend up to 25–30 m. Usually, these trenches are filled with permeable materi-
als such as gravel, polystyrene cores, aerated concrete, or geocomposites (Urciuoli and
Pirone 2013) and at the bottom, a perforated pipe wrapped in geotextile can be placed to
drain the collected water as shown in Fig. 12b. The effectiveness of the deep trench sys-
tems has been studied in Cotecchia et al. (2016); Tagarelli and Cotecchia (2022). Trench
systems have the ability to drop the piezometric head considerably around the portion of
the maximum depth of the slip surface in deep-seated rotational landslides.
The effect of sub-horizontal drains on slope stability has been widely studied (Cai
et al. 1998; Lau and Kenney 1984; Mukhlisin and Aziz 2016; Rahardjo et al. 2003,
2011). From 3D numerical analysis, Cai et al. (1998) concluded that with the increase
in the length of horizontal drains beyond a critical length, the rate of increase in the
factor of safety of the slope decreases. Furthermore, results reflected that the lengthen-
ing of the horizontal drains is more effective compared to decreasing the space between
drains and the drainage is influenced by the ratio between rainfall intensity and saturated
hydraulic conductivity. From both field studies and parametric studies of Rahardjo et al.
(2003), it was evident that the horizontal drains are more effective when placed at the
bottom of the slope and they lower the groundwater table rather than reducing the infil-
tration or forming perched water table in unsaturated soils.
In the construction of horizontal drains, perforated pipes are installed inside pre-
drilled holes as illustrated in Fig. 12c. The perforated pipes are encapsulated with a geo-
textile to prevent clogging of the holes and removal of soil particles. Usually, the diame-
ter of the hole drilled can be around 100–120 mm. With the use of advanced directional
drilling technology, much longer and curved profiles can be implemented when com-
pared to conventional drilling techniques in one direction (Urciuoli and Pirone 2013).
The horizontal drains are usually drilled at a slight angle to the horizontal so that water
flows under gravity. Siphon drainage methods have also been used in slope stabilisation
in some instances (Sun et al. 2019). This method provides a controlled flow velocity
with the change in groundwater level but when the drill hole is arranged vertically, there
can be air accumulations at the siphon top as the maximum siphon head is 10 m (Cai
et al. 2014).
Drainage wells are generally used for landslides with deep slip surfaces as the ground-
water table should be lowered to greater depths. In small diameter wells for draining out
the collected water, pumps, syphoning or self-draining techniques can be used. Medium

13
2210 Natural Hazards (2024) 120:2179–2231

Fig. 13  Schematic illustration of surface and subsurface drainage measures that can be implemented

diameter walls are in the range of 800–1800 mm diameter and gravity drainage using a
base–collector is used. In large diameter drains, mostly sub-horizontal collector drains are
connected to the well and discharged using gravity drainage. Figure 12d and e show the
inside and outside of a large drainage well-constructed using galvanised steel lining.
Tunnels, galleries or adits construction can be expensive and uncommon in landslide
rectifications. There have been studies conducted to evaluate the effectiveness of drainage
tunnels to control landslides using finite element methods as well as monitoring methods
(Lo et al. 2011; Sun et al. 2010; Wei et al. 2019). To regulate the groundwater table in the
Po Shan area in Hong Kong, two drainage tunnels and a system of sub-vertical drains were
constructed in 2009 which was the first of its kind and the system has been effective in
regulating the groundwater table during rainy seasons (Lo et al. 2011). Wang et al. (2020)
compared groundwater table rise before and after the construction of a drainage tunnel and
concluded that it takes 1.34 times longer for the groundwater table to rise and the tunnel
effectively controlled the rise for any rainfall intensity. Similar conclusions have been made
by Wei et al. (2020) after studying the effect of drainage tunnels using numerical model-
ling. For return period storms of 1 to 100 years, Wei et al. (2020) found that the factor of
safety was greater than one after the tunnel construction.
Figure 13 illustrates the drainage measures that can be included in slope rectification.
Subsurface drainage improvement along with subsurface drainage measures, have been
used in many landslide rectification projects in Sri Lanka, considering the low cost com-
pared to other structural measures. For example, subsurface drainage has been effectively
used in major landslide rectification projects in the country such as the Watawala landslide
and Badulusirigama landslide. The drainage measures used at these two sites are shown
in Fig. 14. However, the proper functioning of these drains should be maintained to avoid
slope instability. There have been incidents of slope failures and landslides due to defective
surface drainage systems and lack of maintenance.
One such incident took place in Welipanna area at chainage 42 + 340 km along the
Southern Expressway, Sri Lanka. Due to a crack in one of the cascade drains at the site,
water leaked into the ground during heavy rains. The groundwater table was well below

13
Natural Hazards (2024) 120:2179–2231 2211

Fig. 14  Schematic illustration of a rectification drainage measures at Watawala landslide and, b monitoring
instrumentation and drainage measures at Badulusirigama landslide (Modified after Chandler and Broise
2000; Kankanamage et al. 2022)

13
2212 Natural Hazards (2024) 120:2179–2231

and the surface layers were dry but unfavourably oriented relict discontinuities were
present in the subsurface as revealed during investigations. It was discovered that due to
the loss of matric suction and build-up of pore water pressure inside the relict disconti-
nuities contributed to the failure (Dharmasena et al. 2015; Idirimanna and Kulathilaka
2019). A similar incident occurred in 2015 at chainage 114 km in Kokmaduwa along
the Southern Expressway. Here the cut slope was stabilised for the construction of the
expressway using geometry modifications using berms, shotcreting, surface drainage
and soil nailing. However, the provided drains were not adequate and without proper
maintenance, the drains were blocked and water stagnated. During heavy rainfall, over-
topping of the drains occurred, increasing the water ingestion to the slope and the pore
water pressure inside. With the slope movement, tension cracks appeared in the slope
and water further infiltrated into the ground, finally resulting in a failure.

6.1.3 Retention structures and internal slope reinforcement

In some situations, the use of slope modifications and drainage measures alone would
not be sufficient to stabilise the slopes. In such cases, structural measures such as
retaining walls or internal slope reinforcement should be carried out (Popescu 2001).
Different types of retaining walls and internal slope reinforcement methods are used
in rectification and should be selected according to the site condition and the cost of
construction. Gravity retaining walls at the toe in the form of random rubble masonry
walls, mass concrete walls, reinforced concrete walls, gabion walls, crib block walls and
structures penetrating beyond the failure surface in the form of passive piles, piers, cais-
sons, cast-in situ reinforced concrete walls and rock slope face retention nets are differ-
ent forms of retention structures that are widely in use. As internal slope reinforcement
measures; anchoring, soil nailing, rock bolting, grouting, stone columns or lime/cement
columns can be used. Figure 12f–i illustrate examples of structural and internal slope
reinforcement measures.

6.1.4 Controlling landslide movements

For controlling the movements of debris and rockfalls, rigid and flexible barriers can be
used (Kwan et al. 2015). Barriers which have the structural ability to resist the impact
of landslide debris are called rigid barriers. Rigid barriers such as embankments, rein-
forced concrete dams, ditches and rockfall galleries with cushions have been used for
protection against movements (Mancarella and Hungr 2010; Volkwein et al. 2011).
Flexible barriers such as steel fences have the ability to absorb the kinetic energy from
boulders or debris moving by deformation of the energy dissipation devices used such
as retention cables, steel nets, and steel poles (Song et al. 2019). Hanger nets (drapery
systems) are also used to catch and divert the moving boulders and rocks safely to the
downhill area (Thoeni et al. 2014). Although the cost of construction is low for flexible
barriers, these require regular inspections and maintenance for effective performance
(Volkwein et al. 2011).

13
Natural Hazards (2024) 120:2179–2231 2213

6.2 Monitoring and landslide early warning

Monitoring and warning systems are another treatment option for mitigating the risk.
Landslide monitoring is a key component of identifying the mechanism and occurrence
of landslides and determining suitable thresholds for early warning and forecasting (Chae
et al. 2017). Calvello (2017) has classified the monitoring equipment based on the moni-
toring parameters (i.e. Deformation, Groundwater and Soil moisture, Triggering factors
and Predisposing factors) and based on the monitoring methods (i.e. Geotechnical, Hydro-
logic, Geophysical, Geodetic, Remote sensing and Meteorological). Chae et al. (2017) have
classified monitoring under three categories following the observations made on the field
topography variations, based on the in situ ground-based measurements and remote sens-
ing. Table 4 describes different monitoring equipment, their monitored activity and the
method of monitoring used for rainfall-induced landslides.
Landslide Early Warning Systems (LEWS) are frequently used throughout the world
to communicate the risk to threatened communities and other stakeholders in advance to
reduce the risk of the landslide incident by taking necessary precautions. LEWS are cost-
effective and eco-friendly compared to structural mitigation measures (Intrieri et al. 2012;
Thiebes and Glade 2016). In addition, with the advancement of technology, new monitor-
ing strategies are developed and the existing methods are improved frequently (Chae et al.
2017; Crosta et al. 2017) and verification has gained an interest in literature and among
practitioners (Guzzetti et al. 2020). Calibration of these warning systems has become
simpler with the availability of reliable landslide databases and previous monitoring data
(Calvello and Pecoraro 2018; Froude and Petley 2018; Haque et al. 2016).
LEWS can be categorised as local or regional warning systems according to the refer-
ence scale (Calvello 2017; Calvello and Piciullo 2016; Pecoraro et al. 2019; Thiebes et al.
2012). Local LEWS are systems used within a single landslide scenario and Regional
LEWS are used over a municipal area, a region or a whole nation (Piciullo et al. 2018).
Stähli et al. (2015) have proposed another classification for early warning systems for natu-
ral hazards as Alarm systems, Warning systems and Forecasting systems. Based on func-
tion scale, local LEWS are employed as Alarm systems and Warning systems and regional
LEWS are employed as Warning systems or Forecasting systems. According to Di Biagio
and Kjekstad (2007), LEWS rely on four main actions which are monitoring, analysing
data and forecasting, warning, and response. Considering the forecast models and activities
which lead to warning, forecasting can be considered the core element of LEWS (Calvello
2017). The temporal predictions can be categorised in terms of lead time as long-term,
midterm and short-term predictions. As the lead time increases the accuracy of predicting
the time and the location can be less (Busslin 2009). Site-specific predictions can be made
using both in situ measurements and climate condition measurements such as displace-
ments, acceleration, pore water pressure, acoustic emissions, and rainfall. The most reli-
able method is to use kinematic parameters for time-based forecasting (Lacasse and Nadim
2009). Empirical and Semi-empirical methods have been developed in forecasting the time
of failure using kinematic parameters. However, regional and global scale predictions can
be made only using rainfall monitoring, hydro-meteorological approaches and geomorpho-
logical approaches (Intrieri et al. 2019).
Rainfall threshold-based predictions of landslides are relatively cost-effective and
widely used in LEWS for rainfall-induced landslides. Although this can be effective on a
regional or local scale when it comes to site-specific predictions, reliability can be low as
rainfall is an indirect indicator of instability (Intrieri et al. 2019). Rainfall-based thresholds

13
2214 Natural Hazards (2024) 120:2179–2231

Table 4  Different monitoring equipment, their monitored activities, method of monitoring and equipment
used in Sri Lanka
Monitored Activity Monitoring Method

Instruments used in
Predisposing factor

Remote Sensing

Meteorological
Groundwater

Geotechnical

Sri Lanka
Deformation

Geophysical
Hydrologic

Geodetic
Rainfall
Monitoring Instrument
Inclinometers
Extensometers
Differential monitoring of stability
Tiltmeter
Global Positioning System (GPS)
Interferometer
Total station
Camera
Ground Based Light Detection and Ranging
(GbLiDAR)
Airborne Light Detection and Ranging
(ALiDAR)
Ground Based Synthetic Aperture Radar
(GbSAR)
Interferometric Synthetic Aperture Radar
(InSAR)
Unmanned Air Vehicle (UAV)
Optic Fibre
Geophone
Crack meter
Accelerometer
Seismometer
Ground Penetrating Radar (GPR)
Piezometer
Perforated standpipe
Tensiometer
Thermocouple psychrometer
Electric conductivity sensor
Thermal conductivity sensor
Time Domain Reflectometer (TDR)
Satellite sensor
Rain gauge
Weather station
Hydrometer
Water level meter

can be derived using process-based methods and empirical methods. The process-based
rainfall thresholds can accommodate the site-specific conditions to define the triggering
rainfall intensity. However, the practical usage is limited due to the non-availability of
required detail parameters, the high cost of conducting the detailed investigations and the
high spatial variation in the subsurface. The empirical rainfall thresholds are derived from

13
Natural Hazards (2024) 120:2179–2231 2215

historical rainfall records that may and may not have resulted in slope failure. Thresholds
are usually defined as a lower bound separating rainfall events which triggered landslides
and events which did not cause failure. Although this can be conservative, it can often lead
to false alarms. Guzzetti et al. (2007) have listed 52 different rainfall intensity–duration
thresholds throughout the world and categorised them under the spatial scale.

6.3 Case studies for landslide remediation measures used in rainfall‑induced

landslides

These risk management strategies are used in many countries, but the financial strength of
a country is a key factor which governs landslide mitigation or risk acceptance (Sim et al.
2022). For instance, developed countries spend much more on landslide risk reduction and
monitoring compared to developing tropical countries. Developed countries have a high
willingness to pay and affect the environment to reduce landslide risk and a low tolerance
for risk. Comparatively, developing countries have a higher tolerance for risk as they have
low budgets for landslide risk mitigation. This has been evident in the risk management
projects implemented in developing countries and most of the landslide mitigation projects
are conducted with funding and technical support from developed countries. In addition,
more advanced and sophisticated monitoring such as remote sensing is not widely used
in developing countries for landslide risk management as evident from Table 4. Out of
the many rainfall-induced landslide mitigation projects conducted in tropical countries, an
outline of a few major rectification projects conducted in Sri Lanka is presented below
based on the literature and authors’ experience. Using case studies from Sri Lanka allowed
authors to add new insight from the case studies for the review comparing the effectiveness
of the mitigation measures with respect to monitoring data. The lessons learned from Sri
Lanka’s experience are applicable to a broader range of similar regions.
Watawala landslide is a slow-moving landslide which showed initial signs of movement
in 1957. Within the period 1976 to 1992, this slide moved around 89 m and in June 1992
slide moved 8 m, damaging the railway track and a locomotive engine which were within
the landslide body (Chandler and Broise 2000; Jayarathne et al. 1994). The slide was 60 m
wide and around 500 m in length, and a colluvium layer of a maximum depth of 26 m was
present at the site. Surface and subsurface drainage measures have been effectively used in
the site for mitigation. Drains of length in the order of 480 m were installed in the collu-
vium layer following a profile just underneath the identified failure surface using the direc-
tional drilling technique. To cover the width of the sliding mass, 11 of such drains were
used as shown in Fig. 14a. Due to the high movement accumulated at the failure surface
over the years, the shear strength has reduced to residual values and the prime objective of
the mitigation system was to facilitate rapid dissipation of the pore water pressure in the
vicinity of the failure surface. To withstand the strains due to possible movements, pipes
made of high density polyethylene (HDPE) with slotted drainage zones were installed in
the drilled holes. Six drainage wells of 200 mm diameter were installed up to a depth of
60 m into the firm ground and with automated pumps to pump out water. Surface drainage
measures have been implemented and connected to the adjacent natural valley and diverted
to a natural water stream. Trench drains up to 3 m deep have also been used in the site. Site
displacement monitoring is conducted using inclinometers and survey points. Piezometers
are used to measure the groundwater at different depths in the colluvium and weathered
rock layers and rain gauges to measure the precipitation (Chandler and Broise 2000).

13
2216 Natural Hazards (2024) 120:2179–2231

Fig. 15  Watawala landslide a current condition of the railway track, b water flowing through the directional
drains at the toe and c Regular maintenance work to prevent clogging of the drains

The deformation measurement from 1976 to 1992 was based on the lateral movement
of the railway track. According to Jayarathne et al. (1994), from 1976 to 1992, a rate of 2
to 3 m of lateral deformation per month was inferred, assuming the movements occurred
at a uniform rate over a period of 5 months per year. For the period of July to December
1992, a slope movement of 8 m/ month was observed. From the observations made in April
to May, a lateral deformation of 0.75 m per day was evident and by June 1993 a progres-
sive rate of 2 m/day was observed. During the failure on 3rd July 1993, a movement of
20–25 m was recorded. The subsequent movement for the 9 months after the rectification
work, which was in the range of 10 to 50 mm, was due to the consolidation with the low-
ering of groundwater table. Following the rectifications, no substantial movements were
detected that would impact the railway track. The sub-horizontal drains are functioning
effectively, and routine maintenance has had a positive and significant impact (see Fig. 15).
The Badulusirigama landslide which is a slow-moving deep-seated landslide in Badulla
administrative district in Sri Lanka, was showing accelerations in the movement with
heavy rainfalls. This landslide was 120 m wide and around 500 m in length and the slip
surface was around 10 to 20 m deep. Three independent slip surfaces; upper, middle and
lower could be identified in the narrow and long moving mass which is indicated as A, B
and C, respectively in Fig. 14b. A colluvium layer of 10 to 15 m depth was at the surface
underlain residual soil and weathered rock layers and the groundwater table were at a depth
of 6 to 7 m. For the mitigation of the landslide, 51 sub-horizontal drains of length 45 to
60 m were installed in 6 groups of radial (fan) orientations (to cover the width) at different
levels. These sub-horizontal drains are indicated as 1, 2, 3, 4, 5 and 6 in Fig. 14b. Fur-
thermore, surface drainage measures have also been provided. For monitoring purposes,
inclinometers, extensometers, water level meters, pipe strain gauges and rain gauges have
been provided on the site (Kankanamage et al. 2022; Kankanamge 2020). Instrumentation
locations for monitoring at the Badulusirigama landslide are shown in Fig. 14b.
As evident from Fig. 16a in SB-1 and SB-2 extensometers, a significant movement can
be observed in the slip surfaces B and C during the period of 8th to 17th November 2015.
A deformation of 29.0 mm was recorded for 10 days in SB-2 extensometer. In December
2014, SB-3 and SB-4 extensometers recorded a movement of 8 mm and 13 mm, respec-
tively. Countermeasure works were implemented from June 2016 to July 2017. After the
construction of drainage measures, significant movements were not observed. Although
with rainfall, no immediate drop in groundwater level has been observed in the BB-3 and
BB-5 as shown in Fig. 16b, a drop in the baseline can be observed, indicating that the
sub-horizontal drains are effective. From the BB-2 inclinometer data shown in Fig. 16c,
a cumulative deformation of 45 mm was observed within the two months of October and
November in 2015 at a ground level of 9.5 m to 11.5 m. After rectification measures were

13
Natural Hazards (2024) 120:2179–2231 2217

(a)

Extensometer
data from
2014/12/01 to
2018/03/01

(b)

Groundwater level
& rainfall data from
2014/12/01 to
2018/03/01

Before construction During construction After construction

(c) Inclinometer BB-2 data from 2015/05/28 to 2018/01/21

Vertical profile

Displacement profile

Rainfall profile

Before construction During construction After construction

Fig. 16  Monitoring data for Badulusirigama landslide before and after countermeasures. a Extensometer
readings. b Groundwater level and rainfall. c Inclinometer BB-2 vertical profile, displacement profile and
rainfall and (after Technical Cooperation for Landslide Mitigation Project 2018)

implemented, no such obvious movements were observed, indicating that the rectification
measures were effective.
Ginigathhena landslide is an ancient landslide which was reactivated in 2014 due to an
excavation during a road widening project that happened close to Bridge No 48/2 in the
Avissawella-Nuwara Eliya highway on rainy days. This was rectified using the soil nailing
technique as drainage measures alone cannot control the landslide movement. However, the

13
2218 Natural Hazards (2024) 120:2179–2231

Fig. 17  Rectification measures at Ginigathhena landslide

nailing rectification design was economised with the use of drainage improvement. Surface
drainage has been achieved using berm drains, cut-off drains, trench drains and vegetation
while subsurface drainage has been attained using sub-horizontal drains. The monitoring
has been conducted using surveying with control points, rain gauges and perforated stand-
pipes (Lakruwan and Kulathilaka 2021). Figure 17 depicts the rectification measures con-
ducted to stabilize the landslide.
The landslide which occurred in the Kahagolla area of the Badulla district along the
Beragala-Hali Ela highway which was a deep-seated landslide, was also rectified as a
combination of several rectification measures. The landslide had a length of 600 m and
a width of 150 m. The landslide mass was composed of colluvium and the landslide had
four separate moving blocks along the weathered rock-colluvium interface. At the toe area,
an embankment has been constructed to increase the resisting force. Concrete drains and
vegetation cover has been used for surface drainage control. Subsurface drainage control
has been achieved using six groups of 10 radially drilled sub-horizontal drains and five
drainage wells of 3.5 m diameter with ten sub-horizontal drains connecting radially. All
the drains are operating under gravity. In addition, ground anchors have been installed at
the toe of the landslide for additional support. The rectification measures are illustrated
in Fig. 18a. Monitoring equipment such as inclinometers, extensometers, water level
meters and rain gauges have been installed on the site for monitoring purposes as shown in
Fig. 18b (Karunawardena et al. 2022).
Monitoring data for the extensometers and groundwater level variations with rainfall is
shown in Fig. 18c. For the Kahagolla landslide site, before the design construction work
within the months of November and December of 2015 (T1), high cumulative rainfalls
were received (around 600 mm/month) and all four extensometers show significant defor-
mations indicating creep movement. Moreover, the groundwater level metres indicate a rise
in groundwater level and the time taken for the drawdown of water levels to the equilibrium
is high. During May 2016, short period heavy rainfalls have been received and a small

13
Natural Hazards (2024) 120:2179–2231 2219

Fig. 18  Kahagolla landslide. a Rectification measures implemented at the site. b Locations for the monitor-
ing instruments. c. Monitoring data before and after constructing countermeasures

13
2220 Natural Hazards (2024) 120:2179–2231

creep movement can be observed in all 4 extensometers while a sudden rise and gradual
decrease in the groundwater table was evident from water level metres. Meanwhile, after
construction work on April 2019 (T3), there is a slight movement indicated in extensom-
eter E3 but this rainfall event has not caused much fluctuations in groundwater levels. On
October–November 2019 (T4), an indication of groundwater fluctuation with rainfall can
be observed in B9 water level meter and simultaneous slight deformation in the extensom-
eter. Nevertheless, the groundwater table at B9 has dropped quickly back to the baseline
level within a short-duration. Deformations in both these T3 and T4 events are less than
5 mm. These results indicate that the mitigation work is effective.

7 Conclusion

The majority of the landslides in tropical regions are triggered by the high-intensity or pro-
longed rainfall conditions that exist in these regions. During rainfall, water can percolate to
the ground causing changes in the pore pressure regime and reducing the shear strength. A
rise in fatalities and damage from rainfall-induced landslides can be observed in the recent
past due to climate changes and urbanisation. This review study covers the landslide failure
mechanism and contributing factors for rainfall-induced landslides in tropical regions in
detail. Special attention was given to discuss on the complex geological conditions and
associated landslide hydrology in the tropical region. It is also focused on risk assessment
and management techniques for landslides. Although the techniques discussed under risk
assessment are not specific to the tropical areas, it was necessary to discuss on risk assess-
ment as a preliminary to risk management. Several case studies from a developing tropi-
cal country were reviewed, focussing on how to implement different landslide remediation
measures based on site-specific conditions. The main conclusions that can be drawn from
this study are summarised as follows.

• The landslides in the tropics have a strong correlation with the monsoons, tropical
cyclones and El Nina/La Nino cycles. However, early warning and identifying trigger-
ing rainfall intensities can be conducted to a reasonable accuracy if continuous data are
recorded on landslides and rainfall for long durations which is lacking in most coun-
tries in the region. The baseline information on past landslide are lacking due to lim-
ited financial resources and challenges in collecting complex conditions in the tropics
such as climate characteristics, rapid land use change in the region, complex geological
and hydrological factors, and complex failure mechanisms. To address this issue, it is
essential to promote awareness and invest in data collection and infrastructure with the
collaboration of government agencies, private organisations and the local community.
Maintaining a landslide inventory and systematic data collection can be considered as
the first step for more reliable landslide risk assessments. For most of the developing
tropical countries, international collaborations sharing knowledge and resources can
play a significant role in addressing the challenges in a global scale.
• Further identifying the seasonality and affecting local areas can be vital in resource
allocation and risk management. In landslide databases for rainfall-induced landslides
including the triggering rainfall intensity can be a future development for these data-
bases considering that it is the main external triggering factor. However, there can be
anthropogenic activities which can also contribute to these failures and should be care-
fully studied.

13
Natural Hazards (2024) 120:2179–2231 2221

• The intrinsic factors affecting rainfall-induced landslides in tropical countries are

reviewed in this article and the mechanism of failure due to water percolation is
explained in detail. The main reason for failure is identified as the instability due to the
decrease in shear strength caused by matric suction loss and pore water pressure build-
up in the subsurface due to infiltration. The presence of weak colluvial soils and the
contrast in permeability and shear strength between layers of colluvium, residual soils
and rocks at different levels of weathering make the system highly heterogeneous and
complicated. While potential failure surfaces are often associated with geological layer-
ing, it is important to note that in tropical regions large packages of weathered materi-
als without distinct layering are also common. Furthermore, in seepage and infiltra-
tion analysis, preferential flow paths should be given further attention for more accurate
pore water pressure distributions for stability calculations.
• The landslide risk management frameworks including susceptibility and hazard assess-
ment, risk estimation, and management strategies, are discussed. The available risk
management strategies are appraised as well. The recent trend in quantitative hazard
assessment techniques such as process-based methods for landslides can be advanta-
geous in many ways in predicting, early warning and reducing the likelihood of land-
slides. Nevertheless, the complexity of the internal and external causative factors
related to rainfall-induced landslides in tropical countries can make the deterministic
analysis complicated.
• Landslide remediation using engineering solutions is a key area reviewed in this arti-
cle. It is essential to consider the unique characteristics of the tropical environment and
tailoring the rectifications to local conditions. Several case studies from Sri Lanka are
presented in the article, representing a broader range of similar countries, allowing in-
depth analysis and a comprehensive understanding of landslide risk assessment and
management strategies. Author’s experience contributes to the richness of the findings
and access to data and involvement in these mitigation measures allow the authors to
analyse and compare the effectiveness of these mitigations. For rainfall-induced land-
slides, surface and subsurface drain improvement is the most effective and economical
remediation measure. This would reduce the percolation of water into the subsurface
and drain out already infiltrated groundwater from the landslide body facilitating rapid
dissipation of any pore water pressure increase. It was observed from case studies and
authors’ experience that in some occasions, these drainage measures are effectively
used along with one or two other methods such as; slope modification, retention struc-
tures and internal slope reinforcement.
• Monitoring and early warning can be a cost-effective risk management strategy that
can be used for rainfall-induced landslides. However, it was evident that most of the
advanced monitoring techniques based on remote sensing are still not in use in develop-
ing countries due to the lack of facilities. Recently, the verification of LEWS has gained
interest among practitioners and researchers. The rainfall threshold-based LEWS often
developed for a regional scale and can highlight the requirement of frequently updat-
ing rainfall threshold limits based on the monitoring of local conditions. Dealing with
varying exposure and vulnerability conditions of tropical countries which spread across
different continents, with different cultural, religious, and cosmovisions, is challenging
in landslide risk management. This requires a holistic and culturally sensitive approach
by integrating local knowledge, respecting cultural diversity and customising risk man-
agement strategies. The awareness of the risk among the locals can empower them to
ensure safety within the community with the collaboration of community leaders, reli-

13
2222 Natural Hazards (2024) 120:2179–2231

gious institutes and local authorities. The community engagement in risk reduction is a
feasible method of risk management for developing tropical countries.

Further studies can be recommended to identify the effectiveness of the different miti-
gation measures. Using site-specific conditions and monitoring data, deterministic mod-
els can be developed to identify the effectiveness of rectification measures and compare
them with the designs. This can be used in the economical rectification of designs, optimis-
ing the required factor of safety against failure. The rainfall-induced landslide rectifica-
tion designs can be further optimised by accounting for the matric suction in unsaturated
soils. In addition, the development of a scale to define the magnitude of landslides can
be considered as a requirement for practitioners when presenting the severity of rainfall-
induced landslides. In addition, runout assessment on flow-type landslides due to rainfall
needs improvements with models capable of including complex subsurface and hydrologi-
cal conditions prevailing in tropical areas. Based on analytical techniques, more accurate
predictions for runout distance and detrimental areas can be made, and these areas should
be included in susceptibility maps which are limited in the current practice.
Acknowledgements The authors would like to acknowledge Dr. Asiri Karunawardena, Director General of
National Building Research Organisation (NBRO), Sri Lanka for supporting the research work and the staff
of NBRO for providing assistance in collecting case study details for the review article.

Authors contribution All authors contributed to the study conception and design. Material preparation, data
collection and analysis were performed by MPA and SASK. Study was conceptualized and supervised by
DJR. The first draft of the article was written by MPA, SAS and DJR. All authors reviewed and commented
on previous versions of the manuscript. All authors read and approved the final manuscript.

Funding This work was financially supported by RMIT University, Australia, University of Moratuwa, Sri
Lanka and National Building Research Organisation, Sri Lanka.

Declarations
Conflict of interests The authors have no relevant financial or non-financial interests to disclose.

References
Alcántara-Ayala I (2002) Geomorphology, natural hazards, vulnerability and prevention of natural disasters
in developing countries. Geomorphology. https://​doi.​org/​10.​1016/​S0169-​555X(02)​00083-1
Aleotti P, Chowdhury R (1999) Landslide hazard assessment: Summary review and new perspectives. Bull
Eng Geol Env 58(1):21–44. https://​doi.​org/​10.​1007/​s1006​40050​066
Alimohammadlou Y, Najafi A, Yalcin A (2013) Landslide process and impacts: a proposed classification
method. CATENA. https://​doi.​org/​10.​1016/j.​catena.​2012.​11.​013
Arantes LT, Carvalho ACP, Carvalho APP, Lorandi R, Moschini LE, Di Lollo JA (2021) Surface runoff
associated with climate change and land use and land cover in southeast region of Brazil. Environ-
mental Challenges 3:100054. https://​doi.​org/​10.​1016/j.​envc.​2021.​100054
Arroyo-Solórzano M, Quesada-Román A, Barrantes-Castillo G (2022) Seismic and geomorphic assessment
for coseismic landslides zonation in tropical volcanic contexts. Natural Hazards 114(3). https://​doi.​
org/​10.​1007/​s11069-​022-​05492-8
Australian Geomechanics Society (2000) Landslide risk management concepts and guidelines. Aust
Geomech 35(1):49–92
Australian Geomechanics Society (2002) Landslide risk management concepts and guidelines. Aust
Geomech 37(2):1–44
Australian Geomechanics Society (2007) Guidelines for landslide susceptibility, hazard and risk zoning
for land use management. Australian Geomechanics Society Landslide Taskforce Landslide Zoning
Working Group. Aust Geomech 42(1):13–36

13
Natural Hazards (2024) 120:2179–2231 2223

Bera A, Mukhopadhyay BP, Das D (2019) Landslide hazard zonation mapping using multi-criteria analysis
with the help of GIS techniques: a case study from Eastern Himalayas, Namchi. South Sikkim Nat
Hazards 96(2):935–959. https://​doi.​org/​10.​1007/​s11069-​019-​03580-w
Bogaard TA, Greco R (2016) Landslide hydrology: from hydrology to pore pressure. Wiley Interdiscip Rev
Water. https://​doi.​org/​10.​1002/​wat2.​1126
Borrelli L, Greco R, Gullà G (2007) Weathering grade of rock masses as a predisposing factor to slope
instabilities: reconnaissance and control procedures. Geomorphology 87(3):158–175. https://​doi.​org/​
10.​1016/j.​geomo​rph.​2006.​03.​031
Brand EW, Phillipson HB (1985) Review of international practice for sampling and testing of residual soils.
In: Brand EW, Phillipson HB (eds) Sampling and testing of residual soi. Scorpion Press, London, pp
7–21
Brönnimann C, Stähli M, Schneider P, Seward L, Springman SM (2013) Bedrock exfiltration as a triggering
mechanism for shallow landslides. Water Resour Res 49(9):5155–5167. https://​doi.​org/​10.​1002/​wrcr.​
20386
Busslin M (2009) Landslide time forecast methods
Cai F, Ugai K, Wakai A, Li Q (1998) Effects of horizontal drains on slope stability under rainfall by three-
dimensional finite element analysis. Comput Geotech. https://​doi.​org/​10.​1016/​S0266-​352X(98)​
00021-4
Cai Y, Sun H, Shang Y, Xiong X (2014) An investigation of flow characteristics in slope siphon drains. J
Zhejiang Univ Sci A 15(1):22–30. https://​doi.​org/​10.​1631/​jzus.​A1300​178
Calvello M, Pecoraro G (2018) FraneItalia: a catalog of recent Italian landslides. Geoenviron Disasters.
https://​doi.​org/​10.​1186/​s40677-​018-​0105-5
Calvello M, Piciullo L (2016) Assessing the performance of regional landslide early warning models: the
EDuMaP method. Nat Hazard 16(1):103–122. https://​doi.​org/​10.​5194/​nhess-​16-​103-​2016
Calvello M (2017) Early warning strategies to cope with landslide risk. Riv Ital Geotecn 51(2):63–91.
https://​doi.​org/​10.​19199/​2017.2.​0557-​1405.​063
Carrión-Mero P, Montalván-Burbano N, Morante-Carballo F, Quesada-Román A, Apolo-Masache B (2021)
Worldwide research trends in landslide science. Int J Environ Res Public Health. https://​doi.​org/​10.​
3390/​ijerp​h1818​9445
Casagli N, Guzzetti F, Jaboyedoff M, Nadim F, Petley D (2017) Hydrological risk: landslides. In: Poljanšek
K, Marín Ferrer M, De Groeve T, Clark I (eds) Science for disaster risk management 2017: knowing
better and losing less. Publications Office of the European Union, Luxembourg. https://​doi.​org/​10.​
2788/​688605
Cascini L, Bonnard C, Corominas J, Jibson R, Montero-Olarte J (2005) Landslide hazard and risk zoning
for urban planning and development. In: Hungr O, Fell R, Couture R, Eberhardt E (eds) Landslide
risk management. CRC Press, Boca Raton, pp 199–236
Cascini L, Cuomo S, Pastor M, Sorbino G (2010) Modeling of rainfall-induced shallow landslides of the
flow-type. J Geotech Geoenviron Eng. https://​doi.​org/​10.​1061/​(asce)​gt.​1943-​5606.​00001​82
Cendrero A, Forte LM, Remondo J, Cuesta-Albertos JA (2020) Anthropocene geomorphic change. Climate
or human activities? Earth’s Future 8(5):e2019EF001305. https://​doi.​org/​10.​1029/​2019E​F0013​05
Chae BG, Park HJ, Catani F, Simoni A, Berti M (2017) Landslide prediction, monitoring and early warn-
ing: a concise review of state-of-the-art. Geosci J 21(6):1033–1070. https://​doi.​org/​10.​1007/​
s12303-​017-​0034-4
Chandler KR, Broise M (2000) Remediation of the Watawala landslide, Sri Lanka. In: Bromhead E, Dixon
N, Ibsen ML (eds) Landslides in research, theory and practice. Thomas Telford, Telford. https://​doi.​
org/​10.​1680/​lirta​pv1.​34617.​0041
Che VB, Fontijn K, Ernst GGJ, Kervyn M, Elburg M, Van Ranst E, Suh CE (2012) Evaluating the degree
of weathering in landslide-prone soils in the humid tropics: The case of Limbe, SW Cameroon. Geo-
derma. https://​doi.​org/​10.​1016/j.​geode​rma.​2011.​10.​013
Chen YC, Chang KT, Wang SF, Huang JC, Yu CK, Tu JY, Chu HJ, Liu CC (2019) Controls of preferential
orientation of earthquake- and rainfall-triggered landslides in Taiwan’s orogenic mountain belt. Earth
Surf Proc Land. https://​doi.​org/​10.​1002/​esp.​4601
Chigira M, Yokoyama O (2005) Weathering profile of non-welded ignimbrite and the water infiltration
behavior within it in relation to the generation of shallow landslides. Eng Geol. https://​doi.​org/​10.​
1016/j.​enggeo.​2004.​12.​008
Chigira M, Mohamad Z, Sian LC, Komoo I (2011) Landslides in weathered granitic rocks in Japan and
Malaysia. Bull Geol Soc Malays. https://​doi.​org/​10.​7186/​bgsm5​72011​01
Choi KY, Cheung RWM (2013) Landslide disaster prevention and mitigation through works in Hong Kong.
J Rock Mech Geotech Eng 5(5):354–365. https://​doi.​org/​10.​1016/j.​jrmge.​2013.​07.​007

13
2224 Natural Hazards (2024) 120:2179–2231

Cohen D, Schwarz M, Or D (2011) An analytical fiber bundle model for pullout mechanics of root bundles.
J Geophys Res Earth Surf. https://​doi.​org/​10.​1029/​2010J​F0018​86
Collins BD, Znidarcic D (2004) Stability analyses of rainfall induced landslides. J Geotech Geoenviron Eng
130(4):362–372. https://​doi.​org/​10.​1061/​(asce)​1090-​0241(2004)​130:​4(362)
Cooray PG (1994) Geological factors affecting landslide in Sri Lanka. In: Proceeding of the National Sym-
posium on Landslide in Sri Lanka., pp 12–22
Corominas J, van Westen C, Frattini P, Cascini L, Malet JP, Fotopoulou S, Catani F, Van Den Eeckhaut M,
Mavrouli O, Agliardi F, Pitilakis K, Winter MG, Pastor M, Ferlisi S, Tofani V, Hervás J, Smith JT
(2014) Recommendations for the quantitative analysis of landslide risk. Bull Eng Geol Env. https://​
doi.​org/​10.​1007/​s10064-​013-​0538-8
Cotecchia F, Lollino P, Petti R (2016) Efficacy of drainage trenches to stabilise deep slow landslides in clay
slopes. Géotech Lett 6(1):1–6. https://​doi.​org/​10.​1680/​jgele.​15.​00065
Cristiano E, Bogaard T, Barontini S (2016) Effects of anisotropy of preferential flow on the hydrology and
stability of landslides. Proc Earth Planet Sci 16:204–214. https://​doi.​org/​10.​1016/j.​proeps.​2016.​10.​
022
Crosta GB, Agliardi F, Rivolta C, Alberti S, Dei Cas L (2017) Long-term evolution and early warning strat-
egies for complex rockslides by real-time monitoring. Landslides 14(5):1615–1632. https://​doi.​org/​
10.​1007/​s10346-​017-​0817-8
Crozier MJ (2010) Deciphering the effect of climate change on landslide activity: A review. Geomorphol-
ogy 124(3):260–267. https://​doi.​org/​10.​1016/j.​geomo​rph.​2010.​04.​009
Crozier MJ (1986) Landslides: Causes, Consequences and Environment. In Géographie Physique Et Quater-
naire. https://​doi.​org/​10.​7202/​03270​2ar
Cruden D, Lan HX (2015) Using the working classification of landslides to assess the danger from a natural
slope. Eng Geol Soc Territory. https://​doi.​org/​10.​1007/​978-3-​319-​09057-3_1
Cruden D, Varnes D (1996) Landslide types and processes. In: Turner AK, Schuster R (eds) Landslides:
investigation and mitigation, special Report 247. National Academy Press, Washington
Dahal BK, Dahal RK (2017) Landslide hazard map: tool for optimization of low-cost mitigation. Geoenvi-
ron Disasters. https://​doi.​org/​10.​1186/​s40677-​017-​0071-3
Dahanayake K, Padmore PJ, Kulasena N (2011) Boudinage structures causing instability in some rock cut-
tings along the Southern Expressway, Sri Lanka. Q J Eng Geol Hydrogeol. https://​doi.​org/​10.​1144/​
1470-​9236/​10-​002
Dai FC, Lee CF, Ngai YY (2002) Landslide risk assessment and management: an overview. Eng Geol
64(1):65–87. https://​doi.​org/​10.​1016/​S0013-​7952(01)​00093-X
Debieche TH, Bogaard TA, Marc V, Emblanch C, Krzeminska DM, Malet JP (2012) Hydrological and
hydrochemical processes observed during a large-scale infiltration experiment at the Super-Sauze
mudslide (France). Hydrol Process. https://​doi.​org/​10.​1002/​hyp.​7843
Dewitte O, Depicker A, Moeyersons J, Dille A (2022) Mass movements in tropical climates. In: Shroder JF
(ed) Treatise on geomorphology. Academic Press, Cambridge
Dhakal AS, Sidle RC (2004) Pore water pressure assessment in a forest watershed: simulations and distrib-
uted field measurements related to forest practices. Water Resour Res. https://​doi.​org/​10.​1029/​2003W​
R0020​17
Dharmasena U, Bandara KN, Karunawardena A, Kulathilaka SAS (2015) Back analysis and rectification
of a failed cut slope in the Southern Expressway. In: Proceedings of the International Conference on
Geotechnical Engineering, pp 543–546
Di Biagio E, Kjekstad O (2007) Early warning, instrumentation and monitoring landslides. In: Proc 2nd in
regional training course, Reclaim II, 18–21
Dille A, Kervyn F, Mugaruka Bibentyo T, Delvaux D, Ganza GB, Ilombe Mawe G, Kalikone Buzera C,
Safari Nakito E, Moeyersons J, Monsieurs E, Nzolang C, Smets B, Kervyn M, Dewitte O (2019)
Causes and triggers of deep-seated hillslope instability in the tropics—insights from a 60-year record
of Ikoma landslide (DR Congo). Geomorphology. https://​doi.​org/​10.​1016/j.​geomo​rph.​2019.​106835
Durgin PB (1977) Landslides and the weathering of granitic rocks. GSA Rev Eng Geol 3:125–131. https://​
doi.​org/​10.​1130/​REG3-​p125
Duzgoren-Aydin NS (2002) Comparative study of weathering signatures in felsic igneous rocks of Hong
Kong. Chem Speciat Bioavailab. https://​doi.​org/​10.​3184/​09542​29027​82775​371
Duzgoren-Aydin NS, Aydin A (2006) Chemical and mineralogical heterogeneities of weathered igneous
profiles: implications for landslide investigations. Nat Hazards Earth Syst Sci 6(2):315–322. https://​
doi.​org/​10.​5194/​nhess-6-​315-​2006
Duzgoren-Aydin NS, Aydin A, Malpas J (2002) Distribution of clay minerals along a weathered pyroclastic
profile. Catena, Hong Kong. https://​doi.​org/​10.​1016/​S0341-​8162(02)​00066-8

13
Natural Hazards (2024) 120:2179–2231 2225

Duzgun H, Lacasse S (2005) Vulnerability and acceptable risk in integrated risk assessment framework. In:
Hungr O, Fell R, Couture R, Eberhardt E (eds) Landslide risk management, 1st edn. CRC Press, Boca
Raton, pp 515–526
Eggleton RA (2001) The regolith glossary. In: Surficial geology, soils and landscapes. CRC LEME
Ellefsen KJ, Hsieh PA, Shapiro AM (2002) Crosswell seismic investigation of hydraulically conductive,
fracture bedrock near Mirror Lake, New Hampshire. J Appl Geophys. https://​doi.​org/​10.​1016/​S0926-​
9851(02)​00149-0
Emberson R, Kirschbaum D, Stanley T (2021) Global connections between El Nino and landslide impacts.
Nat Commun. https://​doi.​org/​10.​1038/​s41467-​021-​22398-4
Evans SG, Guthrie RH, Roberts NJ, Bishop NF (2007) The disastrous 17 February 2006 rockslide-debris
avalanche on Leyte Island, Philippines: a catastrophic landslide in tropical mountain terrain. Nat Haz-
ards Earth Syst Sci 7(1):89–101. https://​doi.​org/​10.​5194/​nhess-7-​89-​2007
Fell R, Ho KKS, Lacasse S, Leroi E (2005) A framework for landslide risk assessment and management. In:
Hungr O, Fell R, Couture R, Eberhardt E (eds) Landslide risk management. Taylor and Francis, Boca
Raton, pp 3–26
Fell R, Corominas J, Bonnard C, Cascini L, Leroi E, Savage WZ (2008) Guidelines for landslide suscepti-
bility, hazard and risk zoning for land-use planning. Eng Geol 102(3–4):99–111. https://​doi.​org/​10.​
1016/j.​enggeo.​2008.​03.​014
Fleuty MJ (1964) The description of folds. Proc Geol Assoc 75(4):461–492. https://​doi.​org/​10.​1016/​S0016-​
7878(64)​80023-7
Fontoura TB, Coutinho RQ, da Silva FOT (2023) Geochemical and mineralogical contributions in the study
of sedimentary rock (barreiras formation) soils, Recife/Brazil: implications for landslides. Geotech
Geol Eng 41(1):205–224. https://​doi.​org/​10.​1007/​s10706-​022-​02274-8
Forrester, K. (2001). Subsurface drainage for slope stabilization. ASCE Press.
Fourie AB (1996) Predicting rainfall induced slope instability. Proc Inst Civ Eng Geotech Eng 119(4):211–
218. https://​doi.​org/​10.​1680/​igeng.​1996.​28757
Froude M, Petley D (2018) Global fatal landslide occurrence 2004 to 2016. Nat Hazards Earth Syst Sci Dis-
cuss 2012:1–44. https://​doi.​org/​10.​5194/​nhess-​2018-​49
Galeandro A, Šimůnek J, Simeone V (2013) Analysis of rainfall infiltration effects on the stability of pyro-
clastic soil veneer affected by vertical drying shrinkage fractures. Bull Eng Geol Env 72(3–4):447–
455. https://​doi.​org/​10.​1007/​s10064-​013-​0492-5
García-Soriano D, Quesada-Román A, Zamorano-Orozco JJ (2020) Geomorphological hazards susceptibil-
ity in high-density urban areas: a case study of Mexico City. J S Am Earth Sci 102:102667. https://​
doi.​org/​10.​1016/j.​jsames.​2020.​102667
Gariano SL, Guzzetti F (2016) Landslides in a changing climate. Earth Sci Rev 162:227–252. https://​doi.​
org/​10.​1016/j.​earsc​irev.​2016.​08.​011
Geological Society Engineering Group Working Party (1990) Tropical residual soils geological society
engineering group working party report. Q J Eng Geol. https://​doi.​org/​10.​1144/​GSL.​QJEG.​1990.​023.​
001.​01
Geotechnical Engineering Office (1998) Landslides and boulder falls from natural terrain: interim risk
guidelines. Geo Report No. 75
Geotechnical Manual for Slopes (2nd ed) (1984) Geotechnical Engineering Office, Civil Engineering
Department, Government of Hong Kong
Gerscovich DMS, Vargas EA, de Campos TMP (2006) On the evaluation of unsaturated flow in a natural
slope in Rio de Janeiro, Brazil. Eng Geol. https://​doi.​org/​10.​1016/j.​enggeo.​2006.​07.​008
Guzzetti F, Peruccacci S, Rossi M, Stark C (2007) Rainfall thresholds for the initiation of landslides
in Central and Southern Europe. Meteorol Atmos Phys 98:239–267. https://​doi.​org/​10.​1007/​
s00703-​007-​0262-7
Guzzetti F, Gariano SL, Peruccacci S, Brunetti MT, Marchesini I, Rossi M, Melillo M (2020) Geographical
landslide early warning systems. Earth Sci Rev. https://​doi.​org/​10.​1016/j.​earsc​irev.​2019.​102973
Guzzetti F (2000) Landslide fatalities and the evaluation of landslide risk in Italy. Eng Geol. https://​doi.​org/​
10.​1016/​S0013-​7952(00)​00047-8
Haque U, Blum P, da Silva PF, Andersen P, Pilz J, Chalov SR, Malet JP, Auflič MJ, Andres N, Poyiadji
E, Lamas PC, Zhang W, Peshevski I, Pétursson HG, Kurt T, Dobrev N, García-Davalillo JC, Halkia
M, Ferri S, Keellings D (2016) Fatal landslides in Europe. Landslides. https://​doi.​org/​10.​1007/​
s10346-​016-​0689-3
Hart MW (2000) Bedding-parallel shear zones as landslide mechanisms in horizontal sedimentary rocks.
Environ Eng Geosci 6(2):95–113. https://​doi.​org/​10.​2113/​gseeg​eosci.6.​2.​95
Haugen BD (2017) A design method for landslide surface water drainage control. Environ Eng Geosci
23(4):275–289. https://​doi.​org/​10.​2113/​gseeg​eosci.​23.4.​275

13
2226 Natural Hazards (2024) 120:2179–2231

Hencher SR, Lee SG, Carter TG, Richards LR (2011) Sheeting joints: characterisation, shear strength and
engineering. Rock Mech Rock Eng 44(1):1–22. https://​doi.​org/​10.​1007/​s00603-​010-​0100-y
Huat BBK, Ali FHJ, Low TH (2006) Water infiltration characteristics of unsaturated soil slope and
its effect on suction and stability. Geotech Geol Eng 24(5):1293–1306. https://​doi.​org/​10.​1007/​
s10706-​005-​1881-8
Huggel C, Clague JJ, Korup O (2012) Is climate change responsible for changing landslide activity in high
mountains? Earth Surf Process Landf 37(1):77–91. https://​doi.​org/​10.​1002/​esp.​2223
Hungr O, Clague J, Morgenstern NR, VanDine D, Stadel D (2016) A review of landslide risk acceptabil-
ity practices in various countries. Landslides Eng Slopes Exp Theory Pract. https://​doi.​org/​10.​1201/​
b21520-​135
Hungr O (2016) A review of landslide hazard and risk assessment methodology. Landslides Eng Slopes Exp
Theory Pract. https://​doi.​org/​10.​1201/​b21520-3
Idirimanna IAND, Kulathilaka SAS (2019) Back analysis of slope failure at Welipanna, Southern Express
Way. Geotech J 7(1):1–10
Intrieri E, Gigli G, Mugnai F, Fanti R, Casagli N (2012) Design and implementation of a landslide early
warning system. Eng Geol 147–148:85–90. https://​doi.​org/​10.​1016/j.​enggeo.​2012.​07.​017
Intrieri E, Carlà T, Gigli G (2019) Forecasting the time of failure of landslides at slope-scale: a literature
review. Earth-Sci Rev 193(September 2018):333–349. https://​doi.​org/​10.​1016/j.​earsc​irev.​2019.​03.​019
Irfan TY (1998) Structurally controlled landslides in saprolitic soils in Hong Kong. Geotech Geol Eng
16(3):215–238. https://​doi.​org/​10.​1023/A:​10088​05827​178
IS14243-2 (1995) Guidelines for selection and development of site for building in hill areas, Part 2: Selec-
tion and development (Patent No. 14243–2). In Bureau of Indian Standards, N.Delhi, India (No.
14243–2)
Iverson RM (2000) Landslide triggering by rain infiltration. Water Resour Res. https://​doi.​org/​10.​1029/​
2000W​R9000​90
Iverson RM, Major JJ (1987) Rainfall, ground-water flow, and seasonal movement at Minor Creek landslide,
northwestern California: physical interpretation of empirical relations. Geol Soc Am Bull. https://​doi.​
org/​10.​1130/​0016-​7606(1987)​99%​3c579:​RGFASM%​3e2.0.​CO;2
Jackson L, Bobrowsky P, Bichler A (2012) Identification, maps and mapping—Canadian Technical Guide-
lines and Best Practices related to Landslides: a national initiative for loss reduction: Gelogical Sur-
vey of Canada. Open File. https://​doi.​org/​10.​4095/​292122
Jayarathne VS, Jayarani HM, Dissanayake DMAS, Fernando TRJ (1994) The meaning of surface move-
ments monitored at the Watawala Earthslide. In: Proceedings of National Symposium on Landslides
in Sri Lanka, pp 107–112
Kang X, Wang S, Wu W, Xu G, Zhao J, Liu F (2022) Soil–water interaction affecting a deep-seated land-
slide: from field monitoring to experimental analysis. Bull Eng Geol Env 81(2):82. https://​doi.​org/​10.​
1007/​s10064-​021-​02556-0
Kankanamage L, Kulathilaka A, Wijerathna M (2022) Effectiveness of radial sub-horizontal drainage in
improving the landslide stability: a case study on central highlands, Sri Lanka. In: Rahman MM,
Jaska M (Eds) Proceedings of the 20th international conference on soil mechanics and geotechnical
engineering (pp. 2467–2472). Australian Geomechanics Society
Kankanamge LUM (2020) Effectiveness of subsurface drainage and vegetation in enhancing the slope sta-
bility: a comprehensive case study on Badulusirigama landslide. University of Moratuwa, Moratuwa
Karunawardena A, Toki M, Thilakarathne T, Galhena W (2022) Structural mitigation of a deep-seated slow
moving landslide along a major national road in Sri Lanka. In: Rahman MM, Jaska M (Eds) Pro-
ceedings of the 20th international conference on soil mechanics and geotechnical engineering (pp.
2473–2478). Australian Geomechanics Society
Kirschbaum DB, Adler R, Hong Y, Hill S, Lerner-Lam A (2010) A global landslide catalog for hazard appli-
cations: method, results, and limitations. Nat Hazards. https://​doi.​org/​10.​1007/​s11069-​009-​9401-4
Kirschbaum D, Stanley T, Zhou Y (2015) Spatial and temporal analysis of a global landslide catalog. Geo-
morphology 249:4–15. https://​doi.​org/​10.​1016/j.​geomo​rph.​2015.​03.​016
Kjekstad O, Highland L (2009) Economic and social impacts of landslides. Landslides Disaster Risk
Reduct. https://​doi.​org/​10.​1007/​978-3-​540-​69970-5_​30
Klose M, Damm B, Terhorst B (2015) Landslide cost modeling for transportation infrastructures: a meth-
odological approach. Landslides. https://​doi.​org/​10.​1007/​s10346-​014-​0481-1
Krzeminska DM, Bogaard TA, Malet JP, Van Beek LPH (2013) A model of hydrological and mechanical
feedbacks of preferential fissure flow in a slow-moving landslide. Hydrol Earth Syst Sci. https://​doi.​
org/​10.​5194/​hess-​17-​947-​2013
Kukemilks K, Wagner JF (2021) Detection of preferential water flow by electrical resistivity tomography
and self-potential method. Appl Sci (switzerland). https://​doi.​org/​10.​3390/​app11​094224

13
Natural Hazards (2024) 120:2179–2231 2227

Kwan JSH, Koo RCH, Ng CWW (2015) Landslide mobility analysis for design of multiple debris-resisting
barriers. Can Geotech J. https://​doi.​org/​10.​1139/​cgj-​2014-​0152
Lacasse S, Nadim F (2009) Landslide risk assessment and mitigation strategy. Landslides Disaster Risk
Reduct. https://​doi.​org/​10.​1007/​978-3-​540-​69970-5_3
Lacerda WA (2007) Landslide initiation in saprolite and colluvium in southern Brazil: field and laboratory
observations. Geomorphology 87(3):104–119. https://​doi.​org/​10.​1016/j.​geomo​rph.​2006.​03.​037
Lakruwan SOADM, Kulathilaka SAS (2021) Economizing soil nailing design by drainage improvement—
case history at Ginigathhena. In: Vilímek V, Wang F, Strom A, Sassa K, Bobrowsky PT, Takara K
(eds) Understanding and reducing landslide disaster risk. WLF 2020. ICL contribution to landslide
disaster risk reduction. Springer, Berlin, pp 331–339
Lau KC, Kenney TC (1984) Horizontal drains to stabilize clay slopes. Can Geotech J. https://​doi.​org/​10.​
1139/​t84-​027
Lee ML, Kassim A, Gofar N (2011) Performances of two instrumented laboratory models for the study of
rainfall infiltration into unsaturated soils. Eng Geol. https://​doi.​org/​10.​1016/j.​enggeo.​2010.​10.​007
Lee ML, Ng KY, Huang YF, Li WC (2014) Rainfall-induced landslides in Hulu Kelang area. Natural Haz-
ards, Malaysia. https://​doi.​org/​10.​1007/​s11069-​013-​0814-8
Lee RWH, Law RHC, Lo DOK (2018) Importance of surface drainage management to slope perfor-
mance. HKIE Trans 25(3):182–191. https://​doi.​org/​10.​1080/​10236​97X.​2018.​14994​49
Lee ML, Koo CH, Chong SY, Chin DJ (2022) Laboratory and numerical studies of rainfall infiltration
into residual soil slope improved by biomediated soil cover. Water (switzerland). https://​doi.​org/​
10.​3390/​w1405​0744
Lee LM, Jones DK (2014) Landslide risk assessment, 2nd edn. ICE Publishing, London. https://​doi.​org/​
10.​1680/​lra.​31715
Lo JYC, Chau SF, Cheuk JCY (2011) Performance of a drainage tunnel and subvertical drain system. In:
14th Australasian Tunnelling Conference 2011: Development of Underground Space, Proceedings,
407–416
Luna BQ, Blahut J, Kappes M, Akbas SO, Malet J-P, Remaître A, van Asch T, Jaboyedoff M (2014)
Methods for Debris Flow Hazard and Risk Assessment. In: Van Asch T, Corominas J, Greiving S,
Malet J-P, Sterlacchini S (eds) mountain risks: from prediction to management and governance.
Springer, Netherlands, pp 133–177. https://​doi.​org/​10.​1007/​978-​94-​007-​6769-0_5
Maes J, Kervyn M, de Hontheim A, Dewitte O, Jacobs L, Mertens K, Vanmaercke M, Vranken L, Poesen
J (2017) Landslide risk reduction measures: a review of practices and challenges for the tropics.
Progress Phys Geogr Earth Environ 41(2):191–221. https://​doi.​org/​10.​1177/​03091​33316​689344
Mancarella D, Hungr O (2010) Analysis of run-up of granular avalanches against steep, adverse slopes
and protective barriers. Can Geotech J. https://​doi.​org/​10.​1139/​T09-​143
Manchado MA, Allen S, Ballesteros-Cánovas JA, Dhakal A, Dhital MR, Stoffel M (2021) Three decades
of landslide activity in western Nepal: new insights into trends and climate drivers. Landslides.
https://​doi.​org/​10.​1007/​s10346-​021-​01632-6
Masi EB, Segoni S, Tofani V (2021) Root reinforcement in slope stability models: a review. Geosciences
(switzerland). https://​doi.​org/​10.​3390/​geosc​ience​s1105​0212
McColl ST (2015) Landslide causes and triggers. In: Shroder JF, Davies Risks TBT-LH, Disasters, (eds)
Hazards and disasters series. Academic Press, Cambridge. https://​doi.​org/​10.​1016/​B978-0-​12-​
396452-​6.​00002-1
McDougall S (2017) 2014 Canadian geotechnical colloquium: Landslide runout analysis—current prac-
tice and challenges. Can Geotech J 54(5):605–620. https://​doi.​org/​10.​1139/​cgj-​2016-​0104
Migoń P (2010) Mass movement and landscape evolution in weathered granite and gneiss terrains. Geol
Soc Eng Geol Spec Publ 23:33–45. https://​doi.​org/​10.​1144/​EGSP23.4
Mohamed WNAW, Osman N, Abdullah R (2022) A review of bioengineering techniques for slope stabil-
ity in Malaysia. Int J Environ Sci Technol. https://​doi.​org/​10.​1007/​s13762-​022-​04235-3
Morris CE, Stormont JC (1998) Evaluation of numerical simulations of capillary barrier field tests. Geo-
tech Geol Eng. https://​doi.​org/​10.​1023/A:​10088​53710​339
Mukhlisin M, Aziz NABA (2016) Study of horizontal drain effect on slope stability. J Geol Soc India.
https://​doi.​org/​10.​1007/​s12594-​016-​0417-6
Ng CWW, Shi Q (1998) A numerical investigation of the stability of unsaturated soil slopes subjected to
transient seepage. Comput Geotech 22(1):1–28. https://​doi.​org/​10.​1016/​S0266-​352X(97)​00036-0
Ng CWW, Liu J, Chen R, Xu J (2015) Physical and numerical modeling of an inclined three-layer (silt/
gravelly sand/clay) capillary barrier cover system under extreme rainfall. Waste Manage. https://​
doi.​org/​10.​1016/j.​wasman.​2014.​12.​013
Oilier CD (2010) Very deep weathering and related landslides. Geol Soc Eng Geol Spec Publ. https://​
doi.​org/​10.​1144/​EGSP23.2

13
2228 Natural Hazards (2024) 120:2179–2231

Onda Y, Tsujimura M, Tabuchi H (2004) The role of subsurface water flow paths on hillslope hydrologi-
cal processes, landslides and landform development in steep mountains of Japan. Hydrol Process
18:637–650. https://​doi.​org/​10.​1002/​hyp.​1362
Ozturk U, Bozzolan E, Holcombe EA, Shukla R, Pianosi F, Wagener T (2022) How climate change and
unplanned urban sprawl bring more landslides. Nature 608:262–265
Panwar V, Sen S (2020) Disaster damage records of EM-DAT and desinventar: a systematic comparison.
Econ Disasters Clim Change 4(2):295–317. https://​doi.​org/​10.​1007/​s41885-​019-​00052-0
Pardeshi SD, Autade SE, Pardeshi SS (2013) Landslide hazard assessment: recent trends and techniques.
Springerplus 2(1):1–11. https://​doi.​org/​10.​1186/​2193-​1801-2-​523
Pecoraro G, Calvello M, Piciullo L (2019) Monitoring strategies for local landslide early warning sys-
tems. Landslides 16(2):213–231. https://​doi.​org/​10.​1007/​s10346-​018-​1068-z
Petley DN (2010) On the impact of climate change and population growth on the occurrence of fatal
landslides in South, East and SE Asia. Q J Eng GeolHydrogeol 43(4):487–496. https://​doi.​org/​10.​
1144/​1470-​9236/​09-​001
Petley D (2012) Global patterns of loss of life from landslides. Geology 40(10):927–930. https://​doi.​org/​
10.​1130/​G33217.1
Petley D, Dunning S, Rosser N (2005) The analysis of global landslide risk through the creation of a data-
base of worldwide landslide fatalities. In: Hungr EO et al (eds) Landslide risk management. CRC
Press, Boca Raton
Picarelli L, Lacasse S, Ho KKS (2021) The impact of climate change on landslide hazard and risk BT—
understanding and reducing landslide disaster risk: Volume 1 Sendai landslide partnerships and kyoto
landslide commitment (Sassa K, Mikoš M, Sassa S, Bobrowsky PT, Takara K, Dang K (Eds); pp.
131–141). Springer International Publishing. https://​doi.​org/​10.​1007/​978-3-​030-​60196-6_6
Piciullo L, Calvello M, Cepeda JM (2018) Territorial early warning systems for rainfall-induced landslides.
Earth-Sci Rev. https://​doi.​org/​10.​1016/j.​earsc​irev.​2018.​02.​013
Pollen-Bankhead N, Simon A (2010) Hydrologic and hydraulic effects of riparian root networks on stream-
bank stability: Is mechanical root-reinforcement the whole story? Geomorphology. https://​doi.​org/​10.​
1016/j.​geomo​rph.​2009.​11.​013
Popescu M (2001) A suggested method for reporting landslide remedial measures. Bull Eng Geol Env.
https://​doi.​org/​10.​1007/​s1006​40000​084
Prior DB, Ho C (1972) Coastal and mountain slope instability on the islands of St. Lucia and Barbados. Eng
Geol. https://​doi.​org/​10.​1016/​0013-​7952(72)​90021-X
Puente-Sotomayor F, Egas A, Teller J (2021) Land policies for landslide risk reduction in Andean cities.
Habitat Int. https://​doi.​org/​10.​1016/j.​habit​atint.​2020.​102298
Quesada-Román A (2021) Landslide risk index map at the municipal scale for Costa Rica. Int J Disaster
Risk Reduct 56:102144. https://​doi.​org/​10.​1016/j.​ijdrr.​2021.​102144
Quesada-Román A, Campos-Durán D (2023) Natural disaster risk inequalities in Central America. Papers
Appl Geogr. https://​doi.​org/​10.​1080/​23754​931.​2022.​20818​14
Quesada-Román A, Fallas-López B, Hernández-Espinoza K, Stoffel M, Ballesteros-Cánovas JA (2019)
Relationships between earthquakes, hurricanes, and landslides in Costa Rica. Landslides. https://​doi.​
org/​10.​1007/​s10346-​019-​01209-4
Quesada-Román A, Villalobos-Portilla E, Campos-Durán D (2021) Hydrometeorological disasters in urban
areas of Costa Rica. Central Am Environ Hazards 20(3):264–278. https://​doi.​org/​10.​1080/​17477​891.​
2020.​17910​34
Rahardjo H, Lim TT, Chang MF, Fredlund DG (1995) Shear-strength characteristics of a residual soil. Can
Geotech J. https://​doi.​org/​10.​1139/​t95-​005
Rahardjo H, Hritzuk KJ, Leong EC, Rezaur RB (2003) Effectiveness of horizontal drains for slope stability.
Eng Geol. https://​doi.​org/​10.​1016/​S0013-​7952(02)​00288-0
Rahardjo H, Ong TH, Rezaur RB, Leong EC (2007) Factors controlling instability of homogeneous soil
slopes under rainfall. J Geotech Geoenviron Eng 133(12):1532–1543. https://​doi.​org/​10.​1061/​(asce)​
1090-​0241(2007)​133:​12(1532)
Rahardjo H, Santoso VA, Leong EC, Ng YS, Hua CJ (2011) Performance of horizontal drains in residual
soil slopes. Soils Found. https://​doi.​org/​10.​3208/​sandf.​51.​437
Rahardjo H, Santoso VA, Leong EC, Ng YS, Hua CJ (2012) Performance of an instrumented slope covered
by a capillary barrier system. J Geotech Geoenviron Eng. https://​doi.​org/​10.​1061/​(asce)​gt.​1943-​5606.​
00006​00
Ramberg H (1955) Natural and experimental boudinage and pinch-and-swell structures. J Geol. https://​doi.​
org/​10.​1086/​626293
Ramsay JG (1967) Folding and fracturing of rocks. McGraw–Hil, New York

13
Natural Hazards (2024) 120:2179–2231 2229

Regmi AD, Yoshida K, Dhital MR, Devkota K (2013) Effect of rock weathering, clay mineralogy, and geo-
logical structures in the formation of large landslide, a case study from Dumre Besei landslide. Lesser
Himalaya Nepal Landslides 10(1):1–13. https://​doi.​org/​10.​1007/​s10346-​011-​0311-7
Regmi AD, Yoshida K, Dhital MR, Pradhan B (2014) Weathering and mineralogical variation in gneissic
rocks and their effect in Sangrumba Landslide, East Nepal. Environ Earth Sci 14:8
Revil A, Cary L, Fan Q, Finizola A, Trolard F (2005) Self-potential signals associated with preferential
ground water flow pathways in a buried paleo-channel. Geophys Res Lett. https://​doi.​org/​10.​1029/​
2004G​L0221​24
Ruxton BP, Berry L (1957) Weathering of granite and associated erosional features in Hong Kong. Bull
Geol Soc Am. https://​doi.​org/​10.​1130/​0016-​7606(1957)​68[1263:​WOGAAE]​2.0.​CO;2
Sajinkumar KS, Anbazhagan S, Pradeepkumar AP, Rani VR (2011) Weathering and landslide occurrences
in parts of Western Ghats, Kerala. J Geol Soc India. https://​doi.​org/​10.​1007/​s12594-​011-​0089-1
Scheidl C, Heiser M, Kamper S, Thaler T, Klebinder K, Nagl F, Lechner V, Markart G, Rammer W, Seidl R
(2020) The influence of climate change and canopy disturbances on landslide susceptibility in head-
water catchments. Sci Total Environ 742:140588. https://​doi.​org/​10.​1016/j.​scito​tenv.​2020.​140588
Schulz WH, Lidke DJ, Godt JW (2008) Modeling the spatial distribution of landslide-prone colluvium and
shallow groundwater on hillslopes of Seattle. WA Earth Surf Process Landf 33(1):123–141. https://​
doi.​org/​10.​1002/​esp.​1535
Scott DN, Wohl EE (2019) Bedrock fracture influences on geomorphic process and form across process
domains and scales. Earth Surf Proc Land 44(1):27–45. https://​doi.​org/​10.​1002/​esp.​4473
Sepúlveda S, Petley D (2015) Regional trends and controlling factors of fatal landslides in Latin America
and the Caribbean. Nat Hazard 15(8):1821–1833. https://​doi.​org/​10.​5194/​nhess-​15-​1821-​2015
Shano L, Raghuvanshi T, Meten M (2020) Landslide susceptibility evaluation and hazard zonation tech-
niques—a review. Geoenviron Disasters 7(18):19. https://​doi.​org/​10.​1186/​s40677-​020-​00152-0
Shao W, Bogaard TA, Bakker M, Greco R (2015) Quantification of the influence of preferential flow on
slope stability using a numerical modelling approach. Hydrol Earth Syst Sci. https://​doi.​org/​10.​5194/​
hess-​19-​2197-​2015
Shaw R (1997) Variations in sub-tropical deep weathering profiles over the Kowloon Granite, Hong Kong. J
Geol Soc 154(6):1077–1085. https://​doi.​org/​10.​1144/​gsjgs.​154.6.​1077
Sidle RC, Bogaard TA (2016) Dynamic earth system and ecological controls of rainfall-initiated landslides.
Earth-Sci Rev 159:275–291. https://​doi.​org/​10.​1016/j.​earsc​irev.​2016.​05.​013
Sidle RC, Ochiai H (2013) Landslides: processes, prediction, and land use. AUG J. https://​doi.​org/​10.​1029/​
WM018
Sidle RC, Chigira M (2004) Landslides and debris flows strike Kyushu, Japan. Eos. https://​doi.​org/​10.​1029/​
2004E​O1500​01
Sim KB, Lee ML, Wong SY (2022) A review of landslide acceptable risk and tolerable risk. Geoenviron
Disasters. https://​doi.​org/​10.​1186/​s40677-​022-​00205-6
Sing H, Huat BBK (2012) Formation and classification of tropical residual soils. In: Huat BBK, Toll DG,
Prasad A (eds) Handbook of tropical residual soils engineering. CRC Press, Boca Raton, pp 21–62.
https://​doi.​org/​10.​1201/​b12302-2
Smith E, Smettem RJ, Broadbridge P, Woolhiser DA (2002) Infiltration theory for hydrologic applications.
American Geophysical Union, Washington. https://​doi.​org/​10.​1029/​wm015
Song D, Zhou GGD, Xu M, Choi CE, Li S, Zheng Y (2019) Quantitative analysis of debris-flow flexi-
ble barrier capacity from momentum and energy perspectives. Eng Geol. https://​doi.​org/​10.​1016/j.​
enggeo.​2019.​02.​010
Stähli M, Sättele M, Huggel C, McArdell BW, Lehmann P, Van Herwijnen A, Berne A, Schleiss M, Ferrari
A, Kos A, Or D, Springman SM (2015) Monitoring and prediction in early warning systems for rapid
mass movements. Nat Hazard. https://​doi.​org/​10.​5194/​nhess-​15-​905-​2015
Stead D, Wolter A (2015) A critical review of rock slope failure mechanisms: the importance of structural
geology. J Struct Geol 74:1–23. https://​doi.​org/​10.​1016/J.​JSG.​2015.​02.​002
Strouth A, McDougall S (2021) Societal risk evaluation for landslides: historical synthesis and proposed
tools. Landslides 18(3):1071–1085. https://​doi.​org/​10.​1007/​s10346-​020-​01547-8
Sun H, Wong LNY, Quan Shang Y, Jiang Shen Y, Lü Q (2010) Evaluation of drainage tunnel effectiveness
in landslide control. Landslides. https://​doi.​org/​10.​1007/​s10346-​010-​0210-3
Sun HY, Pan P, Lü Q, Wei ZL, Xie W, Zhan W (2019) A case study of a rainfall-induced landslide involving
weak interlayer and its treatment using the siphon drainage method. Bull Eng Geol Env 78(6):4063–
4074. https://​doi.​org/​10.​1007/​s10064-​018-​1365-8
Tagarelli V, Cotecchia F (2022) Coupled hydro-mechanical analysis of the effects of medium depth drainage
trenches mitigating deep landslide activity. Eng Geol 297:106510. https://​doi.​org/​10.​1016/j.​enggeo.​
2021.​106510

13
2230 Natural Hazards (2024) 120:2179–2231

Tami D, Rahardjo H, Leong EC, Fredlund DG (2004) Design and laboratory verification of a physical
model of sloping capillary barrier. Can Geotech J. https://​doi.​org/​10.​1139/​T04-​036
Technical Cooperation for Landslide Mitigation Project (2018) Monitoring report No.10
Terzaghi K, Peck RB (1967) Soil mechanics in engineering practice. John Wiley, Hoboken
Terzaghi K (1950) Mechanism of landslides. In: Application of geology to engineering practice (pp.
83–123). Geological Society of America. https://​doi.​org/​10.​1130/​BERKEY.​1950.​83
Thiebes B, Glade T, Bell R (2012) Landslide analysis and integrative early warning-local and regional case
studies. In: Eberhardt E (Ed) Landslides and engineered slopes: protecting society through improved
understanding. In: Proceedings of the 11th International and 2nd North American Symposium on
Landslides and Engineered Slopes, 2012 (Vol. 2, pp. 1915–1921). Taylor & Francis Group
Thiebes B, Glade T (2016) Landslide Early Warning Systems-fundamental concepts and innovative applica-
tions. Landslides Eng Slopes Exper Theory Pract 3:1903–1911. https://​doi.​org/​10.​1201/​b21520-​238
Thoeni K, Giacomini A, Lambert C, Sloan SW, Carter JP (2014) A 3D discrete element modelling approach
for rockfall analysis with drapery systems. Int J Rock Mech Min Sci 68:107–119. https://​doi.​org/​10.​
1016/j.​ijrmms.​2014.​02.​008
Thomas MF (1994) Geomorphology in the Tropics. A Study of Weathering and Denudation in Low Lati-
tudes. Wiley, New York
Travelletti J, Sailhac P, Malet JP, Grandjean G, Ponton J (2012) Hydrological response of weathered
clay-shale slopes: Water infiltration monitoring with time-lapse electrical resistivity tomography.
Hydrol Process 26(14):2016–2119. https://​doi.​org/​10.​1002/​hyp.​7983
Turner AK (2018) Social and environmental impacts of landslides. Innov Infrastruct Solut. https://​doi.​
org/​10.​1007/​s41062-​018-​0175-y
Uchida T, Kosugi KNI, Mizuyama T (2001) Effects of pipeflow on hydrological process and its rela-
tion to landslide: a review of pipeflow studies in forested headwater catchments. Hydrol Process
15(11):2151–2174. https://​doi.​org/​10.​1002/​hyp.​281
UNISDR (2022) DesInventar. United Nations Office for Disaster Risk Reduction. https://​www.​desin​ven-
tar.​net
Urciuoli G, Pirone M (2013) Subsurface drainage for slope stabilization. In: Margottini C, Canuti P,
Sassa K (eds) Landslide science and practice. Springer, Berlin, pp 577–585
van Westen CJ, Castellanos E, Kuriakose SL (2008) Spatial data for landslide susceptibility, hazard, and
vulnerability assessment: an overview. Eng Geol. https://​doi.​org/​10.​1016/j.​enggeo.​2008.​03.​010
Vieira BC, Fernandes NF (2004) Landslides in Rio de Janeiro: the role played by variations in soil
hydraulic conductivity. Hydrol Process. https://​doi.​org/​10.​1002/​hyp.​1363
Volkwein A, Schellenberg K, Labiouse V, Agliardi F, Berger F, Bourrier F, Dorren LKA, Gerber W,
Jaboyedoff M (2011) Rockfall characterisation and structural protection—a review. Nat Hazards
Earth Syst Sci. https://​doi.​org/​10.​5194/​nhess-​11-​2617-​2011
Wang G, Sassa K (2001) Factors affecting rainfall-induced flowslides in laboratory flume tests. Geotech-
nique. https://​doi.​org/​10.​1680/​geot.​2001.​51.7.​587
Wang S, Wu W, Wang J, Yin Z, Cui D, Xiang W (2018) Residual-state creep of clastic soil in a
reactivated slow-moving landslide in the Three Gorges Reservoir Region. China Landslides
15(12):2413–2422. https://​doi.​org/​10.​1007/​s10346-​018-​1043-8
Wang D, Xu H, Wang L, Wu X, Sun H (2020) Statistical analyses of the effect of a drainage tunnel on
landslide hydrogeological characteristics. Hydrol Process 34(11):2418–2432. https://​doi.​org/​10.​
1002/​hyp.​13738
Wei ZL, Shang YQ, Sun HY, Xu HD, Wang DF (2019) The effectiveness of a drainage tunnel in increas-
ing the rainfall threshold of a deep-seated landslide. Landslides 16:1731–1744. https://​doi.​org/​10.​
1007/​s10346-​019-​01241-4
Wei Z, Wang D, Xu H, Sun H (2020) Clarifying the effectiveness of drainage tunnels in landslide con-
trols based on high-frequency in-site monitoring. Bull Eng Geol Env 79(7):3289–3305. https://​doi.​
org/​10.​1007/​s10064-​020-​01769-z
Wen BP, Duzgoren-Aydin NS, Aydin A (2004) Geochemical characteristics of the slip zones of a land-
slide in granitic saprolite, Hong Kong: Implications for their development and microenvironments.
Environ Geol. https://​doi.​org/​10.​1007/​s00254-​004-​1145-6
Wesley L (1990) Influence of structure and composition on residual soils. J Geotech Eng 116(4):589–
603. https://​doi.​org/​10.​1061/​(ASCE)​0733-​9410(1990)​116:​4(589)
Wesley L (2009) Fundamentals of soil mechanics for sedimentary and residual soils. Fundam Soil Mech
Sediment Residual Soils. https://​doi.​org/​10.​1002/​97804​70549​056
Wesley L (2011) Stability of slopes in residual soils. Obras y Proyectos 10:47–61. https://​doi.​org/​10.​
4067/​s0718-​28132​01100​02000​05

13
Natural Hazards (2024) 120:2179–2231 2231

Wieczorek GF (1996) Landslide triggering mechanisms. In: Turner AK, Schuster RL (eds) Landslides:
investigation and Mitigation, Special Report 247. National Research Council, Transportation
Research Board, Washington, DC, pp 76–90
WP/WLI, (The International Geotechnical Societies’ UNESCO Working Party on World Landslide
Inventory) (1994) A suggested method for describing the causes of a landslide. Bull Int as Eng
Geol 50(50):71
Xie C, Ni P, Xu M, Mei G, Zhao Y (2020) Combined measure of geometry optimization and vegetation
for expansive soil slopes. Comput Geotech 123:103588. https://​doi.​org/​10.​1016/j.​compg​eo.​2020.​
103588
Zádorová T, Penížek V (2018) Formation, morphology and classification of colluvial soils: a review. Eur
J Soil Sci 69(4):577–591. https://​doi.​org/​10.​1111/​ejss.​12673
Zhang LL, Zhang J, Zhang LM, Tang WH (2011) Stability analysis of rainfallinduced slope failure: a
review. Proc Inst Civ Eng Geotech Eng 164(5):299–316. https://​doi.​org/​10.​1680/​geng.​2011.​164.5.​
299
Zhang L, Li J, Li X, Zhang J, Zhu H (2016) Rainfall-induced soil slope failure. CRC Press, Boca Raton.
https://​doi.​org/​10.​1201/​b20116
Zhang J, Zhu D, Zhang S (2020) Shallow slope stability evolution during rainwater infiltration consid-
ering soil cracking state. Comput Geotech 117:103285. https://​doi.​org/​10.​1016/j.​compg​eo.​2019.​
103285
Zhang JM, Luo Y, Zhou Z, Chong L, Victor C, Zhang YF (2021) Effects of preferential flow induced by
desiccation cracks on slope stability. Eng Geol. https://​doi.​org/​10.​1016/j.​enggeo.​2021.​106164
Ziegler M, Loew S, Bahat D (2014) Growth of exfoliation joints and near-surface stress orientations inferred
from fractographic markings observed in the upper Aar valley (Swiss Alps). Tectonophysics 626:1–20

Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and
institutional affiliations.

Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under
a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted
manuscript version of this article is solely governed by the terms of such publishing agreement and applicable
law.

Authors and Affiliations

M. P. Amarasinghe1,2 · S. A. S. Kulathilaka2 · D. J. Robert1 · A. Zhou1 ·

H. A. G. Jayathissa3

* D. J. Robert
dilan.robert@rmit.edu.au
M. P. Amarasinghe
s3930938@student.rmit.edu.au
S. A. S. Kulathilaka
sas@civil.mrt.ac.lk
A. Zhou
annan.zhou@rmit.edu.au
H. A. G. Jayathissa
jayathissa777@gmail.com
1
School of Engineering, Royal Melbourne Institute of Technology (RMIT) University, Melbourne,
VIC 3001, Australia
2
Department of Civil Engineering, University of Moratuwa, Katubedda, Sri Lanka
3
Landslide and Risk Management Division, National Building Research Organisation, Colombo,
Sri Lanka

13