Geohazards (floods and landslides) in the Ndop Plain, Cameroon Volcanic Line

Abstract

The Ndop Plain, located along the Cameroon Volcanic Line (CVL), is a volcano-tectonic plain, formed by a series of tectonic movements, volcanic eruptions and sedimentation phases. Floods (annually) and landslides (occasionally) occur with devastating environmental effects. However, this plain attracts a lot of inhabitants owing to its fertile alluvial soils. With demographic explosion in the plain, the inhabitants (143,000 people) tend to farm and inhabit new zones which are prone to these geohazards. In this paper, we use field observations, laboratory analyses, satellite imagery and complementary methods using appropriate software to establish hazard (flood and landslide) maps of the Ndop Plain. Natural factors as well as anthropogenic factors are considered.
The hazard maps revealed that 25% of the area is exposed to flood hazard (13% exposed to high flood hazard, 12% to moderate) and 5% of the area is exposed to landslide hazard (2% exposed to high landslide hazard, 3% to moderate). Some mitigation measures for floods (building of artificial levees, raising foundations of buildings and the meticulous regulation of the flood guards at Bamendjing Dam) and landslides (slope terracing, planting of trees, and building retaining walls) are proposed.

Full text

©2016 P. Wotchoko et al., published by De Gruyter Open.
This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivs 3.0 License.
Open Geosci. 2016; 8:429–449
Research Article Open Access
Pierre Wotchoko, Jacques-Marie Bardintze*, Zénon Itiga, David Guimolaire Nkouathio,
Christian Suh Guedjeo, Gerald Ngnoupeck, Armand Kagou Dongmo, and Pierre Wandji
Geohazards (floods and landslides) in the Ndop
Plain, Cameroon Volcanic Line
DOI 10.1515/geo-2016-0030
Received June 22, 2015; accepted January 15, 2016
Abstract: The Ndop Plain, located along theCameroon Vol-
canic Line (CVL), is a volcano-tectonic plain, formed by a
series of tectonic movements, volcanic eruptions and sed-
imentation phases. Floods (annually) and landslides (oc-
casionally) occur with devastating environmental eects.
However, this plain attracts a lot of inhabitants owing to
its fertile alluvial soils. With demographic explosion in the
plain, the inhabitants (143,000 people) tend to farm and
inhabit new zones which are prone to these geohazards.
In this paper, we use eld observations, laboratory analy-
ses, satellite imagery and complementary methods using
appropriate software to establish hazard (ood and land-
slide) maps of the Ndop Plain. Natural factors as well as
anthropogenic factors are considered.
The hazard maps revealed that 25% of the area is exposed
to ood hazard (13% exposed to high ood hazard, 12%
to moderate) and 5% of the area is exposed to landslide
hazard (2% exposed to high landslide hazard, 3% to mod-
erate). Some mitigation measures for oods (building of
articial levees, raising foundations of buildings and the
meticulous regulation of the ood guards at Bamendjing
Dam) and landslides (slope terracing, planting of trees,
and building retaining walls) are proposed.
Keywords: geohazard; ood; landslide; environmental im-
pact; mitigation; Ndop Plain; Cameroon
Pierre Wotchoko: Department of Geology, Higher Teacher’s Train-
ing College, University of Bamenda, P.O. Box 39, Bambili, Cameroon;
Email: pierrewotchoko@yahoo.fr
*Corresponding Author: Jacques-Marie Bardintze: Univ Paris-
Sud, Sciences de la Terre, Volcanologie, Planétologie, UMR CNRS
8148 GEOPS, bât. 504, Université Paris-Saclay, F-91405 Orsay,
France; Email: jacques-marie.bardintze@u-psud.fr; Tel. (33) 1 69 15
67 44
Zénon Itiga: Institute of Geological and Mining Research
(IGRM/ARGV) P.O. Box 4110, Yaounde, Cameroon; Email: zenonit-
iga@yahoo.fr
David Guimolaire Nkouathio: Department of Earth Sciences,
Faculty of Science, University of Dschang, P.O. Box 67, Dschang,
Cameroon; Email: nkouathio@yahoo.fr
1Introduction
Cameroon is exposed to geohazards (ood, landslide, vol-
canic eruption; [1]). The International Disaster Database
(www.emdat.be/ see “Database / Country Prole”) points
out four catastrophic recent (2007-2012) oods that af-
fected between 10 000 and more than 30 000 inhabitants.
The Ndop Plain (35 ×20 km) is situated at a mean alti-
tude of 1150 m along the Cameroon Volcanic Line (CVL). It
is surrounded by a series of escarpments which climb up
to 2151 m, corresponding to the peripheries of the Mounts
Oku and Bamenda, the Mbam and Nkogam Massifs (Fig. 1).
The CVL is interpreted as a mega shear zone that seems
to be structurally subdivided into an important network of
faults associated with a system of alternating horsts and
grabens [2–7]. The morphological formations of numerous
plains all along the CVL have been discussed by [8, 9] and
[10].
Floods are the major hazard in the Ndop Plain due
to its at nature between a combination of both high and
low rise morphologies. Notwithstanding, the Ndop Plain is
well inhabited due to the presence of fertile alluvial soils
resulting from erosion, transport and deposition of materi-
als from the surrounding hills by streams and rivers. More
generally, dierent geohazards have occurred in the West-
ern Cameroon Highlands (WCH), such as landslides, rock
falls and oods [11–16].
In this paper, we shall discuss the topographical and
morphological aspects of the Ndop Plain, using Digital El-
evation Model (DEM), laboratory analyses, eld observa-
Christian Suh Guedjeo: Department of Earth Sciences, Faculty of
Science, University of Dschang, P.O. Box 67, Dschang, Cameroon;
Email: guedjeochristian@yahoo.fr
Gerald Ngnoupeck: Department of Earth Sciences, Faculty of
Science, University of Dschang, P.O. Box 67, Dschang, Cameroon;
Email: ngnoupeckgerald@yahoo.com
Armand Kagou Dongmo: Department of Earth Sciences, Faculty
of Science, University of Dschang, P.O. Box 67, Dschang, Cameroon;
Email: kagoudongmo@yahoo.fr
Pierre Wandji: Laboratory of Geology, Higher Teacher’s Training
College, University of Yaounde I, P.O. Box 47, Yaounde, Cameroon

430 |P. Wotchoko et al.
Figure 1: A: Location of the CVL (Cameroon Volcanic Line in Africa), B: Location of the Ndop Plain in CVL, C: Relief map showing the geologi-
cal setting of the Ndop Plain (elevation in m), source (STRM data).
tions, satellite imagery and computer modeling. The geo-
hazards associated to this plain will also be assessed in
terms of typology, and mapped using the Geographic In-
formation System (GIS) approach (see [17]). The ndings in
this research work will assist to mitigate the eects of geo-
hazards in the Ndop Plain and other plains in Cameroon.
2Description of the study area
2.1 Geography
The Ndop Plain belongs to the Western Cameroon High-
lands (WCH). This sector of the CVL comprises Mounts
Bambouto, Bamenda and Oku. The Ndop Plain is partly
surrounded by Mount Bamenda to the W-SW and Mount
Oku to the N-NW (Fig. 1). The basement is made up of gran-
ites and gneisses.
The Ndop area is characterised by two major relief
features: a mountainous sector with steep hills and a at

Geohazards in the Ndop Plain, Cameroon Volcanic Line |431
plain (Fig. 1). This at relief opens up to Lake Bamend-
jing (a dam lake) in the south. The soils are lateritic, an-
dosol or alluvial types. These soils are hydromorphic ow-
ing to the presence of clays which tend to hold water espe-
cially during oods. The drainage pattern is of two types:
radial and dendritic (Fig. 2). The climate is subtropical:
Annual rainfall and annual average temperature stand at
1700 to 2000 mm and ≈21.3∘C respectively [18]. Upper
Noun Valley Development Authority (UNVDA) annual av-
erage rainfall data (2007-2012) is presented in Fig. 3 and
substantiates the high rainfall in July and August (up to
600 mm each month). The vegetation is of the Sudan Sa-
vannah type [19] and has been greatly modied by anthro-
pogenic activities such as bush re, intensive farming and
overgrazing.
2.2 Geological outlines, tectonic and
morphological evolution
The Ndop Plain was formed during a series of metamor-
phic, tectonic, volcanic, and sedimentation phases, which
succeeded one another [20]. Structural tectonic activity
started with plutonic and metamorphic phases linked
to the Pan African Orogeny during the Cambrian period
(540 Ma) when gneisses and granites were emplaced [21–
23]. The CVL, that was emplaced later, starting at 52 Ma,
consists of a series of volcanoes which are separated
by plains or low lying areas corresponding to collapsed
grabens such as: Tombel, Mbo, Noun, and Tikar [3, 5–10].
Ndop (this paper) is somewhat dierent as it is a low lying
plain.
2.2.1 Pan African plutonic and metamorphic phases
The Pan African Orogeny resulted from the collision of
the Congo Craton and the East-Saharan Metacraton dur-
ing the Neoproterozoic (Ediacarian, 635 to 540 Ma) with
its climax near 600 Ma. During this long period, gneisses
formed from the protolith and emplaced. During the Cam-
brian (540–485 Ma) post-collision phase, late granites em-
placed, associated with ignimbritic volcanic formations.
Then, an intense erosion phase built a peneplain, and this
nished during the Late Ordovician (480 Ma) (see [24] and
references therein).
2.2.2 Cenozoic magmatic phases
The Ndop Plain is surrounded by a series of volcanic
districts such as Mount Bamenda, Mount Oku, and the
Nkogam and Mbam Massifs. The formation of the Ndop
Plain can thus be correlated to the episodes of formation
of these mountains. This magmatic phase is related to the
formation of the Cameroon Volcanic Line which has been
active since 52 Ma until Present [25, 26].
The radiometric analyses of Mount Bamenda lavas
indicate two episodes of felsic volcanism during the
Oligocene and Miocene: a rst volcanic phase from 27.4 ±
0.5 Ma to 18.7 ±0.3 Ma and a second volcanic phase from
13.2 ±0.3 Ma to 12.74 ±0.25 Ma. The Mount Bamenda felsic
volcanism is the oldest of the whole WCH [27]. A mac vol-
canic phase emplaced from the Lower Miocene (17.6 Ma) to
0 Ma and shows that mac volcanism has existed over a
long period of time and is partly coeval to the felsic vol-
canism (between 17.6 and 12.7 Ma) [28, 29].
During the Quaternary up to the Present, weathering
and erosion have been intensive, resulting in the forma-
tion of Quaternary alluvial deposits on basement rocks.
As these materials were transported by streams and rivers,
they were deposited into the plain, forming alluvial de-
posits. These alluvia are rich in the soil nutrients essential
for plant growth. This is the reason why the plain is very
fertile, evidenced by the presence of the Upper Noun Val-
ley Development Agency (UNVDA) rice plantation.
3Cartography
3.1 Methods
Our cartography work involved the use of appropriate soft-
ware such as Surfer 9, MapInfo 8.5, ArcGIS 10.1, Global
Mapper 13 and 3DEM, to realize the various maps required
for this study. The base maps used during this phase were
topographic maps. The procedure employed was as fol-
lows:
Topographic maps (1:50,000) of Nkambe 1b and Foum-
ban 3d from the National Institute of Cartography (NIC)
Cameroon were georeferenced, and vectors such as con-
tour lines, rivers, localities and roads were digitized.
The GeoTi DEM of the Ndop was downloaded from
NASA’s Shuttle Radar Topography Mission (SRTM V2),
with a resolution of 90 m and introduced and imported
into the ArcGIS software.

432 |P. Wotchoko et al.
Figure 2: Detailed hydrographic map of the Ndop Plain area.

Geohazards in the Ndop Plain, Cameroon Volcanic Line |433
Figure 3: Annual average rainfall and temperature data (2007-2012)
(UNVDA, Upper Noun Valley Development Authority).
Figure 4: Panoramic view of the Ndop Plain (a) very steep slopes
surrounding the flat plain, (b) houses along steep slopes.
Surface parameters such as contours, reliefs, slope
dippings, and slope orientations were realised using the
3D spatial analyst tool in the ArcGIS 10.1 software.
With the Global Mapper software, the DEM of the Ndop
Plain was rst introduced. The x, y and z data were ex-
ported to the Surfer software for further modelling, using
the elevation grid and surfer grid in the American Stan-
dard Code for Information Interchange (ASCII) format.
The landslide and ood hazard maps for the Ndop
Plain were realised with combining or weighting the var-
ious parameters in the ArcGIS software. The model em-
ployed in mapping the hazards was a data-driven bivariate
hazard mapping model (see [16, 30]). The various param-
eters (predisposition factors) involved in the realization of
the hazard map are rock type, soil type, land cover, slope
dipping, slope orientation, nature (size) of river and prox-
imity to river (see [31–34], and references therein).
3.2 Characterization of the hazard
parameters
The parameters selected for this study were based on eld
data and on site analysis (Table 1). The choice of variables
that aect landslides is an important step in susceptibility
assessment and the prediction of new events [35, 36]. Geo-
hazards are complex natural processes which are dicult
to model with a few parameters due to variations in insta-
bility over space and time and are conditioned by several
factors [35, 37–40].
The factors chosen here were operational, non-
uniform, non-redundant, measurable, and represented
over the entire area [41]. They include: rock type, soil type,
land cover, slope dipping, slope orientation, distance from
river channel and nature of rivers. Rainfall which is an im-
portant parameter related to the occurrence of landslides
and oods is not used here because it is uniformly dis-
tributed in the Ndop Plain. Thematic maps were prepared
for each of these factors following the methods here de-
scribed.
3.2.1 Slope orientation and slope dipping
Very steep slopes surround the at Ndop Plain (Fig. 4).
As pointed out in many regions worldwide, landslides are
linked to slopes [16, 36, 42–47]. Two characteristics are
classically considered: slope orientation and slope dip-
ping.
These slope parameters were realised with the ArcGIS
10.1 software using the DEM of the area in the 3D spatial
analyst tool.
An aspect map (slope orientation) is essential in slope
failure analysis because of the varying exposure to sun-
light and rainfall. Here, the slope orientations range from
0 to 360∘, and are grouped into 10 classes: at (-1), N (0-
22.5), NE (22.5-67.5), E (67.5-112.5), SE (112.5-157.5), S (157.5-
202.5), SW (202.5-247.5), W (247.5-292.5), NW (292.5-337.5)
and N once more (337.5-360) (Fig. 5a). NW facing slopes re-
ceive higher precipitation more frequently than SW facing
slopes. This is because rainfall is inuenced by the eects

434 |P. Wotchoko et al.
Table 1: Parameters used in flood and landslide mapping in the Ndop Plain.
Data Data source Data type Derived map Parameter class
Material / or rock Field survey Polygon Geologic map –Basalt
–Trachyte
–Rhyolite
–Ignimbrite
–Granite
Soil type Field survey Polygon Soil map –Andosols
–Laterites
–Colluvium
–Alluvium
DEM 1/50,000 topographic map Line vector Slope dipping
(in degree)
–0–5
–5–10
–10–20
–20–35
–>35
Slope orientation
(in degree)
North (0–22.5)
Northeast (22.5–67.5)
East (67.5–125)
Southeast (112.5–157.5)
South (157.5–202.5)
Southwest (202.5–247.5)
West (247.5–292.5)
Northwest (292.5–337.5)
North (337.5–360)
Rivers 1/50,000 hydrographic map
Field survey
Multiple ring
buer
Distance from river
channel (m)
–100 m
–200 m
–400 m
–600 m
Land cover Google Earth 2013 Polygon Land use map –Built up area
–Subsistence farming area
–Plantation area
–Unused land
–Forest area

Geohazards in the Ndop Plain, Cameroon Volcanic Line |435
of the moist southwest monsoon winds originating from
the Atlantic Ocean, and the Harmattan trade winds origi-
nating from the Sahara Desert in the North [48].
The slope dipping was evaluated in degree and
grouped into six classes: 0–5∘, 5–10∘, 10–15∘, 15–20∘, 20–
35∘, and >35∘(up to almost vertical) (Fig. 5b). Generally
the steeper a slope the higher the hazard of a landslide;
however when it becomes too steep, this hazard drops
since soil cannot accumulate on very steep slopes [30])
Two zones of the studied area (one in western part and one
in north-eastern part) are exposed (Fig. 5b).
3.2.2 3-D representation
Precise knowledge of the relief of an area is vital when car-
rying out landslide studies.
The role of relief in slope instability has been disputed
with some authors [49–52] arguing that altitude is a good
indicator conditioning slope movements, while others [53]
do not see any changes in slope movements between low
altitudes and the high altitudes. The three dimensional
representation of the relief of the area was done using the
DEM, from which vertices of points with spatial references
were generated with the Global Mapper 13 software. These
vertices were then exported to the Surfer 9 software to gen-
erate a 3D representation of the area (Fig. 5c). It is con-
rmed that the northern part of the studied area is very un-
even and hilly compared to the central and southern parts
of the area, corresponding to the plain.
3.2.3 Land cover
Land use practice may considerably aect the occurrence
of landslides in an area [38, 54]. The land cover map was
obtained from a SPOT image, map data ©2013, extracted
from Google Earth. This image was imported into ArcGis
10.1; the dierent land uses were manually digitized based
on the textures of objects and then calibrated. This was
later converted to a raster map and a hazard index was at-
tributed to each class. Five main land use patterns were
considered, namely: built up area (25%) close to the town
of Ndop (30 000 inhabitants) and ve villages, forest (10%)
scattered in the plain, subsistence farm area (25%), plan-
tation area (20%) and unused land (20%) corresponding
more or less to the principal slopes (Fig. 5d).
3.2.4 Soil type
The occurrence of landslides within a particular area de-
pends noticeably on the soil type [55]. The soil type de-
pends on the rock type and its morphology, but dierent
soil types may result from the same parent rock, follow-
ing dierential weathering and drainage. Soils were not
mapped in this research; the soil map produced by ISRIC
(International Soil Reference and Information Centre) Li-
brary (65.0), PO Box 353 6700 A.J. Wageningen, The Nether-
lands, was digitized and used for our study.
Five soil type patterns were adopted which are: soils
formed from recent lava ows (12%), andosol (8%), lat-
erite (5%) close to Babungo, colluvium (35%) and allu-
vium (40%) (Fig. 5e). Andosols are located on steep slopes
and are less stable than colluvium and alluvium which
are formed on gentle to at slopes. Hence andosols have
a higher hazard index [20]. These dierent soil types were
manually digitized in the ArcGis 10.1 software and con-
verted into a raster map and index values attributed to it.
3.2.5 Rock type (material)
The studied area is covered by the following rock types:
plutonic rock (20% granite and gneiss), volcanic rock
(15% basalt, 4% trachyte, 5% rhyolite and 5% ignimbrite)
(Fig. 5f). The rest of the area is covered by alluvial mate-
rials. Two representative samples of each rock were col-
lected in the study area for the preparation of thin sections.
Granites show a porphyritic granular texture with an-
gular phenocrysts interlocked. It is made up of quartz,
orthoclase, biotite, microcline, and opaque minerals
(Fig. 6a). Basalts present a microlitic texture; they con-
tain minerals such as olivine, pyroxene, plagioclase and
opaque Fe-Ti oxides which occur as phenocrysts. These
mineral phases also constitute the groundmass (Fig. 6b).
Trachytes present a microlitic porpyhritic texture with
phenocrysts of sanidine and oxides which are automor-
phic and well developed; chlorite is also present in these
rocks (Fig. 6c). The crystalline phases are embedded in a
groundmass made up of microlites of sanidine, biotite and
oxides. The groundmass displays a preferred orientation
of alkali feldspars. This rock evidences re-crystallization
of calcite indicating that weathering is taking place. Rhy-
olites show a microlitic porpyhritic texture, with min-
erals such as alkali feldspar, pyroxene, quartz, and ox-
ides in a glassy groundmass (Fig. 6d). The groundmass
shows a uidal structure with preferred orientation of
the feldspars and devitrication of the quartz. Ignimbrites
have a vitroclastic texture made up of rock fragments, bro-

436 |P. Wotchoko et al.
Figure 5a: Raster maps of parameters used in hazards mapping in the Ndop Plain. (a) slope orientation (aspect).

Geohazards in the Ndop Plain, Cameroon Volcanic Line |437
Figure 5b: Raster maps of parameters used in hazards mapping in the Ndop Plain. (b) slope dipping.

438 |P. Wotchoko et al.
Figure 5c: Raster maps of parameters used in hazards mapping in the Ndop Plain. (c) 3D representation map, elevation in m.

Geohazards in the Ndop Plain, Cameroon Volcanic Line |439
Figure 5d: Raster maps of parameters used in hazards mapping in the Ndop Plain. (d) land cover map.

440 |P. Wotchoko et al.
Figure 5e: Raster maps of parameters used in hazards mapping in the Ndop Plain. (e) soil map.

Geohazards in the Ndop Plain, Cameroon Volcanic Line |441
Figure 5f: Raster maps of parameters used in hazards mapping in the Ndop Plain. (f) geologic map of Ndop [16].

442 |P. Wotchoko et al.
Figure 5g: Raster maps of parameters used in hazards mapping in the Ndop Plain. (g) proximity to river map.

Geohazards in the Ndop Plain, Cameroon Volcanic Line |443
Figure 6: Photomicrography (under crossed-nicols) of rocks from Ndop area; (a) Granite (b) Basalt (c) Trachyte (d) Rhyolite and (e) Ign-
imbrite. Amp = amphibole, Bt = biotite, Cl = chlorite, Mi = microcline, Ol = olivine, Or = orthoclase, Ox = Fe-Ti oxide, Px = pyroxene, Qtz
= quartz, Sa = sanidine. Glass and spherulite are shown in (d) and amme structures are shown in (e) where indicated.

444 |P. Wotchoko et al.
Table 2: Percentage influence on floods of each of the input raster
parameters.
Raster parameter Percentage
influence or hazard
index (%)
Input
eld
1 Land cover 15 Land
cover
2 Reclassed soil type 15 Value
3 Reclassed slope 50 Value
4 Reclassed proximity
to rivers
20 Value
ken pieces of feldspars (sanidine and plagioclase), quartz
and ammes embedded in a glassy groundmass (Fig. 6e).
Locally, groundmass is devitried into small quartz and
feldspar crystals.
The rock type distribution in an area may aect land-
slides at dierent scales. A lithological sketch map wasre-
alized from eld observation data, thin section analysis
(Fig. 6), reading and interpretation of satellite images (i.e.
vegetation cover helped in characterizing unexposed out-
crops as vegetation cover is scarce when growing on thin
layers of soil lying on rock) and DEM (i.e. 3D view helped
to locate domes in the area, regardless if they were covered
by vegetation or not). The diculty to observe the contact
and extension of dierent rock units reduced the accuracy
of this map.
3.2.6 Nature and proximity to river
The nature of the rivers was determined from the dimen-
sion of the river channels. This was obtained by digi-
tizing the 1:50,000 topographic maps of Nkambe 1b and
Foumban 3d using the ArcGIS 10.1 software and also from
eld work. Rivers were grouped into ve classes, from
the largest to the smallest: major river, main river, river,
stream and temporal stream. The dimensions of the rivers
increase as the rivers ow into lowlands and coalesce to-
gether. Generally, the wider and shallower a river channel,
the higher the degree of ood hazard is. Moreover, some
streams may cause severe oods when they receive an ab-
normal inux of water. In case of a ood, the areas close to
a river channel are highly aected compared to distant ar-
eas. A ring buer was realised at a distance of 600 m from
the river at intervals of 100 m, 200 m, 400 m and 600 m and
grouped into four classes (Fig. 5g). Proximity to rivers was
implemented by applying the Euclidean distance function
in ArcGIS using the multiple buer tool.
3.2.7 Combining hazard parameters
Many dierent types of landslide hazard zonation tech-
niques have been developed over the last decades, and the
diculty lies in the weighting the factors [32, 56–60]. In
this paper, the parameters were weighted as follows: the
model builder was used in the ArcGIS 10.1 software and all
the environmental settings were checked such as process-
ing extent, raster analysis and cell size. The parameters
were then introduced into the model builder and reclas-
sied to realize oating points, continuous datasets, cate-
gorize datasets into ranges, and assign each range of val-
ues a discrete integer value. Each parameter was classied
based on its inuence on landslides and oods. Dipping
slopes, for example, were reclassied by assigning new
input eld values to them. Consequently, steeper slopes
were assigned higher values and less steep slopes lower
values. This was done for the other parameters used in the
model. Using the connection tools, the parameters were
connected with their input data, their corresponding tools
and their resultant outputs raster. The model was then run
to ensure functionality.
Using the weighted overlay tool, the values of each
dataset were then weighted [44, 60], and all input pa-
rameters were assigned each a percentage of inuence or
hazard index. The higher the percentage of inuence, the
greater impact a particular input parameter will have on
landslides or oods. The percentage of inuence for both
landslides and oods are presented in Tables 2 and 3. Ac-
cordingly, the weighted overlay operation was done as fol-
lows; 1, 10, and 1 were typed in the From, To, and By elds
in the weighted overlay tool box to avoid having to up-
date the scale values after adding the input datasets. At
this level, some eld values were restricted to give them
a minimum value in the evaluation process, as, for exam-
ple, steep dipping slopes >35∘cannot be exposed to ood
hazards while at areas cannot be exposed to landslides.
4Discussions and conclusions
4.1 Causes of the geohazards and mapping
The causes of oods in the area include natural causes
(siltation, peculiar geomorphology, the nature of soils,
and rainfall) as well as anthropogenic causes (subsis-
tence farming, plantation agriculture, the dam-backing
eect of the Bamendjing Dam, and dumping of refuse
into rivers). For landslides, natural causes include rain-
fall, slope steepness, groundwater, gravity, and the ero-

Geohazards in the Ndop Plain, Cameroon Volcanic Line |445
Table 3: Percentage influence on landslides of each of the input
raster parameters.
Raster parameter Percentage
influence or hazard
index (%)
Input
eld
1 Reclassed slope dip-
ping
50 Value
2 Reclassed soil type 20 Value
3 Reclassed proximity
to rivers
10 Value
4 Reclassed slope ori-
entation
5 Value
5 Land cover 15 Land
cover
Figure 7: Hazard map of the Ndop Plain. Percentages of the surfaces
of exposed areas are listed.
sion of the toe of slopes by rivers, while anthropogenic
causes consist of deforestation, excavation, and anarchi-
cal construction as pointed out by Che et al. (2010). Among
all the mentioned causes, slope steepness is the most im-
portant in this area. Note that anthropogenic factors have
a minor eect compared to natural factors.
In hazard mapping, the main problem resides in the
combination of factors. These factors are not standardized
nor regulated by any international norm. In this study, the
selected factors are based on physical data obtained from
the eld. They do not aect ood and landslide hazards
to the same degree. It is a combination of these parame-
ters acting together which cause the hazard. The realiza-
tion of the hazard map (Fig. 7) consists of computational
weighting all these parameters. The percentage of inu-
ence (hazard index) for each parameter (listed in Table 1)
on the ood and landslide event has been recapitulated in
Tables 2 and 3. It should be noted that hazard indices vary
from one area to another.
From the results obtained it is observed that ood haz-
ard may aect about 25% of the studied area, along a
north-south stripe: 13% are exposed to high ood hazard
and 12% to moderate (Fig. 7). Landside hazard impacts
about 5% of the area, along the borders: 2% are exposed
to high landslide hazard and 3% to moderate. None of the
hazard areas overlap. Thus, about 30% of the area is ex-
posed to a natural hazard (Fig. 7).Although oods are de-
structive, they are also benecial in the agricultural plain:
ooded areas which are swampy are crucial for the growth
of Ndop rice.
More generally, geohazards are common in Cameroon
especially along the CVL [11, 12, 14], with landslides [13, 30,
61] and oods [16] having devastating eects on man and
the environment. Landslides impact areas with slopes of
more than 35∘(see Fig. 5b and 7, 35–80∘according to [62])
while oods are the most widespread hazard in the plain
and aects all localities when it does occur [65].
This is true in all rainy tropical volcanic (active or ex-
tinct) regions with contrasting relief, i.e. on oceanic is-
lands [i.e. Tahiti 31]; [Cape Verde 63] as well as on conti-
nents (Uganda, [64]), with various magnitudes.
Ref. [30] point out the increase of human risks in CVL
as evidenced by the loss of about 30 lives within the last 20
years because of numerous landslides in the Limbe area
on the foot slope of Mount Cameroon. Twenty-four people
died during the 2001 Limbe landslide (2800 people home-
less) and ve others during the 2003 Bambouto-Magha
landslide [13, 62].
Ref. [16] used similar parameters, such as slope, rock
type and soil type, to map the landslide, rock fall and ood
hazards in the environs of Bamenda that have severe en-
vironmental and socioeconomic impacts on the popula-
tion. In the same way, [14] described natural hazards in
the Mount Bambouto caldera, where landslides are most
frequent.
A retrospective analysis of data from the last three
decades clearly indicates an upward trend in the number
of landslides in Cameroon [13]. A proper hazard monitor-
ing and assessment committee needs to set up to man-

446 |P. Wotchoko et al.
Table 4: Some suggestions to mitigate geohazards on the Ndop Plain.
Floods Landslides
–Deepening of the river bed to increase the capacity of
the river;
–Cleaning of gutters especially where they meet rivers
to ease the flow of water;
–Straightening the river channels; this avoids exces-
sive sedimentation of the river bed;
–Building articial levees; this prevents water from
overflowing its banks;
–Raise foundations of buildings: if foundations are
high, water will nd it dicult to enter houses as is
the case in some minor floods;
–Constructing larger and higher bridges: refuse has
the tendency of blocking small and low bridges forc-
ing water to flow out of its normal channel;
–Avoid farming around river channels since this activ-
ity disrupts the soil structure and ne particles are
easily carried into the river bed;
–Planting new trees and preventing the felling of trees;
–Construction in flat areas should be avoided since
these areas are susceptible to floods;
–Meticulous regulation of the flood guards at the Ba-
mendjing Dam so as to avoid water from flooding new
areas.
–The flanks of hills should not be excavated without
any geotechnical measure emplaced to stabilize the
slope;
–Stabilizing structures should be constructed such as
embankments and other retaining walls;
–Very steep slopes should be terraced when a road is
excavated;
–Grass or other plants should be planted along steep
slopes;
–Groundwater should be drained in sensitive areas;
–Very steep slopes can be graded to reduce their gra-
dient.
Table 5: Some suggestions to manage geohazards on the Ndop Plain.
Floods Landslides
–Go to high areas that the rise of water cannot attain;
–Listen for any unusual sounds that might indicate
moving debris, such as trees cracking or boulders
knocking together;
–If trapped in a building climb to the roof of the build-
ing;
–Put on life jackets when available before walking in
flooded areas;
–Progressively remove water as it floods a building ei-
ther by pumping or by using a bucket.
–Immediately evacuate injured or disabled people
from the slide area;
–Do not loiter around a landslide for there may be a
post landslide event;
–Stay alert and awake; many debris-flow fatalities oc-
cur when people are sleeping;
–Stay out of the path of a landslide;
–When driving on a highway be alert of landslide de-
bris blocking the road.
age these hazards better as attention is only paid in cases
where there are casualties or severe destruction. A proper
understanding of geohazards is vital for the management
and understanding of landscape evolution for sustainable
development and a better arrangement of the national ter-
ritory.
4.2 Suggestions for the mitigation and
management of geohazards
To render these hazards less severe or less devastating,
some mitigation and management measures are proposed
to be implemented in the Ndop Plain. Mitigation involves
emplacing measures in the geological context, while man-
agement involves measures in the way that people can
manage themselves and respond to geohazards success-

Geohazards in the Ndop Plain, Cameroon Volcanic Line |447
fully to ultimately survive. These suggestions are pre-
sented in Tables 4 and 5.
Acknowledgement: B. Bonin is thanked for useful re-
marks. Careful reviews by C. Principe and an anonymous
reviewer greatly helped to improve the manuscript.
References
[1] Fogwe Zephania, N., 2010. Mitigating and managing regional
geo-environmental hazards within a decentralisation transition
in Cameroon. J. Hum. Ecol., 30, 3, 187-195.
[2] Gèze, B., 1943. Géographie physique et géologie du Cameroun
occidental. Mém. Muséum Nation. Hist. Natur, Paris, Nlle série,
17, 271 p.
[3] Fitton, J.G., 1987. The Cameroon line, West Africa: A compari-
son between oceanic and continental alkaline volcanism. In: Fit-
ton, J.G. & Upton, B.G.J. (Eds.) Alkaline igneous rocks. Geol. Soc.
London Spec. Publ., 30, 273–291.
[4] Moreau, C., Regnoult, J.M., Déruelle, B. & Robineau, B., 1987.
A new tectonic model for the Cameroon line, Central Africa.
Tectonophysics, 139, 317–334.
[5] Déruelle, B., Moreau, C., Nkoumbou, C., Kambou, R., Lissom, J.,
Njonfang, E., Ghogomu, R.T. & Nono, A., 1991. The Cameroon
Line: A review. In: Kampunzu, A.B., Lubala, R.T. (Eds.) Mag-
matism in Extensional Structural Settings. Springer, Berlin, pp
274–327.
[6] Déruelle, B., Ngounouno,I. & Demaie, D., 2007. The ‘Cameroon
Hot Line’ (CHL): A unique example of active alkaline intraplate
structure in both oceanic and continental lithospheres. C. R.
Geoscience, 339, 9, 589–600.
[7] Fosso, J., Ménard, J.J., Bardintze, J.M., Wandji, P., Tchoua, F.M.
& Bellon, H., 2005. Les laves du mont Bangou: Une première
manifestation volcanique éocène, à anité transitionnelle, de
la Ligne du Cameroun. C. R. Geoscience, 337, 315–325.
[8] Nkouathio, D.G., Ménard, J.J., Wandji, P. & Bardintze, J.M.,
2002. The Tombel graben (West Cameroon): A recent mono-
genetic volcanic eld of the Cameroon Line. J. Afr. Earth Sci., 35,
2, 285–300.
[9] Nkouathio, D.G., Kagou Dongmo, A., Bardintze, J.M., Wandji,
P., Bellon, H. & Pouclet, A., 2008. Evolution of volcanism in
graben and horst structures along the Cenozoic Cameroon Line
(Africa): Implications for tectonic evolution and mantle source
composition. Mineral. Petrol., 94, 3–4, 287–303.
[10] Itiga, Z., Chakam Tagheu, P.J., Wotchoko, P., Wandji, P.,
Bardintze, J.M. & Bellon, H., 2004. La Ligne du Cameroun:
Volcanologie et géochronologie de trois régions (mont Manen-
gouba, plaine du Noun et Tchabal Gangdaba). Géochronique,
91, 13–16.
[11] Tchoua, F.M., 1984. Les coulées boueuses de Dschang (Août
1978). Rev. Géogr. Cameroun 2:25–33.
[12] Tchoua, F.M., 1989. Les matériaux des coulées boueuses de
Dschang (Août 1978). Rev. Géogr. Cameroun, 8, 1, 58–64.
[13] Zogning, A., Ngouanet, C. & Tiafack, O., 2007. The catastrophic
geomorphological processes in humid tropical Africa: A case
study of the recent landslide disasters in Cameroon. Sedimen-
tary Geology, 199, 1–2, 13–27.
[14] Zangmo, T.G., Kagou Dongmo, A., Nkouathio, D.G. & Wandji, P.,
2009. Typology of natural hazards and assessment of associ-
ated risks in the Mount Bambouto caldera (CameroonLine, West
Cameroon). Acta Geologica Sinica, 83, 5, 1008–1016.
[15] Aboubakar, B., Kagou Dongmo, A., Nkouathio, D.G. & Ngapgue,
F., 2013. Instabilités de terrain dans les hautes terres de
l’Ouest Cameroun: Caractérisation géologique et géotechnique
du glissement de terrain de Kekem. Bull. Institut Scientique,
Rabat, Maroc, section Sci. de la Terre, 35, 39–51.
[16] Guedjeo, C.S., KagouDongmo, A., Ngapgue, F., Nkouathio, D.G.,
Zangmo Tefogoum, G., Gountié Dedzo, M. & Nono, A., 2013. Nat-
ural hazards along the Bamenda escarpment and its environs:
The case of landslide, rock fall and flood risks (Cameroon vol-
canic line, North-West Region). Global Adv. Res. J. Geol. Min.
Res., 2, 1, 15–26.
[17] Hamadouche, M.A., Mederbal, K., Kouri, L., Regagba, Z., Fekir,
Y. & Anteur, D., 2014. GIS-based multicriteria analysis: An ap-
proach to select priority areas for preservation in the Ahaggar
National Park, Algeria. Arabian J. Geosci., 7, 2, 429–434.
[18] Ndenecho, E.N., 2009. Cropping systems and post cultivation
succession, agro-ecosystem in Ndop Cameroon. J. Human Ecol-
ogy, 27, 1, 26–34.
[19] Neba, A., 1999.Modern Geography of the Republic of Cameroon.
3rd Edition, Neba Publishers- Bamenda, Cameroon.
[20] Guedjeo, C.S., 2013. Genesis of the Ndop Plain, associated geo-
hazards and environmental impact (North-West Region). HTTC
thesis, University of Bamenda, Cameroon, 70 p.
[21] Tagne-Kamga, G., Mercier, E., Rossy, M. & Nsifa, N.E., 1999.
Synkinematic emplacement of the Pan-African Ngondo igneous
complex (West Cameroon, Central Africa). J. Afr. Earth Sci., 28,
675–691.
[22] Abdelsalam, M.G., Liégeois, J.P. & Stern, R.J., 2002. The Saharan
metacraton. J. Afr. Earth Sci., 34, 119–136.
[23] Njanko, T., Nédélec, A. & Aaton, P., 2006. Synkinematic high-
K calc-alkaline plutons associated with the Pan-African Central
Cameroon Shear Zone (W Tibati area): Petrology and geody-
namic signicance. J. Afr. Earth Sci., 44, 494–510.
[24] Toteu, S.F., Van Schmus, W.R., Penaye, J. & Michard, A., 2001.
New U-Pb and Sm-Nd data from north-central Cameroon and its
bearing on the pre-Pan African history of central Africa. Precam-
brian Res., 108, 1–2, 45–73.
[25] Moundi, A., Wandji, P., Bardintze, J.M., Ménard, J.J., Okomo
Atouba, L.C., Mouncherou, O.F., Reusser, E., Bellon, H. &
Tchoua, F.M., 2007. Les basaltes éocènes à anité transition-
nelle du plateau Bamoun, témoins d’un réservoir mantellique
enrichi sous la ligne volcanique du Cameroun. C. R. Geoscience,
339, 396–406.
[26] Yokoyama, T., Aka, F.T., Kusakabe, M. & Nakamura, E., 2007.
Plume-lithosphere interaction beneath Mt. Cameroon volcano,
West Africa: Constraints from 238U–230Th–226 Ra and Sr–Nd–
Pb isotopes systematics. Geochim. Cosmochim. Acta, 71, 1835–
1854.
[27] Kamgang, P., Njongang, E., Nono, A., Gountie Dedzo, M.
& Tchoua, F., 2010. Petrogenesis of a silicic magma sys-
tem; Geochemical evidence from the Bamenda mountains, NW
Cameroon, Cameroon Volcanic Line. J. Afr. Earth Sci., 58, 285–
304.
[28] Kamgang, P., Njonfang, E., Chazot, G. & Tchoua, F., 2008. Geo-
chemistry and geochronology of mac rocks from the Bamenda

448 |P. Wotchoko et al.
mountains (Cameroon): Source composition and crustal con-
tamination along the Cameroon Volcanic Line. C. R. Geoscience,
340, 850-857.
[29] Kagou Dongmo A., Nkouathio D.G., Pouclet A., Bardintze J.M.,
Wandji P., Nono A. & Guillou H., 2010. The discovery of late
Quaternary basalt on Mount Bambouto: Implications for recent
widespread volcanic activity in the southern Cameroon Line. J.
Afr. Earth Sci., 57, 1–2, 96–108.
[30] Che, V.B., Kervyn, M., Ernst, G.G.J., Trefois, P., Ayonghe, S., Ja-
cobs, P., Van Ranst, E. & Suh, C.E., 2011. Systematic documen-
tation of landslide events in Limbe area (Mt Cameroon Volcano,
SW Cameroon): geometry, controlling, and triggering factors.
Natural Hazards, 59, 47–74.
[31] Guillande, R., Gelugne, P., Bardintze, J.M., Brousse, R.,
Chorowicz, J., Deontaines, B. & Parrot, J.F., 1993. Cartographie
automatique de zones à aléas de mouvements de terrain sur
l’île de Tahiti à partir de données digitales. Bull. Soc. Géol. Fr.,
164, 4, 577–583.
[32] van Westen, C.J., Rengers, N. & Soeters, R., 2003. Use of ge-
omorphological information in indirect landslide susceptibility
assessment. Nat. Hazards, 30, 3, 399–419.
[33] van Westen, C.J., van Asch, T.W.J. & Soeters, R., 2006. Landslide
hazards and risk zonation-Why is it still so dicult? Bull. Engi-
neering Geol. Environ., 65, 2, 167–184.
[34] Guzzetti, F., Ardizzone, F., Cardinali, M., Rossi, M. &Valigi, D.,
2009. Landslide volumes and landslide mobilization rates in
Umbria, central Italy. Earth Planet. Sci. Lett., 279, 3–4, 222–229.
[35] Bateira, C. & Soares, L., 1997. Movimentos em massa no norte
de Portugal. Factores da sua ocorrência. Territorium, 4, 63–77.
[36] Papathanassiou, G., Valkaniotis, S., Ganas, A. & Pavlides, S.,
2013. GIS-based statistical analysis of the spatialdistribution of
earthquake-induced landslides in the island of Lefkada, Ionian
Islands, Greece. Landslides, 10, 771–783.
[37] Irigaray, C., Fernandez, T. & Chacón, J., 1996. Inventory and
analysis of determining factors by a GIS in the northern edge of
the Granada Basin. In: Senneset, K. (Ed.), Landslides – Glisse-
ments de Terrain. Balkema, Rotterdam, vol. 3, 1915–1921.
[38] Zêzere, J.L., Ferreira, A.B. & Rodrigues, M.L., 1999. Landslides
in the north of Lisbon region (Portugal): Conditioning and trig-
gering factors. Physics Chem. Earth, Part A, 24, 10, 925–934.
[39] Nagarajan, R., Roy, A., Vinod Kumar, R., Mukherjee, A. & Khire,
M.V., 2000. Landslide susceptibility mapping based on terrain
and climatic factors for tropical monsoon regions. Bull. Eng.
Geol. and Env., 58, 275–287.
[40] Eze, B. & Ndenecho, E., 2004. Geomorphic and anthropogenic
factors influencing landslides in the Bamenda Highlands, North
West Province, Cameroon. J. Applied Social Sci., 4, 1, 15–26.
[41] Ayalew, L., Yamagishi, H., Marui, H. & Kanno, T., 2005. Land-
slide in Sado Island Japan: GIS-based susceptibility mapping
with comparison of results from two methods and verications.
Engineering Geol., 81, 432–445.
[42] Ayonghe, S.N., Mafany, G.T., Ntasin, E. & Samalang, P., 1999.
Seismically activated swarm of landslides, tension cracks, and
a rockfall after heavy rainfall in Bafaka, Cameroon. Natural Haz-
ards, 19, 13–27.
[43] Guzzetti, F., 2000. Landslide fatalities and the evaluation of
landslide risk in Italy. Engineering Geol., 58, 2, 89–107.
[44] Süzen, M.L. & Doyuran, V., 2004. A comparison of the GIS based
landslide susceptibility assessment methods: Multivariate ver-
sus bivariate. Environmental Geol., 45, 665–679.
[45] Carrara, A., Crosta, G. & Frattini, P., 2008. Comparing models of
debris-flow susceptibility in the alpine environment. Geomor-
phology, 94, 3, 353–378.
[46] Galli, M., Ardizzone, F., Cardinali, M., Guzzetti, F. & Reichen-
bach, P., 2008. Comparing landslide inventory maps. Geomor-
phology, 94, 268-289.
[47] Zêzere, J.L., Henriques, C.S., Garcia, R.A.C., Oliveira, S.C.,
Piedade, A. & Neves, M., 2009., Eects of landslide invento-
ries uncertainty on landslide susceptibility modelling. In: Mal-
let, J.-P., Remaitre, A. & Boggard, T. (Eds.), Landslide Processes:
From Geomorphologic Mapping to Dynamic Modelling. CERG
Editions, Strasbourg, p.81–86.
[48] Afungang, R.N., 2010. Erosion, mass movement and landscape
dynamics: A case study of the Mezam highlands. Master’s the-
sis in landscape dynamics and risk. Department of Geography,
University of Yaounde1-Cameroon, 181 p.
[49] Pachauri, A.K. & Pant, M., 1992. Landslide Hazard Mapping
Based on Geological Attributes. Engineering Geology, 32, 81–
100.
[50] Gokceoglu, C. & Aksoy, H., 1996. Landslide susceptibility map-
ping of the slopes in the residual soils of the Mengen region
(Turkey) by deterministic stability analyses and image process-
ing techniques. Engineering Geol., 44, 1-4, 147-161.
[51] Dai, F.C. & Lee, C.F., 2003. A spatiotemporal probabilistic mod-
elling of storm-induced shallow landsliding using aerial pho-
tographs and logistic regression. Earth Surf. Proc. Landforms,
28, 5, 527–545.
[52] Duncan, J.M. & Wright, S.G., 2005. Soil Strength and Slope Sta-
bility. John Wiley & Sons, New York, 312 p.
[53] Bateira, C., 2010. A avaliação da susceptibilidade natural na
região Norte de Portugal: Análise prospectiva e ordenamento
do território. Prospectiva e Planeamento, 17, 15–32.
[54] Zêzere, J.L., 1997. Movimentos de vertente e perigosidade geo-
morfológica na Região a Norte de Lisboa, Ph.D. Thesis, Univer-
sity of Lisbon, Portugal.
[55] Che, V.B., Kervyn, M., Suh, C.E., Fontijn, K., Ernst G.G.J., Del Mar-
mol M.A., Trefois P., & Jacobs, P., 2012. Landslide susceptibil-
ity assessment in Limbe (SW Cameroon): A eld calibrated seed
cell and information value method. Catena 92, 83–98.
[56] Hansen, A., 1984., Landslide hazard analysis. In: Brunsden, D.
& Prior, D.B. (Eds.), Slope Instability, JohnWiley and Sons, New
York, pp. 523–602.
[57] Varnes, D.J. and IAEG, Commission on Landslides and other
Mass Movements on Slopes, 1984. Landslide hazard zonation:
A review of principles and practice. The UNESCOPress, Paris, 63
p.
[58] Soeters, R. & van Westen, C.J., 1996. Slope instability recog-
nition, analysis and zonation. In: Turner, A.K. & Schuster,
R.L. (Eds.), Landslides Investigation and Mitigation. National
Academy Press, Washington, pp 129–177.
[59] Leroi, E., 1996. Landslide hazard-risk maps at dierent scales:
objectives, tools and developments. In: Senneset, K. (Ed.),
Landslides - Glissements de Terrain. Balkema, Rotterdam, 35–
51.
[60] Aleotti, P. & Chowdhury, R., 1999. Landslide hazard assess-
ment: Summary review and new perspectives. Bull. Eng. Geol.
Env., 58, 21–44.
[61] Ayonghe, S.N., Suh, C.E., Ntasin, E.B., Samalang, P. & Fantong,
W., 2002. Hydrologically, seismically and tectonically triggered
landslides along the Cameroon Volcanic Line, Cameroon. Afr.

Geohazards in the Ndop Plain, Cameroon Volcanic Line |449
Geosci. Rev., 9, 4, 325–335.
[62] Ayonghe, S.N., Ntasin, E.B., Samalang, P. & Suh, C.E., 2004.
The June 27, 2001 landslide on volcanic cones in Limbe, Mount
Cameroon, West Africa. J. Afr. Earth. Sci., 39, 3–5, 435–439.
[63] Ancochea, E., Huertas, M.J., Hernán, F.& Brändle, J.L., 2010. Vol-
canic evolution of São Vicente, Cape Verde Islands: The Praia
Grande landslide. J. Volc. Geotherm. Res., 198, 1–2, 143–157.
[64] Knapen, A., Kitutu, M.G., Poesen, J., Breugelmans, W., Deck-
ers, J. & Muwanga, A., 2006. Landslides in a densely populated
county at the footslopes of Mount Elgon (Uganda): Characteris-
tics and causal factors. Geomorphology, 73, 1–2, 149–165.
[65] Ghogomu, R.T., 2005. Natural and human-induced disasters in
the south continental part of the Cameroon Line: inventory, ge-
ology, and a GIS and radar remote sensing approach. Thèse de
Doctorat d’État, Université Yaoundé I, Cameroun.