Terrain modelling and the impact of lava erosion in a neotectonic landscape: the case of the south west flank of Mount Cameroon, West Africa

E.E. Nama
Telford Institute of Environmental Systems, Salford University, M5 4WT, United Kingdom
E-mail: ernest@nama.freeserve.co.uk

Abstract

The occurrence of crustal movements and frequent seismic events mean that neotectonic terrains are subject to the development of natural hazards. Modelling the irregular shapes of volcanic regions such as the south west flank of Mount Cameroon is a complex problem. The orientation of the landscape has a major impact on the hydrological, geomorphological and tectonic processes active in this volcanic region. The spatial distribution of topographic attributes can often be used as an indirect measure of the spatial variability of these processes and allows them to be mapped and analysed using relatively simple GIS techniques. Many geographic information systems are being developed that store topographic information as the primary data for analysing lava flow and erosion. Digital elevation models are the primary data used in the analysis of the trend of lava flows in this neotectonic terrain.

This paper examines surface elevation based on samples of terrain modelling and field survey. One structure has been used to digitally represent this study: Lattices and Triangular Irregular Networks (TINs). GIS based on digital elevation models, offers the potential to be able to map lineaments and slope angles to predict surface processes such as lava flow and erosion. The prediction of flow directions by the terrain model and subsequent erosion are correlated with field results.

Keywords: Mount Cameroon, lava flow, erosion, lineament, digital elevation model, GIS, neotectonic terrain

1.Introduction

It is doubtful whether the exact time of an outbreak of a volcano such as Mount Cameroon can be exactly predicted except possibly in some cases shortly before its commencement. The main reason for this difficulty is the complex unpredictable interaction of the factors governing the behaviour of a volcano. However, prior to volcanic eruptions in Mount Cameroon, there is often a period of mild to intense earthquake tremors recorded from epicentres within the volcano.

The use of terrain modelling in representing the surface of Mount Cameroon is an appropriate application which is undergoing regular updating to obtain the best possible terrain representation. In this paper terrain modelling is used to demonstrate surface relief features and the direction of lava flow. The accuracy of the relief features, direction of lava flow and extent of lava erosion were investigated in the field.

2. Background

2.1 Hypotheses and terrain modelling

Mount Cameroon is one of Africa’s highest volcanoes and rises more than 4 km above sea level. Numerous small cinder cones, often fissured-controlled, run parallel to the long axis of the volcano, especially on the flanks and surrounding lowlands. The south west slopes of the mountain are steeper and less concave with stony and rocky terrain often with a very irregular topography due to the form of the individual lava flows. There are numerous drainage channels running radially down the mountain, most of them bearing water only after the very heavy rains which also facilitate the process of erosion on steep slopes.

Many hypotheses have been proposed describing the nature and origin of Mount Cameroon: 1.Vertical movements which created either a series of horsts and grabens bounded by SW-NE normal faults (Passarge, 1909; Geze, 1943; Gouhier et al., 1974; Deruelle et al., 1983, 1984) or continental rifting (Lasserre, 1978; Jacquemin, 1981). Such a rift could have been produced directly by a thermal anomaly in the asthenosphere (Dunlop, 1983; Fitton, 1983), a sort of hot spot (Tchoua, 1974; Duncan, 1981). This hypothesis implies a Recent (or Cretaceous – Tertiary) stretching mechanism in the lithosphere. 2.Rejuvenation of a major fault. This fault zone is either supposed to have existed before the opening of the South Atlantic Ocean in the prolongation of the Pernambuco fault (D’Almeida and Black, 1967; Louis, 1970, 1978) or was related to an early stage of seafloor spreading (Upper Cretaceous).

This volcano is the only other volcano outside the Mediterranean to have a documented eruption before the time of Christ – at least 15 eruptions since 450 BC.

Digital elevation model (DEM) is defined here as an orderly array of numbers that present the spatial distribution of elevations above some arbitrary datum in a landscape. It may consist of elevations sampled at discrete points or the average elevation over a specified segment of the landscape, although in most cases it is the former. DEM’s are a subset of digital terrain models (DTM) defined in this paper as ordered arrays of numbers that represent the spatial distribution of terrain attributes. The application of digital elevation modelling in this region to analyse the impact of lava erosion on the terrain is a new concept.

2.2 Chronologic narrative

Geologic studies in this region indicate that Mount Cameroon is volcanically active. The volcanic eruptions commenced in the Upper Cretaceous, but the main eruptions occurred from the Tertiary down to modern times – May to June 2000 saw the last eruption in this region. The mountain is a Hawaiian type of volcano with a huge oval shaped basalt dome without a central crater but instead, flank outflows of lava through radial fissures are typical (Fig. 1). Most of these fissures are found along an oval shaped plateau at 3000-3500 metres, although on the south west flank, presently the most active part of the mountain, fissures extend at low altitudes. Flank eruptions occurred in this region in 1922, 1929, 1982 and 1999. In 1922 almost 10 million m3 ‘aa’ type lava flow from the summit flowed towards Bakingili in the south west flank for seven months. In 1982 5 x 106 m3 of ‘aa’ lava flow and pyroclasts moved 1.5 km for 20 days towards the south west flank. In 1999 lava flow from vents located at 2840 m and 1520 m above sea level flowed towards the south west flank towards Bakingili and stopped approximately 10 metres from the Atlantic Ocean. The lava flows issued from the fissures consist of fine grained basalts. The older basalts are relatively rich in olivine and derived minerals like iddingsite. At the time of writing this paper, there was another eruption near the summit of Mount Cameroon (4000 metres above sea level), with lava flowing towards Bokwango in Buea and Bakingili.

Fig 1. Radial fissures and cones on the SW flank of Mount Cameroon

3.Methods

Digital elevation modelling (DEM) as a GIS technique in generating 3-D terrain models of the south west flank of Mount Cameroon has been important in detecting terrain features and determining direction of lava flow. The underlying concept is that the model will reflect the nature of the terrain for morpho-tectonic identification and isolation (Nama, 2000).

A Triangular Irregular Network (TIN) topology for the area was created by manually digitising a 1 : 50 000 topographic map which approximated the terrain surface by a set of triangular facets. Most TIN models assume planar triangular facets for the purpose of simpler interpolation in applications such as contours (Lee, 1991). The purpose of using a TIN method in this study is to convert the dense vector data to a TIN model in such a way that the surface defined is as close to the DEM surface as possible.

The surface DEM mapping has been applied in this study as an alternative derivative for the detection of brittle tectonic structures such as fractures, faults, large-scale fractures and fracture zones. The scale of the mapped fractures are commonly denoted as lineaments (O’Leary et al., 1976) which control the direction of lava flow downslope and influence the drainage pattern. The extraction of lineaments from a digital elevation model entailed a subjective visual identification and manual digitising. Lineaments in a DEM are identified by a drop in elevation for a short distance (Wladis, 1999), therefore they have been described by a certain frequency in this region in terms of their representation in a digital elevation model. The digital elevation model was transformed into a shaded relief image in which different effects were achieved by varying the value of surface / sun orientation, sun elevation angle and vertical exaggeration to obtain the best brightness information of the shaded relief to identify and map lineaments (Nama, 1999) (Fig. 2).

Field observations undertaken in this region, verified the accuracy of mapped lineaments on DEM and the predicted direction of lava flow.

Fig 2. Mapped Lineaments from DEM of the SW flank

4. Lava Erosion Manifestation

4.1Lava flow

Different ‘aa’ lava types reflect interrelated aspects of magma rheology, gas content, terrain effects and flow rates. Lava viscosity and yield strength generally increase downstream because of decreased gas content, decreased average bulk temperature of the flow, increased crystallisation of microphenocrysts, and incorporation or formation of solids. The net effect is to trigger successive irreversible changes in ‘aa’ texture. In the field, key parameters used in interpreting the evolution of lava flow and channels are variations in lava temperatures and densities in relation to eruption duration, distance from vent and lava type. Temperature determinations on the lava were considered especially important for evaluating the extent of erosion because an increase in temperatures might indicate the arrival of a more fluid magma batch from below the erupting vents.

The details of whether lava channels form by predominantly erosional (thermal or mechanical) or constructional processes or by laminar flow remain debatable. It is likely that volcanic channels form under various non-unique combinations of some or all of these conditions. The goal was to assess field evidence for erosion of substrate material by flowing lava. Mechanical erosion was considered in which preflow material was removed much like bed load in a fluvial stream, and thermal erosion in which preflow material is melted or partly melted and mobilised by flowing lava. In order to understand the conditions under which erosion occurs and to determine the scale of erosion, I investigated basaltic tube-fed flows, basaltic caves, collapsed basaltic tubes and lava excavation of the Limbe – Idenau road.

The criteria used to assess erosion by lava are (1) active lavas observed or inferred to downcut, (2) direct evidence of thermal erosion (i.e., melted substrate), (3) lava undercut into preflow material, (4) geochemical signature showing assimilation of preflow materials into the eroding flow, (5) assimilated fragments of preflow material into the eroding flow, (6) exposed paleosols/scoria in wall of lava tube.

4.1.1 Downcutting

This process was observed in the field in old lava tubes and ducts with collapsed roof segments over lava tubes (Fig. 3). This criterion also included inferred downcutting in active lava tubes determined by geophysical monitoring of the position of the flowing lava in the subsurface. The direction of flow was determined by the proximity to structural discontinuities and the steepness of terrain (Fig. 4).

Fig 3. Collapsed roof segments over lava tubes at 2710m above sea level

Fig 4. Direction of lava flow determined by lineament orientation

4.1.2 Melting

Direct evidence of melting of preflow material such as basalts, regolith and vegetation were observed as result of intense heat. During the 1982 volcano eruption lava flowed from the vent at a velocity of 20 km /hour with at a temperature of 1070° C. The ‘aa’ lava from the volcanic eruption of March 1999 travelled swiftly from the vents but slowed at an estimated 10-25 m/hour on the gentle flanks of the coastal plain in this region. The lava flow temperature measured 972° C at a distance of 300 m from one of the vents. The moderately high viscous mature of the blocky ‘aa’ lava was responsible for the decreased flow velocity. The high temperatures were responsible in melting substrate materials with a relatively lower melting point. This was the case with rocks, regolith and vegetation along the flow paths of the lava.

4.1.3 Material contact

Material contact is based on contact relations between the lava and preflow surfaces. Active lava flow with high temperatues flowed along the valleys to approximately 5 metres from the shores of the Atlantic Ocean dominated by volcanic sediments. The magma that came in contact with regolith and the paved Limbe to Idenau road, close to the shores of the Atlantic Ocean dissolved and eroded these materials. This road was completely eroded and traces of regolith were found attached to consolidated blocky ‘aa’ lava where the flow toe moved more slowly (Fig. 5). Little fluid material was visible at the flow front, and the steep high front advanced by spalling of dense blocks and slabs, largely from the breccia layer riding on top of the advancing flow.

Fig 5. Excavated section of the Limbe – Idenau road

5. Discussion

After the first few days of eruption, the ‘aa’ lava and its distributary channels undergo a complex morphological evolution related to several intertwined changes in lava properties. These include (1) decreases in eruptive volume (2) increased ground mass crystallisation of the magma before venting (3) changes in effective viscosity and yield strength of the lava (4) degassing in the rift magma chamber, at the vent and downchannel which result in density increases of the lava; and physical maturation of the distributary channels.

The use of DEM to determine the flow of lava and subsequent erosion on the slopes of the south west flank highlights three important factors (1) lineament orientations (2) slope facets, and (3) width of depressions. In the field, observations revealed a correlation between the orientation of faults and fractures and the direction of lava flows. In areas where faults lines have been eroded successively by ephemeral flows and runoffs, wide valleys with average widths of 6 metres have been formed increasing the holding capacity of runoff and lava flow downhill.

On the steeper slopes above an altitude of 1200 metres, the velocity of lava flow is high. For example, the 1982 volcanic eruption had a velocity of ‘aa’ lava flow of 4 m/s on the upper steep slopes at an altitude of 2700 metres and reduced to 2 m/s at 2500 metres. During the March – April 1999 eruption, the ‘aa’ lava flow travelled swiftly down the steep upper slopes, but slowed to 10 – 25 m/ hour on the gentle slopes and reduced to 7 m/ hour on the gentle sloping coastal terrain, eroding the Limbe – Idenau road on April 15, 1999. When it eventually ceased, all forward movement on April 17, 1999, the 12 metre thick ‘aa’ blocky lava had traces of consolidated volcanic soil which was evidence of lava erosion in this region. The May – June 2000 eruption at an altitude of 4000 metres flowed at a velocity of 10 m/hour near the vents and increased to 20 – 25 m/hour on the steep upper slopes. The lineament orientation, slope angles and width of depressions have also influenced the flow pattern and erosive potential of this latest eruption.

6. Conclusion

Although Mount Cameroon is a poorly studied volcano, the recent relatively high tectonic events, especially in the south west flank, is generating more interest from researchers. The introduction of terrain modelling in such rugged environment to determine the direction of lava flow and subsequent lava erosion is effective when slope facets, velocity of flow and orientation of lineaments are determined. Nonetheless, observations in the field are vital to correlate the digitally generated data.

References

D’Almeida, F., and Black, R., 1967, Comparison structural entre le Nord-Est du Brésil et l’Ouest africain. Symp. Continental Drift, Montevideo.

Déruelle, B. and Regnoult, J.M, 1983, Carte télé-interprétative des linéaments de la République Unie du Cameroun au Nord du 4ème parallèle (1/1 000 000). Minist. Mines. Energie, Direction des Mines et de la Géologie, Yaounde, Cameroun, 8.

Duncan, R.A., 1981, Hot spots in the southern oceans, an absolute frame of references for motion of the Gondwana continent. Tectonophysics, 74: 29-42.

Dunlop, H.M., 1983, Strontium isotope geochemistry and potassium-argon studies on volcanic rocks from the Cameroon line, West Africa. PhD Thesis, Univ. Edinburgh, 347 pp.

Fitton, J.G., 1983, Active versus passive continental rifting: evidence from the West African rift system. Tectonophysics, 94: 473-481.

Geze, B., 1943, Géographie physique et géologique du Cameroun Occidental. Mémoir du Museum National d’Histoire, Paris, XV11, pp. 12

Gouhier, J; Regnoult, J.M., and Déruelle, B., 1983, Géologie et structure linéamentaires au Cameroun (du lac Tchad a la Sanaga): apport des images Landsat. Annale de Faculté des Sciences, Yaounde, 1, 2, 2, 3-4.

Jacquemin, H., 1981, Contribution a l’étude volcanique du Cameroun (‘Ligne du Cameroun-Adamaoua’). Ann. Fac. Sci., Univ. Yaounde, Cameroun, 17: 3-48.

Lasserre, M., 1978, Mise au point sur les granitoides dits ‘ultimes’ tertiares du Cameroun: gisement, pétrographie et géochronologie. Bull. B.R.G.M., 2e Sér., 4: 148-159.

Louis, P., 1970, Contribution géophysique à la connaissance géologique du bassin du Lac Tchad. Mém. O.R.S.T.O.M., 42: 311pp.

Louis, P., 1978, Gravimétrie et géologie en Afrique occidentale et centrale. Mém. B.R.G.M., 91: 53-61.

Nama E.E., 1999, Lineaments in the Mount Fako region, West Africa: a GIS approach. Annual Conference of the Royal Geographical Society with the Institute of British Geographers, University of Leicester, 4 – 7th January 1999 pp177

Nama E.E., 2000, Modelling variables to predict landslides in the sw flank of the Cameroon Volcanic Line, Cameroon, West Africa, Proceedings of the Asian Conference on Unsaturated Soils, UNSAT – ASIA 2000/ Singapore/ 18-19th May 2000, pp 813 - 817

O’Leary, D.W., Friedman, J.D., and Pohn, H.A., 1976, Lineaments, linear, lineations: Some proposed new standards for old terms. Geological Society of American Bulletin, 87, 1463-1469.

Passarge, S., 1909, Der vulkanische Ausbsuch des Kamerun-berges. Dent. Kolonialblatt., 20, 628 pp.

Tchoua, F., 1974, Contribution à l’étude geologique et pétrographique de quelques volcans de la ligne du Cameroun (Monts Manengouba et Bambouto) Doctoral thesis, Univ. Clermont- Ferrant, 347 pp.

Wladis, D., 1999, Automatic lineament detection using digital elevation models with second derivative filters. Photogrammetric Engineering and Remote Sensing, 65, 4, 453-458.