Modelling Landslide Susceptibility In Parts Of Southeastern Nigeria With Medium Resolution Remotely...

Modelling Landslide Susceptibility In Parts Of Southeastern Nigeria With Medium Resolution Remotely Sensed Images


J. I. Igbokwe1 , K. U. Orisakwe1, J. O. Akinyede2, B. Dang3 and T. Alaga3
1. Department of Surveying and Geoinformatics
Nnamdi Azikiwe University
Awka, Nigeria.

2. Space Applications Department
National Space Research and Development Agency
Abuja, Nigeria

3. National Centre for Remote Sensing

Landslides occur in different parts of Southeastern Nigeria due to widespread impact of gully erosion resulting from annual rainfall and subsequent flooding. In this area landslide occur mostly as earth movement, rock and debris flows on slopes previously weakened by flood water. Remotely sensed images combined with field observation were used in this study to map potential areas of landslides in south – central parts of Anambra State in Southeastern Nigeria. The study generated landslide zonation map highlighting areas of different degrees of susceptibility and confirmed the possibility of using medium scaled remotely sensed data in landslide susceptibility study.

1. Introduction
Landslide is a geological phenomenon, which occurs as a result of ground movement. It can occur as rock falls, failure of unstable slopes, sand and debris flows on slopes (Renwick et al ,1982), etc. Landslides can cause a lot of damages with direct and indirect effects on human settlements and physical infrastructure. Landslides are very predominant where slope stability has been compromised. The Natural causes of landslides include

  • Erosion of the ground/slopes by flood water, rivers or ocean waves
  • Ground water pressure that destabilize sloppy grounds
  • Slopes weakening through heavy rainfalls , snowmelts, etc
  • Earthquakes that destabilize the slopes

In Southeastern Nigeria the major cause of landslides is gully erosion. There is a widespread occurrence of large and deep gullies in the area and the annual impact of rainfall induced floods weaken the ground around the gully sites and slopes resulting in landslides in many locations (Igbokwe et al, 2003, 2008). The porosity of the soil and moisture content as well as slope angles facilitates the occurrence of landslides in the area (Fig.1). Similarly human activities from massive urbanization and agricultural practices alter the terrain and weaken slopes thereby leaving the land prone to landslides. The impact is felt both in the developed and the undeveloped areas causing damages to properties and physical infrastructures and sometimes leading to loss of life.

Fig. 1 . Weakened Slopes collapse under the impact of rainwater in Gullied areas Anambra State in Southeastern Nigeria (Source: Igbokwe et al, 2008)

Remote sensing has been used to study, evaluate and model landslide susceptibility in different parts of the world. For example Gao and Lo (1995) were able to predict landslide probabilities of the mountain terrain of Nelson County, Central Virginia, USA. Similarly, Melzner et al (2006) performed landslide susceptibility analysis using remote sensing derived data and GIS techniques in South Viti Levu, Figi Islands. The study used aerial photography to create a digital landslide inventory map of the area.
The purpose of this study is to map the landslide potential of parts of Anambra State in Southeastern Nigeria, as part of a comprehensive investigation and mapping of gully erosion in Southeastern Nigeria. The study is to classify the area into landslide potentials according to their degree of susceptibility. For this purpose five classes were identified;

  1. Least susceptible Areas
  2. Less
  3. Average
  4. More
  5. Most susceptible Areas

2. The Study Area
The area selected for this study is located within the heavily gullied parts of Anambra State in Southeastern Nigeria. The geographic location is approximately between latitudes 5o 51’ N and 6o 08’N and longitudes 6o 50’E and 7o 10’ E (see fig. 2) . It lies within the tropical region. Early rainfall occur usually in January/February with full commencement of rainy season in March and stopping in November of each year. The dry season lasts between four to five months. The average highest annual rainfall usually recorded around July to October is about 1952 mm.

Fig. 2 Location of the Study Area

The study area is heavily populated and has large network of roads and dense concentration of residential buildings.


3. Materials and Methods
The materials used for the study include Satellite images, topographic data and field data collected from the ground. The satellite images used are Landsat –ETM+ dataset obtained in 2001 and Shuttle Radar Topographic Mission (SRTM) image data of 2002. The method employed for the study is illustrated in figure 3 bellow

Fig. 3 Procedure for Landslide Susceptibility Mapping

The Landsat-ETM+ dataset was georeferenced to Minna national UTM datum and the study area windowed from the georeferenced image. The image window was then subjected to some pre-processing and then demarcated into areas affected by landslides and gullying. The SRTM image was used to generate DTM at 10m interval and then used to extract the terrain elevation, slope gradients, aspects and configuration. These together with the field data were integrated into the demarcated image to generate a detailed classified landslide susceptibility data.

Fig. 4. Landsat –ETM+ Composite of the Study Area

Fig. 5 SRTM Data of the Study Area

4. Results and Discussion
Figure 6 below shows the final landslide susceptibility map of the study area. The result shows that the most susceptible areas to landslides are where the terrain undulation is high. In these areas the impact of rainfall induced flood water is high and slope failures are very predominant. There are also wide spread occurrence of deep and wide gullies in this area, as more and more sloppy grounds cave in after being weakened by the impact of rainfall. Areas that are of average, less or least susceptibility are found mainly in the low lying grounds. These are mostly developed and inhabited areas and human activities contribute to incidences of landslides in the area. Table 1 shows the different areas of susceptibility and the percentage of the total areas occupied. The table revealed that most susceptible areas occupy about 11.98 % of the area studied or 9827.088 Hectares, while more susceptible areas are found in 29.58 % of

Fig. 6. Landslide Susceptibility Map

Table 1. Percentage Coverage of Landslide Susceptible Areas

the study area. Average susceptible areas occupy a total of 25.74 % of the study area or about 21104.772 Hectares. Less and Least Susceptible areas together occupy a total of 31.43 % of the study area or about 25771.475 Hectares of land. The implications of these results are enormous. The area is heavily populated with extensive network of roads and dense concentration of residential houses. As a result, slope failures and landslides have caused extensive damage with each rainy season. Some do result in loss of lives.

The result also confirmed that medium scaled remotely sensed data can be used to analyse areas susceptible to landslide in Southeastern Nigeria. This is important because of difficulty to access very high resolution remote sensing data over these areas.

The fund for this study was provided by the National Space Research and Development Agency (NASRDA), Abuja, Nigeria as part of the comprehensive mapping of impact of gully erosion in Southeastern Nigeria.

Gao, J., and Lo, C. P. (1995). Micro-scaled Modelling of Terrain Susceptibility to Landsliding from a DEM: A GIS approach. Geocarto International, Vol. 10, No. 4, pp 15 – 30

Igbokwe, J. I., Ojiako, J. C. and V. C. Nnodu (2003). Monitoring, Characterisation and Controlling of Floodwater Erosions Using Remote Sensing Techniques. Proceedings of the Technical Session of 38th Annual Conference of Nigerian Institution of Surveyors, Lokoja, Nigeria, pp. 73 –79

Igbokwe, J. I., J. O. Akinyede, B. Dang, T. Alaga, M. N. Ono, V. C. Nnodu, and L. O. Anike (2008). Mapping and Monitoring of The Impact of Gully Erosion in Southeastern Nigeria with Satellite Remote Sensing and Geographic Information System. International Archives of Photogrammetry, Remote Sensing and Spatial Information Sciences. Vol. XXXVII, Part B8, Beijing 2008, pp 865 – 871.

Melzner, S., T. Glade and M. Bonte-Grapentin (2006). Landslide Susceptibility Analysis Using Remote Sensing Derived Data and GIS Techniques – Navua Catchment, South Viti Levu, Figi Islands.

Geophysical Research Abstracts, Vol. 8, 04059.

Renwick, W., Brumbaugh, R. and Loeher, L. (1982). Landslide Morphology and Processes on Santa Cruz Island, California. Geografiska Annaler. Series A, Physical Geography, Vol. 64, No.3 / 4, pp 149 – 159

Rohit Kumer and Sangram Ganguly (1996). An Integrated Approach with GIS and Remote Sensing Techniques for Landslide Hazard Zonation.