- Title
- A characterization of landslide occurrence in the Kigezi Highlands of South Western Uganda
- Creator
- Nseka, Denis
- Subject
- Landslides -- Uganda
- Subject
- Land degradation -- Uganda Earth movements
- Date Issued
- 2018
- Date
- 2018
- Type
- Thesis
- Type
- Doctoral
- Type
- DPhil
- Identifier
- http://hdl.handle.net/10948/33791
- Identifier
- vital:33029
- Description
- The frequency and magnitude of landslide occurrence in the Kigezi highlands of South Western Uganda has increased, but the key underpinnings of the occurrences are yet to be understood. The overall aim of this study was to characterize the parameters underpinning landslide occurrence in the Kigezi highlands. This information is important for predicting or identifying actual and potential landslide sites. This should inform policy, particularly in terms of developing early warning systems to landslide hazards in these highlands. The present study analysed the area’s topography, soil properties as well as land use and cover changes underpinning the spatialtemporal distribution of landslide occurrence in the region. The present study focussed on selected topographic parameters including slope gradient, profile curvature, Topographic Wetness Index (TWI), Stream Power Index (SPI), and Topographic Position Index (TPI). These factors were parameterized in the field and GIS environment using a 10 m Digital Elevation Model. Sixty five landslide features were surveyed and mapped. Soil properties were characterised in relation to slope position. Onsite soil property analysis was conducted within the landslide scars, auger holes and full profile representative sites. Furthermore, soil infiltration and strength tests, as well as clay mineralogy analyses were also conducted. An analysis of the spatial-temporal land use and cover changes was undertaken using satellite imagery spanning the period between 1985 and 2015. Landslides were noted to concentrate along topographic hollows in the landscape. The occurrence is dominant where slope gradient is between 25˚ and 35˚, profile curvature between 0.1 and 5, TWI between 8 and 18, SPI >10 and TPI between -1 and 1. Landslides are less pronounced on slope zones where slope gradient is <15˚ and >45˚, profile curvature <0, TWI <8 and >18, SPI <10 and TPI >1. Deep soil profiles ranging between 2.5 and 7 meters are a major characteristic of the study area. Soils are characterized by clay pans at a depth ranging between 0.75 and 3 meters within the profiles. The study area is dominated by clay texture, except for the uppermost surface horizons, which are loamy sand. All surface horizons analysed had the percentage of sand, silt and clay ranging from 33 to 55%, 22 to 40% and 10 to 30% respectively. In the deeper horizons, sand was observed to reduce drastically to less than 23%, while clay increased to greater than 50%. The clay content is very high in the deeper horizons exceeding 35%. By implication, such soils with a very high clay content and plasticity index are considered as Vertisols, with a profound influence in the occurrence of landslides. The top soil predominantly contains more quartz, while subsurface horizons have considerable amounts of illite/muscovite as the dominant clay minerals, ranging from 43% to 47 %. The liquid limit, plasticity index, computed weighted plasticity index (PIw), expansiveness (ɛex) and dispersion ranging from 50, 22, 17, 10 and 23 to 66, 44,34,54 and 64, respectively also have strong implications for landslide occurrence. Landslides are not normally experienced during or immediately after extreme rainfall events but occur later in the rainfall season. By implication, this time lag in landslide occurrence and rainfall distribution, is due to the initial infiltration through quartz dominated upper soil layers, before illite/muscovite clays in the lower soil horizons get saturated. Whereas forest cover reduced from 40 % in 1985 to 8% in 2015, cultivated land and settlements increased from 16% and 11% to 52% and 25% respectively during the same period. The distribution of cultivated land decreased in lower slope sections within gradient group < 15˚ by 59%. It however increased in upper sections within gradient cluster 25˚ to 35˚ by over 85% during the study period. There is a shift of cultivated land to the steeper sensitive upper slope elements associated with landslides in the study area. More than 50% of the landslides are occurring on cultivated land, 20% on settlements while less than 15 % and 10% are occurring on grassland and forests with degraded areas respectively. Landslides in Kigezi highlands are triggered by a complex interaction of multiple- factors, including dynamic triggers and ground condition variables. Topographic hollows are convergence zones within the landscape where all the parameters interact to cause landslides. Topographic hollows are therefore potential and actual landslide sites in the study area. Characterized by deep soil horizons with high clay content dominated by illite/muscovite minerals in the sub soils and profile concave forms with moderately steep slopes, topographic hollows are the most vulnerable slope elements to landslide occurrence. The spatial temporal patterns of landslide occurrence in the study area has changed due to increased cultivation of steep middle and upper slopes. Characterized by deep soil horizons with high clay content dominated by illite/muscovite minerals in the sub soils and profile concave forms with moderately steep slopes, topographic hollows are the most vulnerable slope elements to landslide occurrence. The spatial-temporal patterns of landslide occurrence in the study area has changed due to increased cultivation of steep middle and upper slopes. A close spatial and temporal correlation between land use/cover changes and landslide occurrence is discernible. The understanding of these topographical, pedological and land use/cover parameters and their influence on landslide occurrence is important in land management. It is now possible to identify and predict actual and potential landslide zones, and also demarcate safer zones for community activities. The information generated about the area’s topographic, pedological and land cover characteristics should help in vulnerability mitigation and enhance community resilience to landslide hazards in this fragile highland ecosystem. This can be done through designating zones for community activities while avoiding potential landslide zones. It is also recommended that, tree cover restoration be done in the highlands and the farmers encouraged to re-establish terrace farming while avoiding cultivation of sensitive steep middle and upper slope sections.
- Format
- xvii, 242 leaves
- Format
- Publisher
- Nelson Mandela University
- Publisher
- Faculty of Science
- Language
- English
- Rights
- Nelson Mandela University
- Hits: 1611
- Visitors: 1846
- Downloads: 365
Collections
NMU School of Environmental Sciences
| Thumbnail | File | Description | Size | Format | |||
|---|---|---|---|---|---|---|---|
| View Details Download | SOURCE1 | DENIS_NSEKA.pdf | 8 MB | Adobe Acrobat PDF | View Details Download |