Assessment of Soil Erodibility in Relation to Soil Degradation and Land Use in Mediterranean Libya
Murad Milad Aburas
A thesis submitted in fulfillment of the requirements for the degree of Doctor of Philosophy in the School of Agriculture, Food and Rural Development
Faculty of Science, Agriculture and Engineering University of Newcastle upon Tyne
June, 2009
ABSTRACT In spite of several studies at different scales in the Mediterranean region of Libya, the extent and nature of soil degradation by water erosion has not been experimentally well quantified, yet erosion risk is nevertheless indicated on soil maps. The objectives of the present study were to: (a) evaluate erosion-driven soil degradation in relation to soil properties and land use impacts; (b) investigate whether the widespread USLE concept of soil erodibility is suited to Mediterranean soil and environmental conditions in Libya; and (c) critically assess whether soil types that have been mapped in this part of northern Libya can be distinguished according to their erodibility? A procedure was developed to assess soil degradation in the field under contrasting land uses on slight to moderate slopes. This involved field-based observations and selected laboratory measurements, including particle size distribution; organic matter content; saturated hydraulic conductivity and aggregate stability, of four soil classes (Calcic Rhodoxeralfs, Typic Rhodoxeralfs, Lithic Rhodoxeralfs and Lithic Rendolls) under three land uses (barley, tomatoes and natural vegetation) at 40 study sites. Soil erodibility was estimated for all soil classes using the predictive K-USLE formula and validated against field observations, erosion plot data, and measured aggregate stability. Measured soil properties and soil loss under simulated rainfall tests were then used to critically examine the concept of soil erodibility and its use under Mediterranean conditions. Typic and Calcic Rhodoxeralfs proved to be more resistant to soil erosion, with lower measured soil erodibility values, fewer field-based indications of sheet and rill erosion, and the smallest amounts of soil loss under simulated rainfall tests compared to Lithic Rhodoxeralfs and Lithic Rendolls. The moderate aggregate stability of the Typic and Calcic Rhodoxeralfs significantly improved their resistance to dispersion by rain drop impact and contributes to their measured higher rates of saturated hydraulic conductivity, resulting in less runoff and reduced soil loss. The Lithic Rhodoxeralfs were more degraded in the field, were more erodible (K-USLE), and showed greater soil loss under simulated rainfall. This can be explained by their smaller organic matter and greater silt and very fine sand contents, which contribute to their lower measured aggregate stability and slower saturated hydraulic conductivity. Contrary to expectations, the calcareous, well aggregated, but shallow, Lithic Rendolls (Red Rendzinas) were also less resistant to erosion than the leached Typic Rhodoxeralfs. Validation of estimated K-USLE erodibility values against field observations and erosion plot data showed that the formula overestimated soil erodibility under Libyan conditions. Simulated rainfall test results confirm that soil properties play an important role in defining erodibility-related processes, however, it was aggregate stability and saturated hydraulic conductivity that together explained about 75% of variation in the measured soil erodibility, rather than soil properties used for K factor estimation. These two parameters are not usually available in soil survey reports, are not used in soil classification, while aggregate stability is not used in the K-USLE index. This highlights the need for revised criteria for soil survey and mapping in the Mediterranean region, if soil erodibility prediction on a spatial basis is to be accurate. It was found that land use, regardless of soil class, is directly related to field-assessed, erosion-driven soil degradation in the following order of severity: barley > tomatoes > natural vegetation.
