Informations et ressources scientifiques
sur le développement des zones arides et semi-arides

Accueil du site → Doctorat → Allemagne → 2012 → Assessment of Land Degradation Patterns in Western Kenya : Implications for Restoration and Rehabilitation

Rheinischen Friedrich-Wilhelms-Universität Bonn (2012)

Assessment of Land Degradation Patterns in Western Kenya : Implications for Restoration and Rehabilitation


Titre : Assessment of Land Degradation Patterns in Western Kenya : Implications for Restoration and Rehabilitation


Université de soutenance : Rheinischen Friedrich-Wilhelms-Universität Bonn

Grade : Doktor der Agrarwissenschaften (Dr. agr.) 2012

Land degradation remains a major threat to the provision of environmental services and the ability of smallholder farmers to meet the growing demand for food. Understanding patterns of land degradation is therefore a central starting point for designing any sustainable land management strategies. However, land degradation is a complex process both in time and space making its quantification difficult. There is no adequate monitoring of many of the land degradation issues both at national and local scale in Kenya. The objective of this study conducted between 2009 and 2012 was to assess the land degradation patterns in Kenya as a basis for making recommendations for sustainable land management. The correlation between vegetation and precipitation and the change in vegetation over the period 2001-2009 was assessed using 250 m resolution Moderate Resolution Imaging Spectroradiometer - Normalized Difference Vegetation Index (MODIS/NDVI) and time-series rainfall data. The assessment at national levels revealed that, irrespective of the direction of change, there was a significant correlation between vegetation (NDVI) and annual precipitation for 32% of the land area. The inter-annual change in vegetation cover, depicted by the NDVI slope, was between -0.067 and +0.068. A negative NDVI slope (indication of degradation) was observed for areas around Lake Turkana and several districts in eastern Kenya. Positive NDVI trends were observed in Wajir and Baringo, which are located in the dry land areas, showing that the vegetation cover was increasing over the years. NDVI difference between the baseline (2001-2003) and end line (2007-2009) showed an absolute change in NDVI of -0.42 to +0.48. But the relative change was between -74% for the degrading areas and +238% for the improving areas with most of the dramatic positive changes taking place in the drylands. Relative to the baseline, 21% of the land was experiencing a decline in the vegetation cover, 12% was improving, while 67% was stable. Classification of Landsat imagery for the period 1973, 1988 and 2003 showed that there were significant changes in land use land cover (LULC) in the western Kenya districts with the area under agricultural activities increasing from 28% in 1973 to 70% in 2003 while those under wooded grassland decreasing from 51% to 11% over the same period. Detailed field observations and measurements showed that over 55% of the farms sampled lacked any form of soil and water conservation technologies. Sheet erosion was the most dominant form of soil loss observed in over 70% of the farms. There was a wide variability in soil chemical properties across the study area with values of most major properties being below the critical thresholds needed to support meaningful crop production. Notable was the high proportion (90%) of farms with slightly acidic to strongly acidic (pH <5.5) soils. Over 55% of the farms had less than 2% soil organic carbon. There was a wide variability in the potential nutrient supply and uptake of the soils with the plots classified as high fertility (HF) having three times higher potential supply of nitrogen and phosphorus compared to the low fertility (LF) plots. The estimated maize yield potential of the soils was between 1.6 t/ha and 2.8 t/ha. However, the actual yield at farm level was less than 1 t/ha. There was a general consensus among the land owners that the productivity of the land, livestock, forests and water resources had declined. Combining methods and approaches for land degradation monitoring and assessment enabled capturing different aspects of the problem of land degradation, and thus important information for the design of sustainable land management strategies. Addressing the multiple nutrient deficiencies and low productivity requires adoption of integrated soil fertility management practices.


Version intégrale (7,7 Mb) **************

Page publiée le 25 janvier 2016, mise à jour le 10 novembre 2021