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

Accueil du site → Doctorat → Allemagne → Assessing trends in land use change in the Borana rangeland Ethiopia as one cause of greenhouse gas emissions and carbon sequestration variations

University of Kassel (2016)

Assessing trends in land use change in the Borana rangeland Ethiopia as one cause of greenhouse gas emissions and carbon sequestration variations

Mgalula, Michael Elias

Titre : Assessing trends in land use change in the Borana rangeland Ethiopia as one cause of greenhouse gas emissions and carbon sequestration variations

Auteur : Mgalula, Michael Elias

Université de soutenance : University of Kassel

Grade : Doktor der Agrarwissenschaften (Dr. agr.). 2016

Résumé partiel
Land use change is one major driver of greenhouse gas (GHG) emissions. When land use changes from ecosystems with permanent natural vegetation to arable farming with the soil lying bare for periods of the year, GHG emissions tend to increase while carbon storage capacity tends to decrease. Crop farming is expanding globally in both smallholder and agro-industrial systems, often onto neighbouring semi-arid to sub-humid rangeland ecosystems. The present study analysed trends in land use change in the Borana rangeland of southern Ethiopia. Increasing population, the emergence of private enclosures, changing land use policies, and increasing climate variability lead to rapid changes in the traditional livestock based pastoral land management system. Based on a literature review on case studies from the East African rangelands, the present study devised a schematic model of land use, greenhouse gas emission and carbon sequestration pathways. Satellite data combined with field survey information were used to analyse the nature and extent of land use and land cover change from 1985 to 2011 in five study areas (Darito/Yabelo District and Soda, Samaro, Haralo, Did Mega/all Dire District). A 12% increase of cultivated area occurred at Darito, mainly at the expense of bushland, while at the other sites cultivation did not expand but grassland increased (+16 to 28%) mainly at the expense of bushland (-23 to 31%). A marked increase in bare land was only observed at the Haralo site (+13%). Factors inducing agricultural expansion were examined in more detail at Darito. GPS measurements and cropping histories of 108 fields of 54 farmers were overlaid in a geographical information system (GIS) with map layers of soil, rainfall, elevation and slope. Multiple linear regression showed that slope and elevation were significant variables for cultivation expansion onto low-lying areas, while soil type, distance from dry-river, and rainfall were not. The model had a low R² (0.154), the explanatory power was low and other variables are likely to explain more of this expansion. Scatterplots of field size and years under cultivation against altitude indicate a recent (since 2000) trend of cultivation toward lower altitudes (below 1620 masl) and larger field sizes (>3ha). Seasonal vegetation changes were analyzed using phenological observations of crops, rainfall, and time series of normalized difference vegetation index (NDVI) data. Phenology records served to identify points in time when the biomass cover of cultivated land was high (greening up phase, pre-harvest) or low (land preparation for planting/seeding). Cropland could be distinguished, best from woodland, not so well from bushland and grassland, using NDVI time series spectral profiles.

Présentation

Version intégrale (6,26 Mb)

Page publiée le 11 janvier 2019