This research addresses a need to provide a better quantitative understanding of how the Southern Sudanian savanna has been actually changing under the combined impact of human induced land use, as a coupling factor of the SES, and climate change impacts. Population growth and expansions of settlements in the Southern Sudanian savanna lead to an increasing demand in agricultural products and land for agricultural production. This demand may lead to changing cropping cycles with shorter fallow periods and as a result amplify land use intensification. Lateral to that, the expansion of croplands into marginal areas not really suitable for cropping, often through the use of unsustainable cropping practices, may result in the loss of soil fertility as well as an increase in surfaces prone to soil erosion.
Deforestation rates in the Southern Sudanian savanna may aggravate the above problems. For instance, according to the Ministry of Environment of Burkina Faso (BF), deforestation rates in BF are in the excess of more than 110,000 ha of forest/year. A total of 75% of this loss is due to the expansion of croplands. However in some West African savannas (e.g. Mali, Burkina Faso and Ivory Coast) an expansion of woody cover through for example, natural regeneration, abandonment of agricultural land and reforestation can also be observed.
Satellite remote sensing data with area-wide repeatable data acquisitions and a long history allow the mapping of the land use and cover changes with a high degree of detail. Additional information can be gained by combining different remote sensing data sets. High temporal data acquisitions resolutions allow the alignment to critical time periods for cropping. The use of time-series data (‘hyper-temporal’ metrics from near to daily moderate resolution imagery) facilitates the derivation of phenology variables and fractional vegetation cover. This information permits the better detection of fine woody cover gradients and subtle tree cover changes related to for instance logging or changes in shrub cover. By linking the land use and land cover maps to macro-economic and farm typology models an improved land use assessments on a local scale is possible.
In this WP we aim to deal with the spatial and temporal changes and assess land use and land cover pathways, as far back in time as possible. An array of high resolution satellite imagery at resolutions from 5-meter (RAPID-Eye) to 15-30s-meter (ASTER, Landsat, possibly Nigeria-SAT 1 and 2) as well as 250-meter MODIS satellite data (‘hyper-temporal’ metrics) is used. Farm-typology models from the ‘Agriculture’ RC are used to cross verify spatial crop patterns within temporal windows. Field derived observations and very high resolution imagery will be collected to validate the LU mapping results. Indicators such as ‘deforestation’ and ‘expansion of croplands’, ‘land cover fragmentation’ will be specifically derived on a local scale from this WP. Lastly the WP aims to investigate driver causalities for the mapped land use change using bio-physical (soil and topography), infrastructural factors such as ‘distances to roads’, ‘access to permanent water’ as well as socio-economic factors i.e. ‘population density’ and ‘distances to towns’. Other socio-economic factors, such as ‘distance to irrigation points’, ‘market access’ and farmers’ ‘perspectives and behavior’ will be investigated in succession in the ‘Agricultural Systems’ RC (WP 3.2, WP 3.3).
Using the driver understanding and historical profiling of land use we aim to ultimately gain an understanding of the characteristics of the coupled SES within the WASCAL watersheds. The land use profiles and indicators from this WP will be used in the ‘Integrative Assessment’ RC to set up basic models that test specific scenarios related to climate change and adapted land use. There is also a strong link to the Competence Center; the research results from this WP will be implemented through the Competence Center using satellite data flows with similar spectral and temporal resolutions (e.g. Landsat Continuity Mission and data from NASAs new NPOESS Preparatory Mission). Staff at the Competence Center will set up process models for the integration of satellite and other data feeds for locally adapted LU mapping. Seamless data flows and processing chains are instigated that would provide consistent land use and land use change data products at regular intervals.