Functional Context Similarity Units for Supporting Out-scaling of Sustainable Land Management in European Mediterranean Bioregion

It is widely recognized that Sustainable Land Management (SLM) are much needed for achieving Land Degradation Neutrality (LDN). Adoption and effectiveness of SLM depend on specific contexts. Contextual diversity makes uniform blanket policies promoting SLM less effective and out-scaling of site-specific SLM difficult. We propose the functional context similar unit (fCSU) approach to overcome these challenges by grouping common biophysical, economic and social contextual factors of land use and land management into distinct context similarity units that shape SLM adoption and resulting LDN. The presented study focuses on the European Mediterranean Bioregion within Mediterranean basin. The socio-ecological contextual factors selected for analysis includes 25 spatial variables of biophysical condition (climate, topography, soil, land cover and livestock density), physical accessibility, demography and economic development status at the spatial resolution of 500 m. We identified and mapped context types using spatially explicit (grid-based) multivariate statistics. Core factors differentiating socio-ecological context were identified using principal component analysis (PCA). Six distinct types of socio-ecological context were defined and mapped by cluster analysis. The empirical nesting of these clusters with aggregated land cover classes resulted in 47 Context Similarity Units (CSU) in the study region. The functionality of the derived SCUs were evaluated by unbalanced, grid-based ANOVA that measured and tested the differences in inter-annual trend of Net Primary Productivity (2001- 2014 period) among the SCU within each aggregated land cover type. The result demonstrates the potential of the function CSU approach to further our understanding LDN in socio-ecological contexts. The result of fCSU mapping can be directly used by LDN-oriented projects/ programs and citizen scientists in the region. For example given limited resource and aims, we can know approximately where efforts should be focused by managing or coping with what drivers. The resulted fCSU is also being implemented as SLM extrapolation domains in the Geoinformatics Options by Context (GeOC) tool tailored to LANDSUPORT's S-DSS: given SLM outcomes in a number of project sites, we can identify where similar intervention options have a potential of success based on contextual similarity. The presented study is under the framework of and supported by LANDSUPPORT project.