Found in Complexity, Lost in Fragmentation: Putting Soil Degradation in a Landscape Ecology Perspective

Abstract

The United Nations Convention to Combat Desertification (UNCCD) assumes spatial disparities in land resources as a key driver of soil degradation and early desertification processes all over the world. Although regional divides in soil quality have been frequently observed in Mediterranean-type ecosystems, the impact of landscape configuration on the spatial distribution of sensitive soils was poorly investigated in Southern Europe, an affected region sensu UNCCD. Our study proposes a spatially explicit analysis of 16 ecological metrics (namely, patch size and shape, fragmentation, interspersion, and juxtaposition) applied to three classes of a landscape with different levels of exposure to land degradation ('non-affected', 'fragile', and 'critical'). Land classification was based on the Environmentally Sensitive Area Index (ESAI) calculated for Italy at 3 time points along a 50-year period (1960, 1990, 2010). Ecological metrics were calculated at both landscape and class scale and summarized for each Italian province-a relevant policy scale for the Italian National Action Plan (NAP) to combat desertification. With the mean level of soil sensitivity rising over time almost everywhere in Italy, 'non-affected' land became more fragmented, the number of 'fragile' and 'critical' patches increased significantly, and the average patch size of both classes followed the same trend. Such dynamics resulted in intrinsically disordered landscapes, with (i) larger (and widely connected) 'critical' land patches, (ii) spatially diffused and convoluted 'fragile' land patches, and (iii) a more interspersed and heterogeneous matrix of 'non affected' land. Based on these results, we discussed the effects of increasing numbers and sizes of 'critical' patches in terms of land degradation. A sudden expansion of 'critical' land may determine negative environmental consequences since (i) the increasing number of these patches may trigger desertification risk and (ii) the buffering effect of neighboring, non-affected land is supposed to be less efficient, and this contains a downward spiral toward land degradation less effectively. Policy strategies proposed in the NAPs of affected countries are required to account more explicitly on the intrinsic, spatio-temporal evolution of 'critical' land patches in affected regions.