Empirical analysis of vegetation dynamics and the possibility of a catastrophic desertification transition

Abstract

The process of desertification in the semi-arid climatic zone is considered by many as a catastrophic regime shift, since the positive feedback of vegetation density on growth rates yields a system that admits alternative steady states. Some support to this idea comes from the analysis of static patterns, where peaks of the vegetation density histogram were associated with these alternative states. Here we present a large-scale empirical study of vegetation dynamics, aimed at identifying and quantifying directly the effects of positive feedback. To do that, we have analyzed vegetation density across 2.5 × 106 km2 of the African Sahel region, with spatial resolution of 30 × 30 meters, using three consecutive snapshots. The results are mixed. The local vegetation density (measured at a single pixel) moves towards the average of the corresponding rainfall line, indicating a purely negative feedback. On the other hand, the chance of spatial clusters (of many “green” pixels) to expand in the next census is growing with their size, suggesting some positive feedback. We show that these apparently contradicting results emerge naturally in a model with positive feedback and strong demographic stochasticity, a model that allows for a catastrophic shift only in a certain range of parameters. Static patterns, like the double peak in the histogram of vegetation density, are shown to vary between censuses, with no apparent correlation with the actual dynamical features. Our work emphasizes the importance of dynamic response patterns as indicators of the state of the system, while the usefulness of static modality features appears to be quite limited.

Clustering and correlations: Inferring resilience from spatial patterns in ecosystems

“…With the increasing availability of high‐resolution spatial datasets, from satellites to drone‐based imagery, of various ecosystems, spatial analyses are likely to be widely deployed in the future. Such data will enable us to quantify not only patterns, as described above, but also dynamics of clusters and correlations (Manor & Shnerb, ; Van Belzen et al, ; Weissmann, Kent, Michael, & Shnerb, ; Weissmann & Shnerb, ). Our study reveals that naive association of observed scale‐free behaviours with either criticality or stability can be misleading.…”

Spatial synchrony in vegetation response

“…For T = 13 we used the Sahel data (EVI index) taken in 2002 and 2015, with 30 × 30 meters resolution. A complete description of these datasets was presented in [20].…”

“…For the Sahel we can study another dataset that was already analyzed (in different context) in [20]. Here we compare the EVI index for 2002 and 2015 (T = 13), in 30 × 30 meters per pixel resolution.…”

Alternative Vegetation States in Tropical Forests and Savannas: The Search for Consistent Signals in Diverse Remote Sensing Data

“…More interesting in the context of this paper, however, is the degree to which spatial variability in a landscape is (a) proportional to the frequency, intensity, and time since disturbance (e.g., a harvest event that removes a fraction of tree cover in a particular location) and subsequent regrowth ("linear patch dynamics"), or is (b) amplified by positive feedbacks that lead to alternative state dynamics in response to disturbance events ("bifurcating patch dynamics"). The classic example of an amplifying feedback producing bifurcating patch dynamics (between forest and savannas in mesic systems, and savanna and grasslands in drier savannas) is the so-called fire-trap, where an initial loss (or gain) of tree cover promotes (or suppresses) grass growth, providing more (or less) fuel for fires and increased (or decreased) tree mortality in a continuing cycle of tree loss (or gain) [16][17][18][19][20][21][22][23].…”

“…Numerous studies have examined satellite observations in search of empirical support for the prevalence of alternative vegetation states [18,19,[24][25][26][27] in spatial and temporal data [23,28]. The majority of studies that presented evidence for alternate states were based on interpretation of frequency distributions of tree cover over large spatial regions with similar mean annual rainfall (implicitly assuming similar local edaphic and ecological conditions).…”