Tree cover change proves stability and instability in tropical ecosystems
Terrestrial tropical ecosystems’ resilience is determined predominantly based on space-for-time substitution, which assumes that the current ‘static’ frequency distribution of ecosystems’ tree cover structure across space also holds across time. However, dynamic and temporal aspects are increasingly important to explicitly account for under ongoing rapid climate change. Here, we empirically study ecosystem stability and instability using remote sensing-derived tree cover change (ΔTC) over the last two decades. We find that considerable ΔTC predominantly takes place in intermediate tree cover ecosystems (i.e., areas with 30-60% tree cover), whereas high (>75%) and low (<10%) tree cover ecosystems only experience limited ΔTC. Our results further suggest that root zone storage capacity, which defines the adaptive capacity of the ecosystem to absorb water stress perturbations, does mediate the relationship between ecosystems’ stability and ΔTC by instigating investment in ecosystems subsoil structure. Based on these analyses, we propose a modified forest resilience metric using both precipitation and root zone storage capacity, which reveals that the Congo rainforests are more resilient than if only precipitation is considered. This study emphasises the importance of temporal dynamics and adaptation of ecosystems in inferring and assessing the risk of forest-savannah transitions under change.