Ecologists have long recognized that biodiversity can play a critical role in stabilizing the functioning of ecosystems over extended periods. However, it remains unclear whether the insights gleaned from small-scale studies can be extended to real-world ecosystems and applied at broader spatial scales. This knowledge gap poses a significant challenge to the application of diversity-stability research, particularly in light of ongoing environmental changes occurring at regional scales and the need for ecosystem management strategies that operate at landscape scales. To address this knowledge gap, we propose a unified mathematical partitioning approach that can be used to investigate the spatial scaling pattern of ecosystem stability (1/CV) and its relationship with that of biodiversity, for example, the stability-area relationships (StAR) and the species-area relationships (SAR). By using a large dataset of terrestrial plants collected from 33 sites of the National Ecological Observatory Network (NEON) between 2013 and 2020, we also estimated the log-log slopes of both SAR and StAR, as well as their empirical relationships and potential driving forces.
Our mathematical approach demonstrated that the stability–area relationship to quantify the scaling of ecosystem stability can be partitioned into spatial scaling of stability and asynchrony at the population level. Among 33 NEON sites, our analysis revealed that both plant species diversity and ecosystem stability exhibited a positive relationship with study area, following power law scaling. We firmly confirmed that the scaling exponent of ecosystem stability with area was primarily driven by the spatial scaling of population asynchrony, rather than population stability. Additionally, we found that sites in warm and humid conditions exhibited fast-increasing ecosystem stability with the area due to a faster increase of population asynchrony, instead of population stability, by a larger species–area accumulation rate. In summary, our study provided a new mathematical framework and empirical evidence to disentangle the underlying mechanisms of the diversity-stability relationships across spatial scales, with significant implications for ecological conservation and management in real landscapes.