Forest fragmentation—the division of large, intact woodlands into smaller patches—has emerged as a critical issue in 21st-century global land cover change, with far-reaching implications for ecosystem health. By increasing forest edges, fragmentation alters forest structure, ecosystem functions, and ecosystem services (the benefits to humans) that potentially reshape resilience. Yet whether this process weakens or strengthens resilience has been a long-standing debate.

On the one hand, environmental stress research suggests that fragmented edges make trees more vulnerable to heat and dryness, reducing resilience. On the other hand, species competition studies argue that fragmentation eases resource competition, favoring fast-growing early successional species that may boost resilience. Until recently, however, these conflicting theories had not been rigorously tested globally.

To test these theories, a research team led by scientists from the Research Center for Eco-Environmental Sciences of the Chinese Academy of Sciences, collaborating with international partners, analyzed satellite data to quantify fragmentation's impact on forest resilience worldwide.

Using satellite-derived tree cover and vegetation indices, the researchers measured fragmentation via edge density—the length of forest edges per unit area, with higher values indicating greater fragmentation. Forest resilience was assessed between 2000 and 2020 using a standard metric: the one-lag temporal autocorrelation (TAC) of the kernel normalized difference vegetation index (kNDVI), derived from Moderate Resolution Imaging Spectroradiometer (MODIS) satellite data. Higher TAC values signal lower resilience.

To isolate fragmentation's effects from tree cover, the researchers applied three methods—partial correlation analysis, partial least squares structural equation modeling (PLS-SEM), and statistical data binning—across two-degree by two-degree global grid cells.

Their findings, published in Nature Ecology & Evolution, revealed statistically significant (P < 0.05) fragmentation–resilience relationships in 77% of fragmented forests, with the effect on resilience varying by biome:

Tropical and temperate forests: Stronger fragmentation (higher edge density) correlated with weaker resilience (higher TAC), particularly in the Brazilian Amazon, Central Africa, and the northern U.S. This pattern stemmed from fragmentation's significant (P < 0.001) reduction of forests' cooling and humidifying effects, raising local temperatures and dryness.

Boreal forests: In contrast, greater fragmentation was linked to enhanced resilience (lower TAC). Here, fragmentation cooled local temperatures, increased soil moisture, reduced dryness, and improved sunlight availability—factors that bolstered resilience.

The study resolves conflicting hypotheses by demonstrating biome-specific impacts, emphasizing fragmentation's role in predicting ecosystem responses to disturbances. It also underscores the need for tailored forest management strategies to mitigate climate change effects.