This study was never meant to be about disturbance. It started as a climate change project.
The original plan seemed straightforward enough. We had access to one of the largest vegetation dataset in the world and wanted to see whether plant populations were already responding to recent climate change. The dataset even contained abundance estimates, which opened the possibility of looking beyond simple range shifts (and their lags) and into changes in population size. I spent a lottt of time developing a fairly complicated modelling framework to do exactly that.
The models failed.
At first, I assumed it was a methodological problem. Then I checked the underlying data. Then I checked the models again. Then I checked the data again. Nothing quite made sense. Eventually, after abandoning the abundance component and focusing on species ranges, things improved. But another problem emerged—most of the climate-based model were still performing surprisingly poorly.
How could this be?
For years, species distribution models (SDMs) have been built around a simple idea: climate largely determines where species occur. When species fail to occupy climatically suitable areas, we often describe this as disequilibrium. Most discussions of disequilibrium focus on climate change, where species lag behind shifting climatic conditions because dispersal, colonisation, and local extinction take time.
That was the story we expected to find.
But hold up. What if climate was only part of the story?
Around the same time, our group was increasingly working on evidence suggesting that Europe’s ecological past looked rather different from the landscapes many of us imagine today. Recent palaeecological studies have challenged this traditional image of a continent dominated by primeval closed-canopy forests, instead pointing towards dynamic landscapes composed of open habitats, light woodland, and denser forest maintained by herbivores, fire, and other disturbances. The work by Elena Pearce et al. (2023) and Szymon Czyżewski et al. (2026) in particular, suggest that disturbance was not simply an occasional event acting upon Europe’s landscape, it was a fundamental process shaping them.
That idea kept nagging at me.
If disturbance helps determine the structure of vegetation, it must also help determine where species occur. Climate might tell us where species could occur, but disturbance regimes might influence where they actually do occur.
So, we tested it.
Using over 1.1 million vegetation plots and SDMs for 3,047 European plant species, we assessed two different types of climate disequilibrium. The first was climate-change disequilibrium, asking whether species distributions were better predicted by past climatic conditions than by standard decadal averages. The second was disturbance-driven disequilibrium, asking whether accounting for disturbance-related habitat indicators improved our ability to predict where species occur.
The results were striking. We found evidence of climate-change disequilibrium in roughly half of all species examined. Species distributions were often better predicted by climatic conditions from 10–18 years before sampling, suggesting that many species are still responding to past rather than present climates. Importantly, these estimates are likely conservative, constrained by the temporal depth of available climate data and vegetation records and the practical realities of modelling thousands of species across an entire continent.
But the bigger surprise came from disturbance. Almost every species we examined—99% of them—showed evidence of disturbance-driven disequilibrium.
This does not mean disturbance is a problem, nor do we imply that European ecosystems should somehow be in equilibrium. In many ways, the opposite may be true. Disturbance regimes generate heterogeneity across space and time. Grazing pressure shifts. Canopy openings appear and disappear. Fire, browsing, and succession constantly reshape habitats. Disequilibrium may therefore be an inherent feature of Europe’s dynamic landscape rather than an exception or deviation from a “norm”.
What surprised us most was where this signal appeared.
We initially expected disturbance-driven disequilibrium to be strongest among species associated with open habitats. Given the growing evidence for historically more open and disturbance-maintained landscapes in Europe, this seemed like a reasonable prediction.
Instead, the strongest disequilibrium occurred at both ends of the disturbance-related gradients.
Species associated with highly open environments showed strong mismatches between climatic suitability and realised distributions. But so did species associated with more shaded, closed-canopy environments. The weakest disequilibrium occurred among species occupying intermediate conditions. This result points towards a story that isn’t one of open versus closed habitats. Instead, it suggests that much of Euope’s flora is shaped by the dynamic disturbance mosaics that generate a shifting spectrum of habitat conditions. To that end, species specialised towards either end of disturbance gradients appear particularly sensitive to how disturbance structures landscapes.
This has important implications for how we think about restoration and conservation. When we talk about restoring ecosystems, it often focuses on restoring habitats. But habitats are ultimately products of ecological processes. If disturbance regimes play such a pervasive role in shaping where species occur, then conservation much consider also the processes that create and maintain them. Rewilding is one possible approach, particularly where it restores natural herbivory and ecological dynamics, but it is far from the only one. More broadly, the challenge is how to maintain or recreate the spatial and temporal heterogeneity that many species appear to depend on.
Perhaps the most important take away is that disequilibrium is not solely a consequence of climate change. Climate change is undoubtedly creating new mismatches and pushing species out of equilibrium. But many species already exist within landscapes structured by disturbance-driven dynamics. Ignoring that reality risks oversimplifying both how species distributions are formed and how ecosystems function.
Climate tells us where species could occur, but disturbance helps determine where they actually do occur. Understanding both is essential if we want to make sense of biodiversity responses in a rapidly changing world.
-Photo by author, Sean E. H. Pang, capturing a semi-managed system in Porto existing (presumably) at disequilibrium