Search for a command to run...
Conservation biology rests on a strong foundational principle of ecology, that there is generally a positive relationship between biodiversity and ecosystem function. In short, diverse systems do more essential “stuff”, for example carbon sequestration, nutrient cycling, and biomass production, than depauperate systems do. A rich literature supports this principle and describes the ubiquity and strength of biodiversity-ecosystem function relationships (BEFs) across varied ecosystems, geographic regions, scales, and categories of functions (Cardinale et al. 2012). Biodiversity-function relationships are fundamental properties of ecological systems and understanding them helps us operationalize some of the ways that biodiversity matters and name some of the things that it does. Ecologists often wonder if we have many, or even any, generalizable laws, but we are at least confident that the natural world does many wonderful things, and it does them better when there are more different kinds of organisms working together. But natural systems experience stressors, like heat, drought, and pollution. In a world in which stress is becoming more frequent and extreme and causes biodiversity loss (Côté et al. 2016; Díaz et al. 2019), knowledge of BEFs helps us predict that biodiversity loss leads to declines in ecosystem function. Unfortunately, this knowledge is not enough. It is not enough because most, if not all, ecosystems will experience multiple stressors simultaneously, a combination of two or more “pushes” against the status quo. We should therefore expect multiple stressors to be the norm rather than an exception. Understanding the effects of multiple stressors on ecosystem function is more nuanced than one might initially assume, however. It makes sense that the addition of multiple stressors would cause additional species loss and therefore push ecosystems farther toward lower biodiversity and concomitantly reduced function, such as less carbon fixed and less biomass produced (Hooper et al. 2012). But a critical question is whether multiple stressors also change the shape of the BEF relationship. In other words, even when stressors do not cause biodiversity loss, does the presence of multiple stressors make ecological communities less efficient and less able to do the work of performing those things we call ecosystem functions? Answering this question is the focus and key innovation of Schäfer et al. (2025). In their meta-analysis of freshwater experiments, Schäfer et al. compiled experimental studies of aquatic ecosystem function conducted under different stressors, and that included measurements of taxonomic richness. A key challenge in these studies is that stressor intensity and the number of stressors may each be confounded with biodiversity: more stressors and greater stressor intensity lead to more biodiversity loss, which itself leads to greater loss of function. By including both richness loss and the number of stressors in their analyses, however, the authors were able to hold richness constant across changes in the number of stressors, allowing them to isolate the effect of the number of stressors on ecosystem function. This approach to disentangling the number of stressors and stressor intensity is important and novel. Further, because of the positive nature of the biodiversity-function relationship, we often assume that loss of function is caused by loss of biodiversity. But Schäfer et al.'s approach opens the door to gathering empirical evidence for the insight that adding stressors changes the very nature of the BEF relationship itself, leading to a loss of functional efficiency, even in high-diversity systems (Rillig et al. 2023). Changes in the shape of the biodiversity-function relationship are precisely what Schäfer et al. document. Across the studies in their analysis, biodiversity loss (measured as richness loss) led to declines in ecosystem function, as would be predicted from positive BEF relationships. The BEF relationships, however, were different in systems experiencing multiple stressors: they were more strongly positive. This critical observation has two important implications. First, at any given level of richness loss, systems experiencing, say, three distinct stressors will experience greater relative reductions in ecosystem function than a system experiencing a single stressor and the same level of richness loss. Second, for systems experiencing multiple stressors, any additional loss in biodiversity will cause a more dramatic reduction in function than would be seen in a single-stressor system. In short, because the presence of multiple stressors changes the shape of the BEF, tilting it more steeply, ecosystem function is far more sensitive to changes in biodiversity when multiple stressors are present. In a world in which stressors are more common, ecosystems and the services they provide are even more sensitive to biodiversity loss. Schäfer et al.'s results are narrow in some senses. Their study focuses strictly on experimental freshwater studies with many of the restrictions typical of rigorous meta-analyses, which require apples-to-apples comparisons. But the results also offer a rare amount of generality. Importantly, neither stressor identity, nor high-level taxonomic identity (e.g., “algae” or “zooplankton”), contributed importantly to the primary result of a steepening of the biodiversity-function relationship as the number of stressors increased. This means that the negative effects on BEF of adding more stressors were not because it became more likely that particularly “bad” stressors were now included, nor that some organismal groups were obviously more sensitive to stressors than others. Based on these results, when more stressors accumulate in an ecosystem, we should expect an increasingly negative effect on ecosystem function, regardless of which stressors are added. The relatively small range (one, two, or three) of stressors and coarse categorization of stressors and organisms in Schäfer et al. (2025) each leave room for future exploration. Further, biodiversity-ecosystem function relationships should be expected to vary in shape for other reasons, for example across ecosystems (freshwater vs. marine vs. terrestrial) and thus may respond differently to the number of stressors. But the lack of obvious contingency in a famously contingent field is encouraging and makes Schäfer et al.'s findings unusually impactful. Schäfer et al.'s recent work adds clarity to the ecological literature, which is replete with examples of the biodiversity-ecosystem function relationship (Tilman et al. 2014). This body of work has been critical in compiling a comprehensive empirical case that biodiversity matters to the existence and functioning of ecosystems and the services they provide. Schäfer et al.'s results advance this line of inquiry by helping us see how BEF research matters to what matters: understanding how the natural world works, what that means for how ecosystems and their functions are changing, and how these new details align, or not, with modern conservation goals and values. Yes, biodiversity matters, but by showing that the biodiversity-function relationship changes depending on the number of stressors, Schafer et al. also show how strongly it matters in different contexts. This key result also represents an important new focus for BEF research: identifying and understanding where, when, and why biodiversity-function relationships become steeper, and therefore more sensitive to biodiversity loss. Perhaps most critically, the “bigness” of Schäfer et al.'s results may help us cut through some of the necessary abstraction inherent to BEF and modern ecological research. In the plainest terms possible: For a long time we have known that biodiversity loss was bad for nature and for the services that nature provides (Daily 1997). We now know that the loss of function caused by biodiversity loss is even greater when ecosystems experience multiple forms of ecological stress. And because multiple forms of stress are a reality of the modern ecological world, we can now see more clearly that biodiversity loss has been degrading ecosystem function even faster than we may have previously predicted. Kevin G. Smith: conceptualization, writing – review and editing. The author declares no conflicts of interest. Data sharing not applicable to this article as no datasets were generated or analysed during the current study.