Why dryland biodiversity matters
Grasslands can support grazing, soil formation, nutrient cycling and carbon storage, but their ability to perform several tasks at once may change as landscapes become drier. A new study of 104 dryland sites examined how plant diversity relates to this combined performance along a gradient of aridity.
The study, published in npj Biodiversity on 12 October 2026, found that positive links between diversity and ecosystem multifunctionality generally became stronger with increasing dryness. However, the pattern weakened once the reported aridity level exceeded 0.83.
That qualification matters. The results do not mean that ever-increasing drought makes biodiversity more effective, or that the driest grasslands automatically gain the most from additional species. Rather, the researchers found that relationships change across environmental conditions.
The research addresses an important question for conservation: should managers prioritise the number of species present, or the differences in ecological traits among those species, when trying to protect multiple ecosystem functions?
What ecosystem multifunctionality means
An ecosystem performs many processes simultaneously. Plants produce biomass, soils retain nutrients, microorganisms transform organic matter and vegetation helps influence water movement. Researchers often combine measurements of several processes into a multifunctionality assessment.
A site can perform well on one measure but poorly on another. High plant production, for example, does not automatically imply equally strong nutrient retention or soil processes. Multifunctionality attempts to capture this broader picture.
Species diversity refers to the variety of species found in a community. Functional diversity instead describes how species differ in traits linked to their ecological roles, such as growth strategies, resource use and tolerance of environmental stress.
These measures overlap but are not interchangeable. Two grasslands may contain the same number of species while differing greatly in the range of strategies those plants employ.
Functional complementarity is one possible reason diversity can help. Plants with different traits may use light, water or soil nutrients in different ways, allowing a community to sustain more processes than a set of very similar species.
Yet the usefulness of that complementarity may depend on environmental conditions. When water becomes exceptionally scarce, survival may depend increasingly on whether a community retains species capable of tolerating severe stress.
How the 104-site analysis worked
Weiling Niu and colleagues examined diversity and ecosystem multifunctionality across 104 dryland sites. They assessed relationships involving species diversity, functional diversity and aridity, then compared how those relationships varied along the dryness gradient.
The research was observational. The authors studied patterns across existing sites rather than randomly assigning entire grasslands to different rainfall regimes or manipulating the species present in a large experiment.
That design offers a valuable comparison across environmental conditions, but it also means that correlations cannot by themselves establish the precise causal contribution of each diversity measure.
Aridity served as a central explanatory gradient. In general terms, higher aridity indicates a greater imbalance between atmospheric demand for water and the water available from precipitation.
The study’s reported threshold of 0.83 is an aridity-index value used in its analysis. It should not be read as a percentage of land lost, a rainfall forecast or a universal drought warning for individual farms.
Researchers compared how the strength of biodiversity and multifunctionality associations changed from relatively wetter drylands to drier settings. Their goal was to understand not merely whether diversity mattered, but when different aspects of diversity mattered most.
The authors also considered whether reduced multifunctionality under increasing aridity was associated with changes in plant diversity, providing insight into potential ecological pathways.
The strongest numerical findings
The headline sample size was 104 dryland sites. Across the gradient, positive relationships between biodiversity and ecosystem multifunctionality strengthened as conditions became drier, before weakening above an aridity level of 0.83.
This pattern suggests that biodiversity can be particularly relevant to maintaining multiple functions across some water-limited environments. It does not show that any individual ecosystem function improves as rainfall declines.
Indeed, the study found that increasing aridity reduced multifunctionality overall, primarily through suppression of plant diversity, particularly under drier conditions.
This is a critical distinction between two statistical statements. Diversity may become a stronger predictor of differences between sites even while the absolute level of ecosystem performance declines with increasing environmental stress.
The authors also found that functional diversity was more closely associated with multifunctionality under relatively wetter conditions. Under drier conditions, species diversity became the stronger predictor.
That switch is the main conceptual result. A broad range of plant traits may matter more where water availability permits different resource-use strategies to coexist. In harsher conditions, retaining enough stress-tolerant species may become especially important.
The publication’s abstract does not report a single universal percentage increase in ecosystem functioning per additional plant species. Assigning such a figure would misrepresent the results.
Likewise, the threshold of 0.83 should not be presented as an exact biological tipping point for every grassland. It describes a change in the relationships observed in this study’s dataset and modelling framework.
Why the relationships change with dryness
Under less severe water limitation, species with different functional traits may divide resources or operate at different times. This creates opportunities for complementarity, with diverse plant strategies collectively supporting ecosystem processes.
As aridity intensifies, environmental filtering becomes more influential. Species unable to cope with low water availability may disappear or become less abundant, leaving communities dominated by plants that tolerate drought.
Under these circumstances, the presence of several suitable species may become more informative than a simple measure of trait variation. Species diversity can reflect the retention of organisms capable of maintaining processes under stress.
The researchers interpret the shift as movement from stronger functional complementarity toward environmental filtering and retention of stress-tolerant taxa.
These are explanations supported by the observed patterns, not proof that one mechanism operates identically at every site. Other factors, including soils, grazing history and local climate, can influence grassland composition and functioning.
The result also challenges a simplistic interpretation of biodiversity protection. A management strategy that adds species without considering their suitability for a drying environment may not provide the intended ecological benefits.
Implications for restoration and conservation
Dryland restoration often involves decisions about which plants to protect, reintroduce or encourage. The findings suggest that those decisions should account for local water constraints rather than relying on a single diversity target across all climates.
In comparatively wetter grasslands, maintaining a variety of functional strategies may support several ecological processes. In drier sites, safeguarding a sufficiently diverse set of stress-tolerant species may be especially important.
This does not mean managers should ignore functional diversity in arid locations. Species identity, traits and richness remain connected, and local management still requires knowledge of the ecosystem being restored.
The work also highlights the potential cost of species losses. Where plant diversity is already being suppressed by dryness, further declines could remove ecological capacity that is difficult to replace.
For policymakers, the findings favour place-specific monitoring. Measuring plant counts alone may miss relevant functional changes, while measuring traits alone may understate the importance of retaining species in harsher environments.
Grasslands support livelihoods as well as biodiversity. In regions dependent on grazing and healthy soils, understanding how water stress alters ecological relationships can help frame more realistic restoration priorities.
However, the study did not experimentally evaluate a particular restoration programme, calculate the economic return on planting species or test management interventions on South African farms. Such claims would require separate evidence.
What the findings do not establish
The analysis involved 104 sites, not every dryland ecosystem on Earth. Its results should not be treated as a direct forecast for each biome, watershed or individual property.
Observational comparisons are vulnerable to unmeasured differences between locations. Soil properties, land use, grazing pressure, historical disturbances and climatic variability may shape both plant communities and measured ecosystem functions.
Multifunctionality is also a constructed research measure. Results can depend on which functions are selected, how they are standardised and how the combined score is calculated.
Functional diversity depends on the plant traits available for analysis. Different trait sets may capture different ecological mechanisms, and trait measurements do not necessarily encompass every interaction among species.
The findings describe relationships across an aridity gradient rather than following the same sites through decades of climate change. Spatial differences between places cannot automatically be equated with the future trajectory of one location.
The study’s authors do not claim that diversity can fully compensate for increasingly severe water scarcity. Their results indicate that drying can suppress biodiversity and multifunctionality even where the diversity-function relationship remains positive.
Future experiments could manipulate species richness and trait composition across controlled water treatments, while long-term monitoring could test whether the observed patterns persist as drylands change.
Research in additional regions could also examine how restoration, grazing and soil management interact with biodiversity under varying levels of drought.
The broader lesson
Related Research Today reporting includes a synthesis of biodiversity and ecosystem services, research on coastal mangrove distributions, and forest land-reform concerns in Mpumalanga. These address different ecosystems and questions, but illustrate why conservation decisions require local evidence.
The study adds nuance to the familiar statement that biodiversity supports healthy ecosystems. Diversity matters, but its contribution to multiple ecological functions can change as environmental conditions become more stressful.
At 104 dryland sites, the researchers observed stronger positive diversity-function relationships with increasing aridity up to a point, followed by weakening above 0.83. They also found a shift from functional diversity toward species diversity as the more informative predictor under drier conditions.
That combination suggests conservation should protect both the range of ecological strategies and the continued presence of plants suited to local water conditions.
The practical message is not that dry grasslands need a single ideal number of species. It is that effective protection and restoration depend on understanding which aspects of biodiversity support ecosystem functions in a changing climate.
Source Information
Title: Diversity–multifunctionality relationships shift with increasing aridity in grasslands
Authors: Weiling Niu, Jingyi Ding, Wenwu Zhao, Yanxu Liu, Shuai Wang, Changjia Li, Yan Li, Xutong Wu, David J. Eldridge and Bojie Fu
Journal: npj Biodiversity
Date: 12 October 2026
Design: Observational multi-site ecological analysis
Sample: 104 dryland sites
