Disturbance regimes and nutrient inputs are changing worldwide, with consequences for the structure and functioning of plant communities. Classical life-history theory predicts that disturbance should shift communities from long-lived perennials toward short-lived annuals, and that nutrient enrichment may amplify this shift. However, these predictions have not been tested experimentally across broad environmental gradients. Here, using a global coordinated grassland experiment spanning 37 sites, we tested how physical disturbance, vegetation removal and shallow soil tillage, and fertilisation reshape annual-perennial balance, and whether disturbance relaxes the climatic limits of annual dominance. Disturbance nearly doubled the proportion of annual species and more than doubled the relative cover of annuals, whereas fertilisation had little influence and did not interact with disturbance. The disturbance-driven shift arose through contrasting pathways: in graminoids and legumes, it reflected the loss of perennial cover, while in forbs, the expansion of annual cover. In the absence of disturbance, annual dominance was restricted to systems with extremely hot and dry summers, but disturbance nearly tripled the extent of climate space in which annuals dominated. By rapidly reassembling after disturbance, annuals may help maintain vegetation cover, but their expansion also signals loss of perennial cover and the long-term ecosystem functions associated with it.

Plant functional traits are fundamental to ecosystem dynamics and Earth system processes, but their global characterization is limited by available field surveys and trait measurements. Recent expansions in biodiversity data aggregation-including vegetation surveys, citizen science observations, and trait measurements-offer new opportunities to overcome these constraints. Here we demonstrate that combining these diverse data sources with high-resolution Earth observation data enables accurate modeling of key plant traits at up to 1 km2 resolution. Our approach achieves correlations up to 0.63 (15 of 31 traits exceeding 0.50) and improved spatial transferability, effectively bridging gaps in under-sampled regions. By capturing a broad range of traits with high spatial coverage, these maps can enhance understanding of plant community properties and ecosystem functioning, while serving as tools for modeling global biogeochemical processes and informing conservation efforts. Our framework highlights the power of crowdsourced biodiversity data in addressing longstanding extrapolation challenges in global plant trait modeling, with continued advancements in data collection and remote sensing poised to further refine trait-based understanding of the biosphere.

Introduction Forest restoration can be achieved by promoting natural regeneration or planting tree seedlings, but the relative benefits of these widely used approaches are questioned. Soil communities may influence restoration outcomes but are usually ignored by monitoring schemes.Objectives We investigated whether analyses of multitrophic soil communities and ecosystem functions can usefully complement conventional plant community measurements for assessing current and potential future states of naturally regenerating and planted native forests.Methods We used DNA metabarcoding to analyze multitrophic soil biodiversity alongside conventional plant community analyses in greater than 20-year-old forests of naturally regenerating and planted Leptospermum scoparium (m & amacr;nuka, Myrtaceae), an important early-successional tree native to Aotearoa-New Zealand, within two similar sites.Results Naturally regenerating forests had higher understory native plant cover and sapling wood density, and more saplings of species that dominate old-growth forests. Planted forests had higher aboveground biomass, richnesses of soil bacteria and protists, and soil pathogen loads. Prokaryote metabolic pathways indicated greater importance of aboveground litter versus belowground plant exudates as soil inputs in planted and naturally regenerating forests, respectively. Soil analyses detected legacy effects of pre-1860 M & amacr;ori land use in several plots.Conclusions Soil biodiversity assessment provided information about past, current, and potential future ecological states that was not apparent from the analysis of aboveground plants alone. Our combined results show that planted forests initially grow faster and sequester more carbon, but naturally regenerating forests might be healthier and able to shift to an old-growth forest state sooner.

Wood density is central for estimating vegetation carbon storage and a plant functional trait of great ecological and evolutionary importance. However, the global extent of wood density variation is unclear, especially at the intraspecific level. We assembled the most comprehensive wood density collection to date, including 109 626 records from 16 829 plant species across woody life forms and biomes (GWDD v.2, available here: doi: 10.5281/zenodo.16919509). Using the GWDD v.2, we explored the sources of wood density variation within individuals, within species and across environmental gradients. Intraspecific variation accounted for c. 15% of overall wood density variation (SD = 0.068 g cm-3). Variance was 50% smaller in sapwood than heartwood, and 30% smaller in branchwood than trunkwood. Individuals in extreme environments (dry, hot and acidic soils) had higher wood density than conspecifics elsewhere (+0.02 g cm-3, c. 4% of the mean). Intraspecific environmental effects strongly tracked interspecific patterns (r = 0.83) but were 70-80% smaller and varied considerably among taxa. Individual plant wood density was difficult to predict (root mean square error > 0.08 g cm-3; single-measurement R2 = 0.59). We recommend: (1) systematic sampling of multiple individuals and tissues for local applications; and (2) expanded taxonomic coverage combined with integrative models for robust estimates across ecological scales.

Wood density influences how quickly woody plants grow, how long they live and how much carbon they store, yet its global variation remains poorly mapped. Here we combined 109,626 wood density measurements from 16,829 species with 300,949 vegetation plots to produce a km-scale map of community-weighted wood density for every woody biome. Our model led to a prediction accuracy 32–51 % higher than previous global products, and a 1.8–3.7-fold wider wood density range (0.28–1.00 g cm−3; global mean: 0.57 g cm−3) than previously assumed. Spatial cross-validation showed low bias (±2.5 % of the mean), and uncertainties decreased from 20% in poorly sampled drylands and boreal regions to 5% in data-rich temperate forests. Mean annual temperature was the best predictor of community-weighted mean wood density, increasing by 0.01 g cm−3 for every 1°C change. We deliver a low-bias, high-resolution wood density layer for Earth system models, together with spatially explicit error maps. This study represents a major step forward for carbon accounting and trait-based forecasts of vegetation change. ### Competing Interest Statement The authors have declared no competing interest. European Research Council, 101001905 Agence Nationale de la Recherche, ANR-10-LABX-25-01, ;ANR-10-LABX-0041 Natural Environment Research Council, NE/S01537X/1, NE/M018458/1, NE/X013766/1, ;NOTALLOWED UK Research and Innovation, EP/Y003810/1 Leverhulme Trust, RPG-2020-341 Agencia Nacional de Investigación y Desarrollo, Anid-Fondecyt 1231025 São Paulo Research Foundation, 13/08722-5 Fondation Pour la Recherche Sur la Biodiversité, BIOSCENEMADA Smithsonian Tropical Research Institute, https://ror.org/035jbxr46 Ministerio de Ciencia, Innovación y Universidades, PID2021-123675OB-C43, TED2021-129770B-C21 Anusandhan National Research Foundation, SRG/2022/002286 SRM University, Andhra Pradesh, SRMAP/URG/E&PP/2022-23/012 Council of Scientific and Industrial Research, no.38(1349)/13/EMR-II National Council for Scientific and Technological Development, 12450/2021-4, 406941/2022-0, 307984/2022-2, PELD 441510/2020-5 Fundação de Amparo à Pesquisa do Estado do Amazonas, 01.02.016301.02529/2024-87 Brazilian Research Network on Climate Change, 01.13.0353-00 Science and Engineering Research Board, PDF/2021/003742/LS Deutsche Forschungsgemeinschaft, CRC TRR-228/1, CRS TRR-228/2 Fundação de Amparo à Ciência e Tecnologia de Pernambuco, https://ror.org/02te5rf52 Royal Society Te Apārangi, RDF-UOA1504, UOA1207 Natural Sciences and Engineering Research Council, https://ror.org/01h531d29 National Science Foundation, 2046768 National Institute of Amazonian Research, https://ror.org/01xe86309 Ministry of Business, Innovation and Employment Universidad de Panamá, https://ror.org/0070j0q91 Ecopetrol (Colombia), 135-2009 Dutch Research Council, NWO-VI.Veni.192.027 Ministerio de Economía, Comercio y Empresa, CGL2013-46808-R Ministerio de Ciencia, Tecnología e Innovación, https://ror.org/03fd5ne08, Convocatoria 860

The spread of invasive plant pathogens is on the rise globally, introducing additional threats to forest ecosystems and their constituent species above those already caused by climate change and other anthropogenic stressors. Austropuccinia psidii is a widespread invasive fungus that infects hundreds of species in the Myrtaceae plant family, causing myrtle rust. It was first detected in New Zealand in 2017, where the threatened endemic tree Lophomyrtus bullata is one of the worst affected hosts. Recent studies have described heavy disease severity on this species but environmental drivers of spatial variation in disease severity have not been explored, including the possible role of edge effects which are prevalent across much of this species' highly fragmented natural range. In this study, we tracked changes in disease incidence and severity across a forest edge gradient over three years. Individuals further from the forest edge, where forest canopies were more intact, forest structure more complex, and with greater understorey humidity, had greater disease severity in the first year of infection. In subsequent years, the influence of edge was no longer evident as myrtle rust severity became more even across the site, but disease was more prevalent in denser L. bullata populations. Only small individuals were asymptomatic across all three years. Since all available myrtle rust control techniques are currently feasible only at small scales, our research implies that their application needs to be as early as possible before infection spreads and across all suitable microclimates, and without bias toward easily accessible forest edges.

Conservation requires aligning the spatial, temporal, and functional scales of environmental management with those of interlinked ecological processes. To assist Indigenous Ngāi Tahu and government-mandated authorities in the management of black swans (kakī anau, Cygnus atratus) in the South Island of Aotearoa New Zealand, we investigated the effects of land cover on space use by swans and identified scale mismatches in contemporary gamebird and wetland management. We GPS-tracked 48 swans over one year and compared home range sizes with spatial scales of gamebird management. We also examined relative preferences of swans for different land-cover types to understand the potential for human-wildlife conflict, an influential component of gamebird and wetland management. Home ranges were typically located within a single waterbody or catchment, whereas gamebird management units each comprised numerous catchments. Within home ranges, swan location records coincided with aquatic vegetation, sandy shorelines, and flooded deciduous hardwoods. Contrary to common perceptions among farmers (primarily European New Zealanders) of swans as widespread pests, swans significantly avoided pasture, except near some shorelines and nesting colonies where wetlands have been converted to pasture. Regional-scale waterfowl management that does not account for habitat selection within home ranges (a spatial scale mismatch) could obscure patterns and determinants of population changes. Moreover, management geared towards suppressing waterfowl numbers in the interests of agricultural land uses that have encroached upon wetlands (a functional scale mismatch) could further degrade waterfowl populations. Mitigating these scale mismatches requires equitable governance arrangements that recognize Indigenous Peoples’ rights and values associated with the environment.

Floral resources are important food resources for pollinators. These resources are produced in different quantities depending on land cover and plant species composition, and the quantity of production varies seasonally. As such, land use change and management of natural resources can have substantial impacts on conservation through resource provision for pollinators, and also commercial enterprises through resources for honeybee hives which require adequate forage to be successful. In New Zealand, locations with vegetation that produce high-value honey also suffer from overcrowding of hives, as beekeepers compete for this valuable resource. At present, there is a lack of quantitative spatial data describing the production of these resources, especially over large spatial scales. Here, using maps of land cover and environment, and a large vegetation plot dataset, we show that the provision of floral resources for pollinators can be estimated spatially at national scales. These maps can be used to estimate the consequences of changing land cover, both historical and with future management actions, and to understand potential threats to floral resource provision. We find that the production of floral resources across New Zealand is highly seasonal, and overwhelmingly produced by indigenous land cover types, especially within public conservation land. Within forests, we show that floral production is dominated by a small number of plant families. Our results show the importance of native land cover for the provision of floral resources for commercial honeybee enterprises and also native pollinators. We anticipate our results will be a starting point to inform management decisions regarding the placement and stocking density of honeybee hives, and also the concession process for honeybee permits on public land. We also show how the restoration of woody ecosystems on cleared land can benefit the conservation of native pollinators by providing abundant and high-quality forage across all seasons.

Plant functional traits are fundamental to ecosystem dynamics and Earth system processes, but their global characterization is limited by the availability of field surveys and trait measurements. Recent expansions in biodiversity data aggregation, including large collections of vegetation surveys, citizen science observations, and trait measurements, offer new opportunities to overcome these constraints. Here we demonstrate that combining these diverse data sources with high-resolution Earth observation data enables accurate modeling of key plant traits at up to 1 km resolution. Our approach achieves high predictive power, reaching correlations up to 0.63 (15 of 31 traits exceeding 0.50) and improved spatial transferability, effectively bridging gaps in under-sampled regions. By capturing a broad range of traits with high spatial coverage, these maps can enhance our understanding of plant community properties and ecosystem functioning globally, and can serve as useful tools in modeling global biogeochemical processes and informing worldwide conservation efforts. Ultimately, our framework highlights the power and necessity of crowdsourced biodiversity data in high-resolution plant trait modeling. We anticipate that advancements in biodiversity data collection and remote sensing capabilities will further refine global trait mapping, fostering a dynamic trait-based understanding of the biosphere. ### Competing Interest Statement The authors have declared no competing interest.

Context Mapping the distribution of species from the genus Metrosideros is crucial for developing surveillance and management plans associated with species conservation in response to issues such as rapid '& omacr;hi'a death spread in the south-central Pacific region.Aims To support this endeavour, we recognised there was a need for open and reliable geographic information system data on island locations, extents, and occurrence data of Metrosideros species.Methods Using an open science framework, we reviewed six sources of island data and five sources of species occurrence data for availability, accuracy, and licencing criteria.Key results OpenStreetMap emerged as the optimal island location data, offering accuracy, precision, and open licencing, with this data improved and reprojected for mapping purposes. The Global Biodiversity Information Facility provided the majority of Metrosideros species occurrence data, but analysis of occurrence data from iNaturalist revealed common mis-identifications with regional biases that were corrected prior to compilation. The occurrence data of Metrosideros species was also supplemented by vegetation plot data, with HAVPlot and sPlotOpen providing key additional data for some species and islands.Conclusions Citizen science data via iNaturalist and OpenStreetMap formed the core of the compiled datasets. While such crowdsourced data can have quality issues, with additional crowdsourced curatorial effort these datasets will be significant and scalable sources of data into the future.Implications All compiled occurrence and GIS data are made openly available via permissive data licences to better support future biogeographical research in the south-central Pacific region.

Although disturbance is considered a major driver of plant invasions across many systems, our understanding of the mechanisms by which disturbance mediates understorey invasions in natural forests is limited. We used a national natural forest inventory dataset spanning New Zealand's wide climatic and soil fertility gradients to disentangle disturbance-mediated community characteristics driving abundance, species richness and functional composition of understorey plant invasions. Disturbance-mediated declines in canopy tree abundance and increases in soil fertility and pH increased non-native plant richness and cover relative to co-occurring native plant assemblages. Cover of non-native species also increased with proximity to anthropogenic land cover. Non-native plant assemblages had higher community-weighted mean (CWM) values for specific leaf area (SLA) but lower CWM values for height and woodiness relative to native assemblages irrespective of disturbance. However, greater nearby anthropogenic land cover drove increased woodiness in non-native assemblages but decreased woodiness in co-occurring native assemblages. Synthesis: Our study provides the first national-scale evidence that canopy disturbance effects on soil properties increase both richness and abundance of non-native plants in natural forest understories. We also revealed functional trait differences between native and non-native assemblages (SLA in particular), which could alter fundamental forest ecosystem processes like litter decomposition and nutrient cycling. Finally, landscape-scale anthropogenic impacts may exacerbate forest invasions by increasing non-native abundance and favouring woody invaders, which may achieve dominance in future forest communities over the longer-term. This study provides evidence that canopy disturbance effects on forest community structure and soil properties increase the richness and abundance of non-native plants in natural forest understories. The authors identified functional trait differences, particularly in specific leaf area, between native and non-native assemblages, potentially altering key ecosystem processes like litter decomposition and nutrient cycling. Additionally, this study reveals that landscape-scale anthropogenic impacts may exacerbate forest invasions, favouring woody invaders and potentially leading to their dominance in future forest communities.image

The density of wood is a key indicator of the carbon investment strategies of trees, impacting productivity and carbon storage. Despite its importance, the global variation in wood density and its environmental controls remain poorly understood, preventing accurate predictions of global forest carbon stocks. Here we analyse information from 1.1million forest inventory plots alongside wood density data from 10,703 tree species to create a spatially explicit understanding of the global wood density distribution and its drivers. Our findings reveal a pronounced latitudinal gradient, with wood in tropical forests being up to 30% denser than that in boreal forests. In both angiosperms and gymnosperms, hydrothermal conditions represented by annual mean temperature and soil moisture emerged as the primary factors influencing the variation in wood density globally. This indicates similar environmental filters and evolutionary adaptations among distinct plant groups, underscoring the essential role of abiotic factors in determining wood density in forest ecosystems. Additionally, our study highlights the prominent role of disturbance, such as human modification and fire risk, in influencing wood density at more local scales. Factoring in the spatial variation of wood density notably changes the estimates of forest carbon stocks, leading to differences of up to 21% within biomes. Therefore, our research contributes to a deeper understanding of terrestrial biomass distribution and how environmental changes and disturbances impact forest ecosystems.

We compared carbon fluxes at locations differing in land-use history and forest restoration strategies, i.e., planted and naturally regenerating forests. We (1) quantified the carbon (C) input from litterfall, (2) measured total soil respiration and its component fluxes, and (3) identified the abiotic and biotic factors influencing C input by litterfall and C release from the soil. We established four plots each in planted and naturally regenerating Leptospermum scoparium (mānuka; Myrtaceae) forests of similar age in warm, temperate New Zealand. Litterfall and total soil respiration were measured over one year. Organic layer and mineral soil samples were analysed for physical, chemical, and biological characteristics. Annual litterfall C input in planted forests on former agricultural land was more than double that of naturally regenerating forests, which was partly explained by higher soil nutrient availability. Annual mean soil respiration was 28

Aim: Woody ecosystems provide critical ecosystem functions and services but are increasingly threatened as invasive pathogens spread globally. Myrtle rust, caused by Austropuccinia psidii, arrived in New Zealand in 2017 and infects at least 12 of 18 species in the susceptible Myrtaceae plant family. Among these are species of structural, successional and cultural importance. We aim to assess whether the functional consequences of Myrtaceae loss could be mitigated if co-occurring species with shared functional attributes are able to replace them. Location: New Zealand (but with concepts and methodologies that apply globally). Methods: Using a nationwide forest and shrubland plot data set, we assessed community vulnerability to the loss of Myrtaceae species by analysing proportional changes in average trait values when they are absent and produced spatial predictions indicating where species loss might have the greatest impact on community functionality. We then assessed whether compensatory infilling by co-occurring species would mediate community vulnerability. Results: Forests and shrublands containing Kunzea ericoides and Leptospermum scoparium are highly vulnerable to their loss. Areas most vulnerable overall are the central and south-eastern North Island, north-eastern South Island and Stewart Island. For all species, compensatory infilling moderated the impact of their loss. However, if co-occurring Myrtaceae were unable to respond, possibly if they were also infected, community vulnerability almost always increased because infilling species had different functional attributes, compounding the functional impact. Main Conclusions: Early successional woody plant communities and Myrtaceae-dominated old-growth forests are at most risk. Our spatial assessment of species-level functional impacts from myrtle rust will facilitate better-informed landscape-level responses. Management actions and monitoring can now be targeted to areas and communities at greatest risk of losing ecosystem-level processes.

Interactions between and within abiotic and biotic processes generate nonadditive density-dependent effects on species performance that can vary in strength or direction across environments. If ignored, nonadditivities can lead to inaccurate predictions of species responses to environmental and compositional changes. While there are increasing empirical efforts to test the constancy of pairwise biotic interactions along environmental and compositional gradients, few assess both simultaneously. Using a nationwide forest inventory that spans broad ambient temperature and moisture gradients throughout New Zealand, we address this gap by analyzing the diameter growth of six focal tree species as a function of neighbor densities and climate, as well as neighbor x climate and neighbor x neighbor statistical interactions. The most complex model featuring all interaction terms had the highest predictive accuracy. Compared with climate variables, biotic interactions typically had stronger effects on diameter growth, especially when subjected to nonadditivities from local climatic conditions and the density of intermediary species. Furthermore, statistically strong (or weak) nonadditivities could be biologically irrelevant (or significant) depending on whether a species pair typically interacted under average or more extreme conditions. Our study highlights the importance of considering both the statistical potential and the biological relevance of nonadditive biotic interactions when assessing species performance under global change.

Ecosourcing seed of 'local genetic stock' for ecological restoration has been practiced in New Zealand for about 50 years. However, we believe that it has become unnecessarily restrictive. Ecosourcing ensures plants used for restoration are adapted to local conditions and maintains current distributional patterns. It also restricts genetic diversity, confines species to their historic range, and reduces the conservation options for threatened species. For example, New Zealand tree species, the life form most frequently used in restoration plantings, have low population genetic differentiation and high net migration of alleles throughout their range. Therefore, very little is gained through restrictive ecosourcing of tree seed. Furthermore, avoidance of the danger of inbreeding depression and widening the scope for closer environmental matching, argues for larger rather smaller source areas. Climate change, extinctions across multiple trophic levels, habitat loss and fragmentation, spread of invasive species, and novel habitats have completely altered the contemporary biotic landscape. Conservation needs to engage with these changes if it is to protect and restore ecosystems. Restrictive ecosourcing is counter-productive as it limits utilising genotypic, phenotypic and ecotypic diversity, and thus the evolutionary potential of indigenous species and ecosystems. It also reduces opportunities to protect biodiversity when populations are small, and limits response to climate change. A new approach is needed. We recommend that phylogeographic patterns and biogeographic boundaries be used to set nine broad ecosourcing regions and, within these regions, phenotypic adaptation to particular environments be used as a guide to seed selection. This more relaxed approach to ecosourcing will improve restoration outcomes through increasing species and genetic diversity, reducing the detrimental effects of inbreeding and promoting the genetic rescue of populations of threatened species. Examples of adopting an eco-evolutionary approach to ecosourcing are provided for the early-successional coloniser Kunzea ericoides and late-successional conifer species.

In vascular plant systematics there are sometimes conflicts between phenotypic and ecotypic variation and genetic differentiation that challenge species concepts, introduce taxonomic confusion, and create nomenclatural uncertainty. Until a 2014 taxonomic revision that segregated Kunzea ericoides into 10 species, it and K. sinclairii were the only species recognised in New Zealand. A recent DNA microsatellite study failed to support any of the new species, instead revealing biogeographic variation. Here we present the results of a genotyping by sequencing study with 1,361 single nucleotide polymorphisms (SNPs), sampling 48 populations representing four Kunzea species from South Island and southern North Island. The SNP study confirms the microsatellite findings: the two widespread species, K. robusta and K. serotina, are indistinguishable and share northern and southern genotypes with other species; a single metapopulation lineage reflects a national north-to-south clinal pattern; and population differentiation is low and net migration high. A significant isolation by distance pattern was revealed with SNPs. The 2014 revision was explicitly based on the unified species concept, but the primary criterion, that each species represents a separate metapopulation lineage, was not demonstrated. Species recognition was based on morphological and ecological criteria that have proved difficult to apply. Applying the unified species concept and the primary criterion of a single metapopulation genetic lineage, we now recognise just a single New Zealand species, K. ericoides, with other species constituting taxonomic synonyms. In doing so, we distinguish a grey zone of taxonomic uncertainty that reflects incomplete lineage sorting, gene flow coupled with a lack of reproductive isolation, and only partial ecotypic and phenotypic differentiation. As demonstrated in the Kunzea revision, there is considerable phenotypic and ecotypic variation in regional populations that is likely to be of ecological and conservation importance. We suggest informal ecotypes are a better way to recognise this level of variation.

The widely observed negative scaling relationship between organism size and abundance is predicted to have a universal -0.75 scaling exponent across all life forms. However, factors influencing frequently observed deviations from this exponent, such as ecosystem succession and organism traits, remain poorly understood. We explore the dependence of size-density scaling on ecosystem succession and organism traits by analysing size-density relationships in trees and soil invertebrates across 183 temperate forest plots comprising urban secondary forests, urban old-growth forests and non-urban natural forests. Exponents of scaling relationships in urban tree and invertebrate communities progressively steepened with increasing restored (planted) forest stand age as small organisms increased in abundance. In contrast, non-urban tree scaling relationships flattened during succession with exponents veering away from -0.75, whereas urban tree and invertebrate communities converged towards this prediction in later successional stages. Our results shed light on how the body size structure of tree and soil invertebrate communities spanning multiple trophic levels shift over successional time as the relative abundances of large versus small-bodied organisms increase. This study emphasises the fundamental influence of organismal traits and ecosystem succession on scaling relationships of organism body size and abundance.Read the free Plain Language Summary for this article on the Journal blog. Read the free Plain Language Summary for this article on the Journal blog.image

1. Species performance in the realised niche is jointly shaped by both abiotic and biotic processes. Moreover, interactions between and within abiotic and biotic processes generate non-additivities, resulting in density dependence that varies in strength or even direction across environments. If ignored, these non-additivities can lead to inaccurate predictions of species responses to changes in environment and community composition. 2. There are increasing empirical efforts to test the constancy of pairwise biotic interactions along environmental and compositional gradients, but rarely along both. We address this gap using nationwide forest inventory data that span broad ambient temperature and moisture gradients throughout New Zealand. 3. We analysed tree diameter growth of six focal tree species as a function of neighbour densities and climate, while accounting for potential abiotic and biotic non-additivities arising from neighbour <m:math xmlns:m="http://www.w3.org/1998/Math/MathML" alttext="\times" display="inline"><m:mo>×</m:mo></m:math> climate and neighbour <m:math xmlns:m="http://www.w3.org/1998/Math/MathML" alttext="\times" display="inline"><m:mo>×</m:mo></m:math> neighbour statistical interactions, respectively. We kept the large number of parameters manageable using Bayesian shrinkage priors and interpretable using average predictive comparisons. 4. We found that the most complex model—featuring biotic interactions that changed with climate and higher-order interactions with intermediary species—had the highest predictive accuracy of tree diameter growth. Compared to climate, biotic interactions typically had stronger effects on tree diameter growth, especially when they were subjected to non-additivities from local climate and the density of a third species. Most non-additivities tended to weakly exacerbate pairwise competition, whereas the few strong non-additivities tended to alleviate pairwise competition or even produce pairwise facilitation. 5. Synthesis: Our study highlights the importance of the interplay between abiotic and biotic processes when predicting how biotic interactions may structure communities under global change. When quantifying the relative importance of biotic and abiotic processes on species performance, we show that the conclusion varies depending on whether we are looking at direct or indirect effects. With accumulating evidence of non-additive biotic interactions, the next crucial step is to uncover their underlying mechanisms.

Evolutionary radiations of woody taxa within arid environments were made possible by multiple trait innovations including deep roots and embolism-resistant xylem, but little is known about how these traits have coevolved across the phylogeny of woody plants or how they jointly influence the distribution of species. We synthesized global trait and vegetation plot datasets to examine how rooting depth and xylem vulnerability across 188 woody plant species interact with aridity, precipitation seasonality, and water table depth to influence species occurrence probabilities across all biomes. Xylem resistance to embolism and rooting depth are independent woody plant traits that do not exhibit an interspecific trade-off. Resistant xylem and deep roots increase occurrence probabilities in arid, seasonal climates over deep water tables. Resistant xylem and shallow roots increase occurrence probabilities in arid, nonseasonal climates over deep water tables. Vulnerable xylem and deep roots increase occurrence probabilities in arid, nonseasonal climates over shallow water tables. Lastly, vulnerable xylem and shallow roots increase occurrence probabilities in humid climates. Each combination of trait values optimizes occurrence probabilities in unique environmental conditions. Responses of deeply rooted vegetation may be buffered if evaporative demand changes faster than water table depth under climate change.