Tropical forest canopies are the biosphere's most concentrated atmospheric interface for carbon, water and energy1,2. However, in most Earth System Models, the diverse and heterogeneous tropical forest biome is represented as a largely uniform ecosystem with either a singular or a small number of fixed canopy ecophysiological properties3. This situation arises, in part, from a lack of understanding about how and why the functional properties of tropical forest canopies vary geographically4. Here, by combining field-collected data from more than 1,800 vegetation plots and tree traits with satellite remote-sensing, terrain, climate and soil data, we predict variation across 13 morphological, structural and chemical functional traits of trees, and use this to compute and map the functional diversity of tropical forests. Our findings reveal that the tropical Americas, Africa and Asia tend to occupy different portions of the total functional trait space available across tropical forests. Tropical American forests are predicted to have 40% greater functional richness than tropical African and Asian forests. Meanwhile, African forests have the highest functional divergence-32% and 7% higher than that of tropical American and Asian forests, respectively. An uncertainty analysis highlights priority regions for further data collection, which would refine and improve these maps. Our predictions represent a ground-based and remotely enabled global analysis of how and why the functional traits of tropical forest canopies vary across space.

Remote sensing is an important tool for monitoring species habitat spatially and temporally. Species distribution models (SDM) often rely on remotely-sensed geospatial datasets to predict probability of occurrence and infer habitat preferences. Lidar measurements from the Global Ecosystem Dynamics Investigation (GEDI) are shedding light on three dimensional forest structure in regions of the world where this aspect of species habitat has previously been poorly quantified. Here we combine a large camera trap dataset of mammal species in Borneo and Sumatra with a diverse set of geospatial data to predict the probability of occurrence of 47 species. Multi-temporal GEDI predictors were created through fusion with Landsat time series, extending back to the year 2001. The availability of these GEDI-based forest structure predictors and other temporally-resolved predictor variables enabled temporal matching of species occurrences and hindcast predictions of species probability of occurrence at years 2001 and 2021. Our GEDI-Landsat fusion approach worked well for forest structure metrics related to canopy height (relative height of the 95th percentile of returned energy R2 = 0.62 and relative RMSE = 41%) but, not surprisingly, was less accurate for metrics related to interior canopy vegetation structure (e.g., plant area volume density from 0 to 5 m above the ground R2 = 0.05 and relative RMSE = 85%). For the SDM analyses, we tested several combinations of predictor sets and found that when considering a large pool of multiscale predictors, the exact composition, and whether GEDI Fusion predictors were included, didn’t have a large impact on generalized linear modeling (GLM) and Random Forest (RF) model performance. Adding GEDI Fusion predictors to a baseline set only meaningfully improved performance for some species (n = 4 for RF and n = 3 for GLM). However, when GEDI Fusion predictors were used in a smaller predictor set that is more suitable for hindcasting species probability of occurrence, more SDMs showed meaningful performance improvements relative to the baseline model (n = 9 for RF and n = 4 for GLM) and the relative importance of GEDI-based canopy structure predictors increased relative to when they were combined with the baseline predictor set. Moreover, as we examined predictor importance and partial dependence, the utility of GEDI Fusion predictors in hindcast models was evident in regards to ecological interpretability. We produced a catalog of probability of occurrence maps for all 47 mammals species at 90 m spatial resolution for years 2001 and 2021, enabling subsequent ecological interpretation and conservation analyses.

Plant-pollinator interactions structure ecological communities and represent a key component of ecosystem functioning. Pollination networks are expected to be more diverse and specialised in the tropics, but pollination ecology in these regions has been understudied in comparison to other areas. We reviewed research on pollination in the tropical Andes, one of the major biodiversity hotspots on Earth, where the uplift of mountains and past climate have resulted in spatiotemporally distinct species interactions. We found 1010 scientific articles on pollination in the Andes, of which 473 included or were carried out in tropical regions. The number of publications on pollination ecology in the tropical Andes has increased exponentially, with Colombia having the most articles, followed by Ecuador and Peru, and with Bolivia and Venezuela having notably fewer studies. More research has been carried out in humid montane forests and agricultural landscapes, and it has predominantly focused on describing diversity of species and interactions while neglecting analyses on the resilience and adaptability of pollinating systems, even though the Andean region is particularly susceptible to the effects of climate change and continues to undergo land conversion and degradation. Remarkably few studies have incorporated local knowledge, thus ignoring connections to human livelihoods and communities. A phytocentric perspective has been predominant, with fewer studies focusing directly on pollinators and a notable lack of articles with a holistic approach to the study of pollination across taxonomic groups at the community or ecosystem level. We propose that future research adopts a cross-scale approach that considers the complexity of the ecological contexts in which plant-pollinator interactions occur, and incorporates long-term monitoring with broader multilayer networks and molecular tools, experiments focused on ecophysiology and behaviour, animal telemetry, process-modelling approaches and participatory science. A stronger field driven by interdisciplinary collaborations will contribute to knowledge about pollination at a global scale, as well as increase our understanding of the diversity and resilience of pollination interactions in this region, thus improving our capacity to predict and avoid ecosystem collapses.

Understanding how the traits of lineages are related to diversification is key for elucidating the origin of variation in species richness. Here, we test whether traits are related to species richness among lineages of trees from all major biogeographical settings of the lowland wet tropics. We explore whether variation in mortality rate, breeding system and maximum diameter are related to species richness, either directly or via associations with range size, among 463 genera that contain wet tropical forest trees. For Amazonian genera, we also explore whether traits are related to species richness via variation among genera in mean species-level range size. Lineages with higher mortality rates—faster life-history strategies—have larger ranges in all biogeographic settings and have higher mean species-level range sizes in Amazonia. These lineages also have smaller maximum diameters and, in the Americas, contain dioecious species. In turn, lineages with greater overall range size have higher species richness. Our results show that fast life-history strategies influence species richness in all biogeographic settings because lineages with these ecological strategies have greater range sizes. These links suggest that dispersal has been a key process in the evolution of the tropical forest flora.

Traditional orangutan distribution and density monitoring requires costly line transect methods on the ground to detect their nests. Recently researchers have started to use unoccupied aerial vehicles, hereafter referred to as drones, to collect such data faster. However, manually inspecting the images acquired by the drone is time-consuming and hence costly. This study explored a deep learning method for the automated detection of orangutan nests in drone-captured aerial images, which can significantly improve the efficiency of orangutan monitoring efforts. The YOLO v10 model was trained using 868 images containing 1568 annotated orangutan nests collected from sites in Sabah, Malaysia, and Sumatra, Indonesia. Images were captured using multirotor and fixed-wing drones at varying altitudes. The model was trained using a transfer learning approach and achieved a mean Average Precision (mAP) of 0.831. The model was subsequently tested on two independent data sets with results showing a precision of 0.98 and recall of 0.88 for a multirotor drone and precision of 0.98 and a recall of 0.71 for a fixed-wing drone which has the benefit of being able to have longer duration flights. The high precision values indicate the model's accuracy in identifying true nest locations, while the recall values demonstrate its ability to detect a significant portion of the nests present in the images. The study highlights how using drones for data collection can reduce survey times compared to ground surveys, and the automation of nest detection further enhances the efficiency of drone surveys. However, the model's recall, especially for fixed-wing drone data, could be improved to ensure accurate population trend analyses. Further research should focus on expanding training data sets and refining models to account for different camera systems and environmental conditions.

AimArbuscular mycorrhizas (AM) and ectomycorrhizas (ECM) have different impacts on nutrient cycling, carbon storage, community dynamics and enhancement of photosynthesis by rising CO2. Recent global analyses have concluded that patterns of AM/ECM dominance in forests worldwide are shaped by climate, with soil nutrients contributing negligible additional explanatory power. However, their reliance on nutrient data from GIS surfaces masks important local influences of parent material, topography and soil age on soil nutrient status. We asked if use of site-specific nutrient data reveals a more important role for nutrients.Time PeriodPresent day.LocationGlobal dataset comprising 703 sites, encompassing forests, savanna/woodlands, shrublands and deserts on all continents except Antarctica.Taxa StudiedArborescent plants, including angiosperms, gymnosperms and tree ferns.MethodsGeneralised Additive Models for Location, Scale and Shape (GAMLSS) to determine the effects of climate variables, soil nitrogen and soil phosphorus on the proportional representation of ECM and of non-mycorrhizal species (NM) in woody vegetation.ResultsGAMLSS showed a strong negative relationship of ECM representation with mean annual temperature (MAT), and a strong negative relationship with soil total nitrogen. NM representation was highest on dry sites and phosphorus-poor sites. Reanalysis showed that GIS-derived soil nutrient data had less explanatory power than site-specific nutrient data, and resulted in poorer model fits.ConclusionsOur results support the long-held belief that soil nutrients as well as climate influence the relative fitness of different mycorrhizal syndromes worldwide, and demonstrate the value of using site-specific nutrient data. Soil nutrients should be considered when predicting the impact of climate change on the mycorrhizal composition of vegetation and resulting shifts in ecosystem processes.

Respiration by trees stems constitutes a substantial proportion of autotrophic respiration in forested ecosystems and has been estimated to contribute 12 – 25 % of total ecosystem respiration, yet little is known about its associated drivers at different spatial scales. Stems are the largest contributor to forest biomass and so the respiratory consumption of stems has the potential to considerably affect carbon budgets in forest communities. As logged and degraded forests are fast becoming the most dominant land-use type throughout the tropics, it is also important to contextualise stem respiration over land use gradients. In this study we quantified stem respiration at individual tree and plot scales in nine 1-ha plots over a gradient of heavily logged to old-growth forest in Malaysian Borneo. We investigated how logging intensity, forest structure, plant functional traits, and soil chemistry influence stem respiration in logged and old-growth forest plots at both scales. We found that, at individual tree level, stem respiration rate per unit stem area was significantly higher in logged than old-growth plots, and this was consistent within most diameter classes. At the 1-ha plot scale, however, total stem respiration did not differ between forest types: the higher stem respiration rate in logged plots was offset by the higher stem area in old-growth plots. At plot level, stem respiration was driven by forest structure and soil chemistry. We found that basal area was a strong predictor of stem respiration within both forest types at plot scale; for a similar basal area, logged plots exhibited a higher stem respiration rate. Partitioning stem respiration into its growth and maintenance components at plot scale highlighted how logged plots prioritise growth in response to intense light competition, as logged plots had significantly higher allocation to growth respiration, whereas old-growth plots prioritised maintenance and cell structure. Our analysis at individual tree scale reinforced these differing priorities, as stem respiration in logged plots was driven by plant traits associated with growth and wood anatomy, as opposed to within old-growth plots where stem respiration was driven by traits associated cell structure and maintenance. These results reflect the different strategies of resource allocation for trees growing in logged and old-growth plots and adds to the growing body of research on autotrophic respiration, the least studied component of forest carbon dynamics within a very understudied yet expanding land use.

Abstract Managing invasive non‐native species is a global challenge, especially for long‐lived trees like Ligustrum lucidum, known for its detrimental effects on invaded ecosystems. Using individual‐based models (IBMs), we simulated different control methods on the population dynamics and range expansion propensity of the established population. Across different sets of simulations, we varied the number of life stages and sites targeted. We additionally investigated how changing the management strategy over time affected outcomes. Controlling all life stages was essential to contain the expansion of L. lucidum. Removing both reproductive and non‐reproductive stages was more than twice as effective as removing either saplings or reproductive stages only, especially if a high number of sites were targeted every year. The method of selecting sites for removal within the population was important if only saplings were removed; in which case targeting the most recently colonized sites was the most effective strategy. Finally, a strategy that switches after 5 years from controlling all stages to focusing exclusively on early life stages could be effective at reducing both the total population size and the area occupied. Practical implication: This approach could be valuable when the availability of long‐term resources for control is limited. The ability of IBMs to simulate various scenarios and assess outcomes at population and landscape levels enhances their utility for predicting invasive non‐native species management success. It can be a solution to reducing the time and cost of fieldwork, helping to identify potential limitations of control actions.

Tea is a globally popular heritage beverage consumed by over three billion people. The unique taste and health benefits of tea are linked to its nutrient composition and antioxidant activity (AOA). As a plant species, tea's nutrient elements and AOA vary based on season, altitude, clone type and leaf age. This study examined the nutrient composition and AOA of young and mature tea leaves from four clones (BC1248, TRI2024, AT53 and TV9) grown at different altitudes under tropical conditions in Malaysia. The results demonstrated that altitude and clone type significantly influenced (p < 0.05) foliar nutrient elements and AOA. Young tea leaves have higher nutrient levels and AOA than mature leaves across all clones. Interestingly, the foliar nutrient availability was higher in the highlands, although the variation across the four clones was insignificant (p > 0.05). On the other hand, foliar nutrient elements varied significantly among lowland tea clones, except for N and Ca. The highest AOA was recorded in young tea leaves of clone BC1248 at the lowland plantations, with total polyphenol contents (TPC), 2,2-diphenyl-1-picrylhydrazylradical (DPPH IC50), and ferric reducing antioxidant power (FRAP) values of 19.60 +/- 0.15 mg GAE/g, 50.70 +/- 1.86 mu g/mL, 2.10 +/- 0.14 mM Fe (II)/g, respectively. The DPPH IC50 and FRAP varied significantly (p < 0.05), except for TPC among the lowland and highland clones. Based on principal component analysis (PCA), we identified that the tropical lowlands of Malaysia were more suitable for growing tea with high AOA. These findings provide valuable insights for growers to develop sustainable tea farming strategies, ensuring optimal yield and targeted quality under tropical conditions.

The impacts of degradation and deforestation on tropical forests are poorly understood, particularly at landscape scales. We present an extensive ecosystem analysis of the impacts of logging and conversion of tropical forest to oil palm from a large-scale study in Borneo, synthesizing responses from 82 variables categorized into four ecological levels spanning a broad suite of ecosystem properties: (i) structure and environment, (ii) species traits, (iii) biodiversity, and (iv) ecosystem functions. Responses were highly heterogeneous and often complex and nonlinear. Variables that were directly impacted by the physical process of timber extraction, such as soil structure, were sensitive to even moderate amounts of logging, whereas measures of biodiversity and ecosystem functioning were generally resilient to logging but more affected by conversion to oil palm plantation.

Spatial heterogeneity in tropical forest productivity and resulting rates of carbon uptake and storage emerge from variation in ecosystem structure and functional traits reflecting differences in climate, edaphic conditions, evolutionary history, and natural and anthropogenic disturbance histories. Yet, models poorly represent this heterogeneity. Remote sensing data provide landscape-scale measures of tropical forest heterogeneity in structure and functional traits that can be used to advance terrestrial biosphere models. To examine whether forest functional traits related to photosynthetic capacity can be used to improve predictions of tropical biomass dynamics and carbon fluxes, we parameterized the Ecosystem Demography model version 2.2 (ED2.2) using canopy traits derived from visible to shortwave infrared (VSWIR) airborne imaging spectroscopy data across an edaphic gradient in Borneo. We find significant site-level differences in relationships between SLA and foliar nutrient concentrations, suggesting that remotely sensed foliar traits can be used to capture variation in photosynthetic capacity at large, edaphically varying spatial scales. We further show that plant functional types parameterized with site-constrained trait values yield more accurate predictions of canopy demography, forest productivity and above-ground biomass dynamics than simulations that depend solely on parameterization of edaphic conditions. However, the most substantial improvements result from allowing for site-level variation in background disturbance rates in the model. Our study reveals the importance of capturing tropical forest heterogeneity in terrestrial biosphere models, particularly as it relates to nutrient availability and disturbance processes.

Demographic advantages over native species are critical for alien tree species to establish and persist in subtropical forests. Yet, we lack an appropriate quantification of these demographic advantages for trees invading forest ecosystems, partly because of the paucity of time series of demographic data that extend beyond the lag phase of invasions. We built Bayesian hierarchical population dynamic models to quantify the changing role of the demographic traits (relative abundance, survival, growth, and recruitment) of alien and native trees in explaining tree invasions through succession in subtropical montane forests in northwest Argentina. We parameterised our model using long-term (30 years) census data for four alien tree species and 21 native tree species in forests undergoing secondary succession on abandoned agricultural land. Among the alien trees, Ligustrum lucidum had the highest rates of survival and growth coupled with high recruitment, and emerged as a late successional invader. Persea americana remained rare throughout our study, while Morus alba and Prunus persica acted as pioneers, with demographic traits that fell within or below the range of variation of native species. The latter two alien species established initially, but declined in abundance later in the succession. Our results indicate that to thrive in the canopy, invasive alien trees should exceed the typical values of survival, growth and/or recruitment for native species, but the specific combination of these traits that is important to invasiveness varies among species and through succession. Alien species that lack any combination of suitable traits do not persist in closed-canopy forests.

Changes in landscape patterns, which refer to the composition and spatial configuration of land use and land cover (LULC) classes in a landscape, can have negative impacts on biodiversity and environmental processes such as carbon cycles. Such impacts are both dependent on the spatial extent of changes and which LULC classes are affected, but previous global-scale landscape pattern assessments have focused on single LULC classes or landscape-level measurements only. A comprehensive, multiscale analysis across multiple LULC types is therefore key for understanding the full impact of landscape pattern change on the environment. We assessed global-scale change in landscape patterns for six LULC classes from the HILDA+ dataset (urban, cropland, pasture/rangeland, forest, unmanaged grass/shrubland, and sparse/no vegetation) between 1992 and 2020. Six class-level landscape metrics with predictable scaling behaviour across landscape extents were calculated at global scale for each LULC class and year. Landscape metrics were quantified for five landscape extents (100, 400, 1600, 6400 and 25,600 km2). Trends in landscape metrics were evaluated and linked to changes in LULC composition (area) and configuration over time. Unmanaged grass/shrubland LULC expanded in area and showed increased number of patches, edge length, and complexity in shapes, while pasture/rangeland and forest LULC tended to decline in area, number of patches, and edge length. Even though there was high spatial heterogeneity in landscape pattern change for all LULC classes, neighbouring 100 km2 landscapes often showed the same directional change in area and fragmentation. Global landscape pattern change was highly variable for all LULC classes between 1992 and 2020, suggesting that drivers of LULC change act on local to regional scales. We expect that the multiscale global dataset of landscape metrics generated here will have future applications in understanding the drivers of landscape pattern change and its environmental impacts.

Tropical forests can vary enormously in their 3D structure and dynamics across surprisingly small spatial scales. However, the drivers that underpin this local-scale variation in forest structure and dynamics remain poorly understood. We acquired repeat airborne laser scanning data across an old-growth tropical forest landscape in Malaysian Borneo, characterized by a steep gradient in soil fertility and topography that gives rise to large variability in canopy 3D structure. Using this unique dataset, we explored how local-scale variation in topography and forest structure shapes rates of gap formation, closure, and canopy growth across the landscape. We found that both canopy gains and losses were 2.5-4.7 times greater in low-lying alluvial forests on fertile soils than in nearby nutrient-depleted kerangas forests on hilltops. Moreover, we found that variation in canopy 3D structure and dynamics was tightly coupled across the landscape, with taller and more structurally heterogeneous canopies also experiencing faster rates of gap dynamics. Our study highlights the key role that soils and topography play in shaping the structural complexity and dynamics of tropical forest landscapes.

Remote sensing is a powerful tool for characterizing ecosystems at large scales. However, the relative importance of leaf traits and canopy structure in characterizing the spatial distribution of functionally distinct tropical forests – the most diverse, structurally complex, and heterogeneous ecosystems on Earth – remains under-explored. Using satellite-resolution LiDAR and imaging spectroscopy metrics, we map spatial turnover in tropical forest function, examine the relative importance of leaf traits and canopy structure, and analyze differences in aboveground carbon and demography. We find that leaf phosphorus, LMA, and canopy height are key distinguishing properties of forest types, achieving accuracies of 85-96% and correspond to differences in community growth and mortality rates. Our remotely sensed forest types align with ground-based forest definitions but enable mapping of their entire extent. At 30 m resolution, our method can be used at large scales with spaceborne data to reveal important differences in structure and function across tropical forests.

Comparing the genetic diversity across different generations within tropical tree populations is an understudied topic. To assess the potential genetic consequences and conservation implications of contemporary disturbances, a population genomic study of Palaquium obovatum across age classes was undertaken. Trees and juveniles were sampled from ten different localities (eight in Singapore, two in Peninsular Malaysia) and subjected to double digest restriction-site associated DNA-sequencing (ddRAD-seq) to assess intergenerational genetic differences and investigate population structure in a hexaploid lineage. Genetic erosion, characterised by reduced heterozygosity, was found to have occurred in almost all wild populations over time, the exceptions being in one isolated coastal population and some areas with cultivated occurences. Population structure was highly localised with the number of genetically distinct populations usually following geographically separated districts, which indicates limitations in pollen and seed dispersal between fragments, possibly due to declines in the associated assemblage of dispersers. For this reason, the germplasm for conserving species diversity in degraded habitats and forest fragments should be selected from a wide range of wild populations across the landscape.

The functional stability of ecosystems depends greatly on interspecific differences in responses to environmental perturbation. However, responses to perturbation are not necessarily invariant among populations of the same species, so intraspecific variation in responses might also contribute. Such inter-population response diversity has recently been shown to occur spatially across species ranges, but we lack estimates of the extent to which individual populations across an entire community might have perturbation responses that vary through time. We assess this using 524 taxa that have been repeatedly surveyed for the effects of tropical forest logging at a focal landscape in Sabah, Malaysia. Just 39 % of taxa – all with non-significant responses to forest degradation – had invariant responses. All other taxa (61 %) showed significantly different responses to the same forest degradation gradient across surveys, with 6 % of taxa responding to forest degradation in opposite directions across multiple surveys. Individual surveys had low power (< 80 %) to determine the correct direction of response to forest degradation for one-fifth of all taxa. Recurrent rounds of logging disturbance increased the prevalence of intra-population response diversity, while uncontrollable environmental variation and/or turnover of intraspecific phenotypes generated variable responses in at least 44 % of taxa. Our results show that the responses of individual species to local environmental perturbations are remarkably flexible, likely providing an unrealised boost to the stability of disturbed habitats such as logged tropical forests.### Competing Interest StatementThe authors have declared no competing interest.

We present an annotated checklist of the 29 species comprising Syzygium subg. Sequestratum. Species in the subgenus were identified through a comprehensive review of herbarium specimens and relevant literature, including both historic taxonomic and regional revisions of Syzygium, and recent molecular phylogenetic analyses. No single character is diagnostic of subg. Sequestratum; instead, a combination of commonly occurring characters is used, including glaucescence in the hypanthium, funnel-shaped flowers <10 mm long and coriaceous leaves with dark drying petioles in contrast to a paler lamina. Species are distributed across East Asia, Malesia and Eastern Australia, with an apparent preference for nutrient-poor habitats. This checklist provides a foundation for future taxonomic revision of this clade, from an early branching node in the tree of life of the world’s largest tropical tree genus.