Higher-order interactions enhance the latitudinal tree diversity gradient

Nature · 2026-04-29

Abstract The global decrease in species diversity from low to high latitudes is among the most robust biogeographic patterns 1,2 . There is continuing debate on the contribution of conspecific negative density dependence (CNDD) to the latitudinal diversity gradient evident for trees 3,4 . Theory suggests that CNDD based on pairwise interactions alone is not sufficient to explain the intricacies of diverse communities, because higher-order interactions (HOIs) may greatly modify these interactions 5,6 . However, there has been a lack of empirical studies investigating how HOIs intertwine with pairwise interactions and how they may contribute to the latitudinal tree diversity gradient. Here we examined both pairwise interactions and HOIs across 32 large permanent forest plots, most in the northern hemisphere. We detected evidence of HOIs in 40% of the 1,543 species–plot combinations for tree growth, and 23% of the 1,340 such combinations for tree survival, with the strength of these interactions declining with latitude. HOIs were found to benefit rare species but disadvantage common species, suggesting a potential mechanism promoting species diversity. This stabilizing effect weakened towards higher latitudes, consistent with the latitudinal tree diversity gradient. Our findings reveal an important interplay between pairwise interactions and HOIs in promoting the latitudinal tree diversity gradient and help to clarify the contribution of CNDD to this biogeographic pattern.

Continental Contrasts in Climate Extremes That Control Tree Fecundity

Global Change Biology · 2026-02-01 · 1 citations

In 2023, more than half of olive harvests (Olea europaea) across Spain, Greece, and Türkiye were lost to drought. The same year late freeze destroyed 90% of the peach crop (Prunus persica) on the Georgia Piedmont and the apple crop (Malus domestica) in central New York, Vermont, and southern Quebec. Climate extremes now rank with the costliest threats to agriculture, but their role in forest recovery from diebacks that are happening globally is unknown for lack of tree fecundity estimates in forests. Tolerance of climate extremes could depend on past exposure but constrained by phylogenetic conservatism. We report a continental scale analysis of climate extremes and forest fecundity across North America and Europe showing that responses to late freeze and drought are happening now. Species differences are not explained by the traits typically included in ecological studies and they are weakly associated with phylogeny. Late freeze, that is, freezing temperatures that follow the onset of flower development in spring, is shown to be "normal" in North America, but not Europe, potentially explaining failed seed production due to delayed onset and the resultant shorter growing period by North American transplants dating back at least to the 18th century. Drought has thus far had the greatest impacts in dry forested regions, but here too, species differences are not explained by traditional trait values. If responses have been buffered from drought and late freeze by past exposure, acclimation and local adaptation prove inadequate as extremes intensify.

NeRF-Accelerated Ecological Monitoring in Mixed-Evergreen Redwood Forest

Forests · 2025-01-17 · 3 citations

Forest mapping provides critical observational data needed to understand the dynamics of forest environments. Notably, tree diameter at breast height (DBH) is a metric used to estimate forest biomass and carbon dioxide (CO2) sequestration. Manual methods of forest mapping are labor intensive and time consuming, a bottleneck for large-scale mapping efforts. Automated mapping relies on acquiring dense forest reconstructions, typically in the form of point clouds. Terrestrial laser scanning (TLS) and mobile laser scanning (MLS) generate point clouds using expensive LiDAR sensing and have been used successfully to estimate tree diameter. Neural radiance fields (NeRFs) are an emergent technology enabling photorealistic, vision-based reconstruction by training a neural network on a sparse set of input views. In this paper, we present a comparison of MLS and NeRF forest reconstructions for the purpose of trunk diameter estimation in a mixed-evergreen Redwood forest. In addition, we propose an improved DBH-estimation method using convex-hull modeling. Using this approach, we achieved 1.68 cm RMSE (2.81%), which consistently outperformed standard cylinder modeling approaches.

Hidden decay of live trees in a tropical rain forest

Ecology · 2025-09-01

The trunks of forest trees store massive amounts of carbon, but fungi actively and invisibly decay wood inside even seemingly healthy trees. Wood-decay fungi are responsible for the loss of stored carbon in living trees, and they make trees susceptible to snapping and uprooting in storms. We used sonic tomography to measure the prevalence and severity of decay in 1744 live trees (≥20 cm diameter) of 171 species on the 50-ha Forest Dynamics Plot on Barro Colorado Island, Panama. A median of <2% of the cross-sectional trunk area showed decay, but 15% of trees had >20% decay. Twenty percent of the combined basal area showed decay, representing a loss of approximately 1% of aboveground biomass. Larger trees more often showed internal decay, with one quarter of trees showing decay before reaching canopy height. Decay severity varied by species; 23% of species showed <2% decay while 9% of species lost over half their basal area. Rare species were more affected than locally abundant species, and species with traits associated with a fast life history were more susceptible to decay. These results suggest that hidden wood decay affects a large proportion of living tropical forest trees.

Machine learning vs. empirical models: Estimating leaf wetness patterns in a wildland landscape for plant disease management

Agricultural and Forest Meteorology · 2025-01-13 · 4 citations

• We compared machine learning and empirical models of leaf wetness duration. • Simple empirical leaf wetness models perform as well as advanced techniques. • Leaf wetness informs wild plant disease dynamics across a coastal climate gradient. This study presents the development and application of models to estimate leaf wetness duration and their integration with drone-based imagery to analyze plant disease patterns across a coastal gradient. By comparing machine learning algorithms with empirical models, we identified that both approaches effectively predict leaf wetness, particularly in a temperate maritime ecosystem. The models were applied to study two manzanita species ( Arctostaphylos tomentosa and A. pumila ), revealing a strong correlation between leaf wetness and disease prevalence. This work highlights the role of microclimate conditions in shaping plant health and disease distribution in coastal shrublands. We compared nine popular machine learning algorithms and four empirical threshold models to characterize leaf wetness patterns in a spatially diverse temperate maritime wildland ecosystem. We suggest that simple empirical leaf wetness models based on dew point depression or relative humidity thresholds perform as well as machine learning techniques and should not be overlooked. The relationship between leaf wetness duration and the spatial distribution of plant disease along a coastal-to-inland climate gradient offers valuable insights into disease dynamics.

A phylogenetic epidemiology approach to predicting the establishment of multi-host plant pests

Communications Biology · 2025-01-24 · 3 citations

Abstract Forecasting emergent pest spread is paramount to mitigating their impacts. For host-specialized pests, epidemiological models of spread through a single host population are well developed. However, most pests attack multiple host species; the challenge is predicting which communities are most vulnerable to infestation. Here, we develop a phylogenetically-informed approach to predict establishment of emergent multi-host pests across heterogeneous landscapes. We model a beetle-pathogen symbiotic complex on trees, introduced from Southeast Asia to California. The phyloEpi model for likelihood of establishment was predicted from the phylogenetic composition of woody species in the invaded community and the influence of temperature on beetle reproduction. Plant communities dominated by close relatives of known epidemiologically critical hosts were four times more likely to become infested than communities with more distantly related species. Where microclimate favored beetle reproduction, pest establishment was greater than expected based only on species composition. We applied this phyloEpi model to predict infestation risk in California using weather data and complete tree inventories from 9262 1-km 2 grids in 170 cities. Regions in the state predicted with low likelihood of infestation were confirmed by independent monitoring. Analysts can adapt these phylogenetic ecology tools to predict spread of any multi-host pest in novel habitats.

Data and scripts for: Airborne DNA reveals predictable spatial and seasonal dynamics of fungi

Zenodo (CERN European Organization for Nuclear Research) · 2024-03-30 · 1 citations

Fungi are among the most diverse and ecologically important kingdoms of life. However, the distributional ranges of fungi remain largely unknown, as do the ecological mechanisms that shape their distributions. To provide an integrated view of the spatial and seasonal dynamics of fungi, we implemented a globally distributed standardised aerial sampling of fungal spores. The vast majority of OTUs were detected only within one climatic zone, and the spatio-temporal patterns of species richness and community composition were mostly explained by annual mean air temperature. Tropical regions hosted the highest fungal diversity except for lichenized, ericoid mycorrhizal, and ectomycorrhizal fungi, which reached their peak diversity in temperate regions. The sensitivity in climatic responses was associated with phylogenetic relatedness, suggesting that large-scale distributions of some fungal groups are partially constrained by their ancestral niche. There was a strong phylogenetic signal in seasonal sensitivity, suggesting that some groups of fungi have retained their ancestral trait of sporulating only for a short period. Overall, our results show that the hyperdiverse kingdom of fungi follows globally highly predictable spatial and temporal dynamics, with seasonality in both species richness and community composition increasing with latitude. Our study reports patterns resembling those described for other major groups of organisms, thus making a major contribution to the long-standing debate on whether organisms with microbial lifestyles follow the global biodiversity paradigms known for macro-organisms. The analyses presented in the paper can be reproduced with the R-script pipeline provided here. The starting point for the scripts is the datafile allData.RData that was published originally by Ovaskainen et al. Data from: Global Spore Sampling Project: A global standardized dataset of airborne fungal DNA. https://doi.org/10.5281/zenodo.10435615 (2024). The datafile allData.RData is provided also here for convenience, and it includes the following three objects: metadata, taxonomy, and otu.table (see Ovaskainen et al. for details). The script pipeline consists of the following elements (for deltails, see the Methods of the paper): Scripts S01: data preparation S01.1_download_clim_data.R. This script downloads daily climatic data for the entire world. S01.2_select_and_preprocess_clim_data.R. This script selects the data relevant for the study locations and preprocesses it. S01.3_add_climatic_data_to_metadata.R. This script adds the preprocessed climatic data to the metadata. S01.4_otu_guild_assignment.R. This script performs the guild assignment to the OTUs. It utilizes the datafiles Fung_LifeStyle_Data.RDS and funguild_db.rds provided here, and it utilizes the taxonomy of ProtaxFungi provided by Ovaskainen et al. Data from: Global Spore Sampling Project: A global standardized dataset of airborne fungal DNA. https://doi.org/10.5281/zenodo.10435615 (2024). Note that while the paper presents analyses and results only for the trait database of Aguilar-Trugueros et al., the scipts repeat the trait analyses also for the FunGuild database. The reason for not showing the results for the FunGuild database in the paper was that the database of Aguilar-Trugueros et al. contains FunGuild as one of the data sources, and that the results were highly coherent between the two databases. S01.5_add_trait_data_to_taxonomy_and_metadata.R. This script adds the guild data and spore size data to taxonomy (taxon-specific traits) as well as to metadata (community-weighted mean traits). It utilizes the datafile Spore_data_12Nov21.RDS provided here. This script can also be used to generated simulated contamination to the OTU table by setting contaminate=TRUE. Scripts S02: exploratory analyses S02.1_show_descriptive_statistics.R. This script outputs basic desriptive statistics from the data. S02.2_make_study_design_maps.R. This script plots the study design map shown in the paper. S02.3_compute_site_and_biome_profiles.R. This script computes site_profiles (needed in ordinations) and biome_profiles (needed to create Venn diagrams). S02.4_make_venns.R. This script produces Venn diagrams. Scripts S03: ordination analyses S03.1_make_ordination_maps.R. This script makes the ordination analyses. Scripts S04: univariate analyses S04.1_conceptualize_univariate_models.R. This script produces a figure that illustrates conceptually the differenent model variants. S04.2_make_univariate_analysis.R. This script implements the univariate analyses. S04.3_show_univariate_results.R. This script summarizes the results of the univariate analyses by producing tables of AIC and R2. S04.4_plot_univariate_results.R. This script plots the univariate models. S04.5_compute_temporal_turnover.R. This script computes site-specific indices of temporal turnover. S04.6_show_temporal_turnover.R. This script generates a plot illustrating temporal turnover. Scripts S05: Hmsc analyses S05.1_define_Hmsc_models.R. This script defines the Hmsc models. It utilizes the R-function as.phylo.formula provided here. S05.2_export_Hmsc_models_for_fitting.R. This script exports the unfitted Hmsc-models for fitting with Hmsc-HPC that operates on python/tensorflow. S05.3_import_fitted_Hmsc_models.R. This script imports the fitted Hmsc-models back to Hmsc-R. S05.4_postprocess_Hmsc_results.R. This script postprocesses the results of the fitted Hmsc model. S05.5_show_Hmsc_results.R. This script generates a plot that illustrates the postprocessed results.

Major axes of variation in tree demography across global forests

Ecography · 2024-05-06 · 3 citations

The future trajectory of global forests is closely intertwined with tree demography, and a major fundamental goal in ecology is to understand the key mechanisms governing spatio‐temporal patterns in tree population dynamics. While previous research has made substantial progress in identifying the mechanisms individually, their relative importance among forests remains unclear mainly due to practical limitations. One approach to overcome these limitations is to group mechanisms according to their shared effects on the variability of tree vital rates and quantify patterns therein. We developed a conceptual and statistical framework (variance partitioning of Bayesian multilevel models) that attributes the variability in tree growth, mortality, and recruitment to variation in species, space, and time, and their interactions – categories we refer to as organising principles (OPs). We applied the framework to data from 21 forest plots covering more than 2.9 million trees of approximately 6500 species. We found that differences among species, the species OP, proved a major source of variability in tree vital rates, explaining 28–33% of demographic variance alone, and 14–17% in interaction with space , totalling 40–43%. Our results support the hypothesis that the range of vital rates is similar across global forests. However, the average variability among species declined with species richness, indicating that diverse forests featured smaller interspecific differences in vital rates. Moreover, decomposing the variance in vital rates into the proposed OPs showed the importance of unexplained variability, which includes individual variation, in tree demography. A focus on how demographic variance is organized in forests can facilitate the construction of more targeted models with clearer expectations of which covariates might drive a vital rate. This study therefore highlights the most promising avenues for future research, both in terms of understanding the relative contributions of groups of mechanisms to forest demography and diversity, and for improving projections of forest ecosystems.

Data from: Global Spore Sampling Project: A global, standardized dataset of airborne fungal DNA

Zenodo (CERN European Organization for Nuclear Research) · 2024-05-07 · 1 citations

Novel methods for sampling and characterizing biodiversity hold great promise for re-evaluating patterns of life across the planet. The sampling of airborne spores with a cyclone sampler, and the sequencing of their DNA, have been suggested as an efficient and well-calibrated tool for surveying fungal diversity across various environments. Here we present data originating from the Global Spore Sampling Project, comprising 2,768 samples collected during two years at 47 outdoor locations across the world. Each sample represents fungal DNA extracted from 24 m3 of air. We applied a conservative bioinformatics pipeline that filtered out sequences that did not show strong evidence of representing a fungal species. The pipeline yielded 27,954 species-level operational taxonomic units (OTUs). Each OTU is accompanied by a probabilistic taxonomic classification, validated through comparison with expert evaluations. To examine the potential of the data for ecological analyses, we partitioned the variation in species distributions into spatial and seasonal components, showing a strong effect of the annual mean temperature on community composition. The database is organized in five datasets in a csv format (columns separated by commas): (1) metadata providing the location, date, and time for each sample, along with sequencing depth and other essential information (metadata.csv); (2) species-level OTU tables per sample describing the number of sequences assigned to each species (otu.table.csv 3); (3) taxonomic classification of each species-level OTU (taxonomy.csv); (4) closest matching sequences and their taxonomy for ASVs in putatively fungal pseudophyla, which are included in (2) and (3) (fungi_pseudophyla.csv); and (5) closest matching sequences and their taxonomy for ASVs in putatively non-fungal pseudophyla, which are not included in the other datasets (nonfungi_pseudophyla.csv). The first four datasets can be linked to each other using the unique sample codes and the unique identifiers for species-level OTUs. The three first datafiles are also provided in allData.RData which can be read into R as load("allData.RData").