Content uploaded by Christopher R Hakkenberg
Author content
All content in this area was uploaded by Christopher R Hakkenberg on Oct 02, 2020
Content may be subject to copyright.
A preview of the PDF is not available
Tree canopy cover constrains the fertility–diversity relationship in plant communities of the southeastern USA
Abstract
The goal of elucidating the primary mechanisms constraining the assembly and distribution of biodiversity remains among the central unresolved challenges facing the field of ecology. Simulation studies and experimental manipulations have focused on how patterns in community assembly result from bivariate relationships along productivity or environmental gradients. However, the joint influence of multiple resource gradients on the distribution of species richness in natural communities remains under‐studied. Using data from a large network of multi‐scale vegetation plots across forests and woodlands of the southeastern US, we find significant evidence for the scale dependent, joint constraints of forest structure and soil resources on the distribution of vascular plant species richness. In addition to their significant partial effects on species richness, understory light levels and soil fertility positively interact, suggesting a trade‐off between the two limiting resources with species richness peaking both in high‐light, low‐fertility conditions as well as low‐light, high‐fertility settings. This finding provides a novel perspective on the biodiversity‐productivity relationship that suggests a transition in limiting resources from soil nutrients to light availability when enhanced productivity results in reduced light resources for subordinate individuals. Results likewise have meaningful implications for our understanding of scale dependent community assembly processes as size‐asymmetric competition replaces environmental filtering as the primary assembly mechanism structuring temperate forest communities along an increasing soil fertility gradient.
Supplementary resource (1)
... β 1 indicates how the diversity effects on spatial stability varied across the stress gradients. Noninformative priors were assigned to all parameters (Hakkenberg et al., 2020;Lin et al., 2022). ...
... We estimated the model parameters using Markov Chain Monte Carlo (MCMC) approach and assessed parameter convergence using Gelman and Rubin's convergence diagnostics after 10,000 iterations from four Markov chains using the brMs package (Hakkenberg et al., 2020). All statistical analyses were carried out using the software R 3.6.2. ...
... This result is due to the fact that facilitation is insufficient to buffer competition for a limiting resource under the harshest conditions. In such very harsh conditions, facilitation may shift towards competition failing to enhance tree resistance to environmental fluctuations and thus failing to increase portfolio effects (Hakkenberg et al., 2020;Holmgren & Scheffer, 2010;Michalet et al., 2021). We also found that the environmental dependence of portfolio effects was me- to enhance the portfolio effects (Michalet et al., 2021). ...
The assumption that greater biodiversity enhances ecosystem stability, commonly known as ‘portfolio effect’, has attracted considerable research attention. However, the potential portfolio effects on spatial stability (the similarity of ecosystem functioning among forest communities) are still poorly examined especially at different spatial scales and under varying environmental stress conditions. Accordingly, this study investigates the biodiversity–spatial stability relationship among regional communities across different spatial scales and environmental conditions in a temperate forest region.
We define spatial stability as the invariability of the productivity of woody plants among plots within a regional community. To test spatial stability, the N closest plots to a given plot were aggregated to form regional communities representing different spatial scales. Structural equation modelling was used to evaluate how biodiversity (including taxonomic [TD] and phylogenetic diversity [PD]) increases spatial stability via species asynchrony and/or population stability across spatial scales. Hierarchical Bayesian modelling was used to evaluate the environmental dependence of the portfolio effects on spatial stability.
TD and PD both increased the spatial stability by increasing asynchrony, but decreased population stability. The portfolio effect of TD on spatial stability became stronger and reached saturation at the intermediate scale and then decreased as regional communities became larger. The portfolio effects of TD were weaker under the stressful conditions of drought, high precipitation seasonality and high elevation but unchanged across temperature seasonality and human disturbance. PD showed no discernable effect on spatial stability and did not change across spatial scale and environmental condition.
Our results suggest that the positive effect of biodiversity on species asynchrony overcomes the negative biodiversity effect on population stability to buffer the spatial change in productivity in diverse communities. Future research of the biodiversity–spatial stability relationship may thus benefit from incorporating different spatial scales and environmental conditions into the analysis.
Read the free Plain Language Summary for this article on the Journal blog.
... Along an elevation gradient, closely representing the beech productivity gradient, conditions sub-optimal for beech are characterised by higher species richness than those close to the optimum. Indeed, the mechanisms driving plant richness have been linked to ecosystem productivity in this way (Hakkenberg et al., 2020). Significantly, this pattern could be driven by not only above-ground mechanisms such as canopy closure and litter mass, but also below-ground competition for soil resources (Mölder et al., 2008). ...
... In addition, C/N and N were also related to species richness in our study. Similarly to other studies, our species richness is higher at lower C/N values, typical for locations with higher soil nutrient availability (Hakkenberg et al., 2020). Poor understorey species richness is usually found at both ends of the fertility gradient; more acidic soil reaction may hamper litter decomposition and nutrient release (Staude et al., 2020). ...
... Poor understorey species richness is usually found at both ends of the fertility gradient; more acidic soil reaction may hamper litter decomposition and nutrient release (Staude et al., 2020). Our results reveal an increasingly negative impact of high C/ N ratio on species richness, where low nutrient availability is exacerbated by competitive pressure of fine roots and symbionts associated with beech (Hakkenberg et al., 2020). We found that fine root mass best predicted the understorey species richness in mature stands. ...
Traditionally focussed on maximising productivity, forest management increasingly has to consider other functions performed by the forest stands, such as biodiversity conservation. Terrestrial plant communities typically possess a hump-back relationship between biomass productivity and plant species richness. However, there is evidence of a reverse relationship in forests dominated by beech, one of the most competitive and widespread tree species in temperate Europe. To fully explore the tree productivity-species richness relationship, we investigated above- and below-ground drivers of understorey plant species richness. We focussed on managed beech forests growing along an elevation gradient in Central Europe. We found that the lowest understorey plant diversity was under conditions optimal for beech. Tree fine root mass, canopy openness, soil C/N ratio, the interaction between tree fine root mass and stoniness, and stand structural diversity explain the variation of understorey species richness. We show that the competition for soil resources is the main driver of plant species diversity in managed forests; maximising beech growth in optimal conditions may thus come at the expense of understorey plant richness.
... Piedmont forest composition and structure have been greatly affected by a century of chronic fire suppression and exclusion, and therefore even modern forests with relatively few human impacts do not necessarily reflect historical vegetation patterns. In areas least altered by humans, most of the present vegetation is closed-canopy forest with limited shrub and herb layers owing to the infertility of the soils (Hakkenberg et al. 2020). The canopy is dominated by several oak species and a scattering of representatives of other genera that vary in abundance with edaphic and moisture characteristics. ...
... Herb layers are usually sparse, with only a few shade-tolerant species frequent, and then usually with low cover. In addition to reduced insolation, the low level of basic cations and typically dry upland soils limit the herb layer cover and diversity (Peet et al. 2014;Hakkenberg et al. 2020); in some places a high population density of white-tailed deer further limits development of both shrub and herb cover due to heavy browsing. ...
The Piedmont (PDMT) ecoregion of the USA stretches from New Jersey to Alabama, nestled between the Coastal Plain and Blue Ridge Mountain physiographic provinces. Many of the notable Piedmont plant communities, including the dominant oak-hickory forests of the region, are reliant upon fire to some degree. Before human settlement, most Piedmont vegetation burned relatively frequently and at low intensities, resulting in extensive closed canopy oak-hickory forests, studded with patches of open woodland and savanna largely defined by unusual soil conditions. Indigenous peoples of the Piedmont used fire as a land management tool for both agriculture and game production. Historical changes in land use throughout the region have altered fire regimes and changed forest dynamics dramatically over the past 400 years. Euro-American settlement led to widespread clearing of land for agriculture and logging; by the early twentieth century, very little old-growth forest remained in the Piedmont. During the mid-twentieth century, the decline of agriculture and the aggressive suppression and exclusion of wildfires brought about the growth of successional forests in the place of older, fire-mediated communities. The Piedmont region is currently experiencing a rapid expansion of the human population and land development, making restoration of the historical fire regime a challenge. However, land managers frequently do use prescribed fire to enhance timberland and restore rare plant communities.
... For small-statured herbaceous plants in nutrient-rich temperate mixed forests, studies have found scale-dependent effects of forest structure (e.g. CC) on plants species richness to peak at intermediate scales (400-1000 m 2 ), though this scale is dependent, critically, on a third interacting factor: soil fertility (Hakkenberg et al., 2020). ...
Lidar‐derived forest structural diversity (FSD) metrics—including measures of forest canopy height, vegetation arrangement, canopy cover (CC), structural complexity and leaf area and density—are increasingly used to describe forest structural characteristics and can be used to infer many ecosystem functions. Despite broad adoption, the importance of spatial resolution (grain and extent) over which these structural metrics are calculated remains largely unconsidered. Often researchers will quantify FSD at the spatial grain size of the process of interest without considering the scale dependency or statistical behaviour of the FSD metric employed.
We investigated the appropriate scale of inference for eight lidar‐derived spatial metrics—CC, canopy relief ratio, foliar height diversity, leaf area index, mean and median canopy height, mean outer canopy height, and rugosity (RT)‐‐representing five FSD categories—canopy arrangement, CC, canopy height, leaf area and density, and canopy complexity. Optimal scale was determined using the representative elementary area (REA) concept whereby the REA is the smallest grain size representative of the extent. Structural metrics were calculated at increasing canopy spatial grain (from 5 to 1000 m) from aerial lidar data collected at nine different forested ecosystems including sub‐boreal, broadleaf temperate, needleleaf temperate, dry tropical, woodland and savanna systems, all sites are part of the National Ecological Observatory Network within the conterminous United States. To identify the REA of each FSD metric, we used changepoint analysis via segmented or piecewise regression which identifies significant changepoints for both the magnitude and variance of each metric.
We find that using a spatial grain size between 25 and 75 m sufficiently captures the REA of CC, canopy arrangement, canopy leaf area and canopy complexity metrics across multiple forest types and a grain size of 30–150 m captures the REA of canopy height metrics. However, differences were evident among forest types with higher REA necessary to characterize CC in evergreen needleleaf forests, and canopy height in deciduous broadleaved forests.
These findings indicate the appropriate range of spatial grain sizes from which inferences can be drawn from this set of FSD metrics, informing the use of lidar‐derived structural metrics for research and management applications.
... This increase in water consumption, consequently decreasing soil water availability, would increase the competition for water between trees and understory plants and would explain the negative effects of specific root length on understory productivity (i.e., plant cover and root biomass). In addition to the belowground competition, our results suggested an aboveground competition for light with (Hakkenberg et al. 2020;Mueller et al. 2016). Besides, we confirmed the role of trees in controlling soil nitrogen and phosphorus contents by modifying litter C:N ratio and root morphological traits related to desiccation and exudation (i.e., N and P-rich compounds, Bardgett, Mommer, and de Vries 2014;Sun et al. 2020). ...
Forests are critical ecosystems to understand the global carbon budget, due to their carbon sequestration potential in both above‐ and belowground compartments, especially in species‐rich forests. Soil carbon sequestration is strongly linked to soil microbial communities, and this link is mediated by the tree community, likely due to modifications of micro‐environmental conditions (i.e., biotic conditions, soil properties, and microclimate). We studied soil carbon concentration and the soil microbial biomass of 180 local neighborhoods along a gradient of tree species richness ranging from 1 to 16 tree species per plot in a Chinese subtropical forest experiment (BEF‐China). Tree productivity and different tree functional traits were measured at the neighborhood level. We tested the effects of tree productivity, functional trait identity and dissimilarity on soil carbon concentrations, and their mediation by the soil microbial biomass and micro‐environmental conditions. Our analyses showed a strong positive correlation between soil microbial biomass and soil carbon concentrations. Besides, soil carbon concentration increased with tree productivity and tree root diameter while it decreased with litterfall C:N content. Moreover, tree productivity and tree functional traits (e.g. root fungal association and litterfall C:N ratio) modulated micro‐environmental conditions with substantial consequences for soil microbial biomass. We also showed that soil history and topography should be considered in future experiments and tree plantations, as soil carbon concentrations were higher where historical (i.e., at the beginning of the experiment) carbon concentrations were high, themselves being strongly affected by the topography. Altogether, these results imply that the quantification of the different soil carbon pools is critical for understanding microbial community–soil carbon stock relationships and their dependence on tree diversity and micro‐environmental conditions. This article is protected by copyright. All rights reserved.
... Although structure exerts primary controls on understorey light levels, canopy density and spacing likewise affect root competition for spatially heterogeneous water and nutrient resources (Hakkenberg et al., 2020) and indirectly affect microclimate and soil water availability by constraining ground-level solar radiation (Chen et al., 1999). ...
Aim
Empirical biodiversity – forest structure relationships (BSRs) underlie the use of forest structure as a remotely sensible proxy of biodiversity. However, little is known about how BSRs generalize to continental scales or how climate interacts with structure to drive local patterns in plant diversity. Resolving these research gaps in macrosystems ecology will strengthen our understanding of the biogeography of plant diversity, with implications for global-scale biodiversity mapping.
Location
USA.
Time period
Contemporary.
Major taxa studied
Vascular plants and trees.
Methods
We combined climate variables with field measurements and airborne lidar from all forest and woodland plots in the National Ecological Observatory Network (NEON) to characterize the role of climate in constraining BSRs across the United States. Spatial generalized linear mixed models were used to quantify the individual and joint effects of structure and climate on vascular plant and tree diversity.
Results
These findings provide evidence for broad-scale BSRs across the USA; namely, between plant/tree diversity and forest structural metrics in the vertical and horizontal planes. Vascular plant diversity was positively related to horizontally heterogeneous and bottom-skewed canopies, whereas tree diversity was positively associated with canopy cover and structural heterogeneity. In addition, climate variables related to stress (negatively), energy (positively) and seasonality (negatively) affected broad-scale patterns in diversity, with water availability (but not temperature) exerting significant effects on structural conditions. Importantly, climate and structure interact to explain variance in BSRs along coldness, mean temperature and evapotranspiration gradients.
Main conclusions
Our findings reinforce the importance of local context dependence in assessing non-stationary biogeographical patterns in forest biodiversity, as well as the unique ways that forest structure and climate interact to constrain plant diversity. Beyond these theoretical insights, this study provides an empirical foundation for generalizing remotely sensed estimates of climate and forest structure to model biodiversity over vast extents.
Mixed-species forests are known to provide enhanced ecosystem functioning when compared with monocultures. The strength of the positive diversity effect is, however, largely dependent on environmental conditions and the human footprint. The specific patterns of context-dependence effects of diversity on productivity and their causal relationships are not well known, especially regarding large-scale natural plant communities. Accordingly, the focus of this study is to evaluate effects of environmental variables on biodiversity-productivity relationships (BPR) in the temperate forest region of North-Eastern China. We used a structural equation modeling and hierarchical Bayesian model to assess individual and interactive effects of human disturbance, environmental factors, and species diversity on productivity. The SEMs showed that the negative interaction effect of human disturbance and mean annual temperature on species diversity was greater than that of human disturbance on species diversity. The positive diversity effects on forest production were severely reduced under stressful soil water conditions due to increased competition for soil water resources. The positive diversity effects were greater under harsh annual mean temperature conditions. The biodiversity effects on productivity were influenced by negative interactive effect of the human footprint and soil water availability. Our results suggest that current management strategies for diversity conservation may be countered by robust negative interactive effects and thus, may be inefficient in temperate forest regions. The variations in the strength of the relationships between tree species diversity and forest productivity are critically dependent on several environmental stressors. Our results highlight the important role of human disturbance in mediating the response of BPR to soil water deficit, and the crucial role of management strategy to prepare for future climatic challenges in our temperate forest region.
Although forest disturbances have become more frequent and severe due to ongoing climate change, our understanding of post-disturbance development of vegetation and tree–herb layer interactions remains limited. An extreme windstorm, which occurred on 19 November 2004, destroyed icea abies (L.) H. Karst dominated forests in the High Tatra Mts. Here, we studied short-term changes in diversity, species composition, and aboveground biomass of trees and herb layer vegetation, including mutual relationships that elucidate tree–herb interactions during post-disturbance succession.
Assessment of species composition and tree biomass measurements were performed at
50 sample plots (4 � 4 m) along two transects 12, 14, and 16 years after the forest destruction. Heights and stem base diameters of about 730 trees were measured and subsequently used for the calculation of aboveground tree biomass using species-specific allometric relationships. Aboveground biomass of herb layer was quantified at 300 subplots (20 � 20 cm) by destructive sampling. Species richness and spatial vegetation heterogeneity did not significantly change, and species composition
exhibited small changes in accordance with expected successional trajectories. While aboveground tree biomass increased by about 190%, biomass of annual herb shoots decreased by about 68% and biomass of perennial herb shoots was stable during the studied period. The contribution of trees to total aboveground biomass increased from 83% to 97%. After 16 years of forest stands recovery, tree biomass represented approximately 13% of forest biomass before the disturbance. Herb layer biomass, particularly the biomass of annual herb shoots, was more closely related to tree cover than to tree biomass and its decline could be assigned to gradual tree growth. Our study provides clear evidence that short-term successional processes in post-disturbance vegetation are much better detectable by biomass than by diversity or compositional measures and emphasized the importance of light conditions in tree–herb competitive interactions.
Aims
Competitive inhibition of temperate forest tree recruits by herbs is likely important on sites with high fertility owing to faster height growth and consequent pre‐emption of light. We explored the site conditions and stand structure under which herbaceous growth has an impact on tree regeneration.
Location
Plot data from 610 forest sites were collected from five areas across the southern Appalachian Mountains.
Methods
Several plant guilds were distinguished based on various biological traits. Deterministic models of forest understorey were validated through recursive path analysis. The numerical analyses were performed both on all plots and on a subset of 150 plots free of evergreen shrubs.
Results
In general, total herb cover increased with soil fertility, but in sites without evergreen shrubs no relationship emerged. Total herb cover varied inversely with woody stem density (saplings excluded), but the slope was much less steep in the absence of evergreen shrubs. Tree sapling density displayed a left‐tailed, asymmetric response with respect to total herb cover, but a symmetric unimodal response against tall herb cover. The shape of the distribution of tree stems by diameter class shifted from unimodal under a very sparse herbaceous layer to negative exponential in stands with mid to high herb cover. This was due to the suppressive impact of evergreen shrubs on understorey vegetation, which led to a positive covariance between total herb cover and tree sapling density. These two understorey variables became unrelated in the path model built on the subset without evergreen shrubs, but a similar model involving tall herbs revealed a direct negative effect of tall herb cover on tree sapling density.
Conclusions
Our results provide evidence of tree recruits exclusion by tall herbs on fertile sites but not on acidic sites, where herb interference is much reduced by the suppressive effect of evergreen shrubs and trees on herbaceous layer vegetation.
Understanding biodiversity-productivity relationships (BPRs) is of theoretical importance, and has important management implications. Most work on BPRs has focused on simple and/or experimentally assembled communities, and it is unclear how these observed BPRs can be extended to complex natural forest ecosystems. Using data from over 115,000 forest plots across the contiguous United States, we show that the bivariate BPRs are positive in dry climates and hump-shaped in mesic climates. When considering other site characteristics, BPRs change to neutral in dry climates and remain hump-shaped in humid sites. Our results indicate that climatic variation is an underlying determinant of contrasting BPRs observed across a large spatial extent, while both biotic factors (e.g., stand age and density) and abiotic factors (e.g., soil properties) can impact BPRs within a given climate unit. These findings suggest that tradeoffs need be made when considering whether to maximize productivity vs. conserve biodiversity, especially in mesic climates.
Because biodiversity is multidimensional and scale‐dependent, it is challenging to estimate its change. However, it is unclear (1) how much scale‐dependence matters for empirical studies, and (2) if it does matter, how exactly we should quantify biodiversity change. To address the first question, we analysed studies with comparisons among multiple assemblages, and found that rarefaction curves frequently crossed, implying reversals in the ranking of species richness across spatial scales. Moreover, the most frequently measured aspect of diversity – species richness – was poorly correlated with other measures of diversity. Second, we collated studies that included spatial scale in their estimates of biodiversity change in response to ecological drivers and found frequent and strong scale‐dependence, including nearly 10% of studies which showed that biodiversity changes switched directions across scales. Having established the complexity of empirical biodiversity comparisons, we describe a synthesis of methods based on rarefaction curves that allow more explicit analyses of spatial and sampling effects on biodiversity comparisons. We use a case study of nutrient additions in experimental ponds to illustrate how this multi‐dimensional and multi‐scale perspective informs the responses of biodiversity to ecological drivers.
The central role of floristic diversity in maintaining habitat integrity and ecosystem function has propelled efforts to map and monitor its distribution across forest landscapes. While biodiversity studies have traditionally relied largely on ground-based observations, the immensity of the task of generating accurate, repeatable, and spatially-continuous data on biodiversity patterns at large scales has stimulated the development of remote-sensing methods for scaling up from field plot measurements. One such approach is through integrated LiDAR and hyperspectral remote-sensing. However, despite their efficiencies in cost and effort, LiDAR-hyperspectral sensors are still highly constrained in structurally- and taxonomically-heterogeneous forests - especially when species’ cover is smaller than the image resolution, intertwined with neighboring taxa, or otherwise obscured by overlapping canopy strata. In light of these challenges, this study goes beyond the remote characterization of upper canopy diversity to instead model total vascular plant species richness in a continuous-cover North Carolina Piedmont forest landscape. We focus on two related, but parallel, tasks. First, we demonstrate an application of predictive biodiversity mapping, using nonparametric models trained with spatially-nested field plots and aerial LiDAR-hyperspectral data, to predict spatially-explicit landscape patterns in floristic diversity across seven spatial scales between 0.01m2–900m2. Second, we employ bivariate parametric models to test the significance of individual, remotely-sensed predictors of plant richness to determine how parameter estimates vary with scale. Cross-validated results indicate that predictive models were able to account for 15-70% of variance in plant richness, with LiDAR-derived estimates of topography and forest structural complexity, as well as spectral variance in hyperspectral imagery explaining the largest portion of variance in diversity levels. Importantly, bivariate tests provide evidence of scale-dependence among predictors, such that remotely-sensed variables significantly predict plant richness only at spatial scales that sufficiently subsume geolocational imprecision between remotely-sensed and field data, and best align with stand components including plant size and density, as well as canopy gaps and understory growth patterns. Beyond their insights into the scale-dependent patterns and drivers of plant diversity in Piedmont forests, these results highlight the potential of remotely-sensible essential biodiversity variables for mapping and monitoring landscape floristic diversity from air- and space-borne platforms. This article is protected by copyright. All rights reserved.
Net primary productivity (NPP) is a variable of primary interest to ecologists, as it is related both to resource availability, potentially affecting biological diversity, and to the dynamics of the carbon cycle. However, there are alarming discrepancies in NPP estimates as well as in the reported form of the relationship between NPP and species richness. Such discrepancies could be due to the different and often simplified assumptions of various global NPP models and the heterogeneity of field NPP measurements that comprise a mix of natural vegetation and plantations. Here we review different global models of NPP and available original sources of NPP field measurements in order to examine how their geographic patterns are affected by various assumptions and data selection, respectively. Then we review studies dealing with diversity-productivity relationships in view of different NPP estimates. We show that although NPP does generally decrease with increasing latitude, geographic NPP patterns considerably differ between individual models as well as between the models and field NPP data. Such inconsistencies might be partially responsible for discrepancies in productivity-richness relationships, although these are also driven by other factors that covary with productivity and affect diversity patterns. To reconcile the discrepancies between various NPP measures, it is necessary to (1) standardize field NPP data, (2) develop scaling techniques that bridge the gap between the scale of field NPP measurements and NPP models, and (3) build global NPP models that account for nutrient limitation (especially concerning phosphorus in the tropics) and are parameterized by field measurements. Also, (4) a better theory needs to be developed to distinguish the effect of productivity from the effects of other environmental variables on diversity patterns. Improving our ability to estimate NPP will help us predict future NPP changes and understand the drivers of species richness patterns. This article is protected by copyright. All rights reserved.
Significance
Global environmental change involves many factors that occur simultaneously, yet they are usually studied in isolation. Here we report a long-term global change experiment that subjected California grassland to multiple individual and simultaneous changes in temperature, precipitation, carbon dioxide, and nitrogen. Our analysis revealed nonlinear and interactive effects of temperature and precipitation on grassland net primary production (NPP), which defined a ridge-shaped NPP response surface to these two variables. Added nitrogen raised the peak of the NPP response surface, and added CO 2 shifted the peak to lower temperatures. Our approach was validated by tests showing an absence of progressive effects over the years. In other ecosystems, our approach may be similarly powerful for probing the effects of multifactor global change.
Environmental filtering, where the environment selects against certain species, is thought to be a major mechanism structuring communities. However, recent criticisms cast doubt on our ability to accurately infer filtering because competition can give rise to patterns identical to those caused by environmental filtering. While experiments can distinguish mechanisms, observational patterns are especially problematic. The environment determines community composition not only directly via survival, but also by influencing competition. If species population growth rates covary with environmental gradients, then outcomes of competitive exclusion will also vary with the environment. Here, we argue that observational studies remain valuable, but inferences about the importance of the environment cannot rely on compositional data alone, and that species abundances, population growth, or traits must be correlated with the environment.
