Phenotypic plasticity masks range-wide genetic differentiation for vegetative but not reproductive traits in a short-lived plant.

Genetic differentiation and phenotypic plasticity jointly shape intraspecific trait variation, but their roles differ among traits. In short-lived plants, reproductive traits may be more genetically determined due to their impact on fitness, whereas vegetative traits may show higher plasticity to buffer short-term perturbations. Combining a multi-treatment greenhouse experiment with observational field data throughout the range of a widespread short-lived herb, Plantago lanceolata, we (1) disentangled genetic and plastic responses of functional traits to a set of environmental drivers and (2) assessed how genetic differentiation and plasticity shape observational trait-environment relationships. Reproductive traits showed distinct genetic differentiation that largely determined observational patterns, but only when correcting traits for differences in biomass. Vegetative traits showed higher plasticity and opposite genetic and plastic responses, masking the genetic component underlying field-observed trait variation. Our study suggests that genetic differentiation may be inferred from observational data only for the traits most closely related to fitness.

Predicting the responses of subalpine forest landscape dynamics to climate change on the eastern Tibetan Plateau.

Subalpine vegetation across the Tibetan Plateau is globally one of the most sensitive to climate change. However, the potential landscape-scale effects of climate change on subalpine forest dynamics remain largely unexplored. Here, we used a forest landscape model (LANDIS-II) coupled with a forest ecosystem process model (PnET-II) to simulate forest dynamics under future climate change in Jiuzhaigou National Nature Reserve in the eastern subalpine region of the Tibetan Plateau. We examined changes in the composition, distribution and aboveground biomass of cold temperate coniferous forests, temperate coniferous forests, deciduous broad-leaved forests and redwood forest under four climate change scenarios (RCP2.6, RCP4.5, RCP8.5 and the current climate) from 2016 to 2096. Our model predicts that by 2096, (i) cold temperate coniferous forests will expand and increase by 7.92%, 8.18%, 8.65% and 7.02% under current climate, RCP2.6, RCP4.5 and RCP8.5 scenarios, respectively; (ii) distribution of forests as a whole shows upward elevational range shift, especially under RCP8.5 scenario and (iii) total aboveground biomass slowly increases at first and then decreases to 12%-16% of current distribution under RCPs. These results show that climate change can be expected to significantly influence forest composition, distribution and aboveground biomass in the subalpine forests of eastern Tibetan Plateau. This study is the first to simulate forest dynamics at the landscape scale in subalpine areas of the Tibetan Plateau, which provides an important step in developing more effective strategies of forest management for expected climate change, not only in China but also around the world.

Altered climate leads to positive density-dependent feedbacks in a tropical wet forest.

Climate change is predicted to result in warmer and drier Neotropical forests relative to current conditions. Negative density-dependent feedbacks, mediated by natural enemies, are key to maintaining the high diversity of tree species found in the tropics, yet we have little understanding of how projected changes in climate are likely to affect these critical controls. Over 3 years, we evaluated the effects of a natural drought and in situ experimental warming on density-dependent feedbacks on seedling demography in a wet tropical forest in Puerto Rico. In the +4°C warming treatment, we found that seedling survival increased with increasing density of the same species (conspecific). These positive density-dependent feedbacks were not associated with a decrease in aboveground natural enemy pressure. If positive density-dependent feedbacks are not transient, the diversity of tropical wet forests, which may rely on negative density dependence to drive diversity, could decline in a future warmer, drier world.

Disturbances Can Promote and Hinder Coexistence of Competitors in Ongoing Partner Choice Mutualisms.

Ecosystems are under threat from anthropogenic and natural disturbances, yet little is known about how these disturbances alter mutualistic interactions. Many mutualistic interactions are highly context dependent and dynamic due to "ongoing" partner choice, impeding our understanding of how disturbances might influence mutualistic systems. Previously we showed that in the absence of additional known mechanisms of competitive coexistence, mutualistic fungi can coexist in a system where the plant community associates dynamically with two empirically defined arbuscular mycorrhizal fungal types: a cheap kind that provides low nutrient benefits, and an expensive type that provides high nutrient benefits. We built on this framework to ask how disturbances of different types, frequencies, amplitudes, and predictabilities alter ongoing partner choice and thereby influence the coexistence of mutualists. We found that the effects of disturbances depend on the type, amplitude, and predictability of disturbances and, to a lesser extent, on their frequency. Disturbance can disrupt mutualist coexistence by enabling hosts more efficiently to exclude partners that behave as parasites. Disturbance can also promote coexistence by altering the strength and direction of consumer-resource interactions. Predicting the effects of disturbance on the mutualist community therefore requires us to understand better the consumer-resource relationships under various environmental conditions. We show how, through such context-dependent effects, disturbance and ongoing partner choice can together generate relative nonlinearity and investment in future benefit, introducing fluctuation-dependent mechanisms of competitive coexistence. Our findings support a broadening of the conceptual framework regarding disturbances and competition to include fluctuation-dependent mechanisms alongside the spatiotemporal intermediate disturbance hypothesis.

Symbiont community diversity is more variable in corals that respond poorly to stress.

Coral reefs are declining globally as climate change and local water quality press environmental conditions beyond the physiological tolerances of holobionts-the collective of the host and its microbial symbionts. To assess the relationship between symbiont composition and holobiont stress tolerance, community diversity metrics were quantified for dinoflagellate endosymbionts (Family: Symbiodiniaceae) from eight Acropora millepora genets that thrived under or responded poorly to various stressors. These eight selected genets represent the upper and lower tails of the response distribution of 40 coral genets that were exposed to four stress treatments (and control conditions) in a 10-day experiment. Specifically, four 'best performer' coral genets were analyzed at the end of the experiment because they survived high temperature, high pCO2 , bacterial exposure, or combined stressors, whereas four 'worst performer' genets were characterized because they experienced substantial mortality under these stressors. At the end of the experiment, seven of eight coral genets mainly hosted Cladocopium symbionts, whereas the eighth genet was dominated by both Cladocopium and Durusdinium symbionts. Symbiodiniaceae alpha and beta diversity were higher in worst performing genets than in best performing genets. Symbiont communities in worst performers also differed more after stress exposure relative to their controls (based on normalized proportional differences in beta diversity), than did best performers. A generalized joint attribute model estimated the influence of host genet and treatment on Symbiodiniaceae community composition and identified strong associations among particular symbionts and host genet performance, as well as weaker associations with treatment. Although dominant symbiont physiology and function contribute to host performance, these findings emphasize the importance of symbiont community diversity and stochasticity as components of host performance. Our findings also suggest that symbiont community diversity metrics may function as indicators of resilience and have potential applications in diverse disciplines from climate change adaptation to agriculture and medicine.

Understanding the recruitment response of juvenile Neotropical trees to logging intensity using functional traits.

Selective logging remains a widespread practice in tropical forests, yet the long-term effects of timber harvest on juvenile tree (i.e., sapling) recruitment across the hundreds of species occurring in most tropical forests remain difficult to predict. This uncertainty could potentially exacerbate threats to some of the thousands of timber-valuable tree species in the Amazon. Our objective was to determine to what extent long-term responses of tree species regeneration in logged forests can be explained by their functional traits. We integrate functional trait data for 13 leaf, stem, and seed traits from 25 canopy tree species with a range of life histories, such as the pioneer Goupia glabra and the shade-tolerant Iryanthera hostmannii, together with over 30 yr of sapling monitoring in permanent plots spanning a gradient of harvest intensity at the Paracou Forest Disturbance Experiment (PFDE), French Guiana. We anticipated that more intensive logging would increase recruitment of pioneer species with higher specific leaf area, lower wood densities, and smaller seeds, due to the removal of canopy trees. We define a recruitment response metric to compare sapling regeneration to timber harvest intensity across species. Although not statistically significant, sapling recruitment decreased with logging intensity for eight of 23 species and these species tended to have large seeds and dense wood. A generalized linear mixed model fit using specific leaf area, seed mass, and twig density data explained about 45% of the variability in sapling dynamics. Effects of specific leaf area outweighed those of seed mass and wood density in explaining recruitment dynamics of the sapling community in response to increasing logging intensity. The most intense treatment at the PFDE, which includes stand thinning of non-timber-valuable adult trees and poison-girdling for competitive release, showed evidence of shifting community composition in sapling regeneration at the 30-yr mark, toward species with less dense wood, lighter seeds, and higher specific leaf area. Our results indicate that high-intensity logging can have lasting effects on stand regeneration dynamics and that functional traits can help simplify general trends of sapling recruitment for highly diverse logged tropical forests.

Dynamic preferential allocation to arbuscular mycorrhizal fungi explains fungal succession and coexistence.

Evidence accumulates about the role of arbuscular mycorrhizal (AM) fungi in shaping plant communities, but little is known about the factors determining the biomass and coexistence of several types of AM fungi in a plant community. Here, using a consumer-resource framework that treats the relationship between plants and fungi as simultaneous, reciprocal exploitation, we investigated what patterns of dynamic preferential plant carbon allocation to empirically-defined fungal types (on-going partner choice) would be optimal for plants, and how these patterns depend on successional dynamics. We found that ruderal AM fungi can dominate under low steady-state nutrient availability, and competitor AM fungi can dominate at higher steady-state nutrient availability; these are conditions characteristic of early and late succession, respectively. We also found that dynamic preferential allocation alone can maintain a diversity of mutualists, suggesting that on-going partner choice is a new coexistence mechanism for mutualists. Our model can therefore explain both mutualist coexistence and successional strategy, providing a powerful tool to derive testable predictions.

Associations among arbuscular mycorrhizal fungi and seedlings are predicted to change with tree successional status.

Arbuscular mycorrhizal (AM) fungi in the soil may influence tropical tree dynamics and forest succession. The mechanisms are poorly understood, because the functional characteristics and abundances of tree species and AM fungi are likely to be codependent. We used generalized joint attribute modeling to evaluate if AM fungi are associated with three forest community metrics for a sub-tropical montane forest in Puerto Rico. The metrics chosen to reflect changes during forest succession are the abundance of seedlings of different successional status, the amount of foliar damage on seedlings of different successional status, and community-weighted mean functional trait values (adult specific leaf area [SLA], adult wood density, and seed mass). We used high-throughput DNA sequencing to identify fungal operational taxonomic units (OTUs) in the soil. Model predictions showed that seedlings of mid- and late-successional species had less leaf damage when the 12 most common AM fungi were abundant compared to when these fungi were absent. We also found that seedlings of mid-successional species were predicted to be more abundant when the 12 most common AM fungi were abundant compared to when these fungi were absent. In contrast, early-successional tree seedlings were predicted to be less abundant when the 12 most common AM fungi were abundant compared to when these fungi were absent. Finally, we showed that, among the 12 most common AM fungi, different AM fungi were correlated with functional trait characteristics of early- or late-successional species. Together, these results suggest that early-successional species might not rely as much as mid- and late-successional species on AM fungi, and AM fungi might accelerate forest succession.

Nitrogen-fixing trees inhibit growth of regenerating Costa Rican rainforests.

More than half of the world's tropical forests are currently recovering from human land use, and this regenerating biomass now represents the largest carbon (C)-capturing potential on Earth. How quickly these forests regenerate is now a central concern for both conservation and global climate-modeling efforts. Symbiotic nitrogen-fixing trees are thought to provide much of the nitrogen (N) required to fuel tropical secondary regrowth and therefore to drive the rate of forest regeneration, yet we have a poor understanding of how these N fixers influence the trees around them. Do they promote forest growth, as expected if the new N they fix facilitates neighboring trees? Or do they suppress growth, as expected if competitive inhibition of their neighbors is strong? Using 17 consecutive years of data from tropical rainforest plots in Costa Rica that range from 10 y since abandonment to old-growth forest, we assessed how N fixers influenced the growth of forest stands and the demographic rates of neighboring trees. Surprisingly, we found no evidence that N fixers facilitate biomass regeneration in these forests. At the hectare scale, plots with more N-fixing trees grew slower. At the individual scale, N fixers inhibited their neighbors even more strongly than did nonfixing trees. These results provide strong evidence that N-fixing trees do not always serve the facilitative role to neighboring trees during tropical forest regeneration that is expected given their N inputs into these systems.

Arbuscular mycorrhizal fungal diversity and natural enemies promote coexistence of tropical tree species.

Negative population feedbacks mediated by natural enemies can promote species coexistence at the community scale through disproportionate mortality of numerically dominant (common) tree species. Simultaneously, associations with arbuscular mycorrhizal fungi (AMF) can result in positive effects on tree populations. Coupling data on seedling foliar damage from herbivores and pathogens and DNA sequencing of soil AMF diversity, we assessed the effects of these factors on tree seedling mortality at local (1 m2 ) and community (16 ha plot) scales in a tropical rainforest in Puerto Rico. At the local scale, AMF diversity in soil counteracted negative effects from foliar damage on seedling mortality. At the community scale, mortality of seedlings of common tree species increased with foliar damage while rare tree species benefited from soil AMF diversity. Together, the effects of foliar damage and soil AMF diversity on seedling mortality might foster tree species coexistence in this forest.

Long-lasting effects of land use history on soil fungal communities in second-growth tropical rain forests.

Our understanding of the long-lasting effects of human land use on soil fungal communities in tropical forests is limited. Yet, over 70% of all remaining tropical forests are growing in former agricultural or logged areas. We investigated the relationship among land use history, biotic and abiotic factors, and soil fungal community composition and diversity in a second-growth tropical forest in Puerto Rico. We coupled high-throughput DNA sequencing with tree community and environmental data to determine whether land use history had an effect on soil fungal community descriptors. We also investigated the biotic and abiotic factors that underlie such differences and asked whether the relative importance of biotic (tree diversity, basal tree area, and litterfall biomass) and abiotic (soil type, pH, iron, and total carbon, water flow, and canopy openness) factors in structuring soil fungal communities differed according to land use history. We demonstrated long-lasting effects of land use history on soil fungal communities. At our research site, most of the explained variation in soil fungal composition (R2  = 18.6%), richness (R2  = 11.4%), and evenness (R2  = 10%) was associated with edaphic factors. Areas previously subject to both logging and farming had a soil fungal community with lower beta diversity and greater evenness of fungal operational taxonomic units (OTUs) than areas subject to light logging. Yet, fungal richness was similar between the two areas of historical land use. Together, these results suggest that fungal communities in disturbed areas are more homogeneous and diverse than in areas subject to light logging. Edaphic factors were the most strongly correlated with soil fungal composition, especially in areas subject to light logging, where soils are more heterogenous. High functional tree diversity in areas subject to both logging and farming led to stronger correlations between biotic factors and fungal composition than in areas subject to light logging. In contrast, fungal richness and evenness were more strongly correlated with biotic factors in areas of light logging, suggesting that these metrics might reflect long-term associations in old-growth forests. The large amount of unexplained variance in fungal composition suggests that these communities are structured by both stochastic and niche assemblage processes.

Ontogenetic shifts in trait-mediated mechanisms of plant community assembly.

Identifying the processes that maintain highly diverse plant communities remains a central goal in ecology. Species variation in growth and survival rates across ontogeny, represented by tree size classes and life history stage-specific niche partitioning, are potentially important mechanisms for promoting forest diversity. However, the role of ontogeny in mediating competitive dynamics and promoting functional diversity is not well understood, particular in high-diversity systems such as tropical forests. The interaction between interspecific functional trait variation and ontogenetic shifts in competitive dynamics may yield insights into the ecophysiological mechanisms promoting community diversity. We investigated how functional trait (seed size, maximum height, SLA, leaf N, and wood density) associations with growth, survival, and response to competing neighbors differ among seedlings and two size classes of trees in a subtropical rain forest in Puerto Rico. We used a hierarchical Bayes model of diameter growth and survival to infer trait relationships with ontogenetic change in competitive dynamics. Traits were more strongly associated with average growth and survival than with neighborhood interactions, and were highly consistent across ontogeny for most traits. The associations between trait values and tree responses to crowding by neighbors showed significant shifts as trees grew. Large trees exhibited greater growth as the difference in species trait values among neighbors increased, suggesting trait-associated niche partitioning was important for the largest size class. Our results identify potential axes of niche partitioning and performance-equalizing functional trade-offs across ontogeny, promoting species coexistence in this diverse forest community.

Negative density-dependent mortality varies over time in a wet tropical forest, advantaging rare species, common species, or no species.

Although one of the most widely studied hypotheses for high tree diversity in the tropics, the Janzen-Connell hypothesis (JC), and the community compensatory trend upon which it is based, have conflicting support from prior studies. Some of this variation could arise from temporal variation in seedling survival of common and rare species. Using 10 years of data from La Selva Biological Station in Costa Rica, we analyzed annual seedling survival and found that negative density-dependence (negative DD) was significantly stronger for rare species than for common species in 2 years and was significantly stronger for common species than for rare species in 4 years. This temporal variation in survival was correlated with climatic variables: in warmer and wetter years, common species had higher negative DD than rare species. The relationship between climate and variation in JC effects on seedling survival of common and rare species could have important consequences for the maintenance of tree species diversity in Central America, which is predicted to experience warmer and wetter years as global change proceeds.