Search

Ecological models predict that the effects of mammalian herbivore exclusion on plant diversity depend on resource availability and plant exposure to ungulate grazing over evolutionary time. Using an experiment replicated in 57 grasslands on six continents, with contrasting evolutionary history of grazing, we tested how resources (mean annual precipitation and soil nutrients) determine herbivore exclusion effects on plant diversity, richness and evenness. Here we show that at sites with a long history of ungulate grazing, herbivore exclusion reduced plant diversity by reducing both richness and evenness and the responses of richness and diversity to herbivore exclusion decreased with mean annual precipitation. At sites with a short history of grazing, the effects of herbivore exclusion were not related to precipitation but differed for native and exotic plant richness. Thus, plant species’ evolutionary history of grazing continues to shape the response of the world’s grasslands to changing mammalian herbivory.

The effects of altered nutrient supplies and herbivore density on species diversity vary with spatial scale, because coexistence mechanisms are scale dependent. This scale dependence may alter the shape of the species–area relationship (SAR), which can be described by changes in species richness (S) as a power function of the sample area (A): S = cAz, where c and z are constants. We analysed the effects of experimental manipulations of nutrient supply and herbivore density on species richness across a range of scales (0.01–75 m²) at 30 grasslands in 10 countries. We found that nutrient addition reduced the number of species that could co-occur locally, indicated by the SAR intercepts (log c), but did not affect the SAR slopes (z). As a result, proportional species loss due to nutrient enrichment was largely unchanged across sampling scales, whereas total species loss increased over threefold across our range of sampling scales.

Wisconsin's plant communities are responding to shifting disturbance regimes, habitat fragmentation, aerial nitrogen deposition, exotic species invasions, ungulate herbivory, and successional processes. To better understand how plant functional traits mediate species' responses to changing environmental conditions, we collected a large set of functional trait data for vascular plant species occupying Wisconsin forests and grasslands. We used standard protocols to make 76,213 measurements of 34 quantitative traits. These data provide rich information on genome size, physical leaf traits (length, width, circularity, thickness, dry matter content, specific leaf area, etc.), chemical leaf traits (carbon, nitrogen, phosphorus, potassium, calcium, magnesium, ash), life history traits (vegetative and flower heights, seed mass), and traits affecting plant palatability (leaf fiber, fat, and lignin). These trait values derive from replicate measurements on 12+ individuals of each species from multiple sites and 45+ individuals for a selected subset of species. Measurements typically reflect values for individuals although some chemical traits involved composite samples from several individuals at the same site. We also qualitatively characterized each species by plant family, woodiness, functional group, and Raunkiaer lifeform. These data allow us to characterize trait dimensionality, differentiation, and covariation among temperate plant species (e.g., leaf and stem economic syndromes). We can also characterize species' responses to environmental gradients and drivers of ecological change. With survey and resurvey data available from400 sites in Wisconsin, we can analyze variation in community trait distributions and diversity over time and space. These data therefore allow us to assess how trait divergence vs. convergence affects community assembly and how traits may be related to half-century shifts in the distribution and abundance of these species. The data set can be used for non-commercial purposes. The data set is licensed as follows: CC-By Attribution 4.0 International. We request users cite both the OSF data set and this Ecology data paper publication.

Aim Biotic homogenization – the tendency for communities to converge in species composition – has occurred in many ecosystems, creating management challenges. The extent to which this convergence in species composition is related to convergence in trait composition (‘functional homogenization’), however, remains unresolved. Location North America, Wisconsin.

Plant traits are particularly important in determining plant community structure. However, how can one identify which traits are the most important in driving community assembly? Here we propose a method 1) to quantify the direction and strength of trait selection during community assembly and 2) to obtain parsimonious lists of traits that can predict species relative abundances in plant communities. We tested our method using floristic data from 32 plots experiencing different treatments (fertilisation and grazing) in southern France. Twelve functional traits were measured on 68 species. We determined the direction and strength of selection on these 12 traits using a metric derived from a maximum entropy model (i.e. lambda). We then determined our parsimonious list of traits using a backward selection of traits based on these lambda values (for all treatments and in each treatment separately). We finally compared our method to two other methods: one based on iterative RLQ and the other based on an entropy-based forward selection of traits. We found major differences in the direction and strength of selection across the 12 traits and treatments. From the 12 traits, plant vegetative and reproductive heights, leaf dry matter content leaf nitrogen content, specific leaf area, and leaf phosphorus content were particularly important for predicting species relative abundances when considering all treatments together. Our method yielded results similar to those produced by the entropy-based approach but differed from those produced by the iterative RLQ, whose selected traits could not significantly predict species relative abundances. Together these results suggest that the assembly of these communities is primarily driven by a small number of key functional traits. We argue that our method provides an objective way of determining a parsimonious list of traits that together accurately predict community structure and which, despite its complementarities with entropy-based method, offers significant advantages.

Summary1. Al th o ug h h ab itat frag me n tatio n is reco gn ized as a ma jor th reat to bio dive rsity, few stu dies h av eex am ined th e relative imp o rtan ce o f loca l, land scap e an d histo rica l facto rs in co nt ro lling loca lsp ec ies as semb lage s, an d h o w th ese facto rs intera ct,in pa tchy ec o sy stem s. We q ua n tiÞe d th e directan d ind irec t effects o f pa tch size ,p atch h etero ge n eity,ag ricu ltu ra linten sity an d p atch ag e o n plan tsp ec ies rich n ess an d co mp o sitio n o ffo rest pa tche s em b ed d ed in ag ricu ltu ra lland scap es.2. In six 5 a 5 k m- sa mp ling lan ds ca pe wi n do ws , we su rv ey ed ea ch fo rest p atch fo r va scular p lantsp ec ies an d co llec ted th ree sets of ind ep en d en t va riab les, d escribing p atch size an d h etero ge n eity,lan ds ca pe co mp o sition an d h isto ry .T he six wi n do ws w ere ar ra ng ed alon g a gr ad ien t o f ag ricu ltu reint en sity in ru ra llan ds ca pe s of th e P icar d y region (N Fr an ce ).3. We u sed n on -m etric mu ltidim en sion al scaling (N MS ) to d etec t ma jor en viro n me n tal gr ad ient su n de rlying va riation in sp ec ies co mp o sition am on g pa tche s. We th en co n stru cted stru ctu ra l eq u a-tion mo de ls (S E M) to q ua nt ify th e direct an d ind irec t effects of th e th ree sets of va riab les o n loca lp lant d iversity,wh ich wa s su cc essive ly inco rp or ated as pa tch scor es alon g th e Þ rst th ree NM S ax es,w oo dy sp ec ies rich ne ss,fo rest he rb sp ec ies rich ne ss,an d no n- fo rest h erb sp ec ies rich ne ss.4. A ma jor insu en ce o f th e lan ds ca pe ma trix on loca l sp ec ies co mp os ition wa s reve aled b y NM San d su bs eq ue nt S EM ,ma inly th ro u gh no n -for esthe rb sp ec ies, w hich ex plain ed mo st of th e be tw ee np atch s or istic d issimi larity.S pe cies rich ne ss increa sed wi th pa tch he tero ge n eity,wh erea s pa tch ar ean ev er h ad a direct effect. Fo rest he rb s we re mo re resp o ns ive to pa tch ag e an d co nn ec tivity tha no th er sp ec ies, wh erea s n on -for est an d wo o d y sp ec ies we re mo re insu en ce d by ag ricu ltu re inten sityin th e su rro u nd ing ma trix.5. S ynthesis.We us ed o ne o f th e larg est da ta sets ev er co llec ted in temp erate frag me n ted fo rests tob u ild, fo r th e Þ rsttime ,a stru ctur al mo de linco rp or ating allsu sp ec ted dr ivers of loca lplan t co mm u -n ities. We sh o we d th at th e n u mb er an d iden tity of loca lsp ec ies co ex istin g in su cc essio na lfrag me n tso f a fo rest me taco mm un ity at a give n time is co nt ro lled by a u n iqu e co mb ination o f int erac tin gloc al, lan ds ca pe an d histo rica l factor s. Pr eservin g the larg est, old est frag me n ts an d fav o ur ing sp e-cies mo ve me n ts in th e su rro un d ing ma trix is th e be st wa y to co ns erve fo rest sp ec ialists in ch an gingru ra llan ds ca pe s.Key-wor ds: agricu ltural intens ity, disper sal lim itation, habita t fragm entat ion, he dgerow net-work, metacommun ity, plant communi ty divers ity, specie sÐarea relationshi p, specie sÐtimerelationshi p, struc tural equati on modellingIntroduc tion

We evaluate the predictive power and generality of Shipley's maximum entropy (maxent) model of community assembly in the context of 96 quadrats over a 120-km2 area having a large (79) species pool and strong gradients. Quadrats were sampled in the herbaceous understory of ponderosa pine forests in the Coconino National Forest, Arizona, U.S.A. The maxent model accurately predicted species relative abundances when observed community-weighted mean trait values were used as model constraints. Although only 53% of the variation in observed relative abundances was associated with a combination of 12 environmental variables, the maxent model based only on the environmental variables provided highly significant predictive ability, accounting for 72% of the variation that was possible given these environmental variables. This predictive ability largely surpassed that of nonmetric multidimensional scaling (NMDS) or detrended correspondence analysis (DCA) ordinations. Using cross-validation with 1000 independent runs, the median correlation between observed and predicted relative abundances was 0.560 (the 2.5% and 97.5% quantiles were 0.045 and 0.825). The qualitative predictions of the model were also noteworthy: dominant species were correctly identified in 53% of the quadrats, 83% of rare species were correctly predicted to have a relative abundance of < 0.05, and the median predicted relative abundance of species actually absent from a quadrat was 5 x 10(-5).