Community‐level vs species‐specific approaches to model selection

A topic of particular current interest is community‐level approaches to species distribution modelling (SDM), i.e. approaches that simultaneously analyse distributional data for multiple species. Previous studies have looked at the advantages of community‐level approaches for parameter estimation, but not for model selection – the process of choosing which model (and in particular, which subset of environmental variables) to fit to data. We compared the predictive performance of models using the same modelling method (generalised linear models) but choosing the subset of variables to include in the model either simultaneously across all species (community‐level model selection) or separately for each species (species‐specific model selection). Our results across two large presence/absence tree community datasets were inconclusive as to whether there was an overall difference in predictive performance between models fitted via species‐specific vs community‐level model selection. However, we found some evidence that a community approach was best suited to modelling rare species, and its performance decayed with increasing prevalence. That is, when data were sparse there was more opportunity for gains from “borrowing strength” across species via a community‐level approach. Interestingly, we also found that the community‐level approach tended to work better when the model selection problem was more difficult, and more reliably detected “noise” variables that should be excluded from the model.     

A meta‐analysis of species–abundance distributions

The species–abundance distribution (SAD) describes the abundances of all species within a community. Many different models have been proposed to describe observed SADs. Best known are the logseries, the lognormal, and a variety of niche division models. They are most often visualized using either species richness – log abundance class (Preston) plots or abundance – species rank order (Whittaker) plots. Because many of the models predict very similar shapes, model distinction and testing become problematic. However, the variety of models can be classified into three basic types: one that predicts a double S‐shape in Whittaker plots and a unimodal distribution in Preston plots (the lognormal type), a second that lacks the mode in Preston plots (the logseries type), and a third that predicts power functions in both plotting types (the power law type). Despite the interest of ecologists in SADs no formal meta‐analysis of models and plotting types has been undertaken so far. Here we use a compilation of 558 species–abundance distributions from 306 published papers to infer the frequency of the three SAD shapes in dependence of environmental variables and type of plotting. Our results highlight the importance of distinguishing between fully censused and incompletely sampled communities in the study of SADs. We show that completely censused terrestrial or freshwater animal communities tend to follow lognormal type SADs more often than logseries or power law types irrespective of species richness, spatial scale, and geographic position. However, marine communities tend to follow the logseries type, while plant communities tend to follow the power law. In incomplete sets the power law fitted best in Whittaker plots, and the logseries in Preston plots. Finally our study favors the use of Whittaker over Preston plots.     

Community‐level spatial structure supports a model of stochastic geometry in species‐rich shrublands

In some ecosystems a small suite of species can determine community‐level patterns of species richness by acting as either ‘accumulators’ or ‘repellers’; that is, the richness of the immediate neighbourhood of such species departs from that expected on the basis of a given null model. Using the individual species‐area approach, we evaluated community‐level spatial pattern in four species‐rich shrublands (two 40 × 40 m and two 30 × 30 m plots) by assessing the frequency of accumulator and repeller species and whether any such species were associated with specific life‐history characteristics. Few species departed from the expectations of the null spatial model that we used, although, at three of the four sites, accumulators were more common than repellers. Departures from the null model we assessed were most prevalent within just 1 m of focal individuals and were not consistently associated with specific life‐history traits. Model‐based clustering suggests that there are distinct sub‐communities in each of the four communities, but while internally spatially cohesive, these sub‐communities intermingle and their membership is not predictable from the life‐history traits of their constituent species. Comparable analyses in other species‐rich systems have also detected a similar absence of spatial interactions. The disturbance regime in the shrubland communities we consider is markedly different from those in the forest ecosystems where previous studies have been concentrated (recurrent fire versus infrequent gap‐phase dynamics) and resources are more limiting. Thus, our results provide further support for the generality of a model of stochastic geometry, likely underpinned by stochastic dilution effects, in species‐rich plant communities.     

Variation in species‐level plant functional traits over wetland indicator status categories

Wetland indicator status (WIS ) describes the habitat affinity of plant species and is used in wetland delineations and resource inventories. Understanding how species‐level functional traits vary across WIS categories may improve designations, elucidate mechanisms of adaptation, and explain habitat optima and niche. We investigated differences in species‐level traits of riparian flora across WIS categories, extending their application to indicate hydrologic habitat. We measured or compiled data on specific leaf area (SLA ), stem specific gravity (SSG ), seed mass, and mature height of 110 plant species that occur along the Colorado River in Grand Canyon, Arizona. Additionally, we measured leaf δ¹³C, δ¹⁵N, % carbon, % nitrogen, and C/N ratio of 56 species with C3 photosynthesis. We asked the following: (i) How do species‐level traits vary over WIS categories? (ii) Does the pattern differ between herbaceous and woody species? (iii) How well do multivariate traits define WIS categories? (iv) Which traits are correlated? The largest trait differences among WIS categories for herbaceous species occurred for SSG , seed mass, % leaf carbon and height, and for woody species occurred for height, SSG , and δ¹³C. SSG increased and height decreased with habitat aridity for both woody and herbaceous species. The δ¹³C and hence water use efficiency of woody species increased with habitat aridity. Water use efficiency of herbaceous species increased with habitat aridity via greater occurrence of C4 grasses. Multivariate trait assemblages differed among WIS categories. Over all species, SLA was correlated with height, δ¹³C, % leaf N, and C/N; height was correlated with SSG and % leaf C; SSG was correlated with % leaf C. Adaptations of both herbaceous and woody riparian species to wet, frequently inundated habitats include low‐density stem tissue. Adaptations to drier habitats in the riparian zone include short, high‐density cavitation‐resistant stem tissue, and high water use efficiency. The results enhance understanding about using traits to describe plant habitat in riparian systems.     

An attempt to restore a central European species‐rich mountain grassland through grazing

Abstract. This paper describes the effects of re‐establishing seasonal cattle grazing by 0.7 animal.ha^‐1 on vegetation in a long‐term abandoned, and partly degraded, semi‐natural mountain pasture in the ?umava National Park, Czech Republic. There was very uneven grazing intensity inside the locality, and grazing preference changed during the season: cattle grazed most of the time in productive but species‐poor Deschampsia cespitosa swards, but changed to a species‐rich Violion caninae stand in the middle of the summer. A species‐rich Carex rostrata community was only grazed at the end of the season. Species‐poor swards dominated by Nardus stricta and Carex brizoides were mainly used as resting areas. Both grazing and excluding from grazing had a negative effect on species diversity of the Deschampsia cespitosa swards. The soil seed bank contained only few species that are characteristic of mountain grassland communities, and seed dispersal of the target species by cattle dung was also found to be very limited. Thus both grazing and exclusion from grazing are probably of limited value for the restoration of species‐rich grasslands from species‐poor Deschampsia cespitosa swards in this case.     

Toward accurate species‐level metabarcoding of arthropod communities from the tropical forest canopy

相似文献   

Scaling phenology from the local to the regional level: advances from species‐specific phenological models

Plant phenology, the study of seasonal plant activity driven by environmental factors, has found a renewal in the context of global climate change. Phenological events, such as leaf unfolding, exert strong control over seasonal exchanges of matter and energy between the land surface and the atmosphere. Phenological models that simulate the start of the growing season should be efficient tools to predict vegetation responses to climatic changes and related changes in energy balance. Species‐specific phenological models developed in the eighties have not been used for global‐scale predictions because their predictions were inaccurate in external conditions. Recent advances in phenology modelling at the species level suggest that prediction at a large scale may now be possible. In the present study, we tested the performance of species‐specific phenological models in time and space, looking at their ability (i) to predict regional phenology when previously fitted at a local scale, and (ii) to predict phenological trends, linked to climate changes, observed over a long‐term. For that task we used an historical phenological dataset from Ohio from the late ninetieth century and an airborne pollen dataset from Ontario, Québec and Maryland from the late twentieth century. The results show that the species‐specific phenological models used in this study were able to predict regional phenology even though they were fitted locally. The reconstruction of a phenological time series over the twentieth century showed a significant advancement of 0.2 days per year in the date of flowering of Ulmus americana, but very weak trends for Fraxinus americana and Quercus velutina.     

Comparative trophic impacts of two globally invasive cyprinid fishes reveal species‐specific invasion consequences for a threatened native fish

相似文献   

Seabird species‐ and assemblage‐level isotopic niche shifts associated with changing prey availability during breeding in coastal Newfoundland

Shifting prey availability can lead to altered species interactions, indicated by variation in the dietary niche breadth and position of species within an assemblage. On the Newfoundland coast, annual inshore spawning migration of the dominant forage fish, Capelin Mallotus villosus, provides an excellent opportunity to investigate the influence of varying prey availability on dietary niche breadth and position among species. During June–August 2017, we investigated species‐ and assemblage‐level dietary responses to shifting Capelin availability of three Capelin‐eating, sympatrically breeding auk species, the Atlantic Puffin Fratercula arctica, Razorbill Alca torda and Common Murre Uria aalge. The diet of Leach's Storm Petrels Oceanodroma leucorhoa, which breed alongside the three auk species but are not known to rely on Capelin, was also examined to determine dietary shifts throughout breeding that were unrelated to Capelin availability. We quantified stable isotope ratios (δ¹⁵N, δ¹³C) in seabird blood components (plasma, cellular component) collected both before and after spawning Capelin arrived in the study area and compared isotopic niche breadth within a Bayesian framework. At the species level, auk trophic position increased and isotopic niche breadth narrowed after Capelin arrived, suggesting a more Capelin‐based diet. Simultaneously, trophic diversity of the auk assemblage, reflecting the extent of spacing among niches of species, decreased after spawning Capelin arrived inshore. Contrastingly, increased trophic position but broader isotopic niche breadth during higher relative to lower Capelin availability for Leach's Storm Petrel confirm that this species is probably not affected by the inshore arrival of Capelin, but instead that isotopic changes may be more related to a shift in breeding stage to chick‐rearing. Overall, our findings reiterate the importance of Capelin as a prey resource for breeding auks in coastal Newfoundland, but that the degree of reliance on Capelin varies among species, possibly allowing coexistence of these ecologically similar species. The findings highlight potential changing species interactions, such as increased competition, under declines in Capelin biomass.     

Determinants of species‐area relationships for marsh‐nesting birds

ABSTRACT To clarify the underlying causes of the species‐area relationship in marsh‐nesting birds, I studied eight freshwater tidal marshes of the Connecticut River that differed in area, degree of isolation, mudflat cover, water cover, tidal regime, and extent of individual plant communities. I measured these habitat variables on aerial infrared photos, and surveyed bird populations by mapping the distribution of all birds in marshes under 5 ha in area and establishing 50‐m radius plots in marshes over 5 ha. From surveys, I determined species richness, population densities, and total populations. Analysis revealed a positive relationship between species richness and area, but no correlation between area and habitat heterogeneity. Other habitat variables were poor predictors of species richness. The lack of a relationship between habitat and species richness appeared to be a consequence of most vegetation types present not being sufficiently distinct for birds to differentially associate with them. I also found no relationship between bird population density and area, suggesting that habitat quality in marshes did not improve with increasing size, and species evenness declined with increasing richness because greater richness was associated with the presence of more rare species. Larger marshes had more rare species, species with larger populations, and species with a minimum threshold area for occurrence. Thus, my results are consistent with theoretical predictions that larger populations are less prone to local extinction and, as individuals are added to a community, more rare species are present.