Summary Species distribution models (SDMs) have become a standard tool in ecology and applied conservation biology. Modelling rare and threatened species is particularly important for conservation purposes. However, modelling rare species is difficult because the combination of few occurrences and many predictor variables easily leads to model overfitting. A new strategy using ensembles of small models was recently developed in an attempt to overcome this limitation of rare species modelling and has been tested successfully for only a single species so far. Here, we aim to test the approach more comprehensively on a large number of species including a transferability assessment. For each species, numerous small (here bivariate) models were calibrated, evaluated and averaged to an ensemble weighted by AUC scores. These ‘ensembles of small models’ (ESMs) were compared to standard SDMs using three commonly used modelling techniques (GLM, GBM and Maxent) and their ensemble prediction. We tested 107 rare and under‐sampled plant species of conservation concern in Switzerland. We show that ESMs performed significantly better than standard SDMs. The rarer the species, the more pronounced the effects were. ESMs were also superior to standard SDMs and their ensemble when they were evaluated using a transferability assessment. By averaging simple small models to an ensemble, ESMs avoid overfitting without losing explanatory power through reducing the number of predictor variables. They further improve the reliability of species distribution models, especially for rare species, and thus help to overcome limitations of modelling rare species.
Ensembles of Small Models (ESM) represent a novel strategy for species distribution modelling with few observations. ESMs are built by calibrating many small models and then averaging them into an ensemble model where the small models are weighted by their cross‐validated scores of predictive performance. In a previous paper (Breiner, Guisan, Bergamini, & Nobis, Methods in Ecology and Evolution, 6, 1210–1218, 2015), we reported two major findings. First, ESMs proved largely superior to standard models in terms of model performance and transferability. Second, ESMs including different modelling techniques did not clearly improve model performance compared to single‐technique ESMs. However, ESMs often require a large computation effort, which can become problematic when modelling large numbers of species. Given the appealing new perspectives offered by ESMs, it is especially important to investigate if some techniques yield increased performance while saving computation time and thus could be predominantly used for building ESMs. Here, we present results from a reanalysis of a subset of the data used in Breiner et al. (2015). More specifically, we ran ESMs: (1) fitted with 10 modelling techniques separately (in Breiner et al., 2015 we used only three techniques); and (2) using various parameter options for each modelling technique (i.e., model tuning). We show that ESMs vary in model performance and computation time across techniques, and some techniques are advantageous in terms of optimizing model performance and computation time (i.e., GLM, CTA and ANN). Including one of these modelling techniques could thus optimize computation time compared to using more computing‐intensive techniques like GBM. Next, we show that parameter tuning can improve performance and transferability of ESMs, but often at the cost of computation time. Parameter tuning could therefore be used when computing resources are not a limiting factor. These findings help improve the applicability and performance of ESMs when applied to large numbers of species.
Aims Phytosociological classification of fen vegetation (Scheuchzerio palustris‐Caricetea fuscae class) differs among European countries. Here we propose a unified vegetation classification of European fens at the alliance level, provide unequivocal assignment rules for individual vegetation plots, identify diagnostic species of fen alliances, and map their distribution. Location Europe, western Siberia and SE Greenland. Methods 29 049 vegetation‐plot records of fens were selected from databases using a list of specialist fen species. Formal definitions of alliances were created using the presence, absence and abundance of Cocktail‐based species groups and indicator species. DCA visualized the similarities among the alliances in an ordination space. The ISOPAM classification algorithm was applied to regional subsets with homogeneous plot size to check whether the classification based on formal definitions matches the results of unsupervised classifications. Results The following alliances were defined: Caricion viridulo‐trinervis (sub‐halophytic Atlantic dune‐slack fens), Caricion davallianae (temperate calcareous fens), Caricion atrofusco‐saxatilis (arcto‐alpine calcareous fens), Stygio‐Caricion limosae (boreal topogenic brown‐moss fens), Sphagno warnstorfii‐Tomentypnion nitentis (Sphagnum‐brown‐moss rich fens), Saxifrago‐Tomentypnion (continental to boreo‐continental nitrogen‐limited brown‐moss rich fens), Narthecion scardici (alpine fens with Balkan endemics), Caricion stantis (arctic brown‐moss rich fens), Anagallido tenellae‐Juncion bulbosi (Ibero‐Atlantic moderately rich fens), Drepanocladion exannulati (arcto‐boreal‐alpine non‐calcareous fens), Caricion fuscae (temperate moderately rich fens), Sphagno‐Caricion canescentis (poor fens) and Scheuchzerion palustris (dystrophic hollows). The main variation in the species composition of European fens reflected site chemistry (pH, mineral richness) and sorted the plots from calcareous and extremely rich fens, through rich and moderately rich fens, to poor fens and dystrophic hollows. ISOPAM classified regional subsets according to this gradient, supporting the ecological meaningfulness of this classification concept on both the regional and continental scale. Geographic/macroclimatic variation was reflected in the second most important gradient. Conclusions The pan‐European classification of fen vegetation was proposed and supported by the data for the first time. Formal definitions developed here allow consistent and unequivocal assignment of individual vegetation plots to fen alliances at the continental scale.
Changing land use has a major impact on lichen diversity. This study attempts to identify patterns or trends of lichen functional groups along a land use gradient, ranging from natural forests to open agricultural landscape. In eight countries, covering six main European biogeographic regions, lichen vegetation was assessed according to a standardized scheme. Data on reproductive, vegetative and ecological traits was compiled and relative species richness for all classes of all traits calculated. Relationships between the land use gradient and relative species richness of trait classes were analysed. Open and intensively managed landscapes harbour more fertile species while sterile species are relatively more important in forests. This finding is also supported by analyses of different classes of dispersal propagules. The importance of species with the principal photobiont Trebouxia s.l. increases linearly with intensification of land use. A converse pattern is revealed by species with Trentepohlia. Concerning substratum specialization only generalists show an effect along the land use intensity gradient. Their relative species richness decreases from landscapes dominated by forests to open agricultural landscape. A considerable decline in the rare lichen species richness as a result of land intensification is predicted.
Fig. 2. Spatial coverage of GrassPlot data from Morocco to Japan. Currently, the majority comes from sub-Mediterranean to hemiboreal Europe (black = multi-scale plots, grey = other plots). Current content v. 1.00 (January 2018) • 126 datasets • 198 data owners • 36 countries • 168,997 plots, among them 14,064 with data also for non-vascular plants • 66,000 0.01-m² plots, 17,206 1-m² plots, 5,520 10-(or 9-) m² plots, 2,545 100-m² plots • 2,797 nested-plot series (with at least 4 grain sizes)
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
customersupport@researchsolutions.com
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.
Copyright © 2025 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.