National‐scale predictions of plant assemblages via community distribution models: Leveraging published data to guide future surveys

Abstract

1. Species distribution models (SDMs) have been widely used to create maps of expected species incidence, often using citizen science (CS) occurrence data as inputs. Environmental policy is informed by knowledge of community distributions, but there have been fewer attempts to utilise the potential of community distribution models (CDMs) to predict these. Many countries have vegetation community classification systems which include phytosociological information on individual species.Within Great Britain, the National Vegetation Classification (NVC) is the primary standard for vegetation communities, and while maps have been produced at regional scales, cost-effective techniques are required for national scales.2. Published NVC occurrence records of 22 upland NVC communities in England and Wales were used as observed occurrences (presence-only data). Predictors for the CDMs were enhanced vegetation index (EVI), elevation, slope, aspect, temperature and rainfall. Five modelling methods were investigated: generalised linear models (GLMs), support vector machines (SVMs), random forests (RFs), maximum entropy (MaxEnt) and maximum likelihood (MaxLike). Model quality was assessed via bootstrapping via area under the curve (AUC), true skill statistic (TSS) and Kappa index.3. There were only small differences in the accuracy of the models (median TSS model accuracy 0.742; range 0.280 to 0.873) with RF models the best overall CDM method. Across all NVC communities, summer and winter maximum temperatures and annual rainfall were the most important predictor variables.NVCs with spatially disjunct distributions in both lowlands and uplands, or that responded to localised management or environmental conditions, were poorly predicted.4. Synthesis and applications. Vegetation communities can be reliably predicted at large spatial scales using CDMs from extant datasets. Management practitioners

Synergistic use of Sentinel-2 and UAV-derived data for plant fractional cover distribution mapping of coastal meadows with digital elevation models

“…Secondly, the RF algorithm used in this study assessed the accuracy of non-parametric-based regression models to predict the distribution of plant communities, as commonly used in Earth observation studies (Ferreira et al, 2022). In the literature, random forest is acknowledged as a robust algorithm for plant distribution at large scales (Maxwell et al, 2018;Butler and Sanderson, 2022), despite the challenges in its interpretability (Simon et al, 2023). RF averages the predictions of individual trees, a process that contributes to the model's robustness and ability to generalize; however, it underpredicts samples in either of the extremes, contributing to the model's uncertainties (Kuhn and Johnson, 2013).…”

“…Globally, coastal wetlands provide a wide range of ecosystem services including flood and wave attenuation (Möller et al, 2014), estuarine filtration (Celis-Hernandez et al, 2022;de Lacerda et al, 2022), biodiversity maintenance (Sutton-Grier and Sandifer, 2019), and climate mitigation through carbon sequestration and storage (Martinetto et al, 2023;Maxwell et al, 2023). However, these ecosystems are under threat from a range of stressors including climate change (Ward et al, 2016b;Mafi-Gholami et al, 2019;Ward, 2020), pollution (Celis-Hernandez et al, 2020;Li et al, 2021) and direct losses through conversion to other land uses (de Lacerda et al, 2021).…”

Modeling the rarest of the rare: a comparison between multi‐species distribution models, ensembles of small models, and single‐species models at extremely low sample sizes

“…Although there are a variety of situations that give rise to 'rarity' (Rabinowitz 1981), as the term is used here, 'rare' species have few known occurrences and are especially challenging to model (Jeliazkov et al 2022). Indeed, many critically endangered species have very few occurrence records (Lomba et al 2010, Zizka et al 2018) and low sample sizes have been identified as one of the factors most often found to reduce model accuracy (Stockwell and Peterson 2002, Hernandez et al 2006, Wisz et al 2008, Thibaud et al 2014, Santini et al 2021, Butler and Sanderson 2022. The suggested minimum numbers of occurrences range from approximately 5 to > 200 for 'traditional' single-species SDMs (Wisz et al 2008, van Proosdij et al 2016, Santini et al 2021.…”

The Habitat Map of Switzerland: A Remote Sensing, Composite Approach for a High Spatial and Thematic Resolution Product

“…Natura 2000 habitat types were mapped nationwide for Germany with MODIS imagery [17], and WorldView2/3 imagery was used to develop the Hong Kong habitat map [18]. In contrast, distribution modelling approaches based on topographical, climate, geological and land use predictors rather than remote sensing information have been successfully applied in Norway and Britain [19,20]. These maps differentiate a limited number of habitat types/classes, e.g., 17 classes for England [14], 18 for Germany [17], 31 for Norway [19] and 21 for Hong Kong [18], and either have higher uncertainty/less detail around areas with finely resolved habitat mosaics (i.e., in urban areas, [14]) or incorporate higher resolution ancillary data in a composite approach [18].…”