This perspective addresses the nature of the past and current relationships between the ecological and biogeographical regionalization of plants and plant communities. It uses four examples (two related to continental scale and two related to regional/local scale) to document the cases of cross-pollination between both approaches in the past and the need for concerted use of both ecological and distributional data in formulating robust regional spatial classification systems of biotic assemblages.
The continuous development of statistical tools applied to ecology has contributed to great advances for modelling species' niches and distributions from opportunistic observations. However, as these observations are subject to biases caused by spatial variation in sampling effort, ecological niche models (ENMs) are also frequently biased. Among several bias correction methods that have been proposed, spatial filtering—imposing a minimum distance between occurrences—is widely used, yet lacks clear guidelines for choosing the filtering distance. Here, we aimed to explore the impact of spatial filtering distances on the performance of ENMs.
Europe.
Virtual species.
We applied ENMs to two virtual species with contrasting levels of specialisation, across a spectrum of modelling conditions, bias types and sample sizes.
Models applied to the specialist species had on average a lower performance than those applied to the generalist species. Using a biased sample reduced model performance, especially when the bias was strong, and when the sample size was large. In many cases, spatial filtering failed to improve model performance or even reduced it. We did find an improvement for the generalist species modelled with large and strongly biased datasets. However, there was no optimal filtering distance, as this improvement was linearly and positively associated with filtering distance. Moreover, because the initial bias was strong and the filtered dataset became very small, the resulting models had only very low accuracy.
Our results suggest that there is no optimal filtering distance for dealing with sampling bias in ENMs, and that spatial filtering never improves model performance enough to draw accurate predictions. We therefore recommend spatial filtering to be employed cautiously, only when enough data are available, and bearing in mind that its effectiveness remains highly uncertain.
Sampling bias and gaps have a direct influence on the perceived patterns of biodiversity, hence limiting our ability to make well-informed decisions about biodiversity conservation. Yet most methods either disregard or underestimate the effects of sampling bias and gaps in modelling biodiversity patterns. Our objective is to test the sensitivity of commonly used methods for modelling biodiversity dimensions (richness, endemism, and beta diversity) to sampling bias and collection gaps, and as a way to mitigate those effects we introduce a novel approach that employs the sampling effort to minimize the effects of collection bias and gaps in biodiversity models.
South America.
Here, we use controlled simulations of virtual species distribution and sampling effort to test the sensitivity to sampling bias and collection gaps by commonly used methods, that is, species distribution models (SDMs), spatial interpolation (SI), and environmental prediction (EP), for estimating species richness, endemism, and beta diversity. Our research contributes to advancing biodiversity modelling by introducing a novel approach, named uniform sampling from sampling effort (USSE), that employs the sampling effort to minimize the effects of collection bias and gaps.
EP with USSE has proven effective in accurately predicting species richness, especially in scenarios in which the sampling effort does not coincide with the biodiversity niches. It outperformed SI and SDMs. The latter performed poorly, yielding the lowest predictive score. In estimating endemism and beta diversity, all methods yielded similar results, without statistically significant differences. For estimating beta diversity, the generalized dissimilarity model proved to be a robust method, even in face of biased sampling. Controlled simulations are key to testing biodiversity methods. These tests can isolate confounding factors inherent to real-world data, enabling robust methodological assessments. Although fieldwork and curation of collections must remain indispensable, novel biodiversity methods could help overcome the limitations of sampling biases, helping expedite conservation actions much needed.
The distribution of within-species lineages has been affected by Quaternary climate changes, and population differentiation has been influenced by species life history traits. We investigated whether the distribution of individual mitochondrial genetic lineages reflects the constituent species' traits. Using the functionally diverse group of butterflies, we examined which lineages are present in Central Europe, an important suture zone.
Czech Republic and Western Palearctic.
A total of 140 butterfly species.
We sequenced DNA barcodes (cytochrome c oxidase 1) (959 sequences) of the entire Czech Republic butterfly fauna and used Barcoding of Life Data System to visualise the species' biogeographic patterns across Europe. We categorised the distribution patterns of lineages inhabiting the Czech Republic, and used multivariate statistics to interpret these categories by the butterflies' habitats, life history traits and threat levels.
Open habitat dwellers with specialist traits belonged to Eastern, Southern and temperate lineages. Habitat generalists and woodland dwellers belonged to the Western lineage, formed several lineages or displayed low genetic diversity; they often developed on woody plants, were large-winged and had long flight periods. The most threatened species were the specialists of Southern, Eastern and temperate lineages.
The distribution of lineages in Central Europe reflects the history of Quaternary ecosystems: during cold periods of the Pleistocene, the diverse open habitats prevailed and species could expand westwards. Such species also suffer the most under the current anthropogenic habitat alteration. On the other hand, the mobile generalists and woodland dwellers expanded to Central Europe during the Holocene. Our approach of linking the distribution of lineages with species traits can be transferred to other study systems, and we show that DNA barcoding of under-sampled areas represents a powerful tool for discovering the driving forces of biogeography.
On the cover: Pleroma stenocarpum (Melastomataceae), a key tree species in secondary Yungas montane humid forests, La Paz, Bolivia. Photo credit: Alfredo Fuentes.
Answering many fundamental and applied scientific questions relies on accurate geographic range maps for species, such as those compiled by experts working with the International Union for Conservation of Nature (IUCN). However, these maps are resource demanding to produce and only available for a limited number of organisms. Here, we test to what extent standardized, data-driven methods based on publicly available occurrences can reproduce expert-based IUCN range maps and the macroecological conclusions drawn from them.
Global.
Present.
Birds.
We estimated the geographic ranges for 7385 non-marine bird species which either were non-migratory or had spatially connected breeding and wintering ranges from publicly available, georeferenced point occurrences. We then quantified the spatial overlap between these range estimates and the IUCN expert-derived range estimates. Finally, we compared global species richness patterns and the environmental correlates that emerge from both approaches.
We find that range estimates based on point occurrence records overlap on average 52% with expert range estimates for the same species. The global species richness patterns estimated under both approaches are overall similar but show local and regional differences, for example, in the tropical Andes of northern South America and the Central Arc region of Africa. The estimated global drivers of richness are similar.
Expert-derived estimates of species distributions are not reproducible by data-driven approaches relying on currently available public records, even for well-documented taxa such as birds. However, these discrepancies do not substantially change our macroecological understanding of global drivers of bird diversity.
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：