Using open-source biodiversity data to predict present and future macrofungal diversity and distribution in Europe

Exploring species distribution patterns and understanding the environmental factors determining their distribution are essential for biodiversity conservation, especially in the context of global change. However, this knowledge is still lacking for many species, particularly at large spatial extents. Fungi, as one of the most diverse groups of organisms, play key roles in regulating global carbon and nutrient cycling. However, compared to plants or animals, studying the diversity and distribution of fungal species at large spatial extents is rather rare, partly due to the cryptic characteristics of fungal species, which result in many challenges with sampling. The ongoing digitization of natural history museum collections and the development of citizen science means more biodiversity data are becoming available and published by various cloud platforms, such as Global Biodiversity Information Facility (GBIF). Furthermore, the development of species distribution models and the availability of spatial environment data have further stimulated the use of opensource biodiversity data to inform governments and societies about the status of biodiversity and the impact of development. However, doubts about sampling bias, such as spatial bias and taxon bias of these open-source biodiversity data, have raised questions about the use of these data when drawing inferences about fungal diversity and distribution.

By using macrofungal occurrence data integrated by Global Biodiversity Information Facility (GBIF) in Europe, this thesis aims to investigate the utility of open-source biodiversity data for understanding macrofungal diversity and distribution and to predict macrofungal distribution patterns under both now and future climate change over large spatial extents, using species distribution models. The research in this thesis firstly examined three widely used bias correction methods, i.e., geographical filtering of occurrence data, environmental filtering of occurrence data, and weighted background data selection, in predicting diversity patterns of macrofungal species. The results showed that the weighted background data selection method is an effective approach to mitigate the effects of sampling bias, which is useful for predicting macrofungal species diversity and distribution at a large spatial extent using MaxEnt, especially when there are species with a small sample size or with an unknown sampling effort. Secondly, using the weighted background data selection method, this thesis investigated whether long-term accumulated open biodiversity data could be used to reveal macrofungal diversity patterns at a large spatial extent. The results showed that the cumulative number of macrofungal species stabilized into distinct distribution patterns with localized hotspots of predicted macrofungal diversity with sampling efforts greater than approximately 30 years. Thirdly, this thesis predicted macrofungal diversity and distribution patterns and explored the determining factors at the continental level using long-term accumulated macrofungal data and weighted background data selection method in SDM. Analysis of the modelling results showed that eastern Denmark and southern Sweden are biodiversity hotspots for both functional groups of macrofungal species. Tree species and human disturbance (i.e., the human footprint index) were found to be the two most important predictor variables explaining the distribution of ectomycorrhizal and saprotrophic macrofungi. Lastly, this thesis predicted macrofungal response under future climate and tree distribution change. Overall, the models projected that large areas would exhibit increased macrofungal species richness under future climate change. However, the projected tree species distribution could restrict future macrofungi distributional shifts.

This thesis contributes insights into the utility of open-source biodiversity data for understanding species diversity and distribution at large spatial extents. Our findings also provide information about macrofungal diversity and distribution patterns under current and future scenarios at a large spatial extent. Furthermore, the thesis highlights the importance of integrating open-source biodiversity data using SDMs and spatial data layers of geophysical variables to implement the post-2020 global biodiversity framework of the Convention on Biological Diversity and the 2030 EU Biodiversity Framework.