Continental risk assessment for understudied taxa post‐catastrophic wildfire indicates severe impacts on the Australian bee fauna

Abstract

The 2019-2020 Australian Black Summer wildfires demonstrated that single events can have widespread and catastrophic impacts on biodiversity, causing a sudden and marked reduction in population size for many species. In such circumstances, there is a need for conservation managers to respond rapidly to implement priority remedial management actions for the most-affected species to help prevent extinctions. To date, priority responses have been biased towards high-profile taxa with substantial information bases. Here, we demonstrate that sufficient data are available to model the extinction risk for many less well-known species, which could inform much broader and more effective ecological disaster responses. Using publicly available collection and GIS datasets, combined with life-history data, we modelled the extinction risk from the 2019-2020 catastrophic Australian wildfires for 553 Australian native bee species (33% of all described Australian bee taxa). We suggest that two species are now eligible for listing as Endangered and nine are eligible for listing as Vulnerable under IUCN criteria, on the basis of fire overlap, intensity, frequency, and life-history traits: this tally far exceeds the three Australian bee species listed as threatened prior to the wildfire. We demonstrate how to undertake a wide-scale assessment of wildfire impact on a poorly understood group to help to focus surveys and recovery efforts.We also provide the methods and the script required to make similar assessments for other taxa or in other regions.

Accounting for the neglected: Invertebrate species and the 2019–2020 Australian megafires

“…We show that, potentially, far more invertebrate than vertebrate species were affected by the 2019–2020 fires, with many more of these invertebrate species having all or most of their known range burnt. These results are consistent with, and provide a substantial expansion from, previously published studies on smaller subsets of the Australian invertebrate fauna (Dorey et al, 2021; Hyman et al, 2020). Furthermore, given that most Australian invertebrates are undescribed, hence not included in our assessment, it is likely that the number of invertebrate species affected by these fires is much higher than the tally we derived.…”

“…There have been some published accounts of the intersection of the 2019-2020 megafires on Australian invertebrates; however, these have been restricted thus far to assessing fire overlap on limited subsets of invertebrate fauna, such as: 553 species of Australian bees (<.2% of the Australian invertebrate fauna) (Dorey et al, 2021); one genus of spiny crayfish, Euastacus spp. (57 species, <.02% of the Australian invertebrate fauna), within an assessment that focused mainly on vertebrates (Legge et al, 2021(Legge et al, , 2022; 733 species across several invertebrate groups in New South Wales (.2% of the Australian invertebrate fauna) (Hyman et al, 2020); and 451 species listed as threatened under Australian national and state/territory legislation or by the IUCN (<.2% of the Australian invertebrate fauna) (Legge et al, 2021).…”

Making choices: prioritising the protection of biodiversity in wildfires

“…A further critical consideration should be irreplaceability (Pressey et al 1994): the extent to which an asset is either localised or has multiple occurrences, with prioritisation given to the former. Although the Black Summer wildfires were so extensive that they caused major population losses for many widespread species, impacts were most pronounced for some highly localised species, where fire impacts affected the entire population (Dorey et al 2021;Gallagher et al 2021;Moir 2021).…”

A globally synthesised and flagged bee occurrence dataset and cleaning workflow

“…The primary repositories that we sourced data from were (i) GBIF, (ii) SCAN, (iii) Integrated Digitized Biocollections (iDigBio), (iv) the United States Geological Survey (USGS), and (v) the Atlas of Living Australia (ALA). Additionally, we have sought out and incorporated smaller public or private data sources from (i) Allan Smith-Pardo (ASP), (ii) Robert Minckley (BMin), (iii) Elle Pollination Ecology Lab (EPEL 26 ), (iv) Bombus Montana (Bmont; Casey Delphia 27 ), (v) Ecdysis (Ecd 28 ), (vi) Gaiarsa, et al 29 (Gai), (vii) the Connecticut Agricultural Experiment Station (CAES 30,31 ), (viii), USDA ARS South-eastern USA (Parys), (ix) Eastern Colorado (Arathi Seshadri), (xi) Florida State Collection of Arthropods (FSCA), (xi) Armando Falcon-Brindis (Arm), and (xii) four more publicly available bee datasets from the literature (SMC, Bal, Lic, Dor, VicWam) [3][4][5][32][33][34][35][36][37][38][39] . These datasets were collated by directly engaging and collaborating with their owners, particularly where data gaps were perceived, but with a particular emphasis on the Americas.…”

“…Asterisks next to function names indicate that they are parallel-ready (can be run across multiple threads). 30,31 ), Ecdysis (Ecd 28 ), Gaiarsia (Gai 29 ), Elinor Lichtenberg (Lic 37 ), Victorian and Western Australian Museum (VicWam 5,39 ), and the United States Geological Survey (USGS) data. The size of each ring and inner linkage (chord) is relative to the number of occurrences.…”