Range expansion of the Asian native giant resin bee Megachile sculpturalis (Hymenoptera, Apoidea, Megachilidae) in France

Abstract In 2008, a new species for the French bee fauna was recorded in Allauch near Marseille: the giant resin bee, Megachile sculpturalis (Smith, 1853). This was the first European record of this species that is native to East Asia. To our knowledge, it is the first introduced bee species in Europe. Here, we provide an overview of the current distribution of M. sculpturalis in France and we describe the history of its range expansion. Besides our own observations, information was compiled from literature and Internet websites, and by contacting naturalist networks. We collected a total of 117 records (locality × year combinations) for the 2008–2016 period. The geographical range of M. sculpturalis has extended remarkably, now occupying a third of continental France, with the most northern and western records located 335 and 520 km from Allauch, respectively. Information on its phenology, feeding, and nesting behavior is also provided. We report several events of nest occupation or eviction of Osmia sp. and Xylocopa sp. individuals by M. sculpturalis. Our results show that M. sculpturalis is now well established in France. Given its capacity to adapt and rapidly expand its range, we recommend amplifying the monitoring of this species to better anticipate the changes in its geographical range and its potential impacts on native bees.

& Felicioli, 2014). The increasing use of these structures as a conservation tool for bees (MacIvor & Packer, 2015) could have promoted its spread.
Due to their crucial role as pollinators, bees are perceived to be beneficial and, comparatively to other biological groups, few concerns have been paid to the introduced bees. Nevertheless, in Oceania and North and South America, many examples of bee introductions and their subsequent deleterious consequences for native bee and plant species have been described in the literature (Geslin et al., 2017;Goulson, 2003). Introduced bees could either compete directly for nesting or feeding resources or indirectly by modifying the whole plant-pollinator network through enhancing the reproduction of exotic plant species (e.g., Lupinus arboreus in Tasmania subsequently to the invasion of Bombus terrestris, Stout, Kells, & Goulson, 2002).
Regarding M. sculpturalis, competition for nesting resources and aggressive behaviors toward other bee species have been previously reported. Indeed, in the United States, Laport and Minckley (2012) and Roulston and Malfi (2012) described aggressive evictions and occupations of nests of a native bee species (Xylocopa virginica L.) by

M. sculpturalis.
In contrast with North and South America, where many introduced bees have been detected, bee introductions have been rare in European countries. To our knowledge (Goulson, 2003;Russo, 2016), M. sculpturalis is the first introduced bee species in Europe, where it has been detected for the first time in 2008. The observation occurred in Allauch near Marseille, France (Vereecken & Barbier, 2009). Then, the species was reported in Italy in 2009 (Quaranta et al., 2014), in Switzerland in 2012 (Amiet, 2012), in Germany in 2015 (Westrich, Knapp, & Berney, 2015), and in Austria in 2017 (P. Westrich, unpublished data 2017).
Since the first observation in 2008, no review of the range expansion of M. sculpturalis in France has been reported. Here, we gathered all available data on the presence of M. sculpturalis in France. We documented its nesting and feeding behaviors, and the interactions with French native bees. We discuss avenues and potential consequences of this invasion and present monitoring measures.

| Megachile sculpturalis
Bee identification at the species level usually requires a high level of taxonomic expertise and is mostly impossible without collecting specimens and identifying them using both relevant literature and large reference collections. However, in our studied area, M. sculpturalis can be easily differentiated from other bee species: it is one of the largest bee species in France, with a size range from 22 to 27 mm long for females and from 14 to 19 mm long for males (Paiero & Buck, 2003), it has infuscated wings, and its thorax is covered with orange hair (Figure 1). Therefore, this species is easily recognizable for an informed naturalist in the field and in pictures.

| Data gathering
In order to assess the current distribution as well as the history of the range expansion of M. sculpturalis in France, we looked for all available occurrence data. Our study covered the 2008-2016 period, and we gathered data until 31 October 2016.

| RESULTS
We collected 117 records, that is, locality × year combinations, from 70 different observers. The two main sources of data were the "Apoidea-Gallica" forum and naturalists we directly contacted (respectively, 28.2% and 19.6% of records, Table 1). Then, the websites "Le Monde des insectes" and "Spipoll" represented, respectively, 17.1% and 12.8% of records. Overall, M. sculpturalis was recorded in 72 French cities ( Figure 3 and Table 2). In some cities, the species was recorded at several localities. The total number of localities was 83 ( Table 2).   -Ponel, et al., 2016;Westrich et al., 2015). Both studies suggested that the larval pollen provision contained a majority of pollen from S. japonica.
We compiled 39 nesting events (locality × substrate combinations) in 35 different localities. In 26 cases, nesting occurred in a humanmade nesting structure ("bee hotels") ( Figure 4). Stems were used in 12 cases (Arundo sp. were used four times and Sambucus sp. were used twice). Logs drilled with holes were used in nine cases (Quercus suber and Pinus sp. were both recorded once). For the remaining "bee hotel" cases, we did not have information about the type of cavity used. In 11 cases, nests were located in an old tree. The species Quercus pubescens and Quercus ilex were both reported twice, and we did not have information about the tree species used for the other events. In two cases, nesting occurred in a wooden beam.
Events of nest occupation or eviction of native wild bees from their nests were recorded in four localities (two times for both Osmia sp. and Xylocopa sp.).

| DISCUSSION
Since its first record in 2008 (Vereecken & Barbier, 2009), the geographical range of M. sculpturalis has extended remarkably, now occupying a third of continental France. In addition, M. sculpturalis is now present in Austria, Italy, Germany, and Switzerland and is therefore well established in mainland Europe. Several bee species have been introduced, accidentally or deliberately, outside their native range due to human activities, mainly in North and South America and in Oceania (Goulson, 2003;Russo, 2016). In the northeastern USA, for example, Bartomeus et al. (2013) identified 20 bee species not native to the USA. To our knowledge, M. sculpturalis is the first introduced bee species in France and, more globally, in Europe.
Our study illustrates the power of citizen science, defined as the involvement of volunteers in research (Dickinson, Zuckerberg, & Bonter, 2010), to assess the ongoing range expansion of M. sculpturalis. Citizen science offers several advantages to collect data on species distribution, such as an extension of spatial and temporal sampling effort (including data collection on private lands such as gardens) and time and cost reduction (Dickinson et al., 2010). Citizencollected data can make a major contribution to understand changes in species' distributions and, more specifically, biological invasions (see Kamenova et al.,  for the study on bee species recently introduced to Australia). In our study, the data we collected fall into the "opportunistic data" category, that is, data collected without standardized protocol and sampling effort. Interest of such datasets is often limited by the lack of "absence data" (e.g., Sequeira, Roetman, Daniels, Baker, & Bradshaw, 2014).
Although some entomologists contacted us to report the nonobserva-  Megachile sculpturalis is known to develop aggressive behavior toward other species and compete for their nesting sites (Laport & Minckley, 2012;Roulston & Malfi, 2012). We also report several events of nest occupation or eviction of Osmia sp. and Xylocopa sp.
individuals by M. sculpturalis. Few concerns are generally paid to bee invasions due to their crucial role as pollinators. But our study shows that competition for nesting sites between M. sculpturalis and native bee species could occur. In France, species such as Xylocopa spp., Lithurgus spp., Osmia spp., Megachile lagopoda, and some Anthidium spp. could be negatively affected. Moreover, introduced bees may also spread diseases to native bees (e.g., Graystock, Blane, McFrederick, Goulson, & Hughes, 2015;Singh et al., 2010).
Taken together, M. sculpturalis presence and spread may have deleterious consequences for native bees and its progression should be carefully monitored. Its high detectability (large species occurring in many habitats, including urban areas) and its easy identification make this species particularly appealing and appropriate for citizen science.
We suggest continuing and amplifying the monitoring of the species.
In parks and gardens, the visual inspection of S. japonica flowers and bee hotels could provide an effective means of detecting the species in an area but prospecting in natural areas (e.g., forests) should not be neglected in order to obtain the most accurate picture of the species' spread and its impacts.

ACKNOWLEDGMENTS
We are grateful to all the persons that provided records of Megachile sculpturalis in France or that helped us to get records: Romain Baghi,