key: cord-0295413-7p5z1g79 authors: Noël, Grégoire; Van Keymeulen, Violette; Barbier, Yvan; Smets, Sylvie; Van Damme, Olivier; Colinet, Gilles; Ruelle, Julien; Francis, Frédéric title: Nest aggregations of wild bees and apoid wasps in urban pavements: a “street life” to be promoted in urban planning date: 2021-12-16 journal: bioRxiv DOI: 10.1101/2021.12.15.472743 sha: f62b74561e87cd9a39e00dd0c0d82c67f3412a44 doc_id: 295413 cord_uid: 7p5z1g79 In the last 10 years, knowledges of wild bees and apoid wasps’ community dynamics have gained interest in urban ecology focusing on the availability of floral resources in cities. Although understudied, the urban environment impacts the conditions of their nesting sites. Recent observations in the Brussels-Capital Region (Belgium) showed that urban pavements can be a novel nesting opportunity for Hymenoptera ground-nesting species such as wild bees and apoid wasps. Here, using citizen science, we investigated the richness of ground-nesting species living under urban pavements, the preferences of the sidewalk joint size related to ground-nesting species size and for sidewalk type or for soils texture under the pavements on the nesting site selection. A total of 22 species belonging to 10 Hymenoptera families of wild bees and digger wasps with their associated kleptoparasites were identified on 89 sites in Brussels. Sandstone setts or concrete slabs with an unbound joint size around 1 cm were found to be best suitable urban pavements for the ground-nesting species. The soil texture under the pavement was highly sandy among our samples. Finally, we also suggest engineering management guidelines to support bee and wasp species nesting under urban pavement in highly urbanized areas. Such observations pave the way for much research in the field of urban ecology to conceive multifunctional pavement promoting biodiversity. capture insect or spider prey as predator to feed their offspring (Bitsch and Leclercq 1993) . In 84 Belgium, 199 species of apoid wasps were recorded and more than half nests under the 85 ground (i.e., 107 spp.) (Pauly 1999 ; Rasmont and Haubruge 2002) . Furthermore, some taxa Moreover, their ability to locate preys in large landscape is disrupted by the habitat 91 fragmentation caused by urban matrix (Kareiva 1987) leading to change in the community 92 structure of apoid wasps in cities (Christie and Hochuli 2009; Burkman and Gardiner 2014) . 93 However, to our knowledge, there is still no study mentioning nesting adaptation regarding to 94 urban matrix for apoid wasps. 95 Urbanization leads to a considerable increase in impervious surfaces (buildings, sidewalks, 96 roads…), which limit nesting opportunities on natural substrates (Cane 2005) and would lead 97 to competition for these sites among ground-nesting bees and apoid wasps. Nevertheless, 98 recent observations in the Brussels-Capital Region (BCR, Belgium) showed that urban 99 pavements have become a novel nesting opportunity for certain ground-nesting species (Pauly 100 2019a). A sandy mound indicates the presence of these ground-nesting species (Fig. 1a) . 101 Indeed, the drainage of flexible pavements, the recurrent use of sandy substrate under the 102 paving materials and the thermal capacity of the pavements could probably contribute to the 103 potential of pavements to host ground-nesting bees and wasps, providing them an interesting 104 shelter in an urban environment. The joint size could have a direct impact on the nesting 105 ability of the bees and wasps: when joint size is lower than their thorax size, it does not allow 106 digging a gallery. Based on the scientific literature on variables that influence nest site 107 selection in natural environments (Stephen 1960 (Stephen , 1965 of the jointing material, the type of pavement and the location of the nest on the pavement. As 112 an entry zone to the subsurface material, the nature of the jointing material defines the 113 hardness of the substrate and would therefore influence the ability of bees to tunnel into it, 114 acting as a filter in nest site selection. As a pioneering study in urban ecology of Hymenoptera ground-nesting species, we 116 addressed the following questions regarding wild bee and apoid wasp community according 117 to their nesting preferences up and under the pavements: (i) what is the ground-nesting 118 community living under the BCR pavements ? (ii) Is there a link between ground nesting 119 species size and joint size ? (iii) Does the pavements type have an impact on it ? (iv) How is 120 the soil texture under pavement selected by ground-nesting species ? In an attempt to answer 121 these questions, we developed a citizen science pipeline to identify their nesting sites 122 followed by an appropriate methodology to characterize the selected nesting sites. After 123 species identification, an analysis of associated edaphic preferences was performed for 124 pavements selection. Finally, the implications of our results were discussed regarding to the 125 new challenges in urban pavement design that should promote ground-nesting insects. Identification of the potential study sites 128 Only 4 nesting sites on the pavements had been identified and confirmed in BCR by Pauly in 129 6 for subsequent field validations, (iii) the date of the last observation of the nest, (iv) the 141 potential picture of the nest or the insect in presence to pre-validate that the observation took 142 place on a pavement and is related to our taxa of interest. Validation of the study sites 144 The dataset curation from the coding form was started in mid-March 2020 and continued 145 throughout the participatory survey until mid-July 2020. We removed observations unrelated 146 to taxa and location of our study sampling strategy (e.g., ants, cavity-nesting bees, outside of 147 BCR ...). We validated study sites on the field from April 05, 2020 to July 20, 2020 during specimens were stored in a freezer at -20°C directly after field survey. The joint size was measured on 6 nests randomly chosen within a site using a millimetre lath 167 perpendicular to the joints and passing through the nest entrance. When more than one 168 ground-nesting species was on site, we took measurements randomly on site, without 169 distinction between species. The joint structure variable was implemented in the database as a Sand of the mounds was collected from random 3-10 nest entrances (max. 50g) on pavements 178 as corresponding to the soil layer excavated by insects digging activity under pavements ( Fig. 179 1a). Regarding to preliminary results of the substrate texture under pavements, sandy mounds 180 seem to be a good proxy of soil texture (see in supplementary information S1). Laboratory data collection 182 All collected specimens were prepared for identification using Mouret et al., (2007) protocol. 183 We used several identification keys to define bee and wasp species (Bitsch and Leclercq (Table 1) . A post-hoc Tukey test with adjustment of multiple comparisons was applied 210 to compare mean pair of joint size. We also measured the species size as well as the inter- to detect whether there were similarities in particle size preferences the ground-nesting 219 species and between their respective families (as shown in Table 1) Schaerbeek form the pack head, while Molenbeek-Saint-Jean, Jette and Ganshoren completed 228 it (Fig. S1 ). There was no encoding from citizens of Saint-Josse ten-Noode and Koekelberg. Between April 5 and July 31, 2020, we characterized 89 sites throughout BCR that fullfilled 230 the validation criteria (Fig. 2) . Ixelles, Watermael-Boitsfort and Uccle municipalities were the 231 most sampled (Fig. S2 ). Species recorded 235 We collected 153 specimens belonging to 22 species including 11 solitary bee species, 9 236 apoid wasp species and 2 chrysid species (Table 1) . only 10 ground-nesting species (Fig. 3) . The average joint size for all species was 1.08 cm ± 252 0.57 cm with a maximum measured at 3.00 cm and a minimum at 0.20 cm. Details by species 253 were given in Table S1 . A significant difference among the mean joint size between different 254 ground-nesting species was observed (df = 9; F = 1.97; P = 0.041). However, after adjustment As some data were not taken into account due to misfit with our classification (e.g., hybrid 269 pavement …), a total of 79 sites were characterized. The joints were mostly unbound and 270 account for 80% of the sites encountered whereas 20% of the joints were characterized as 271 rigid and degraded, leaving openings for ground-nesting species to dig and nest (Fig. 5a) . As for grain size. The particle size analysis method did not allow us to distinguish between clay 288 and silt composition. The sandy fraction of the samples was always higher than 85% and the 289 silt and clay fractions were always lower than 10%, which placed all the samples from 290 mounds as sandy and homogeneous texture (Fig. S4) . their respective families (Fig. 6) . (Table 1) with their respective sandy mounds were more prone to be detected. While small species nesting within the pavements such as Lasioglossum spp. or the other 326 small-sized apoid species (Table 1) will be easily undetected because they were located in 327 little frequented streets, sheltering small nest-aggregations with invisible sandy mounds. We This species was observed on 4 different sites in BRC, which would indicate that it nests 384 commonly in BRC pavements and could therefore be the subject of a conservation project. Another finding of our study was that the investigated nest aggregation is species-mixed such 386 as C. arenaria and D. hirtipes in a large series of sites (Fig. S3 ). From our sample protocol, it 387 was difficult to observe nesting distinctions or sharing between ground-nesting species. implying that ground-nesting species did not have a preferential size according to their size. Probably, the morphological limit for Apoid to pass through a given diameter is the distance with an unbound jointing size around 1cm on an unbound foundation. However, this 518 pavement hosting ground-nesting species was only an archetype (Fig. 7) from the results of 519 our study to BCR specifications. We considered other alternatives for urban pavements that 520 would probably host urban insect community. For potential conservation, it is important to 521 provide a structure that is favourable for new nesting process. In general, common practices 522 recommend that joint size should be as small as possible, which should limit the access for the peripheral to the pavements as they can exist in BCR allotments (Fig. S6) . In addition, we recommend the implementation of localized campaigns to raise awareness of 542 urban ground-nesting bees and wasps among resident citizens in order to avoid nest 543 destruction due to a lack of knowledge of the ecology of these species as they are solitary and 544 not dangerous to humans. Similarly, it would be interesting to ensure continuous actions in 545 order to initiate and reinforce a change of paradigm with regard to the city's aesthetic criteria: "clean" and "unfunctional" pavements in human-centered city vs. "multifunctional" This study pointed out that pavements, previously perceived as inappropriate for biodiversity, 571 could serve as refuge for some insect populations. Therefore, by modulating pavement design 572 and proposing alternative construction models, we could turn the city into a more welcoming 573 place for biodiversity. 574 575 The accuracy of citizen science data: A 577 quantitative review Bumble bees (Bombus spp) along a gradient of 580 increasing urbanization Eco-evolutionary dynamics in an urbanizing planet Global urban signatures of phenotypic change 585 in animal and plant populations The effects of seed mix diversity on soil conditions 588 and nesting of bees in prairie restorations Nesting habitat of ground-nesting bees: a review mapview: Interactive Viewing of 592 Spatial Data in R Biodiversity in the city: key challenges 594 for urban green space management Supporting bees in cities: How bees are influenced by local and 597 landscape features Managing urban areas for insect 599 pollinators: As town and cities continue to grow how can land managers help insect 600 pollinators in urban areas? Living With Environmental Change Policy and Practice 601 Notes Nesting Biology of the Solitary Digger Bee Habropoda 604 depressa ( Hymenoptera : Anthophoridae ) in Urban and Island Environments Charte sur les revêtements piétons en Région 607 de Hyménoptères Sphecidae d'Europe Occidentale 609 Hyménoptères Sphecidae d'Europe Occidentale Second Edi. Fédération Française des Sociétés de Hyménoptères Sphecidae d Fédération Française des Sociétés de Biologie vybraných druhů kleptoparazitických včel [Biology of selected cuckoo bee species (Hymenoptera: Apocrita Cahier des charges type relatif aux voiries en Région 618 de Functional ecology of wild bees in cities: towards a better 620 understanding of trait-urbanization relationships The effect of urbanization on ant abundance and 623 diversity: A temporal examination of factors affecting biodiversity Urban greenspace composition and landscape context 626 influence natural enemy community composition and function Bees, Pollination, and the Challenges of Sprawl. In: Nature in Fragments The 629 Legacy of Sprawl Landscaping pebbles attract nesting by the native ground-nesting bee 631 Halictus rubicundus (Hymenoptera: Halictidae) Soils of Ground-Nesting Bees (Hymenoptera: Apoidea): Texture, Moisture Cell Depth and Climate Complex responses within a desert bee guild 636 (Hymenoptera: Apiformes) to urban habitat fragmentation Responses of wasp communities to urbanization: Effects on 639 community resilience and species diversity Evaluation of nest-site selection of 642 ground-nesting bees and wasps (Hymenoptera) using emergence traps Impact of urbanization on predator and parasitoid 645 insects at multiple spatial scales Revêtements modulaires en pierre naturelle R95 Code de bonne pratique pour la conception et l'exécution de revêtements en 649 pavés de béton The Solitary Bees Pour la conservation des Hyménoptères aculéates en The little things that run the world revisited: A 654 review of ant-mediated ecosystem services and disservices (Hymenoptera: Formicidae) Accounting for imperfect detection and survey bias in statistical analysis 657 of presence-only data Urban bumblebees are smaller and more 661 phenotypically diverse than their rural counterparts Changes in integument structure during 664 the imaginal molt of honey bee Field Guide to the Bees of Great Britain and Ireland, 1st edn Urbanisation drivers and underlying 668 mechanisms of terrestrial insect diversity loss in cities Ant biodiversity and its relationship to ecosystem functioning: A review Decreasing abundance, increasing diversity and 673 changing structure of the wild bee community (Hymenoptera: Anthophila) along an 674 urbanization gradient Use of human-made nesting structures by wild 677 bees in an urban environment L'importance biologique et la conservation des 680 The proportion of impervious surfaces at 682 the landscape scale structures wild bee assemblages in a densely populated region Capitalizing on opportunistic data for 685 monitoring relative abundances of species Effects of spatial autocorrelation and imperfect detection on species 688 distribution models In: Rote Liste des 691 gefährdeten Tiere und Pflanzen in der Bundesrepublik Deutschland The city as a refuge for insect pollinators Influence of Nesting Characteristics on Health of Wild Bee 695 Urban drivers of plant-pollinator interactions Human influences on rates of phenotypic 700 change in wild animal populations Contrasting the effects of natural 703 selection, genetic drift and gene flow on urban evolution in white clover (Trifolium 704 repens) Habitat fragmentation and the stability of predator-prey interactions 2020) factoextra: Extract and Visualize the Results of Multivariate Pollinator size and its consequences: Robust 710 estimates of body size in pollinating insects Trend estimation in populations with imperfect 713 detection {FactoMineR}: A Package for Multivariate Analysis The nesting habits of solitary bees : A comparative study 718 The Bees of the World Monographic revision of the Melittidae s Mise en place d'une collection d'abeille 723 (Hymenoptera, Apoidea) dans le cadre d'une étude de la biodiversité Leveraging the power of place in citizen 725 science for effective conservation decision making Methods for creating bare ground on farmland in 728 UK, and their effectiveness at recruiting ground-nesting solitary bees Soil characteristics of nesting sites of native bees associated with low-732 bush blueberry in Maine Eco-evolutionary dynamics: Intertwining ecological and 734 evolutionary processes in contemporary time Catalogues des hyménoptères aculéates de Belgique Contribution à l'inventaire des abeilles sauvages de la Région de Bruxelles-739 Capitale et de la Forêt de Soignes (Hymenoptera: Apoidea) Abeilles de Belgique et des régions limitrophes (Insecta: Hymenoptera: 741 Apoidea). Famille Halictidae Nest sharing and provisioning activity 743 of females of the digger wasp, Cerceris rubida (Hymenoptera, Crabronidae) Abiotic and biotic factors influencing nest-site selection by 746 Halictus rubicundus, a ground-nesting halictine bee 748 R Core Team (2020) R: A Language and Environment for Statistical Computing Point process models for presence-only analysis A roadmap for urban evolutionary ecology Modelling occurrence and abundance of species when 756 detection is imperfect Phylogenomic analysis of Apoidea sheds new 759 light on the sister group of bees Evaluating nesting microhabitat for ground-nesting bees using 762 emergence traps Invasion, competition, and biodiversity loss 765 in Urban ecosystems Artificial Bee Beds for the Propagation of the Alkali Bee, Nomia 767 melander1 Effects of Soil Moisture on Survival of Prepupae of the Alkali Bee1 Urbanisation is associated with shifts in 771 bumblebee body size, with cascading effects on pollination Urban areas as hotspots for bees and 774 pollination but not a panacea for all insects World Urbanization Prospects: The Gastheren van enige soorten Sphecodes in Drenthe (Hymenoptera: Apidae) Observations relatives à la biologie et 781 la nidification de quelques abeilles sauvages psammophiles d'intérêt en Wallonie. 1. 782 Observations printanières Model-Based Control of Observer Bias for the 787 Analysis of Presence-Only Data in Ecology How urbanization is driving pollinator 789 diversity and pollination -A systematic review 2016) ggplot2: Elegant Graphics for Data Analysis Thermal ecology, size effects, and the origins of communal behaviour in 793 Cerceris wasps Zur Bionomie der Sandbiene Andrena barbilabris (Kirby 1802) und ihrer 795 Kuckucksbienen Nomada alboguttata Herrich-Schäffer 1839 und Sphecodes pellucidus 796 Smith 1845 Nest site preference and success in a gregarious Effects of urbanization on Neotropical wasp and 801 bee assemblages in a Brazilian metropolis This study was made possible with the financial support from Bruxelles Environnement (BE) 806 through the STREETBEES project 2019G0250. GN & FF have received research support 807 from BE Thanks to Julie Bonnet (ULiège, Functional and Evolutionary Entomology) for the insect 819 preparation and some identifications. The R code and the data used in this paper are available 820 at https://github.com/gregnoel/Street_Apoid.git 809 The authors have no relevant financial or non-financial interests to disclose