Circling back:Vespulaspp. re-invasion after knockout from volcanic ash fall in Patagonia

2 1) Effective management of invasive species requires understanding re-invasion dynamics, 3 to prevent control failures, refine methodologies, and enhance protocols. Re-invasion is 4 the process by which eradicated or naturally disappeared exotic species establish 5 themselves again, in previously occupied areas. Re-invasion often bypasses initial 6 invasion phases such as transport and introduction. During re-establishment, changes in 7 the invader, the invaded environment, and species interactions can influence the 8 success of invasion. 9 2) We investigated re-invasion patterns of Vespula germanica and Vespula vulgaris wasps 10 in NW Patagonia (Argentina) following a significant natural disturbance: the 2011 11 Puyehue-Cordón Caulle volcanic eruption, which caused unprecedented arthropod 12 defaunation. By establishing 23 sampling sites in heavily affected regions, we conducted 13 sampling events over three years (2012-2014). 14 3) Our findings show that re-invasion of both Vespula spp. occurs fast, with minimal 15 interspecific competition among them observed. The proximity to urban centers, acting 16 as refuges, further facilitated their establishment. 17 4) This unique case study highlights the adaptability of these invasive wasps in the face of 18 extreme environmental disruptions. By describing the importance of considering re-19 invasion routes and refuges, this research may help develop more effective 20 management strategies to control Vespula populations. 21

22 Biotic interaction, Invasive species, social wasps, Vespula germanica, Vespula vulgaris, 23 yellowjacket. 24 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted December 10, 2024. ; https://doi.org/10.1101/2024.12.08.626997doi: bioRxiv preprint 2

25 Biological re-invasion refers to the process by which an exotic species re-establishes in an area 26 where it was previously established and had been eradicated (Banks et al. 2018). This could 27 occur on islands cleared of pests and mainland sites where an invader has been controlled 28 (Pluess et al. 2012). Unlike the initial invasion, re-invasion often bypasses the transport and 29

phases, as it usually comes from nearby invaded sites (Blackburn et al. 2011). An 30 example is the re-invasion of termites ( Coptotermes formosanus) in New Orleans parks, where 31 re-invasion pressure originated from surrounding populations (Mullins et al. 2011). 32 The mechanisms underlying re-invasion may differ from those governing the initial invasion 33 (Hansen et al. 2020). Changes in the re-invader, in the invaded environment or novel species 34 interactions may promote or hinder the re-invasion process (Banks et al. 2018) . For example, in 35 several invaded regions, brown rats ( Rattus norvegicus) and black rats (Rattus rattus ) often re-36 invade due to increased genetic diversity and behavioral flexibility compared with the initial 37 invasion, dispersal from nearby populations and re-establishment following the elimination of 38 competition (Fraser et al. 2015, Sjodin et al. 2020). Also, facilitations between re-invaders could 39 promote re-invasion. In Australia, the eradication of red foxes ( Vulpes vulpes ) allowed 40 populations of its prey, the European rabbit ( Oryctolagus cuniculus) to increase, which in turn 41 facilitated the rapid re-establishment of foxes (Saunders et al. 2010). 42 The order of species introduction also plays a crucial role in community composition (Chase 43 2003). Interactions among invasive species become particularly significant when one species is 44 already established and holds an incumbent advantage (or priority effect) over a newly arriving 45 invader (Duncan and Forsyth 2006). Local eradication of competitors can “reset” the process, 46 facilitating novel interactions. In Argentina, two invasive wasp species, Vespula vulgaris and 47 Vespula germanica, display this pattern. The former was observed in Patagonia in 2010, many 48 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted December 10, 2024. ; https://doi.org/10.1101/2024.12.08.626997doi: bioRxiv preprint 3 years after V. germanica invaded the region, following a sequence also observed in other parts 49 of the world (Olafsson 1979; Harris et al. 1991; Medina and Muñoz 2013). 50 Competition and coexistence are well-documented processes affecting these wasp species. 51 Notably, some invaded areas exhibit coexistence of both species, while others show dominance 52 by a single species (Harris et al. 1994; Pereira et al. 2022). In the Nothofagus spp. forests of 53 New Zealand (i.e., an invaded area), V. vulgaris was more abundant than V. germanica, despite 54 arriving later. However, this did not occur in urban and other forested habitats suggesting a 55 significant role of the environment in modeling the outcome of competition (Harris et al. 1994 ; 56 Badejo et al. 2020). 57 The Puyehue-Cordón Caulle volcanic complex erupted in June of 2011, dispersing ash over 7.5 58 million hectares in southern Argentina (Gaitán et al. 2011). This eruption significantly affected 59 insect populations in the area (Elizalde 2014). The impact of the ash on Vespula spp. 60 populations provides a unique opportunity to assess their re-invasive capacity of these wasps . 61 This "natural defaunation experiment" eliminates biases related to priority effects and re-62 invasion sequence. Our aim was to study the re-invasion patterns of V. germanica and V. 63 vulgaris in areas affected by the eruption. We hypothesize that both species successfully re-64 invade under equal initial conditions. Also, we expect the abundance of V. germanica to be 65 negatively impacted by the presence of V. vulgaris in natural areas, while V. vulgaris is not 66 affected. 67

68 The Puyehue volcano, part of the larger Puyehue-Cordón Caulle volcanic complex, is situated in 69 Chile (2236 m a.s.l., 40.5°S, 72.2°W). Our research was conducted in Argentina, in region that 70 experienced the heaviest ashfall following the eruption (Gaitán et al. 2011). This geographical 71 zone encompasses urban, suburban, and rural areas in NW Patagonia (40°- 41°S, 71°-72°W; 72 Fig. 1). 73 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted December 10, 2024. ; https://doi.org/10.1101/2024.12.08.626997doi: bioRxiv preprint 4 74 Fig. 1. Map of the region affected by the ashfall from the eruption of the Puyehue volcano. Purple circles 75 represent the 23 sampling sites, while the light blue area indicates lakes, and the grey area indicates the 76 affected zone. 77 To evaluate the recolonization of V. germanica and V. vulgaris populations, we placed traps 78 along ash deposition isocurves (Gaitán et al. 2011) at 23 sampling sites over a large area, with 79 an average distance of 5 km and a minimum of 500 m between sites (Fig. 1) . Each trap 80 consisted of a plastic bottle featuring entry holes and baited with beef to capture wasps 81 efficiently. Two traps were placed at each site and operated for 7 days. We collected and 82 recorded the number of wasps captured of each species, and then re-baited the traps. This 83 process was repeated twice, completing three weeks of sampling between March and April 84 each year (2012, 2013 and 2014) (i.e., 9 data points per site) . Also, the distance from each site 85 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted December 10, 2024. ; https://doi.org/10.1101/2024.12.08.626997doi: bioRxiv preprint 5 to the nearest urban center (defined as localities with more than 15000 inhabitants) was 86 measured. 87 To investigate the reintroduction of V. germanica and V. vulgaris into de-faunated areas, we 88 modelled wasp abundance using a generalized mixed effect model (GLMM). Fixed effects 89 included the “distance to the nearest urban center ”, “the sampling year ” and “species”, while 90 “site” were treated as a random effect . We assumed a negative binomial distribution with linear 91 parameterizations and log-link function to address the high number of zeros (75% of 92 observations) and resulting overdispersion. Overdispersion was modeled with the “year” in the 93 dispformula function. Model selection was based on Akaike information criteria. Analyses were 94 conducted using R statistical environment version 4.2.3 (packages: glmmTMB, performance, 95 fitdistrplus, DHARMa, car, bbmle, and emmeans). 96

and discussion 97 In total, 860 wasps of V. germanica and 940 of V. vulgaris were trapped, with the highest 98 catches occurring in 2014 (84% and 91% respectively). No significant differences were found 99 between the abundance of Vespula species captured (GLMM, Z=0.19, d.f.= 1, p=0.84). 100 However, their abundance could be explained by the year (GLMM, Z=9.98, d.f.= 2, p<0.001) 101 and distance to the nearest urban center (GLMM, Z=-3.7, d.f.= 1, p<0.001). 102 In 2012 Vespula spp. were captured at only two sites (out of a total of 23). Over the following 103 years, captures increased, but no wasps were found at sites farthest from urban centers. Wasp 104 abundance decreased as the distance from urban center increased, with all captures occurring 105 within 30 km of an urban center. The highest abundance was recorded in 2014, with an average 106 of 23.3 wasps per trap, peaking at 53.9 wasps at sites within 5 km of urban areas (Fig. 2). No 107 wasps were captured at sites 30 to 50 km away from urban center (Fig. 2). These results show 108 that both species successfully re-invaded, reaching similar abundances in urban and non-urban 109 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted December 10, 2024. ; https://doi.org/10.1101/2024.12.08.626997doi: bioRxiv preprint 6 areas. However, contrary to our expectations, the abundance of V. germanica was not 110 negatively affected by V. vulgaris, as wasp populations increased independently of congeneric 111 presence. 112 113 Fig. 2. Abundance of Vespula wasps’ populations (mean ± standard error) at 23 sites over 3 years, in 114 relation to the distance to the nearest urban center, fitted by the chosen model. Significance differences 115 were found (GLMM, year=p<0.001; distance=p<0.001). 116 The years since the eruption and proximity to urban center explained the abundance of both 117 species. The significant number of traps with no captures in our results, mainly during the first 118 years, confirms the devastating effect of volcanic ashfall on Vespula spp. In 2012, only a few 119 wasps were found in one urban center. Considering the eruption occurred in late autumn when 120 queens were in hibernation in sheltered sites, it is possible that some were buried beneath the 121 ashes. Additionally, those that hibernated in protected sites faced difficulties in finding suitable 122 nesting sites. 123 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted December 10, 2024. ; https://doi.org/10.1101/2024.12.08.626997doi: bioRxiv preprint 7 Our findings show that populations of both wasps noticeably recovered by 2014. For some 124 species such as many social insects, urban environments may facilitate establishment and 125 population growth by increasing food availability or shelter (Sol et al. 2013). Our study suggests 126 that urban centers could have served as sources for re-invasion by offering human-led ash-127 cleared habitats such as gardens or buildings. Added to this, t he rapid recovery may stem from 128 the high propagule pressure exerted by surrounding areas where both Vespula species had 129 been previously established. This enables re-invaders to more swiftly overcome Allee effects 130 compared to the challenges encountered during an initial invasion (Melo et al. 2023). 131 Biological invasions are influenced, in part, by the native community resistance and prior 132 invasion events. For exotic species to thrive, they must demonstrate competitive superiority over 133 resident species (native or prior invaders) with which they share one or more dimensions of their 134 ecological niche (Elton 1958). W e conclude that the re-establishment of both species is not 135 adversely impacted by the presence of the other species, at least during the initial three years. 136 The observed coexistence among these generalist invaders at certain locations may be linked 137 to distinct interaction mechanisms across various spatial scales (Harris et al. 1991; Masciocchi 138 et al. 2019; Masciocchi et al. 2023). 139 There is limited theory on re-invasion process, which is crucial to develop optimal management 140 strategies (Banks et al. 2018) . Understanding the factors that allow exotic species to reinvade 141 and spread, and predicting their behavior in the new habitat, also contributes to our growing 142 knowledge on invasion ecolog y. With the unprecedent rate of increase in biological invasions 143 globally, such knowledge is important to guide prevention and management strategies. 144

145 Badejo O, Leskinen JT, Koistinen A, Sorvari J (2020) Urban environment and climate condition-146 related phenotypic plasticity of the common wasp Vespula vulgaris. Bull Insectology 73: 285-147 294 148 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted December 10, 2024. ; https://doi.org/10.1101/2024.12.08.626997doi: bioRxiv preprint 8 Banks PB, Byrom AE, Pech RP, Dickman CR (2018) Reinvasion is not invasion again. 149 BioScience 68:792–804 150 Blackburn TM, Pyšek P, Bacher S, et al (2011) A proposed unified framework for biological 151 invasions. Trends Ecol Evol 26:333–339 152 Chase JM (2003) Community assembly: when should history matter? Oecologia 136:489-498. 153 DOI: 10.1007/s00442-003-1311-7 154 Duncan R, Forsyth D (2006) Competition and the assembly of introduced bird communities. I n 155 Cadotte MW, McMahon SM, Fukami T (eds). Conceptual ecology and invasion biology: 156 reciprocal approaches to nature. Springer, Dordrecht, The Netherlands. pp. 405-421 157 Elizalde L (2014) Volcanism and arthropods: a review. Ecol Austral 24:3–16 158 Elton CS (1958) The ecology of invasions by animals and plants. Methuen & Co, London 159 Fordham RA, Craven AJ, Minot EO (1991) Phenology and population structure of annual nests 160 of the German wasp Vespula germanica (Fab.) in Manawatu, New Zealand, with particular 161

to late summer and autumn. N Z J Zool 4223:127 –137. DOI: 162 10.1080/03014223.1991.10757959 163 Fraser CI, Banks SC, Waters JM (2015) Priority effects can lead to underestimation of dispersal 164 and invasion potential. Biol Invasions 17:1–8. DOI: 10.1007/s10530-014-0714-1 165 Gaitán JJ, Ayesa JA, Umaña F, et al (2011) Cartografía del área afectada por cenizas 166 volcánicas en las provincias de Río Negro y Neuquén. INTA SC Bariloche Argentina. 167 Hansen N, Hughes NK, Byrom AE, Banks PB (2020) Population recovery of alien black rats 168 Rattus rattus : A test of reinvasion theory. Austral Ecol 45:291 –304. 169 https://doi.org/10.1111/aec.12855\ 170 Harris RJ, Moller H, Winterbourn MJ (1994) Competition for honeydew between two social 171 wasps in South Island beech forests, New Zealand. Insectes Sociaux 41:379 –394. DOI: 172 10.1007/BF01240641 173 Masciocchi M, Unelius CR, Buteler M (2019) Foraging niche separation of social wasps in an 174 invaded area: Implications for their management. J Appl Entomol 143:1115 –1121. DOI: 175 10.1111/jen.12708 176 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted December 10, 2024. ; https://doi.org/10.1101/2024.12.08.626997doi: bioRxiv preprint 9 Masciocchi M, Mattiacci A, Villacide JM, Buteler M, Porrino AP & Martínez AS. (2023). Sugar 177 responsiveness could determine foraging patterns in yellowjackets. Sc ientific Reports, 13, 178 20448. 179 Medina RB, Muñoz CV (2013) Primer reporte de Vespula vulgaris (Linnaeus, 1758) 180 (Hymenoptera: Vespidae) en Chile. Bol Soc Entomológica Aragon 52:277–278 181 Melo R, Masciocchi M, Corley JC (2023) Allee effects in an invasive social wasp: an 182 experimental study in colonies of Vespula germanica. Sci Rep 13:16323 183 Olafsson E (1979) A review of the wasps (Hymenoptera, Vespidae) recorded in Iceland. 184 Natturufraedingurinn 49:27–40 185 Pereira AJ, Masciocchi M, Corley JC (2022) Long-term coexistence of two invasive vespid 186 wasps in NW Patagonia (Argentina). Oecologia 199:661–669 187 Pluess T, Cannon R, Jarošík V et al (2012) When are eradication campaign successful? A test 188 of common assumptions. Biol Invasions 14:1365-1378. DOI: 10.1007/s10530-011-0160-2 189 Saunders GR, Gentle MN, Dickman CR (2010) The impacts and management of foxes Vulpes 190 vulpes in Australia: Impact and management of foxes in Australia. Mammal Rev 40:181 –211. 191 DOI: 10.1111/j.1365-2907.2010.00159.x 192 Sjodin BMF, Irvine RL, Ford AT, et al (2020) Rattus population genomics across the Haida 193 Gwaii archipelago provides a framework for guiding invasive species management. Evol Appl 194 13:889–904. DOI: 10.1111/eva.12907 195 Sol D, Lapiedra O, González-Lagos C (2013) Behavioural adjustments for a life in the city. Anim 196 Behav 85:1101–1112 197 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted December 10, 2024. ; https://doi.org/10.1101/2024.12.08.626997doi: bioRxiv preprint