Sample collection

A total of six wasp collection events were conducted on Ulleungdo. Wasp traps were installed at 12 locations from May 18 to 20, 2022, and collected monthly thereafter. The traps were removed after the last collection between October 22 and 24, 2022.

Regarding survey points, one or two traps were installed for each ri to evenly survey the entire area, including Seonginbong Peak, Ulleungdo’s iconic mountain. Initially, four traps (Trap No. 1-4) were installed around the trail from Dodong to Seonginbong, and the remaining traps (Trap No. 5-12) were allocated to each ri (Fig. 1).

Collection method

The wasp traps were installed by drilling a square hole (5×5 cm) in the upper half of a 2 L plastic barrel, into which a mixture of grape juice (500 mL) and yeast was added. Captured wasps were washed with water, immersed in 95% alcohol for preservation, and subsequently identified. Moreover, after collecting the wasp traps, any wasps not captured by the traps were collected through a 30-minute sweeping effort using an insect net in the grassy areas surrounding the traps. Furthermore, we scoured for wasp nests by visually inspecting the eaves of nearby houses, cracks in walls, grassy terrain, and cliffs.

Emergency call data

Information regarding calls for wasp nest removal in the Ulleungdo area was provided by the ESCCs under the National Fire Service. The data encompassed the past five years (2018-2022) and was categorized by year, month, and location. This data was subsequently analyzed and mapped. Additionally, the number of reports was analyzed in relation to unit area and population. Reports concerning other bees, such as bee swarms, were excluded from the dataset, focusing solely on reports pertaining to the removal of social wasp nest.

Identification

Wasps collected via traps and insect nets were identified using identification keys from Archer (2012), Choi (2004), Kim (2001), Kim et al. (2006), and Matsuura and Yamane (1990). Additionally, as the information on the number of wasp nest removals was obtained from telephonic data, it was not used for species identification. However, some nests were reported via calls to 119 rescuers, and information on species at the genus level was secured through photo collation.

Diversity of social wasps collected from Ulleungdo

A total of 5,068 individuals belonging to two genera and two species were collected through five trappings at 12 locations on Ulleungdo Island. Vespa simillima simillima Smith exhibited the highest count, with 3,961 individuals (78.2%) collected (Fig. 2A), followed by Vespula flaviceps flaviceps (Fig. 2C) with 1,107 individuals (21.8%; Table 1).

By location, 942 and 936 individuals were collected from traps 1 and 3, respectively (the largest number of individuals collected), whereas only 42 and 39 individuals were collected from traps 8 and 9, indicating a significant variation. Furthermore, the majority of individuals collected by sweeping were either V. simillima simillima or Vl. flaviceps flaviceps, although five individuals of Polistes snelleni de Saussure (Fig. 2D) were collected at location 5. Although nests were searched for under house eaves, between wall gaps, and on cliffs, no nest traces were found except one V. simillima simillima nest found on a cliff (Fig. 2B). Consequently, this survey yielded a total of 5,132 wasps representing three genera and three species.

Emergency call data for social wasp nest removal in Ulleungdo

Over the past five years (2018-2022), a total of 60 emergency calls were made for wasp nest removal on Ulleungdo. By year, in 2018, 2019, and 2022, the number of cases appeared similar (17-20), but it notably decreased in 2020 and 2021 with only two and four cases, respectively (Fig. 3).

Examining the monthly records, emergency calls initiated in June, peaked in August, and then decreased. In particular, the number of calls in July (18 cases), August (23 cases), and September (11 cases) accounted for 87% of the total reports (Fig. 4).

In terms of regions, the Dodong-ri area reported the most cases with 22, followed by Jeodong-ri with 10 cases. All other regions reported fewer than 10 cases, with Namseo-ri and Hyeonpo-ri recording the fewest cases, with two each (Fig. 5). When evaluating the number of reports per unit area, Dodong-ri had the highest ratio of 3.55, followed by Jeodong-ri with 1.35. However, in terms of the number of reports per population, Na-ri had the highest ratio of 3.00, followed by Namseo-ri with 1.00. Conversely, Dodong-ri and Jeodong-ri, which had the highest number of reports, exhibited very low numbers, with 0.51 and 0.53, respectively (Table 2).

Diversity of social wasps in Ulleungdo

Since Cho (1955), when Vespidae species were first recorded on Ulleungdo, a total of nine species of three genera in two subfamilies have been documented in the list of social wasps on Ulleungdo to date (Table 3; Cho, 1955; Choi, 2023; Lee & Kwon, 1981; Lee et al., 2006; Lim et al., 2013; Won et al., 2023). However, as stated by Choi et al. (2013), the Vespidae species list in Korea has undergone significant revisions, with the inclusion of synonyms, new species, unrecorded species, and invasive species.

Consequently, when examining social wasp checklists in a specific area, entomologists and the general public (except some wasp taxonomists) often rely on outdated taxonomic information, potentially leading to errors in identification. This undermines the reliability of many of the recorded data for existing social wasps.

For instance, when reviewing the species list in Table 3, Vespula flaviceps lewisii Smith was reidentified as Vl. flaviceps flaviceps by Choi et al. (2013). Moreover, Vespula rufa schrenckii Radoszkowski is the rarest species in Korea, primarily found in the highlands of Gangwon-do and northern Gyeongbuk. Therefore, the presence of this species on Ulleungdo is highly unlikely.

Additionally, Polistes jadwigae jadwigae Dalla Torre was once considered a synonym for Polistes yokahamae Radoszkowski by Kojima and Hagiwara (1998). However, in the past, it was often misidentified as Polistes rothneyi koreanus van der Vecht (Choi et al., 2013). Therefore, most of the species categorized under the subgenus Gyrostoma are highly likely to be P. rothneyi koreanus.

Hence, despite considerable efforts made throughout Ulleungdo to confirm the distribution of P. rothneyi koreanus, neither individuals nor any signs of nests were discovered. Although some information on its existence was obtained through photo comparisons of nest removal by rescue teams at the Ulleungdo fire station, further verification is necessary.

Polistes olivaceus De Geer is a species not naturally found in Korea, but its continuous recording for approximately 50 years poses questions regarding the reliability of species identification in the genus Polistes. Furthermore, as Polistes japonicus japonicus de Saussure and Polistes mandarinus de Saussure are representative species that are commonly misidentified by domestic entomologists, their records in Ulleungdo are also highly dubious.

In contrast, it appears quite unusual that V. simillima simillima and Vl. flaviceps flaviceps, which dominated the results of this survey, are absent from the collection records of Ulleungdo in the past. It remains uncertain whether this absence is due to past misidentifications by entomologists or if these species have been recently introduced from the mainland and have since thrived. Additionally, there is currently no way to ascertain whether the Polistes species recorded in the past were indeed misidentified or have vanished for specific reasons.

Therefore, the findings of this survey indicate that there are currently three species of social wasps belonging to three genera in Ulleungdo. Furthermore, additional species will need to be confirmed through further investigations.

Removal of wasp nests on Ulleungdo

Over the past five years, there have been 60 reports for wasp nest removals on Ulleungdo, averaging 12 per year. This number does not significantly differ from the mainland (Choi et al., 2012; 2019). The limited inflow and outflow of external wasp populations due to the geographical characteristics of islands likely contribute to the consistent number of wasp colonies that occur each year.

When we examine the yearly data, the numbers for 2018, 2019, and 2022 appear relatively high, while in 2020-2021, they dropped significantly. By region, it was observed that Dodong-ri and Jeodong-ri reported the highest numbers. As the number of wasp nest removals is not a measure of the number of natural occurrences of wasp nests, but the number of calls made, it is inevitably influenced by the activity of the individuals encountering the wasp nest. Consequently, given that the number of reports tends to be higher in areas with larger populations (Choi et al., 2019), it is reasonable to see higher reports in Dodong-ri and Jeodong-ri, which have the highest populations on Ulleungdo. Additionally, the sharp decline in 2020-2021 may be attributed to a significant reduction in tourist numbers and a decrease in opportunities to encounter wasp nests, as locals limited their outdoor activities because of COVID-19.

According to the life cycle of social wasps, queens emerge from winter hibernation in April to form embryonic nests, with the first workers appearing in June. The number of workers increases rapidly in July as the nest size increases. By August and September, the nest reaches its maximum size. As these nests, initially inconspicuous, become visible to the naked eye, the number of reports surges, particularly from July onwards.

Despite no reported cases of wasp stings in Ulleungdo over the past five years, it is important to note that V. simillima simillima and Vespula koreensis koreensis being the most populous native wasps in Korea, can engage in group attacks. These attacks typically occur when humans inadvertently disturb their nests or act threateningly near the nest entrance.

Therefore, if a nest is discovered in spring or early summer, it is necessary to ensure its swift removal by reporting it to the 119 rescue team. While the number of human casualties from wasp stings remains low in Ulleungdo, damage to marine products caused by yellowjackets is frequently reported. Ulleungdo is a significant hub for squid production in Korea, accounting for approximately 750 tons of squid as of 2018, which amounts to nearly 94% of Ulleungdo’s seafood sales (Kim, 2019). Yellowjackets contribute to the supply of protein to their larvae by hunting Lepidoptera larvae in the forests. However, they also collect the remains of terrestrial animals such as mammals, amphibians, fish, and annelids, as are the flesh of aquatic products such as fish, squid, and clams left on land (Matsuura & Yamane, 1990). In Ulleungdo, captured squids are left outdoors to naturally air-dry, attracting numerous yellowjackets that consume the squids’ flesh. This significantly diminishes the quality of squid products and leads to substantial economic losses (Cho, 2011). Given that these damages can persist as long as yellowjackets are present, preventive measures are imperative.

Furthermore, since 2017, only Jangwon honey bees, known as Korea’s premier beekeeping breed, have been reared, replacing the existing honey bees on Ulleungdo (Hong, 2017; Kim et al., 2015). V. simillima simillima has been documented as a major pest that preys on bees in apiaries (Choi & Kwon, 2015). However, the damage caused by this species is less severe than that inflicted by Vespa mandarinia Smith or V. velutina nigrithorax, which are typically found inland. Therefore to prevent negative impacts on beekeeping, it is necessary to continuously monitor the possible invasion of V. mandarinia or V. velutina nigrithorax, which have not been detected in Ulleungdo until now.