Climate change impacts the current and potential distribution of many insects, since temperature is often a limiting factor to where the insects can survive. The Formosan subterranean termite, Coptotermes formosanus Shiraki (Blattodea: Rhinotermitidae), has never been reported in South Korea despite its close proximity to 2 countries (China and Japan) where this economically important pest has been reported. This may be due to the average winter temperature in South Korea which is below 4 °C, the lower limit of the current distribution range of Formosan subterranean termite. However, with climate change leading to increased temperatures, South Korea may be susceptible to successful invasion by Formosan subterranean termite. The objective of this study is to estimate the future possible distribution of Formosan subterranean termite in Korea based on temperature. Climate data from Korea showed a significant increase of 2.19 °C per 100 yr in average annual temperature from 1910 to 2018. Previous and current average winter temperatures were higher than 4 °C only in Jeju, and most provinces did not exceed 4 °C, except for some southern cities such as Busan in 2000 to 2019. With the estimated rate of temperature rises, winter temperatures in Gyeongsangnam-do will exceed 4 °C starting from 2020, and Jeollanam-do will exceed 4 °C from 2060. Coupled with the statistically significant, increased annual trade between Korea and other countries (China, Japan, Taiwan, and the USA) where C. formosanus is currently distributed, we predict that Formosan subterranean termite will become established in South Korea, probably starting from a southern trade port such as Busan.
Climate change is an inevitable phenomenon and has had a significant impact on species extinction and the introduction of invasive species. Invasions by alien species in non-native areas due to climate change are important driving factors that affect ecosystems and biodiversity (Vitousek 1994; Dukes & Mooney 1999). These invasive species often have an enormous economic impact on agriculture, the environment, and structures, as well as losses from regulatory actions and trade restrictions (Holmes et al. 2009). The total economic loss by invasive species in the USA, for example, is estimated around USD $120 billion (Pimentel et al. 2005).
In addition to climate change, human activities translocate species beyond their native range, which has increased substantially during the last centuries. Hence, globalization and increase of trade are responsible for spreading alien species into non-native areas. This is related to propagule pressure, defined as a measure of the number of individuals released into a non-native region, and consists of the propagule size, number of individuals per release event and number, and number of discrete release events (Lockwood et al. 2005). Propagule pressure is not easy to measure directly, but can be estimated by using international trade as a proxy (Westphal et al. 2008).
The Formosan subterranean termite, Coptotermes formosanus Shiraki (Blattodea: Rhinotermitidae), is one of the world's 100 worst invasive alien species (Lowe et al. 2000), and is the most-studied termite species due to its economic importance in temperate and subtropical regions. The nest of C. formosanus may be found either underground or inside trees, and the mature colony contains millions of individuals (King & Spink 1969; Su & Scheffrahn 1988). The foraging distance of C. formosanus may be up to 100 m using underground tunnels or aboveground shelter tubes (King & Spink 1969).
The global economic loss, including control and repair costs, caused by subterranean termites is USD $32 billion annually (Rust & Su 2012). Formosan subterranean termite is endemic to southern China and Taiwan (Kistner 1985; Su & Tamashiro 1987; Li 2000; Su 2003; Li et al. 2009), and has been introduced to multiple locations, including Japan (Su & Tamashiro 1987), South Africa (Coaton & Sheasby 1976), Marshall Island, Midway Island, and the US (Su & Tamashiro 1987; Su 2003; Evans et al. 2013). Past and current distribution of C. formosanus in the world showed that it is restricted to an area about 35° north and south of the equator (Su & Tamashiro 1987).
Despite the fact that South Korea is located between ca. 33° to 38° north and between China and Japan, there is no official record of Formosan subterranean termite in South Korea. These 2 neighboring countries have known distributions of Formosan subterranean termite and have a long history of damage by this pest (Mori 1987; Wang et al. 2002). The distribution of C. formosanus in South Korea is controversial. Early work by Lee et al. (1998) reported finding C. formosanus from 3 southern locations: Busan, Jinju, and Geoge Island. However, a later extensive survey of the same areas failed to find C. formosanus in Korea (Lee et al. 2015). The reason it is controversial is that firstly, Lee et al. (1998) did not include any pictures of the specimens. Secondly, they stated that they confirmed the presence of C. formosanus based on damage, which is unreliable. Lastly, if it was present in 1998, the damage by C. formosanus also should have been reported elsewhere. However, there was no such reports yet. Officially, only 2 species of subterranean termites were found, Reticulitermes speratus kyushuen-sis Morimoto (Becker 1969) and Reticulitermes kanmonensis Takematsu (both Blattodea: Rhinotermitidae) (Lee et al. 2015).
Temperature is the crucial factor responsible for the distribution of C. formosanus (Kofoid 1934; Shimizu 1962; Coaton & Sheasby 1976; Su & Tamashiro 1987). The distribution of this species in Japan is limited to regions where the monthly mean temperature of winter is above 4 °C (Abe 1937; Mori 1987). However, C. formosanus also was discovered in Kanagawa Prefecture (Japan), where the temperature drops lower than 4 °C in the winter (Mori 1987).
Considering the geographical proximity of South Korea to China and Japan, we hypothesize that the temperature may not have been favorable to Formosan subterranean termite in the past, but is likely to become suitable in the future due to climate change, especially in the southern part of South Korea. To project the potential future areas that may be threatened by Formosan subterranean termite in South Korea, we surveyed the amount of imported wooden materials and the amount of trade between South Korea and other countries where Formosan subterranean termite is currently distributed (i.e., Japan, China, Taiwan, and the US) as a proxy of propagule pressure. Further, we analyzed past and current temperatures to find the trend of temperature due to climate change and average winter temperatures to determine the possible distribution limits of Formosan subterranean termite in South Korea if it is introduced. Furthermore, the future distribution limit with the consideration of temperature increase rate was estimated.
Materials and Methods
STUDY AREA
Republic of Korea (South Korea) occupies the southern half of the Korean Peninsula, which was divided by the Korean Demilitarized Zone after the Korean War in 1953. The land mass of South Korea is approximately 100,000 km2, and it has about 2,500 km of coastline (Fig. 1). There are 9 provinces, the first-level divisions within the Republic of Korea. In this study, we assigned letters to the provinces for simplicity (A to I, from south to north and from east to west) (Table 1).
South Korea has a temperate climate with 4 distinct seasons (winter: Dec–Feb; spring: Mar–May; summer: Jun–Aug; fall: Sep–Nov). The annual precipitation of South Korea varies yr to yr; usually, it is about 1,200 mm. During the summer, there are summer monsoons with occasional typhoons.
Fig. 1.
Geographic location of the Republic of Korea, South Korea, in Far East Asia. The name of the provinces are presented on the map as A to I, from south to north and from east to west. The map was produced using data from GADM and USGS (Projection: WGS1984 (horizontal) and EGM96 (vertical)) in ArcMap 10.5.1.

Table 1.
Information of climate data stations including latitude, longitude, and elevation used in this study. For simplicity, we assigned provinces letters from A to I.

INDIRECT MEASUREMENT OF PROPAGULE PRESSURE
We surveyed the amount of imported wooden materials, including raw logs, lumber, and timber, and possible transportation sources of termites, from 1995 to 2017 (Korea Forest Service, Statistical Yearbook of Forestry [ https://www.forest.go.kr/]). We were not able to acquire the exact amounts of imported wooden materials from China, Japan, Taiwan, and the US. Instead, we examined the amount of annual trade from 2000 to 2018 between Korea and these other countries where C. formosanus is currently distributed, because they are potential sources of future introduction events into the Korean Peninsula (Korea Customs Service trade statistics [ https://unipass.customs.go.kr]).
TEMPERATURE DATA
There are 95 land-based stations for climate measurement in South Korea. All data are accessible and available on the internet ( https://data.kma.go.kr/). Using the climate data from 3 stations, we calculated the average, minimum, and maximum annual temperatures from 1910 to 2018. We also analyzed the average winter temperature, defined as the average temperature in Dec through Feb from 1960 to 2019 using data from 3 to 7 stations for each province (Table 1).
We estimated future temperatures based on the increase rate of the average annual temperatures. On average, the temperature increased 2.19 °C per 100 yr; therefore, the increase rate for 20 yr was determined to be 0.44 °C. Therefore, the future temperature was calculated by adding 0.44 °C to previous temperatures.
MAPPING AND STATISTICAL ANALYSIS
Country/province shapefiles from GADM ( https://gadm.org/download_country_v3.html) and elevation raster imagery from the USGS EROS Archive - Digital Elevation - Shuttle Radar Topography Mission (SRTM) 1 Arc-Second Global ( https://dds.cr.usgs.gov/srtm/version2_1/SRTM3/Eurasia/) were used to create the map (ArcMap 10.5.1, Environmental Systems Research Institute, Redlands, California, USA).
Pearson's correlation coefficient test was used to determine the statistical correlations between the amount of imported wooden materials, amount of trade between the Republic of Korea and other countries, average annual temperatures, average annual minimum temperatures, average annual maximum temperatures, and yr. All statistical analyses were carried out using SPSS v19.0 (IBM Corp. 2010).
Results
The amount of imported wooden materials significantly increased between 1995 and 2017 (Pearson's correlation coefficient r = 0.764; P < 0.001; Fig. 2A). In addition, the amount of trade between South Korea and China (r = 0.964; P < 0.001), Japan (r = 0.629; P = 0.004), Taiwan (r = 0.982; P < 0.001), and the US (r = 0.922; P < 0.001) statistically increased over time (Fig. 2B).
Overall, there were significant increases in average annual (r = 0.804; P < 0.0001), minimum (r = 0.859; P < 0.0001), and maximum (r = 0.658; P < 0.0001) temperatures from 1910 to 2018 (Fig. 3). In South Korea, previous and current average winter temperatures were higher than 4 °C, which is the northern temperature limit of C. formosanus only in location A. Most locations (i.e., B to I) do not exceed 4 °C, except some cities in location B, such as Busan, Namhae, Geoje, and Tongyeong between 2000 and 2019 (Fig. 4). Future estimated winter temperatures of location B, however, exceed 4 °C after 2020, and those of location C will exceed 4 °C by 2060 (Fig. 5).
Fig. 2.
(A) Bars indicate amounts of imported wooden materials into Korea from 1995 to 2017, and the dotted line on top of the bars represents the trend line for imported wooden materials. The data were obtained from the Statistical Yearbook of Forestry, Korea Forest Service. (B) Amounts of trade goods (USD billions) between South Korea and China (dotted), Japan (dashed), Taiwan (solid with square) and US (solid). Arrows indicate a linear equation and R-square of the trend line.

Discussion
In this study, we revealed that the Formosan subterranean termite, which is the most economically important termite, possibly will be found in southern regions of the Korean Peninsula due to climate change. Coptotermes formosanus is known as an urban pest due to its propensity to be found in man-made wooden structures such as houses and historical properties (Lowe et al. 2000; Rust & Su 2012; Evans et al. 2013). This invasive termite has never been completely eradicated once it becomes established in an area (Su 2003).
To predict the distribution of species, several climate variables have been used (Guisan & Zimmermann 2000; Beaumont et al. 2005). The variables, minimum, mean, and maximum temperatures are some of the basic components to project future distribution. Accordingly, most studies that predict the distribution of subterranean termites also applied air temperatures, although subterranean termites are soil-dwelling insects (Li et al. 2013; Tonini et al. 2014; Buczkowski & Bertelsmeier 2017). This is because soil temperatures are closely related to air temperatures, and may be predicted from air temperatures (Toy et al. 1978; Zheng et al. 1993).
In temperate latitudes, low temperature has a significant impact on range expansion, and affects insect survival and development (Robinet & Roques 2010). For instance, the northern distribution limit of the southern pine beetle, Dendroctonus frontalis Zimmermann (Coleoptera: Curculionidae), is determined by lower winter temperature (Ungerer et al. 1999). Studies on the distribution of Formosan subterranean termite indicate that the lower limit in winter temperature (about 4 °C) is the most important factor responsible for its distribution (Kofoid 1934; Abe 1937; Shimizu 1962; Coaton & Sheasby 1976; Mori 1987; Su & Tamashiro 1987; Su 2003). In the past (i.e., 1960 to 1999), there was no area with suitable tempertaure in Korea except Jeju Island (location A, Fig. 4). The rise in winter temperatures in southern parts of Korea, such as coastal areas in southern regions (locations B and C), may allow Formosan subterranean termite to become established in the future (Fig. 5). Additionally, we speculate that if the temperature rises at the same rates that we observed for the past 100 yr, Formosan subterranean termite also may expand its distribution farther north in the future (Figs. 5, 6), with the possible exception of mountainous sections of the central and northeast regions (Fig. 6), due to the relatively low temperatures at high elevations.
Fig. 3.
Annual temperature of Korea based on 3 stations (Busan, Daegu, and Seoul) from 1910 to 2018. Solid, dashed, and dotted lines are the average, minimum, and maximum, respectively. Arrows indicate the linear equation and R-square.

Aside from rising temperatures, anthropogenic activities may provide favorable conditions for establishment of Formosan subterranean termite colonies in urban areas. For instance, heating systems in structures, as well as moisture-related conditions around structures, such as rain seepage, plumbing leaks, condensation, and poor ventilation, may supply enough heat and moisture for termites to survive (Mallis & Story 2003). Accordingly, it is possible that these kinds of favorable conditions may alter our projections.
Fig. 4.
Average winter temperature of each province (A to I). Black, white, and gray bars indicate the average winter temperature of each time period, and the dashed line indicates the northern distribution temperature (4 °C) limit of Coptotermes formosanus. The map in the figure shows the location of A to I.

The natural rate of spread of termites is relatively slow because of the time required for colony maturity (Chouvenc & Su 2014) and low distance of the dispersal flights (about 0.2 to 1 km) by alates (winged reproductives) (Messenger & Mullins 2005; Mullins et al. 2015). The minimum distances from areas in China and Japan where Formosan subterranean termite is known to occur to South Korea are 543 km (from Jiangsu, China, to Jeju, Korea) and 244 km (from Yamaguchi, Japan, to Busan, Korea), respectively. Considering these distances, we can exclude the possibility of natural dispersal from either China or Japan to South Korea. Therefore, Formosan subterranean termite most likely will be introduced either by trade of goods, such as in shipping containers, raw logs, timber, and lumber, or by boats to port cities (Su 2003), which is the same way Formosan subterranean termite has invaded other countries. Incipient colonies of Formosan subterranean termite are able to seal off the nest after tunneling into wet wood, allowing them to survive during transportation. We showed that the amount of imported wooden materials significantly increased over time, as well as the amount of trade between Korea and countries where Formosan subterranean termite is currently distributed (Fig. 1A, B).
Fig. 5.
Past and estimated average winter temperature of each province (A to I) according to the temperature increase trend of Korea. Each bar indicates the average winter temperature of each time period. The dashed line in the figure shows the northern distribution temperature limit of Coptotermes formosanus.

Although environmental factors, such as temperature, are suitable, invasive species need to survive in interspecific competition with native species (Su 2013). Therefore, competition can be another limiting factor to the establishment of invasive species in non-native areas. However, it is believed that Formosan subterranean termite is more aggressive and usually out-competes different species such as Reticulitermes (Su 2003). Formosan subterranean termite has not shown antagonistic behavior towards colonies of other conspecific colonies in introduced areas, e.g., Florida, USA (Su & Haverty 1991), and C. formosanus may fuse with other colonies (Lee et al. 2019).
Busan, the second largest city in Korea, located in the southern tip of the Korean Peninsula, has the largest port in Korea and acts as a point of entry for invasive alien species. The pine wood nematode, Bursaphelenchus xylophilus (Steiner and Buhrer) Nickle (Secernentea: Parasitaphelenchidae), which is the agent for pine wilt disease, was first found in Busan in 1988, and spread to most mountains in South Korea, resulting in an average USD $8 million per yr in control programs (Shin 2008). A tropical/subtropical species of hornet, Vespa velutina Lepeletier (Hymenoptera: Vespidae) also was found first in Busan in 2003 (Kim et al. 2006), and is distributed in locations B and C (Fig. 1) as of 2010 (Choi et al. 2012). The hornet is now well-adapted to and has become a pest in urban environments (Choi et al. 2012). Another invasive species, the red imported fire ant, Solenopsis invicta Buren (Hymenoptera: Formicidae), native to central South America, has been found recently in Busan Port in 2018 (Choi et al. 2018). Furthermore, 10 different macro-moths of subtropical origin were found in Korea in 2006 (Park et al. 2006).
Buczkowski and Bertelsmeier (2017) showed that under the RCP 4.5 scenario in 2050, the future distribution range of C. formosanus will expand north to North Korea. However, our study showed that C. formosanus may not survive if it is introduced in the northern region of South Korea due to low winter temperatures. Buczkowski and Bertelsmeier (2017) incorrectly assumed that C. formosanus is currently distributed in South Korea.
In conclusion, with the rise of temperature due to climate change and the increase of trade between South Korea and other countries where Formosan subterranean termite is currently distributed, Formosan subterranean termite may be introduced and be able to establish in the southern part of the Korean Peninsula if it survives during transportation.
Fig. 6.
Potential future distribution of Coptotermes formosanus based on the estimated winter temperature according to Figure 5. The dotted lines indicate the estimated northern distribution limits of 2020, 2060, and 2100, respectively.

A targeted quarantine inspection in the southern part of South Korea is required to prevent the introduction of Formosan subterranean termite in Korea, and baiting programs using chitin synthesis inhibitors may be needed to eradicate the initial small population. Otherwise, Formosan subterranean termite will cause damage in the southern region of South Korea as they already have done in other countries.
Acknowledgments
We are grateful to Thomas Chouvenc (University of Florida) for providing constructive comments. This work was supported in part by the USDA National Institute of Food and Agriculture, Hatch project number FLA-FTL-005865.