INTRODUCTION
In regards to the life history of animals, moving from one habitat to another habitat is related to the securing of resources including food and water, mating, sunbathing, hibernation, nest, hideout, survival and reproduction. Most animals move only when it is necessary, because their movement is associated with energy loss, unfavorable environmental conditions and exposure to predators. Also, how often and how far animals move differ greatly from one species to another (Pough et al., 2004).
In case of the movement patterns of snakes, they are either (1) difficult to be predicted because of random movement, or (2) sedentary whether they are concentrated or not concentrated in one area (Seigel et al., 2001). Recently, it was revealed that snakes' moving distance and home range differ by species, time slot, gender and seasonal factors (Shine and Lambeck, 1985; Shine, 1987; Fitzgerald et al., 2002; Brito, 2003; Shine et al., 2003; Whitaker and Shine, 2003). In particular, the red-tongued viper snakes, which belong to the Gloydius, are very sedentary like Gloydius shedaoensis that inhabit an islet in the northeast of China (Shine et al., 2003). On the other hand, some snakes like Crotalus adamanteus, a type of rattlesnake that inhabits the United States, move relatively long distances (Waldron et al., 2006). Red-tongued viper snakes belong to Reptilia, Squamata, Serpentes, Crotalinae and Gloydius (O’Shea, 2005); they are distributed in Korea, northeast of China and far east of Russia (Zhao and Adler, 1993). Until now, in regards to the ecology of red-tongued viper snakes, studies have been conducted to identify their reproductive cycle and litter size (Kim and Oh, 2014a), sexual size dimorphism (Kim and Oh, 2014a) and prey use (Kim and Oh, 2014c). Also, a study on capture-recapture was conducted for reptile resource management in national parks by using a PIT tag (Lee et al, 2009). However, there is no information on snakes' movement and home range.
Accordingly, this study was carried out to identify the moving distance and home range of the red-tongued viper snakes that inhabit Gapado, an islet of Jeju Island, by using the mark-recapture method, and to provide data necessary for ecological studies on other types of snakes that inhabit Korea.
MATERIALS AND METHODS
This study was carried out in Gapado, the annexed island located in the southwest of Jeju Island. The island's area is about 0.87 km2 or 1/2100 of Jeju Island. It belongs to Gapa-ri, Daejeong-eup, Seogwipo-si, Jeju Special Self-Governing Province in terms of administrative district. It is situated at latitude 33° 09′ 35″~ 33° 10′ 30″ and longitude 126° 15′ 56″~126° 16′ 57″. Gapado is 20 m above sea level and its ground is flat. Most of the island is farmland except some areas where there are villages. Grasslands or simple habitats are developed at some places at the edge of the island, and this study was carried out in the grasslands, which is located in the southwest of Gapado, where goats are grazed (Figure 1).
This study used the mark-recapture method to find out the movement and home range of red-tongued viper snakes from June 2006 to June 2009. Because the number of surveys is determined by whether it is possible to enter the island depending on weather conditions, a survey was carried out 1~4 times a month. In regards to the marking of individuals, two methods were used: the method of ventral clipping abdominal scales and the method of inserting a PIT tag (Passive Integrated Transponder tag; ID 162A, TROVAN) under the skin (Figure 2). Figure 2a shows the location where ventral clipping is conducted to the individual no. 155, and Figure 2b shows the image of the individual no. G001 when it was recaptured. Figure 2c shows how to insert the PIT tag under the skin.
The capturing location was identified using a GPS (eTrex Vista® C, Garmin), and the moving distance and home range were calculated using MapSource program (ver. 6.8, Garmin). The MCP (Minimum Convex Polygon) method was used to determine the size of home range of the individuals that were captured more than three times. SPSS (ver. 12.0) statistical program was used for a statistical analysis. χ2 -test was used to analyze sex ratio of the captured individuals, and t-test was used to analyze the difference in home range between male and female redtongued viper snakes. Also, a correlation between the moving distance and the interval between capture- recapture and a correlation between the body length and the home range size were analyzed through Pearson's correlation.
RESULTS AND DISCUSSION
In terms of the number of marked individuals by year, 13 individuals were marked in 2006, 42 individuals in 2007, 61 individuals in 2008 and 15 individuals in 2009. Among the 131 marked individuals, except two individuals of which gender marking was omitted, 56 were females and 73 were males. The ratio of males is higher (Males : Females = 1.30 : 1, n = 129), but there was no statistically significant difference (χ2-test, χ2 = 2.240, P> 0.1). However, the ratio of males was higher than the sex ratio of the offspring of the individuals that inhabit Jeju Island (Males : Females = 1.15 : 1). There is no difference in the sex ratio of the offspring of most snakes (Wang et al., 2003; Taylor and Denardo, 2005), but because females face many difficulties due to reproduction (Shine, 1980), their mortality is higher. A good example is Trimeresurus stejegeri stejegeri that inhabits Taiwan: the observed frequency of males was significantly higher than that of females when they become adults (Wang et al., 2003).
Among the marked individuals, 22 (16.8 %) were recaptured, and 8 individuals (6.1 %) were recaptured more than twice (Table 1). The recapture rates after inserting the PIT tag were 16.67 % and 8.33 % regarding the surveys that were carried out in Mt. Odae National Park and Mt. Wolak National Park (Lee et al., 2009), Accordingly, the recapture rate of this study (16.8 %) was similar to the result of the survey conducted in Mt. Odae National Park but higher than that of Mt. Wolak National Park. Compared to the recapture rates of Vipera berus 45 % (females 54 %, males 33 %), which inhabits Switzerland, and Drymarchon coupri 43 % (Stevenson et al., 2009), which inhabits Georgia, the United States, recapture rates of red-tongued viper snakes were very low. The durations from capture to recapture varied from 1 to 710 days, but there was no correlation between the moving distance and the interval between capture-recapture (n = 33, r = 0.173, P> 0.1). In case of G015, it moved 44 m during 35 days and 22 m during 14 days, but G001, G009, G018, G023, G072, G091 and G104 moved only 3-32 m during more than 300 days. Also, the interval between capturerecapture of other individuals varied but there was no difference in their locations (Table 1).
The movement of snakes is associated with the securing of resources including water and food, mating, sunbathing, hibernation, nest, hideout and various factors required for reproduction (Pough et al., 2004), and snakes' mortality may go up by getting exposed to many risks such as energy loss and encountering with natural enemies. In general, there may be a difference in the mortality by age due to the moving distance: young snakes primarily die when dispersing after hatching out, and the mortality of juvenile snakes that stay in one place is low. Also, male snakes may die during the mating season because they need to move a long distance to find their mate, and oviparous adult females may die while moving to lay eggs. Also, the mortality of ambush foragers, animals which are sedentary and capture their prey by ambush, is lower than that of active foragers, animals that hunt preys while moving a long distance (Bonnet et al., 1999).
As it is shown in the mark-recapture results, the red-tongued viper snakes living in Gapado tend to stay in one place and barely move long distances. This can have a positive effect in decreasing their mortality. Also, because most of the red-tongued viper snakes that inhabit Gapado prey on small lizards like Scincella vandenburghi and large centipedes such as Scolopendra subspinipes mutilans, they can minimize the loss in energy used while moving to other places where the availability of food resources is low (Kim, 2011). In fact, there is a study that reported that the energy consumption of Crotalus cerastes, the rattlesnake that inhabits the American Continent and captures preys by ambush, is lower than that of Masticophis flasgellum, a snake that moves around to capture preys (Secor and Nagy, 1994). Also, it was reported that the moving distance of other snakes that belong to the Viperidae, the snakes that capture preys by ambush, is short: Vipera latastei. which inhabits northern Portugal, moves an average 5.4 m per day (Brito, 2003), and Gloydius shedaoensis, which inhabits a small island situated at the northeast of China, moves an average 2 m per day (Shine et al., 2003).
The results of calculating the home ranges of eight individuals that were captured more than three times using the MCP (Minimum Convex Polygon) method show that the home ranges of the red-tongued viper snakes in Gapado were 8~167 m2 (64.0m2 ± 57.0 m2), considerably smaller than that of Vipera latastei in northern Portugal, which is 2,400 m2 (Table 1). Their home range is usually identified by attaching a signal transmitter on their body. Home range sizes vary among snakes (Pough et al, 2004), and it is known to be affected by their body size, seasonal factors, gender, reproduction and ecological characteristics (Shine, 1979; Shine and Lambeck, 1985; Fitzgerald et al., 2002; Brito, 2003; Shine et al., 2003; Whitaker and Shine, 2003; Waldron et al., 2006).
Some snakes like G. shedaoensis, which inhabits an islet situated at the northeast of China (area 0.73 km2) and has average home ranges of 2,429 m2 and 2,851 m2 (< 0.3ha) for females and males respectively, are very sedentary and move less than 2 m a day. On the other hand, the male Hoplocephalus stephensii (Stephen's banded snake), which belongs to the Elapidae and inhabits the forests formed in the east coast of Australia, has a large home range of an average 202,000 m2 (Fitzgerald et al., 2002). The male Crotalus adamanteus (Eastern diamondback rattlesnake) and the male C. horridus (Canebrake rattlesnake), the kinds of rattlesnakes that inhabit same places in South Carolina, the United States, also have very large home ranges of 848,200 m2(169,100~3,104,800 m2) and 483,800 m2(86,000~1,075,000 m2) respectively (Waldron et al., 2006). The red-tongued viper snakes that inhabit Gapado seem to be staying in one place, and their home range was smaller than that of G. shedaoensis, which is very sedentary. The island where G. shedaoensis inhabit is isolated, but because it is situated at the migratory route of birds, many birds pass through this island in spring and autumn. Because G. shedaoensis preys on migratory birds by ambushing them from trees or the ground (Shine et al., 2002a, 2002b, 2002c), it does not move a long distance to capture preys. Many red-tongued viper snakes that inhabit Gapado were also observed to be hiding behind stone walls or nearby them to ambush preys as shown in Figure 3. They chose stone walls as their ambush sites because they can hide right away when in danger. Also, the size of the home range of red-tongued viper snakes that inhabit Gapado is much smaller than that of G. shedaoensis, which inhabits in a similar environment, because there may be a difference in the size of available habitats. G. shedaoensis can use the whole island as its habitat, but as for red-tongued viper snakes, except the village and farmland, they can only use the southwestern part of Gapado, the survey site, and the grasslands formed at the edge of Gapado as their habitats.
As such, the red-tongued viper snakes that inhabit Gapado are highly sedentary and their habitats are limited, a fact that contradicts their very low recapture rates. This may be so because the size of red-tongued viper snakes that inhabit Gapado is smaller than those living in Jeju Island (Kim and Oh, 2014b). The red-tongued viper snakes that inhabit Jeju Island are also easily found as they usually hide behind stone walls.
Home ranges of snakes also vary according to their gender. While there is a small difference between the home ranges of male and female G. shedaoensis, a very sedentary snake which inhabits an islet in the northeast of China (Shine et al., 2003), the average home range of female Epicrates inornatus, a snake that belongs to the Boidae and inhabits Puerto Rico, 7,890 m2, is larger than that of its male (5,000 m2) counterparts (Puente-Rolón and Bird-Picó, 2004). Also, there is a significant difference between the home ranges of female and male Pseudonaja textilis (Brown snake) and Hoplocephalus stephensii, very venomous snakes that belong to the Boidae and inhabit Australia. The home ranges of male and female P. textilis, which inhabit agricultural areas in the northeast part of Australia, are 117,900 m2 and 14,600 m2, males showing eight times larger home ranges than that of females (Whitaker and Shine, 2003). Also, the home ranges of male and female H. stephensii, which live in the forests in the east coast of Australia, are 202,000 m2 and 54,000 m2, males showing a four times larger home range than that of females (Fitzgerald et al., 2002). Also, the home range of male C. adamanteus, a kind of rattlesnake, is three times broader than that of nonpregnant females (males: 842,000 m2, nonpregnant females: 286,300 m2) (Waldron et al., 2006). According to the study result of calculating the home range of the serpent Elaphe schrenckii, which inhabits Korea, using the MCP method based on the radio tracking result, the home range of males (389,600 m2) was two times larger than that of females (162,500 m2) (Lee et al., 2009). As such, the difference between the home ranges of females and the males are more evident in snakes with large home ranges. On the other hand, because the red-tongued viper snakes in Gapado mostly stay in one place, their home range was very small: there was no significant difference between the home ranges of females, 62.0 m2 (n = 4), and males 66.0 m2 (n = 4). (t=0.092, df=6, P>0.05).
Sometimes there is a correlation between the size of the home range and the size of a snake such as Pseudonaja textilis, a big snake with a large home range which lives in Australia (Whitaker and Shine, 2003), but sometimes there is no correlation between them as in the case of Gloydius shedaoensis which inhabits an islet in the northeast of China (Shine et al., 2003). The size of the home range and the body length of the red-tongued viper snakes living in Gapado did have a correlation but it was not statistically significant (Figure 4).
Home ranges of snakes also vary according to seasonal factors, ecological characteristics and the pregnancy status for females. In other words, the size of the home range of the male Vipera latastei, a type of European Viperidae, tends to get larger during September (Brito, 2003); the home range of G. shedaoensis, a type of Viperidae which inhabits an islet in the northeast of China and captures preys through ambush, is very small (Shine et al., 2003). Home range sizes also vary depending on the pregnancy of females: the home range of pregnant snakes is smaller than that of nonpregnant ones (Whitaker and Shine, 2003; Waldron et al., 2006). Accordingly, because there are various factors that affect the home ranges of snakes, more studies need to be carried out to verify the details of the ecological characteristics of snakes.
This study was conducted in a limited area, but can serve as precious data in understanding the Viperidae which play an important role in maintaining biodiversity. There aren't any ecological studies on snakes that inhabit Korea, and data from this study can serve as models for conducting ecological studies on other snakes that are necessary in improving the diversity.
