Journal Search Engine
Search Advanced Search Adode Reader(link)
Download PDF Export Citaion korean bibliography PMC previewer
ISSN : 2288-1115(Print)
ISSN : 2288-1123(Online)
Korean Journal of Ecology and Environment Vol.54 No.1 pp.61-70

Influences of Forest Type and Fragmentation by a Road on Beetle Communities in the Gwangneung Forest, South Korea

Tae-Sung Kwon, Jong-Kook Jung1, Young-Seuk Park2*
Alpha Insect Diversity Lab, Nowon, Seoul 01746, Republic of Korea
1Division of Forest Insect Pests and Diseases, National Institute of Forest Science, Dongdaemun, Seoul 02455, Republic of Korea
2Department of Biology, Kyung Hee University, Dongdaemun, Seoul 02447, Republic of Korea
*Corresponding author: Tel: +82-2-961-0946, Fax: +82-2-961-0244 E-mail:
19/03/2021 23/03/2021 24/03/2021


The effects of forest type and fragmentation of forests by a road on the beetle community were investigated in the Gwangneung Forest, South Korea. Beetles were collected monthly using pitfall traps and by sweeping at 16 sites (eight in coniferous forests and eight in deciduous forests) for one year from April 1993 to April 1994, excluding winter. A total of 17,616 beetles belonging to 271 species from 39 families were collected. Among them, Synuchus cycloderus was the dominant species, with 14,060 individuals accounting for 80% of the total population. The influence of forest type (coniferous and deciduous) or sampling region (fragmented by a road) on species richness (number of species) was observed. We found that species richness was substantially different depending on the sampling method and taxa used. Beetles collected using pitfall traps responded more sensitively to habitat types than those collected by sweeping. Four dominant families differently responded to forest fragmentation and forest types. Carabidae was influenced by forest fragmentation, whereas Staphylinidae and Curculionidae were influenced by forest types. Chrysomelidae was not influenced.


    National Research Foundation of Korea(NRF)


    Beetles are the most diverse taxa, with an estimated 300,000~450,000 species (Nielsen and Mound, 1999). They are very important insects that play a crucial role in maintaining the ecosystem and have various ecological roles (Bouchard et al., 2017). Therefore, many studies have been conducted on the diversity of beetles and the influence of environmental factors on beetle communities. However, owing to the large number of beetle species, studies on all beetles are rare, and are usually conducted on specific taxa that serve as potential indicators such as Carabidae (Rainio and Niemelä, 2003), Scarabaeidae (Spector, 2006), and Staphylinidae (Pohl et al., 2007) for environmental changes. Studies with selected taxa can induce biased responses to environmental changes because some taxa have unique ecological niches, resulting in a biased response to the environment. Several studies have focused on the usefulness of higher taxa as a surrogate for species richness (Ricotta et al., 2002;Báldi, 2003;Kwon et al., 2016;Lee et al., 2020), although there is an opposite view (Rosser and Eggleton, 2012). It appears that the whole fauna of the order beetles is more likely to show the influence of the environment than the specific taxa (de Oliveira et al., 2020).

    Forests in Korea are mainly deciduous, such as oaks and coniferous, such as red pine, Korean pine, and Rigida pine (KFS, 2021). Generally, the diversity of insects in deciduous forests is higher than that in coniferous forests because of the higher diversity of understory vegetation in deciduous forests. However, few studies have investigated this hypothesis. In Korea, 63% of the land is covered by forests, a large proportion of which is fragmented by roads and urbanization. Although forest fragmentation caused by roads has a remarkable influence on beetle diversity, few studies have been conducted in South Korea, especially on carabid beetles (Do and Joo, 2013;Jung et al., 2018).

    The questions in this study are as follows: 1) Are the beetles that live in deciduous and coniferous forests different? 2) Are there more beetle species living in deciduous forests? 3) Are the beetles living in forests fragmented by roads any different? 4) Does forest fragmentation by roads and different forest types impact the entire Coleoptera and low-level taxa?


    1. Field sampling

    This study was conducted in the Gwangneung Forest, an old forest preserved as a tomb forest of King Sejo of the Joseon dynasty. As a result, a wide variety of plants, animals, and insects live here (KNA, 2021). Field sampling was conducted monthly at 16 sampling sites from April 1993 to April 1994, excluding the winter (Table 1). The Gwangneung Forest was divided by a road into two regions: Soribong and Jukyupsan, each of which had eight sampling sites (four in the deciduous forest and four in the coniferous forest). Details of the sampling sites have been reported by Kwon (1995).

    The beetles were collected using pitfall traps and sweeping at each sampling time. In the sweeping sampling, 33 sweeps were repeated three times at each site using an insect net (diameter 40 cm). In the pitfall sampling, five cellophane cups (7 mm in mouthpart diameter, 7 mm in height) were set up on the ground at 1~1.5 m intervals, and 20% formalin was used as a preservative in the pitfall traps. The collected samples were preserved in 70% ethyl alcohol for identification. After splitting the specimens based on morphological groups, three or four specimens in each group were dried to be identified at the species or genus level. Species that could not be identified were classified as the morphological species. The specimens are currently preserved in the Insect Specimen Room of the National Institute of Forest Science, Korea.

    2. Data analysis

    The differences in beetle communities were analyzed using different sampling methods at different taxonomic levels. Nonmetric multidimensional scaling (NMDS) was conducted to reveal the differences in beetle communities among sampling sites and to determine the relationship between beetle communities and environmental factors. NMDS was analyzed using “envfit” in “vegan” package (Oksanen et al., 2020) in R (R Core Team, 2020). Species richness was considered in all samples, sampling methods, and the four selected families (Carabidae, Staphylinidae, Chrysomeridae, and Curculionidae). These were compared between forest types (deciduous forests and coniferous forests) and sampling regions separated by roads (Jukyupsan and Soribong) using a regression model with categorical variables (forest type and sampling region). Species richness of the four families was used for samples collected with both pitfall traps and sweeping. A t-test was conducted to compare the differences in species richness between forest types, as well as between sampling regions. Statistical analyses were conducted with the ‘stat’ package in R (R Core Team, 2020). Using the collected data, Kwon (1996) reported the diversity and abundance of Carabidea and found that habitat fragmentation by roads did not influence the distribution of carabid beetles.


    1. Beetle assemblages

    From the field survey, 271 taxa from 39 families with 17,616 beetles were collected (Table 2). Among them, 45 taxa (16%) were identified at the family level, 97 (35%) at the genus level, and 105 species (38%) were identified at the species level. The number of individuals collected with the two different sampling methods was very different because of the differences in habitats and feeding habits of each taxon. The species collected using sweeping showed higher diversity than those collected by pitfall traps (Table 2, Fig. 1). Beetles belonging to Carabidae and Staphylinidae which forage on surfaces were collected mainly with pitfall traps, whereas beetles feeding on vegetation such as Chrysomelidae and Curculionidae were collected mainly by sweeping (Table 2).

    Synuchus cycloderus was the dominant species, with 14,060 individuals accounting for 80% of the total population (Table 3). However, this species was collected only with pitfall traps, indicating that it mainly forages on the ground. Other abundant species (1% or more), including Chlaenius naeviger, Eucarabus sternbergi, Harpalus sp. 1, Anatis halonis, and Psammoecus triguttatus were also collected using pitfall traps. Pseudocneorhinus sp., Enaptorrhinus granulatus, and Phyllolytus variabilis of Curculionidae were the dominant species collected during sweeping. In contrast, only one individual was collected from 119 taxa, accounting for 44% of the total collected taxa. It is a common phenomenon that many species are rare.

    It is necessary to clarify why S. cycloderus is extremely abundant in the Gwangneung Forest. Lee and Kwon (2013) argued that this dominance occurs as the resource monopoly by the dominant species intensifies in a stable environment, like in old forests. The population size of S. cycloderus is relatively smaller in mountains in other regions (Jung et al., 2018, 2020) than in the present study. The difference could partly be due to the forest environment depending on forest type, soil properties, disturbance, etc. In fact, S. cycloderus is known to prefer dry forests (Fujita et al., 2008;Jung et al., 2018, 2020). However, limited species that are tolerant to disturbance can prevail in severely disturbed areas (Connell, 1978). Therefore, an intermediate disturbance hypothesis presenting high species richness with an intermediate disturbance has been proposed (Connell, 1978;Osman, 2015), although conflicting results have been reported (Fox, 2013).

    NMDS showed differences in beetle community composition according to forest type, sampling region (habitat fragmentation), sampling methods, and among the four selected families (Fig. 2, Table 4). The beetle communities collected using pitfall traps were significantly different with respect to the forest type and between the two regions divided by a road (Fig. 2a, b, Table 4). However, the beetles collected by sweeping did not differ with respect to the forest types and regions (Table 4). Meanwhile, Carabidae was not different between forest types, but differed between the two regions in the NMDS ordination (Fig. 2c, d, Table 4). In contrast, Staphylinidae and Curculionidae were not different between the two regions, but differed between the two forest types (Fig. 2e-h, Table 4). Chrysomelidae was not different between the two forest types or between the two regions (Table 4). These results indicate that the dispersion capability of species may be crucial in understanding the species composition in two regions fragmented by roads. Unlike Staphylinidae, Curculionidae, and Chrysomelidae, many carabid beetle species that prefer forest environments cannot fly because their hindwings are generally reduced or lacking (Jung et al., 2018).

    In general, predatory insects such as Staphylinidae have a low dependence on plants, but leaf-feeders such as Chrysomelidae depend on plants living in forests. Therefore, it is highly likely that Chrysomelidae should respond to forest type, but Staphylinidae should not. This study showed opposite responses in both these families. However, Curculionidae, another plant feeder, responded significantly to forest type, but not to forest fragmentation. Carabidae responded significantly to forest fragmentation, but not to forest type. Thus, the response to the environment (type of forest, area cut off by roads) was very different among taxa. Therefore, using one specific family as an indicator to evaluate the entire fauna of Coleoptera may be problematic. In the data pooled from the two collection methods, pitfall traps showed a significant response to forest type and forest fragmentation, but sweeping showed no response. Beetles collected using pitfall traps are mainly predators or decomposers, and those collected by sweeping are mainly plant feeders. Therefore, the latter are expected to respond to forest type, contrary to the former. However, this was not observed in the present study.

    2. Diversity

    The total of 271 taxa collected in this survey was relatively lower than those reported in other studies. For example, a total of 350 taxa were collected at seven sampling sites in the Wando Arboretum Forest, South Korea, in 2015 (Kwon et al., 2018), and 310 taxa were identified at five sites in Gariwangsan Mountain in 2011 (Lee et al., 2014). In the present study, 141 taxa with 17,060 individuals and 166 taxa with 556 individuals were collected using pitfall traps and a sweeping method, respectively. In contrast, Lee et al. (2014) collected more taxa using pitfall traps (110 taxa) than by the sweeping method (33 taxa) in Gariwangsan Mountain.

    The effects of forest type or region on species richness were substantially different depending on the sampling method (Fig. 3a, Table 5). The total taxa richness was higher in Jukyupsan than in Soribong; although the difference was not statistically significant between regions (t=1.125, p=0.286) (Fig. 3a). Meanwhile, the total taxa richness was slightly different between deciduous and coniferous forests with low statistical significance (t= -1.991, p=0.067). This is also supported by the results of previous studies. For example, Moon et al. (2018) reported that moth diver- sity was higher in Mongolian oak forests than in Japanese larch forests, while Jung et al. (2020) reported that species richness of carabid beetle pine forests in Korea is generally lower than that in deciduous forests. However, in our study, species richness displayed higher variation in coniferous forests than in broad-leaved forests (Fig. 3a). This may be related to the degree of openness of canopy. In deciduous forests, this degree was usually constant, whereas in coniferous forests, there was a large difference between the sampling sites (Table 1). A high degree of canopy openness induces a high diversity of substratum vegetation, resulting in a high diversity of insects (Moon et al., 2018). In addition, some taxa, such as carabid beetles, are affected by forest fragmentation when there is a decrease in forest patch size (Jung et al., 2018).

    Species richness was different between the two forest types in samples collected using pitfall traps, although the difference was not statistically significant (t= - 2.05, p=0.060) (Fig. 3b). The difference was more clearly explained by the addition of forest types and regions in the ANOVA model (F=4.1, p=0.041) (Table 5). However, no difference was observed between the sampling regions.

    Species richness of Staphylinidae was significantly different between forest types, being plentiful in deciduous forests (t= -3.379, p=0.005) (Fig. 3e), which can be explained by considering both, forest types and sampling regions (F=6.80, p=0.010) (Table 5). However, no significant difference was detected between the sampling regions (F=0.92, p=0.352) (Fig. 3e). Meanwhile, species richness was not significantly different between the sampling regions and forest types in the other three families, Carabidae, Chrysomelidae, and Curculionidae (Fig. 3d, f, g). In the case of vegetation foraging (plant feeder) insects, the dependence on plants was higher than that in predatory insects (Sobek et al., 2009). Therefore, it was expected that there would be significant differences according to forest type. However, the present study does not support this hypothesis. Staphylinidae, a predator, is expected to be relatively independent of forest type, which was clearly not the case since this family showed a sensitive response to forest types.

    In conclusion, our study observed the influence of forest type and sampling region (fragmented by a road) on species richness. The differences were significant depending on the sampling method and taxa. Beetles collected using pitfall traps responded more sensitively to habitat type than those collected by sweeping. Four dominant families differently responded to forest fragmentation and forest types. Carabidae was influenced by forest fragmentation, whereas Staphylinidae and Curculionidae were influenced by forest types. Chrysomelidae was not influenced.

    Author information

    Tae-Sung Kwon (Alpha Insect Diversity Lab, Principal Researcher), Jong-Kook Jung (National Institute of Forest Science, Researcher), Young- Seuk Park (Kyung Hee University, Professor)

    Author contribution statement

    Conceptualization T.- S. Kwon, Y.-S. Park, Field survey and data collection: T.-S. Kwon, Data analysis and writing: T.-S. Kwon, J.-K. Jung, Y-S. Park, Review and editing: T.-S. Kwon, J.-K. Jung, Y-S. Park

    Conflict of interest

    The authors declare no conflict of interest. The sponsors had no role in the design, execution, interpretation, or writing of the study.


    This work was supported by the National Research Foundation of Korea (NRF) funded by the Korean government (MSIP, grant number NRF-2019R1A2C 1087099).



    Species rank abundance at two different study regions with two different sampling methods.


    NMDS ordination with beetle communities. Beetle assemblage collected using pitfall traps (a and b), assemblages of Carabidae (c and d), assemblages of Staphylinidae (e and f), and assemblages of Curculionidae (g and h). DF: deciduous forest, CF: coniferous forest, J: Jukyupsan, and S: Soribong. Sweeping and Chrysomelidae were not presented because factor analysis on the nonmetric multidimensional scaling displayed no statistical significance as shown in Table 4.


    Comparison of species richness according to forest type and sampling regions. (a) Total species richness of beetles collected using both pitfall trap and sweepings, (b) species richness with pitfall trap, (c) species richness with sweeping, and species richness of Carabidae (d), Staphilinidae (e), Chrysomelidae (f), and Curculionidae (g).


    Characteristics of study sites in Gwangneung.

    Number of species and individuals in each family collected using two different sampling methods in Gwangneung in 1993~1994.

    Ten most dominant species and their abundance at two different sampling regions using two different sampling methods.

    Factor analysis on the nonmetric multidimensional scaling of beetle communities.

    Regression analysis with categorical variables on species richness with one or two factors (forest type and study region). F_type+ Region: regression with two variables, F_type+Region+F_type×Region: regression with two variables by considering interactions between two variables


    1. Báldi, A. 2003. Using higher taxa as surrogates of species richness: a study based on 3700 Coleoptera, Diptera, and Acari species in Central-Hungarian reserves. Basic and Applied Ecology 4: 589-593.
    2. Bouchard, P. , A.B.T. Smith, H.B. Douglas, M.L. Gimmel, A.J. Brunke and K. Kanda.2017. Biodiversity of Coleoptera, p. 337-417. In: Insect Biodiversity: Science and Society (Foottit, R.G. and P.H. Adler, eds.). 2nd ed. Wiley, Chichester.
    3. Connell, J.H. 1978. Diversity in tropical rain forests and coral reefs. Science 199(4335): 1302-1310.
    4. de Oliveira, Jr S.S. , J.C. Ortega, L.G. dos Santos Ribas, V.G. Lopes and L.M. Bini.2020. Higher taxa are sufficient to represent biodiversity patterns. Ecological Indicators 111: 105994.
    5. Do, Y. and G.J. Joo.2013. The effect of fragmentation and intensive management on carabid beetles in coniferous forest. Applied Ecology and Environmental Research 11(3): 451-461.
    6. Fox, J.W. 2013. The intermediate disturbance hypothesis should be abandoned. Trends in Ecology & Evolution 28(2): 86- 92.
    7. Fujita, A. , K. Maeto, Y. Kagawa and N. Itô.2008. Effects of forest fragmentation on species richness and composition of ground beetles (Coleoptera: Carabidae and Brachinidae) in urban landscapes. Entomological Science 11: 39-48.
    8. Jung, J.K. and J.H. Lee.2020. Trait-specific responses of carabid beetle diversity and composition in Pinus densiflora forests compared to broad-leaved deciduous forests in a temperate region. Diversity 12(7): 275-286.
    9. Jung, J.K. , S.K. Lee, S.I. Lee and J.H. Lee.2018. Trait-specific response of ground beetles (Coleoptera: Carabidae) to forest fragmentation in the temperate region in Korea. Biodiversity and Conservation 27: 53-68.
    10. KFS.2021. Korea Forest Service. (accessed on 15 Feb, 2021).
    11. KNA.2021. Korea National Arboretum. (accessed on 15 Feb, 2021).
    12. Kwon, T.-S. 1995. Structures of hemipterous insect communities in Kwangnung Experimental Forest. FRI Journal of Forest Science 51: 45-52.
    13. Kwon, T.-S. 1996. Diversity and Abundance of Ground Beetle (Coleoptera; Caraboidae) in the Kwangnung Experimental Forest. Korean Journal of Entomology 26: 351-361.
    14. Kwon, T.-S. , Y.S. Kim, S.W. Lee and Y.-S. Park.2016. Changes of soil arthropod communities in temperate forests over 10 years (1998-2007). Journal of Asia-Pacific Entomology 19(1): 181-189.
    15. Kwon, T.-S. , Y.-K. Park, S.-S. Kim, J.-S. An, S.-M. Lee and G.-H. Kim.2018. Beetles of Wando-Arboretum. Jeolanam-do, Korea. Wando-Arboretum 258p.
    16. Lee, C.M. and T.S. Kwon.2013. Community structure, species diversity of insects (ants, ground beetles), and forest health in the Hongneung forest. Journal of Korean Forest Society 102: 97-106.
    17. Lee, C.M. , T.S. Kwon, Y.K. Park, S.-S. Kim, J.H. Sung and Y.K. Lee.2014. Effect of forest management on diversity and abundance of beetles (Coleoptera) in Mt. Gariwang-san. Research Report 14-06. Korea Forest Research Institute, Seoul.
    18. Lee, D.S. , T.S. Kwon, S.S. Kim, Y.K. Park, H.M. Yang, W.I. Choi and Y.-S. Park.2020. Changes of Ground-dwelling Arthropod Communities for 10 Years after Thinning in a Pinus koraiensis Plantation. Korean Journal of Ecology and Environment 53(2): 208-219.
    19. Moon, M.Y. , S.-S. Kim, D.-S. Lee, H.M. Yang, C.-W. Park, H.S. Kim and Y.-S. Park.2018. Effects of Forest Management Practices on Moth Communities in a Japanese Larch (Larix kaempferi (Lamb.) Carrière) Plantation. Forests 9(9): 574.
    20. Nielsen, E.S. and L.A. Mound.1999. Global diversity of insects: the problems of estimating numbers, p. 213-222. In: Nature and Human Society: the Quest for a Sustainable World (Raven, P.H. and T. Williams, eds.). National Acedemy Press, Washington, DC.
    21. Oksanen, J. , F.G. Blanchet, M. Friendly, R. Kindt, P. Legendre, D. McGlinn, P.R. Minchin, R.B. O’Hara, G.L. Simpson, P. Solymos, M.H.H. Stevens, E. Szoecs and H. Wagner.2020. vegan: Community Ecology Package. R package version 2.5-7.
    22. Osman, R.W. 2015. The Intermediate Disturbance Hypothesis, p. 441-450. In: Encyclopedia of Ecology (Fath, B. ed.). 2nd ed. Elsevier, Oxford.
    23. Pohl, G.R. , D.W. Langor and J.R. Spence.2007. Rove beetles and ground beetles (Coleoptera: Staphylinidae, Carabidae) as indicators of harvest and regeneration practices in western Canadian foothills forests. Biological Conservation 137(2): 294-307.
    24. R Core Team2020. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. URL
    25. Rainio, J. and J. Niemelä.2003. Ground beetles (Coleoptera: Carabidae) as bioindicators. Biodiversity and Conservation 12: 487-506.
    26. Ricotta, C. , M. Ferrari and G. Avena2002. Using the scaling behaviour of higher taxa for the assessment of species richness. Biological Conservation 107: 131-133.
    27. Rosser, N. and P. Eggleton.2012. Can higher taxa be used as a surrogate for species-level data in biodiversity surveys of litter/soil insects? Journal of Insect Conservation 16: 87- 92.
    28. Sobek, S. , M.M. Gobner, C. Scherber, I. Steffan-Dewenter and T. Tscharntke.2009. Tree diversity drives abundance and spatiotemporal beta-diversity of true bugs (Heteroptera). Ecological Entomology 34: 772-782.
    29. Spector, S. 2006. Scarabaeine dung beetles (Coleoptera: Scarabaeidae: Scarabaeinae): an invertebrate focal taxon for biodiversity research and conservation. The Coleopterists Bulletin 60: 71-83.