Next Article in Journal
Phenolic Leaf Compounds in Ash Trees (Fraxinus excelsior L.) in the Context of Ash Dieback
Previous Article in Journal
Fire Effects on Lichen Biodiversity in Longleaf Pine Habitat
Previous Article in Special Issue
Fragmentation, Ecological Assessment, and Diversity of EU Forest Habitat Types: A Case Study in the Calabria Region Oak Woodlands (Southern Italy)
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Forests
Volume 16
Issue 9
10.3390/f16091386
forests-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Butterfly Community Responses to Urbanization and Climate Change: Thermal Adaptation and Wing Morphology Effects in a Conserved Forest, South Korea

by
Tae-Sung Kwon
1,
Sung-Soo Kim
2,
Ilju Yang
3,
A Reum Kim
4 and
Young-Seuk Park
3,5,*
1
Alpha Insect Diversity Lab, Nowon, Seoul 01746, Republic of Korea
2
Research Institute for East Asian Environment and Biology, Gangdong, Seoul 05236, Republic of Korea
3
Department of Biology, Kyung Hee University, Dongdaemun, Seoul 02447, Republic of Korea
4
Division of Forest Ecology, National Institute of Forest Science, Dongdaemun, Seoul 02445, Republic of Korea
5
Korea Institute of Ornithology, Kyung Hee University, Dongdaemun, Seoul 02447, Republic of Korea
*
Author to whom correspondence should be addressed.
Forests 2025, 16(9), 1386; https://doi.org/10.3390/f16091386
Submission received: 20 July 2025 / Revised: 21 August 2025 / Accepted: 25 August 2025 / Published: 28 August 2025
(This article belongs to the Special Issue Biodiversity in Forests: Management and Monitoring for Conservation—2nd Edition)
Browse Figures
Versions Notes

Abstract

Habitat and climate changes driven by human activities are altering the distribution of organisms globally. In South Korea, recent temperature increases have exceeded twice the global average, and habitats have markedly changed and shrunk due to urban development driven by population growth and economic expansion. Despite its high biodiversity and over 500 years of preservation, Gwangneung Forest in South Korea has experienced habitat alterations due to the urbanization of surrounding rural areas since the 1990s. In this study, we aimed to evaluate how butterfly communities respond to urbanization and climate change using long-term monitoring data (1998–2015) from the conserved Gwangneung Forest. We considered the thermal adaptation types (cold-, warm-, and moderately adapted species), habitat types (forest edge, forest inside, and grassland), diet breadth (monophagous, oligophagous, and polyphagous), and wingspan of butterflies. Linear regression analysis of the abundance trends for each species revealed that cold-adapted species experienced population declines, while warm-adapted species showed increases. Changes in butterfly abundance were associated with both thermal adaptation type and wingspan, with larger, more mobile species showing greater resistance to habitat loss in surrounding areas. To preserve butterfly diversity in Gwangneung Forest and across South Korea, it is crucial to conserve open green habitats—such as gardens, small arable lands, and grasslands—within urban areas, especially considering the impacts of climate change and habitat loss, which disproportionately affect smaller species with limited mobility.

1. Introduction

Habitat and climate changes caused by human activities induce alterations in the distribution areas of organisms on various scales (local, regional, continental, and global) [1,2,3,4,5]. Since 1850, global temperatures have increased by an average of 0.06 °C per decade, with 2023 being approximately 1.36 °C warmer than the preindustrial average of the late 19th century (1850–1900) [6,7]. Meanwhile, temperatures in South Korea have recently surpassed twice the global average [8], and the habitats of many species have been markedly altered and reduced due to human population growth and economic development [9]. This has induced changes in the distribution area and a decline in population size in many taxa.
Climate and land use strongly influence biodiversity, and habitat connectivity has a positive impact on ecologically specialized, sedentary, and endangered species. In particular, land-use heterogeneity promotes the diversity of ecological niches, which subsequently accelerates species diversity, including that of specialist species [10]. Liivamägi et al. [11] revealed that the surrounding landscape and habitat characteristics are crucial for butterfly species richness. Understanding the butterfly–surrounding landscape relationship is relevant to butterfly conservation [12]. Butterflies living in a certain area are not fixed to that area, but population and gene exchange occur through immigration to and exodus from surrounding areas [13]. If nearby habitats are reduced, these populations and gene flow are restricted, resulting in biodiversity decline [14]. Therefore, many butterfly species may have declined due to the changes in metapopulation dynamics resulting from habitat reduction in the surrounding area. The decline in open-habitat butterflies is expected to exceed that in forest-residing butterflies because the forested area is legally protected but the agricultural area with open butterfly habitats is sharply declining due to urbanization in South Korea.
Climate change-induced northward shifts in southern butterfly distributions were observed in South Korea, with vegetation and climate changes triggering butterfly community changes [4,5,15]. Based on over 40 years of personal observations of butterflies across various regions of South Korea, their populations have markedly declined in all areas compared to those in the past (Kim S_S-., personal observation). In South Korea, the abundance of many butterfly species is expected to decrease because the green spaces used as butterfly habitats, such as grasslands, shrub lands, and arable lands, have been changed into urban areas, such as apartment complexes, buildings, industrial factories, and roads, resulting in habitat declines.
Recent studies have reported a global decline in the abundance and distribution of insects, including Lepidoptera [16,17,18], with habitat degradation and climate change being major drivers of insect decline [19]. In particular, large moths displayed greater declines than small moths in the Mediterranean region [20] and in South Korea [21], indicating that morphological traits, such as wingspan, are related to these declines. The moth wingspan is closely linked to ecological traits and habitat requirements [21]. Large moths often have broader wingspans, which may confer advantages such as increased flight endurance and dispersal ability [22].
Gwangneung Forest, one of the few old forests in South Korea, has been well preserved thanks to the government’s continuous efforts and is home to various organisms, making it a UNESCO Biosphere Reserve (https://www.gfbr.kr/, accessed on 23 June 2024). However, habitat isolation and fragmentation due to the development of surrounding areas are affecting the forest’s biodiversity. A total of 148 butterfly species have been recorded in the Gwangneung Forest region, representing 66% of South Korea’s 225 species [23]. However, Kwon et al. [24] estimated that only 128 species currently inhabit the area, indicating that around 20 species (13%) have disappeared. For instance, Argynnis nerippe was once widespread but now only exists in the central regions of South Korea, likely due to habitat loss and climate change. Other species like Leptidea morsei and Maculinea teleius face local extinction due to similar factors [25]. Conversely, Bibasis aquilina, newly discovered in this study, displays an expanding distribution [25], reflecting broader shifts in habitat distribution patterns due to environmental changes.
Despite the effective conservation of Gwangneung Forest, the surrounding butterfly habitats are facing severe threats due to a marked reduction in agricultural land and a significant expansion of urban areas. This shift in land use is likely to disrupt metapopulation dynamics, leading to a greater decline in species numbers than any potential increases, ultimately reducing overall butterfly abundance. This habitat reduction may disproportionately affect populations of smaller butterfly species. Generalist species, characterized by their broad ecological niches, are thought to withstand disturbances from climate and habitat changes more effectively [26,27].
Therefore, in this study, we aimed to assess how butterfly communities respond to urbanization and climate change using long-term monitoring data (1998–2015) from the conserved Gwangneung Forest. Specifically, we examined the influence of thermal adaptation types (cold-, warm-, and moderately adapted species), habitat preferences (forest edge, forest interior, and grassland), diet breadth (monophagous, oligophagous, and polyphagous), and wingspan on butterfly population trends.

2. Materials and Methods

2.1. Study Area

The field survey was carried out in Gwangneung Forest (37.748419 N, 127.153548 E), located near Seoul, in South Korea (Figure 1). Gwanneung Forest comprises natural stands (836 ha) and plantations (1387 ha), which are primarily composed of Pinus koraiensis, P. rigida, Larix leptolepis, and Abies holophylla [28]. Trees were planted on 605 ha of the plantations from 1914 to 1970. The natural stands are primarily composed of Quercus serrata, Q. mongolica, Carpinus laxiflora, and Pinus densiflora. Gwangneung Forest has been preserved by the government for more than 500 years and is home to various wildlife [29]. Although Gwangneung Forest is well preserved, the rapid urbanization of surrounding areas has caused large-scale habitat changes. The surrounding areas were rural with small villages and agricultural areas even in the early 1990s but have now been altered into urban areas, including apartment complexes and industrial complexes (Figure 1, Supplementary Material Figure S1). The habitat in South Korea, including Gwangneung Forest, has experienced an increase in temperature. The annual mean temperature and annual precipitation are 12.8 °C and 88.0 mm, respectively [30], and the annual mean temperature in the recent 30 years increased 1.6 °C compared to that in the previous 30 years [31].
Butterfly monitoring was conducted using the line transect method by walking along the forest road in Gwangneung Forest twice a month between April and October for 18 years from 1998 to 2015, 11−13 times a year [32]. However, the surveys were conducted 3−7 times a year from 1999 to 2004 because of limited field survey resources. The survey route length was approximately 4 km: 2 km of a plantation area and 2 km of a natural broad-leaved forest (Figure 1). Butterfly surveys were conducted on clear days using the line transect method. Observers proceeded at a consistent pace of 2 km/h along designated transects, recording all individuals of each butterfly species observed within a 10 m wide transect band directly ahead. Details of the environmental factors and sampling methods of the survey site were outlined by Kwon et al. [24].

2.2. Meteorological Data

To evaluate the effects of meteorological factors on butterfly population changes, we obtained monthly average temperatures data, such as mean temperature, minimum temperature, and maximum temperature, measured at Pocheon weather station, which is the nearest station to the study area and is operated by the Korea Meteorological Administration (http://www.kma.go.kr, accessed on 23 June 2024).

2.3. Biological Traits

Butterfly community alterations were evaluated based on biological traits, such as the thermal adaptation types, habitat types, diet breadth, and wingspan of each species (Supplementary Material Table S1). Thermal adaptation types were categorized into three classes in Korea—species with a northerly distribution (hereafter, cold-adapted species), species with a southerly distribution (warm-adapted species), and miscellaneous species (moderately adapted species)—depending on their thermal adaptation and distribution areas [25]. Cold-adapted species are defined as those whose southern distribution boundary in East Asia is located within Korea. Conversely, warm-adapted species are identified by a northern distribution boundary. Cold-adapted species typically thrive in areas cooler than Korea, while warm-adapted species are more suited to warmer regions. Moderately adapted species lack distinct distributional boundaries, suggesting they may be well suited to the thermal conditions prevalent in Korea.
Habitat types were classified into three groups according to habitat preferences: forest interior, forest edge, and grassland in an open area [25,33]. Diet breadth was categorized into three groups based on feeding preferences: monophagous (feeding on plants of one genus), oligophagous (feeding on plants of one family), and polyphagous (feeding on two or more plant families) [25,33]. It should be noted that the diet breadth is primarily determined by caterpillar preferences. Wingspan data were obtained from Joo et al. [34].

2.4. Data Analysis

Among the butterflies observed in the survey, the abundant species observed for more than 5 years were used in the analysis to evaluate the abundance shift for each species. The number of surveys was slightly different each year. Therefore, the standardized abundance was used by calculating the abundance in each year divided by the number of surveys in the corresponding year (Supplementary Material Table S2). The standardized abundance was log-transformed (ln(abundance + 1)) to reduce variations before the analyses. The abundance change for each species was determined using the slope of a linear regression analysis against years. A one-sample t-test was conducted to test whether the slope was equal to 0 or not. A χ2-test was used to determine whether the number of species with a slope exceeding 0 differed from that of species with a slope below 0.
A backward stepwise regression analysis was performed with the slope values of selected abundant species as dependent variables and with thermal adaptation types, habitat types, diet breadth, wingspan, and abundance (abundance per survey) as explanatory variables. The model performance was estimated with the coefficient of determination (R2) and Akaike information criterion (AIC). The abundance change for each category of biological traits was evaluated through linear regression, and the relationships between the abundances in each category were estimated using a Spearman rank correlation coefficient. In addition, the interactive effects of habitat change (represented by habitat types or wingspan) and climate change (thermal adaptation types) were analyzed using regression analysis, incorporating interaction terms such as habitat types × thermal adaptation types or wingspan × thermal adaptation types. To assess multicollinearity in the regression analyses, correlation coefficients between quantitative variables were examined. Since all were below 0.1 for the correlation coefficient, no variables were removed.
The Kruskal–Wallis test was conducted to compare the differences in the regression slopes in each species in each biological trait category, and the Dunn test, as a multiple comparison test, was carried out when the slope significantly differed among categories. An F-test linear regression was used to evaluate the relationship between the slope of each species and the wingspan. The Wilcoxon test was conducted to compare differences in abundance between two study periods that showed large differences from each other.
The lm function in the package stats in R [35] was used for linear and backward stepwise regression analyses. The krustal.test function in the package stats was used for the Kruskal–Wallis test, and the function dunntest in the package FSA [36] in R was used for the Dunn test with Bonferroni correction. The wilcox.test function in the package stats in R was used for the Wilcoxon test. The Spearman rank correlation was calculated with the function rcorr in the package Hmisc [37] in R.

3. Results

A total of 100 species (11,867 individuals) in five families were observed for 18 years in Gwangneung Forest (Table 1, Supplementary Material Table S2). The family Nymphalidae had the highest species richness with 48 species (7373 individuals) followed by Lycaenidae with 25 species (684 individuals). Pieridae had only six species, but their abundance was high, with 2418 individuals following Nymphalidae. The more often species were observed among the years, the greater their abundance (Figure 2a); however, the rate of increase in abundance declined as the frequency increased. The overall butterfly abundance displayed decreasing trends by survey year, although the statistical evidence was marginal (y = 14.773 − 0.322 × year, R2 = 0.2019, F1,16 = 4.047, p = 0.061) (Figure 2b). A marked decrease in 2003 and 2004 recovered in the following two years, and abundance during the 2003–2005 period was significantly lower than those during the preceding (2000–2002) (7.67 ± 1.81, mean ± standard deviation) and subsequent (2006–2008) (14.55 ± 0.907) periods (Wilcoxon test, W = 0.00, p = 0.028). After its recovery in 2006, the abundance decreased gradually during the survey period. This population decline had a strong positive correlation with low temperature in May (r = 0.709, p < 0.001) (Figure 2b,d). Species richness displayed high fluctuation and had a positive correlation with abundance (r = 0.65, p < 0.01) (Figure 2b,c). Most biological traits displayed similar patterns in abundance shifts (Figure 3). Meanwhile, the abundance of species inhabiting grasslands was lower after 2007 compared to before 2007 (Figure 3b).
The dominant species observed throughout the survey period exhibited extremely high abundance compared to other species (Table 1). The dominant species was Libythea celtis in Nymphalidae with 3540 individuals (29.8% of the total abundance), followed by Pieris melete in Pieridae with 1989 individuals (16.8% of the total abundance). These two species accounted for almost half of the overall abundance, showing a high dominance of butterflies in the study area. Meanwhile, 8 species, such as Artopetes pryeri, were observed with only one individual, and 16 species, such as Favonius yuasai, were observed at only one survey time (Supplementary Material Table S2). Bibasis aquilina was newly recorded in this study.
Most categories in the trait types had positive correlations with each other (Figure 4). In particular, the abundance of moderately-adapted species not belonging to cold- or warm-adapted species had a strong positive correlation with species inhabiting the forest edge and with monophagous and oligophagous species among the diet types (r > 0.90, p < 0.05).
Among the 100 butterfly species observed in this survey, a subset of 67 species consistently recorded over a five-year period was analyzed using a linear regression to assess changes in their abundance over time relative to the sampling years. The mean slope for the selected species was −0.004 ± 0.009 (mean ± standard deviation), indicating that it was not zero (one sample t = −3.77, df = 66, p < 0.001), and the frequency of the slopes followed a normal distribution (Shapiro–Wilk test, W = 0.9642, p > 0.05) (Figure 5). Among the 67 species, 46 species (68.7%) decreased in their abundance by showing negative slopes, while 21 species increased with positive slopes (χ12 = 9.3, p < 0.01). The backward stepwise regression analysis clearly showed the influence of the explanatory variables on the linear regression slope. In the backward stepwise regression, abundance was dropped first, followed by dietary breadth and habitat type (Table 2). The final model retained only two variables: thermal category and wingspan. The interactive influences of habitat change and climate change (e.g., habitat types × thermal adaptation types and wingspan × thermal adaptation types) were not significant (p > 0.05). Warm-adapted species had positive slope values of 0.002 ± 0.007, whereas cold-adapted species had negative slope values of −0.003 ± 0.009, although they were not significantly different from each other (Tukey’s HSD test, p > 0.05) (Figure 6a). However, the slopes of the warm-adapted species significantly differed from those of the moderately adapted species (Tukey’s HSD test, p < 0.05). Meanwhile, the slope increased as a function of the wingspan (y = −0.01108 + 0.000134x, Adj. R2 = 0.0641, F1,65 = 5.52, p < 0.02) (Figure 7). The interaction between the two variables was not significant in determining the slopes (p > 0.31). The wingspan was not significantly different among the thermal adaptation types (Figure 6b).

4. Discussion

4.1. Butterfly Decline Driven by Surrounding Habitat Loss

The number of butterfly species showing a negative population trend was twice that of species with a positive trend, and the average slope value was below zero, indicating a general decline in species in Gwangneung Forest. Although the statistical significance was marginal (p = 0.061), overall butterfly abundance exhibited a declining trend over the survey period, with marked decreases observed in 2003 and 2004 in this study. This decline in abundance might have been caused by the low temperature in this period. The minimum temperature in May had a strong positive correlation with abundance (r = 0.709, p < 0.001) (Figure 2d). Similarly, ant assemblages in a Gwngneung LTER site, located in the central part of the butterfly monitoring route, showed lower diversity in 2004 compared to 2003 and 2005, and this decline in 2004 was attributed to the heavy rainfall in 2003 [38]. Therefore, the additive influence of cold stress and heavy rainfall in the previous year might have led to the great decline in butterfly populations in 2004.
The observed decrease in butterfly abundance in Gwangneung Forest over 18 years aligns with the global trend of biodiversity decline [39,40]. Various factors influence insect biodiversity decline [19,39]. In particular, habitat loss and climate change could be the key drivers of butterfly abundance decline in Gwangneung Forest. As Gwangneung Forest is a well-preserved area with limited habitat changes, especially during our survey period, the local habitat conditions in the study area might not have contributed to abundance decline. Nevertheless, the surrounding landscape experienced substantial changes from natural or green lands to urban areas. Habitat reduction in the surrounding area, specifically the reduction in open habitats rather than forests, led to a decline in species inhabiting open habitats compared with those inhabiting forests. However, an influence of habitat type—such as a greater decline in grassland-inhabiting species—was not observed in this study (Figure 3b). The significant influence of wingspan might suggest that mobility plays a critical role in butterfly conservation within urbanizing areas. In such landscapes, maintaining open green habitats, including parks, small agricultural plots, and grasslands, is essential for preserving butterfly diversity. Meanwhile, in a long-term study, Kwon et al. [4,15] reported an increase in forest species and a decrease in open-habitat species by comparing butterflies in Gwangneung over 50 years, from the late 1950s to the early 2000s. This coincides with the period of rapid reforestation in South Korea from 1960 to 2000 [41].
This study explored the potential influence of surrounding habitat loss on butterfly declines. Nonetheless, we acknowledge that our evaluation relied on inferred data rather than direct measurements. Accordingly, future field studies are necessary to empirically validate these relationships.

4.2. Wingspan and Surrounding Area

The positive correlation between butterfly wingspan and abundance changes contrasts with findings in moths [20], where moths with larger wingspans showed a greater decline in occurrence compared to past surveys [21]. In contrast, a comparison of the entire butterfly fauna in South Korea over consecutive 10-year periods (2007–2009 vs. 2017–2019) revealed that species with smaller wingspans experienced a more significant decline than those with larger wingspans (Kwon et al., unpublished), which supports the results of our current study. Therefore, the observed positive correlation between the wingspan and population change appears to be a consistent phenomenon in Gwangneung Forest and across South Korea.
The extensive urbanization of areas surrounding butterfly habitats, such as Gwangneung Forest, poses a marked threat in South Korea. Habitat loss and the resulting fragmentation increase the distances between habitats, rendering species mobility crucial for population maintenance [13,42,43]. Therefore, species with larger wingspans have a survival advantage due to their enhanced ability to migrate to distant habitats [43,44], and they may also be better equipped to cope with climate change [45]. Butterfly communities are impacted by the broad-scale effects of urbanization and fine-scale local processes [46,47,48]. Liivamägi et al. [11] highlighted that the surrounding landscape and habitat characteristics are crucial for butterfly species richness. Therefore, insights into the butterfly habitat–surrounding landscape relationship are essential for butterfly conservation [12,42]. In contrast, for moths, factors such as light pollution or changes in ecosystem function (e.g., changes in bird feeding) have a greater impact on population changes than habitat reduction or urbanization [22]. This difference might result from habitat requirement variations: butterflies primarily inhabit open habitats, which are directly reduced by urbanization, whereas moths primarily inhabit forests, where the impact of urbanization is relatively minor because of the legally protected forests in South Korea. For butterflies, even if the larvae’s habitat is a forest, the adults mainly occupy open habitats. Therefore, the impact of open habitats is more severe for butterflies than for moths.

4.3. Thermal Adaptation

Beyond wingspan, the next major factor influencing changes in butterfly abundance is their thermal adaptation type, based on temperature preferences. Our results showed a weak signal for an increase in abundance of warm-adapted compared to moderately adapted and cold-adapted species, indicating the impact of climate changes on butterfly distribution. This was partially supported by Kwon, Kim, Chun, Byun, Lim, and Shin [15], who reported that the butterfly population increase was greatest in warm-adapted species. However, our study found no differences between the cold-adapted species and moderately adapted species. Most moderately adapted species are cold-adapted, with their main distribution areas located north of Korea [36]. Meanwhile, Kwon et al. [4] reported that the species (Melitaea britomartis, Neptis rivularis, Gonepteryx maxima, Nymphalis xanthomelas, and Coenonympha oedippus) with the most notable population declines were cold-adapted species in the same study area, while those with the most marked increases (Pieris canidia and Zizeeria maha) were warm-adapted species. However, the declining cold-adapted species (M. britomartis, G. maxima, and C. oedippus) were not observed in our study, with P. canidia and Z. maha exhibiting a slight increasing trend in their abundance during the study period.
Thermal adaptation is a critical aspect of maintaining physiological function in ectothermic organisms and is closely linked to climate change. Mutamiswa et al. [49] reviewed the mechanisms and patterns of thermal adaptation in the context of shifting climates, with a focus on Lepidoptera. They summarized thermal adaptation across several dimensions, including behavioral, evolutionary, morphological, and physiological aspects. Similarly, Tsai et al. [50] investigated the physical and behavioral adaptations of butterflies that prevent overheating in their living wings. Their findings highlighted the physiological importance of wing temperature and how it is regulated through structural and behavioral strategies.
Climate change influences biological traits, which in turn affects the distribution and abundance of species [51]. Na et al. [52] found a positive correlation between butterfly wingspan and temperature in Korea, suggesting a morphological response to thermal conditions. Lee et al. [53] analyzed the distribution patterns of butterflies in Korea and found that warm-adapted species, although able to expand their temperature tolerance and habitat range, are more vulnerable to climate change due to their localized adaptations. Similarly, Adhikari et al. [54] reported that warm-adapted species are particularly susceptible to climate change, as evidenced by a northward shift in their northern range margins. In Japan, various butterfly species have also exhibited a northeastward expansion of their ranges in response to climate change [55,56,57].

4.4. Food Niche and Population Change

Although diet breadth is considered a potential driver of the extirpation and population decline of butterflies [26,27], its effect on population changes in this study was not significant. Kwon, Lee, Kim, Won, Kim, and Park [5] analyzed changes in butterfly populations in Korea over 60 years and reported an increase in species with a narrow diet breadth—a phenomenon that contradicts general expectations [58]. In South Korea, forest restoration has led to an increase in monophagous species (those that feed on plants of a single genus, Quercus) that consume the leaves of broad-leaved trees, such as oaks. However, this study found significant declines in monophagous species like Favonius orientalis, Favonius ultramarinus, Dilipa fenestra, and Hestina persimilis, which feed on the leaves of large trees, such as oak and Celtis jessoensis. Changes in the habitat environment could not explain the decline in these species, as Gwangneung Forest is well preserved. Among the declining species, Erynnis montanus is particularly notable, as it was a dominant species commonly observed in this survey area. Despite its prevalence in the central regions of the Korean Peninsula, such as Gyeonggi-do and Gangwon-do, it is rarely seen in southern regions [33]. These distribution characteristics suggest that these species could gradually decrease in the central region due to climate change. Meanwhile, Luehdorfia puziloi was the only species whose abundance increased significantly. This monophagous species primarily feeds on plants of the Aristolochiaceae family, such as Asarum sieboldii and Asarum maculatum. It is a cold-adapted species found in northeastern China and the Russian Far East, and in Gyeonggi-do and Gangwon-do in South Korea [25,33]. The increase in this species’ population is presumed to be related to an increase in its food plants. Therefore, investigating a possible increase in Aristolochiaceae plants in Gwangneung Forest is necessary.

4.5. Biological Traits and Population Change

Most trait categories showed positive correlations with one another (Figure 3). Notably, the abundance of species not classified as cold- or warm-adapted exhibited a strong positive correlation with species inhabiting forest edges, as well as with dietary specialists—specifically monophagous and oligophagous species (r > 0.90, p < 0.05). We acknowledge that such strong correlations may raise concerns about potential redundancy among trait categories. However, we interpret these associations not as statistical artifacts but as ecologically meaningful patterns that reflect co-occurring traits shaped by shared environmental preferences or strategies. For instance, species lacking strong thermal adaptations may be more flexible in their habitat use and, thus, more likely to occur in edge environments, where microclimatic conditions tend to be variable. Similarly, dietary specialists such as monophagous and oligophagous species may be more dependent on host plants commonly found along forest edges. These co-occurrences suggest that certain trait combinations are favored under specific environmental conditions, such as habitat fragmentation or edge effects. Rather than indicating redundancy, these patterns offer valuable insights into how species’ functional traits align with ecological filters and landscape features.

5. Conclusions

An analysis of 18 years of butterfly survey data from Gwangneung Forest revealed that more species experienced population declines than increases. This suggests that butterfly populations in the region may be declining, potentially due to habitat loss disrupting metapopulation dynamics, particularly in the face of ongoing surrounding development. However, the expected pattern of greater decline among open-habitat species compared to forest species was not observed, indicating that additional factors may be at play. While habitat loss is likely a major driver of butterfly decline across South Korea, few studies have quantitatively confirmed this. Another significant finding of this study is the climate change-induced increase in warm-adapted species within Gwangneung Forest—a trend likely to persist. Moreover, wingspan had a strong association with population changes, possibly reflecting the effects of habitat fragmentation in the surrounding landscape. Overall, our results underscore the need for targeted conservation strategies to preserve butterfly diversity in Gwangneung Forest and across South Korea. In particular, conserving open green habitats should be prioritized. Effective management must also consider climate change, regional habitat loss, and key biological traits such as wingspan and diet breadth.

Supplementary Materials

The following supporting information can be downloaded at https://www.mdpi.com/article/10.3390/f16091386/s1, Figure S1: Changes in land cover from 1990s to 2010s in the surrounding area (Gyeonggi-do and Seoul) of Gwangneung Forest in South Korea; Table S1: Slope in a linear regression analysis and biological traits of butterfly species based on Joo et al. (2021) [34], Kim and Seo (2012) [25], Kwon et al. (2017) [59], and Kim et al. (2012) [33]; Table S2: Abundance (number of individuals) and frequency (number of recorded surveys) of butterfly species found in Gwangneung Forest from 1998 to 2015.

Author Contributions

T.-S.K.: Conceptualization, Data curation, Methodology, Data analysis, Writing—original draft; Writing—review and editing. S.-S.K.: Data curation, Writing—review and editing. I.Y.: Data curation, Writing—review and editing. A.R.K.: Data curation, Writing—review and editing. Y.-S.P.: Conceptualization, Methodology, Visualization, Data analysis, Writing—original draft, Writing—review and editing. All authors have read and agreed to the published version of the manuscript.

Funding

This work was supported by a National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIT) (RS-2024-00456138) and the National Institute of Forest Science, Republic of Korea (FE0100-2024-04).

Data Availability Statement

The original contributions presented in this study are included in the article/Supplementary Material. Further inquiries can be directed to the corresponding author.

Conflicts of Interest

The authors declare no conflicts of interest.

References

  1. Parmesan, C.; Phyholm, N.; Tefanescus, C.; Hill, J.K.; Thomas, C.D.; Descimon, H.; Huntley, B.; Kalla, L.; Kullberg, J.; Tammaru, T.; et al. Poleward shifts in geographical ranges of butterfly species associated with regional warming. Nature 1999, 399, 579–583. [Google Scholar] [CrossRef]
  2. Parmesan, C.; Yohe, G. A globally coherent fingerprint of climate change impacts across natural systems. Nature 2003, 421, 37–42. [Google Scholar] [CrossRef]
  3. Roth, T.; Plattner, M.; Amrhein, V. Plants, birds and butterflies: Short-term responses of species communities to climate warming vary by taxon and with altitude. PLoS ONE 2014, 9, e82490. [Google Scholar] [CrossRef]
  4. Kwon, T.-S.; Kim, S.-S.; Lee, C.M. Local change of butterfly species in response to global warming and reforestation in Korea. Zool. Stud. 2013, 52, 47. [Google Scholar] [CrossRef]
  5. Kwon, T.-S.; Lee, C.M.; Kim, E.-S.; Won, M.; Kim, S.-S.; Park, Y.-S. Habitat change has greater effects than climate change on butterfly occurrence in South Korea. Glob. Ecol. Conserv. 2021, 26, e01464. [Google Scholar] [CrossRef]
  6. Lindsey, R.; Dahlman, L. Climate Change: Global Temperature. (Online Resource). 2024. Available online: https://www.climate.gov/news-features/understanding-climate/climate-change-global-temperature (accessed on 17 July 2024).
  7. NOAA National Centers for Environmental Information. Monthly Global Climate Report for Annual 2023, Published online January 2024. Available online: https://www.ncei.noaa.gov/access/monitoring/monthly-report/global/202313 (accessed on 17 July 2024).
  8. Kwon, W.T. Current status and perspectives of climate change sciences. Korean J. Amos. Sci. 2005, 41, 325–336. [Google Scholar]
  9. Oh, J.-H.; Kim, Y.-K.; Kwon, J. An analysis of landcover change and temporal landscape structure in the main ridge area of the Baekdu Daegan mountain system. J. Korean Assoc. Geogr. Inf. Stud. 2007, 10, 49–57. [Google Scholar]
  10. Habel, J.C.; Teucher, M.; Gros, P.; Schmitt, T.; Ulrich, W. Land use and climate change affects butterfly diversity across northern Austria. Landsc. Ecol. 2021, 36, 1741–1754. [Google Scholar] [CrossRef]
  11. Liivamägi, A.; Kuusemets, V.; Kaart, T.; Luig, J.; Diaz-Forero, I. Influence of habitat and landscape on butterfly diversity of semi-natural meadows within forest-dominated landscapes. J. Insect Conserv. 2014, 18, 1137–1145. [Google Scholar] [CrossRef]
  12. Pendl, M.; Hussain, R.I.; Moser, D.; Frank, T.; Drapela, T. Influences of landscape structure on butterfly diversity in urban private gardens using a citizen science approach. Urban Ecosyst. 2022, 25, 477–486. [Google Scholar] [CrossRef]
  13. Wood, B.C.; Pullin, A.S. Persistence of species in a fragmented urban landscape: The importance of dispersal ability and habitat availability for grassland butterflies. Biodivers. Conserv. 2002, 11, 1451–1468. [Google Scholar] [CrossRef]
  14. Ponce-Reyes, R.; Nicholson, E.; Baxter, P.W.J.; Fuller, R.A.; Possingham, H. Extinction risk in cloud forest fragments under climate change and habitat loss. Divers. Distrib. 2013, 19, 518–529. [Google Scholar] [CrossRef]
  15. Kwon, T.S.; Kim, S.S.; Chun, J.H.; Byun, B.K.; Lim, J.H.; Shin, J.H. Changes in butterfly abundance in response to global warming and reforestation. Environ. Entomol. 2010, 39, 337–345. [Google Scholar] [CrossRef]
  16. Dar, A.; Jamal, K. The decline of moths globally: A review of possible causes. Munis Entomol. Zool. J. 2021, 16, 310–319. [Google Scholar]
  17. Sánchez-Bayo, F.; Wyckhuys, K.A.G. Worldwide decline of the entomofauna: A review of its drivers. Biol. Conserv. 2019, 232, 8–27. [Google Scholar] [CrossRef]
  18. Wagner, D.L. Moth decline in the Northeastern United States. News Lepid. Soc. 2012, 54, 52–56. [Google Scholar]
  19. Fox, R. The decline of moths in Great Britain: A review of possible causes. Insect Conserv. Divers. 2013, 6, 5–19. [Google Scholar] [CrossRef]
  20. Wölfling, M.; Uhl, B.; Fiedler, K. Ecological drift and directional community change in an isolated Mediterranean forest reserve-Larger moth species under higher threat. J. Insect Sci. 2020, 20, 7. [Google Scholar] [CrossRef]
  21. Kwon, T.-S.; Kim, S.-S.; Lee, D.-S.; Park, G.E.; Park, Y.-S. Large moths facing steeper decline than small moths in South Korea: Implications for ecosystem dynamics and conservation. Glob. Ecol. Conserv. 2024, 55, e03234. [Google Scholar] [CrossRef]
  22. Wagner, D.L.; Fox, R.; Salcido, D.M.; Dyer, L.A. A window to the world of global insect declines: Moth biodiversity trends are complex and heterogeneous. Proc. Natl. Acad. Sci. USA 2021, 118, e2002549117. [Google Scholar] [CrossRef]
  23. Byun, B.-K.; Lee, Y.J.; Weon, G.-J.; Sohn, J.-D. Butterfly Fauna of the Gwangneung Forest, Korea. J. Asia-Pac. Entomol. 2005, 8, 199–210. [Google Scholar] [CrossRef]
  24. Kwon, T.-S.; Byun, B.-K.; Lee, B.-W.; Lee, C.-Y.; Shon, J.-D.; Kang, S.-H.; Kim, S.-S.; Kim, Y.-K. Estimation of species richness of butterfly community in the Gwangneung forest, Korea. Korean J. Appl. Entonol. 2009, 48, 439–445. [Google Scholar] [CrossRef]
  25. Kim, S.-S.; Seo, Y.-H. Life Histories of Korean Butterflies; Sakyejul: Paju, Republic of Korea, 2012. [Google Scholar]
  26. Koh, L.P.; Soghi, N.S.; Brook, B.W. Ecological correlates of extinction proneness in tropical butterflies. Conserv. Biol. 2004, 18, 1571–1578. [Google Scholar] [CrossRef]
  27. Palash, A.; Paul, S.; Resha, S.K.; Khan, M.K. Body size and diet breadth drive local extinction risk in butterflies. Heliyon 2022, 8, e10290. [Google Scholar] [CrossRef]
  28. Korea Forest Research Institute. Experimental Forest of Gangneung; Udgomunhwa Publishing: Seoul, Republic of Korea, 1994. [Google Scholar]
  29. KNA. Korea National Arboretum. 2023. Available online: http://www.kna.go.kr/ (accessed on 11 October 2023).
  30. Cho, Y.-C.; Cho, H.-J.; Kim, J.-S.; Cho, J.-H.; Jung, S.-H.; Kim, H.-G.; Sim, H.-S.; Lee, D.; Kim, H.-C. Forest of Korea (V) Gwangreung Forest Biosphere Reserve; Korea National Arboretum: Pocheon, Republic of Korea, 2020. [Google Scholar]
  31. Korea Meteorological Administration (KMA); National Institute of Meteorological Sciences (NIMS). Analysis of Climate Change in South Korea Over 109 Years (1912–2020); Korea Meteorological Administration, and National Institute of Meteorological Sciences: Seoul, Republic of Korea, 2021. [Google Scholar]
  32. Pollard, E. A method for assessing changes in the abundance of butterflies. Biol. Conserv. 1977, 12, 115–134. [Google Scholar] [CrossRef]
  33. Kim, S.-S.; Lee, C.M.; Kwon, T.-S.; Joo, H.Z.; Sung, J.H. Korean Butterfly Atlas 1996-2011; Korea Forest Research Institute: Seoul, Republic of Korea, 2012. [Google Scholar]
  34. Joo, H.Z.; Kim, S.-S.; Kim, H.C.; Sohn, J.D.; Lee, Y.J.; Ju, J.S. The Butterflies of Korea; Geobook Pub. Co.: Seoul, Republic of Korea, 2021. [Google Scholar]
  35. R Core Team. R: A Language and Environment for Statistical Computing; R Foundation for Statistical Computing: Vienna, Austria, 2023; Available online: https://www.R-project.org/ (accessed on 23 May 2023).
  36. Ogle, D.H.; Doll, J.C.; Wheeler, A.P.; Dinno, A. FSA: Simple Fisheries Stock Assessment Methods. R Package Version 0.9.5. 2023. Available online: https://CRAN.R-project.org/package=FSA (accessed on 23 May 2024).
  37. Harrell, F., Jr. Hmisc: Harrell Miscellaneous_. R Package Version 5.1-3. 2024. Available online: https://CRAN.R-project.org/package=Hmisc (accessed on 22 November 2024).
  38. Kwon, T.-S.; Lee, C.M.; Sung, J.H. Diversity decrease of ant (Formicidae, Hymenoptera) after a forest disturbance: Different responses among functional guilds. Zool. Stud. 2014, 53, 37. [Google Scholar] [CrossRef]
  39. Seibold, S.; Gossner, M.M.; Simons, N.K.; Blüthgen, N.; Müller, J.; Ambarlı, D.; Ammer, C.; Bauhus, J.; Fischer, M.; Habel, J.C.; et al. Arthropod decline in grasslands and forests is associated with landscape-level drivers. Nature 2019, 574, 671–674. [Google Scholar] [CrossRef] [PubMed]
  40. van Klink, R.; Bowler, D.E.; Gongalsky, K.B.; Swengel, A.B.; Gentile, A.; Chase, J.M. Meta-analysis reveals declines in terrestrial but increases in freshwater insect abundances. Science 2020, 368, 417–420. [Google Scholar] [CrossRef]
  41. Bae, J.-S.; Lee, K.-B. Impacts of the substitution of firewood for home use on the forest greening after the 1945 liberation of Korea. J. Korean For. Soc. 2006, 95, 60–72. [Google Scholar]
  42. Dover, J.; Settele, J. The influences of landscape structure on butterfly distribution and movement: A review. J. Insect Conserv. 2009, 13, 3–27. [Google Scholar] [CrossRef]
  43. Sekar, S. A meta-analysis of the traits affecting dispersal ability in butterflies: Can wingspan be used as a proxy? J. Anim. Ecol. 2012, 81, 174–184. [Google Scholar] [CrossRef] [PubMed]
  44. Nieminen, M.; Rita, H.; Uuvana, P. Body size and migration rate in moths. Ecography 1999, 22, 697–707. [Google Scholar] [CrossRef]
  45. Shirey, V.; Neupane, N.; Guralnick, R.; Ries, L. Rising minimum temperatures contribute to 50 years of occupancy decline among cold-adapted Arctic and boreal butterflies in North America. Glob. Change Biol. 2024, 30, e17205. [Google Scholar] [CrossRef]
  46. Adams, B.J.; Li, E.; Bahlai, C.A.; Meineke, E.K.; McGlynn, T.P.; Brown, B.V. Local- and landscape-scale variables shape insect diversity in an urban biodiversity hot spot. Ecol. Appl. 2020, 30, e02089. [Google Scholar] [CrossRef]
  47. Bang, C.; Faeth, S.H. Variation in arthropod communities in response to urbanization: Seven years of arthropod monitoring in a desert city. Landsc. Urban Plan. 2011, 103, 383–399. [Google Scholar] [CrossRef]
  48. Niemelä, J.; Kotze, D.J.; Venn, S.; Penev, L.; Stoyanov, I.; Spence, J.; Hartley, D.; de Oca, E.M. Carabid beetle assemblages (Coleoptera, Carabidae) across urban-rural gradients: An international comparison. Landsc. Ecol. 2002, 17, 387–401. [Google Scholar] [CrossRef]
  49. Mutamiswa, R.; Mbande, A.; Nyamukondiwa, C.; Chidawanyika, F. Thermal adaptation in Lepidoptera under shifting environments: Mechanisms, patterns, and consequences. Phytoparasitica 2023, 51, 929–955. [Google Scholar] [CrossRef]
  50. Tsai, C.-C.; Childers, R.A.; Nan Shi, N.; Ren, C.; Pelaez, J.N.; Bernard, G.D.; Pierce, N.E.; Yu, N. Physical and behavioral adaptations to prevent overheating of the living wings of butterflies. Nat. Commun. 2020, 11, 551. [Google Scholar] [CrossRef]
  51. Solomon, S. Climate Change 2007-the Physical Science Basis: Working Group I Contribution to the Fourth Assessment Report of the IPCC; Cambridge University Press: New York, USA, 2007; Volume 4. [Google Scholar]
  52. Na, S.; Lee, E.; Kim, H.; Choi, S.; Yi, H. The relationship of mean temperature and 9 collected butterfly species’ wingspan as the response of global warming. J. Ecol. Environ. 2021, 45, 21. [Google Scholar] [CrossRef]
  53. Lee, S.; Jeon, H.; Kim, M. Spatial distribution of butterflies in accordance with climate change in the Korean Peninsula. Sustainability 2020, 12, 1995. [Google Scholar] [CrossRef]
  54. Adhikari, P.; Jeon, J.-Y.; Kim, H.W.; Oh, H.-S.; Adhikari, P.; Seo, C. Northward range expansion of southern butterflies according to climate change in South Korea. J. Clim. Change Res. 2020, 11, 643–656. [Google Scholar] [CrossRef]
  55. Kitahara, M. Poleward range shift of butterflies under global warming. Nat. Insects 2008, 43, 19–23. [Google Scholar]
  56. Inoue, T. Spread of the distributional range of Narathura bazalus (Hewitson)(Lepidoptera, Lycaenidae) in Ibaraki prefecture, central Japan. Trans. Lepid. Soc. Jpn. 2005, 56, 287–296. [Google Scholar]
  57. Ogawa-Onishi, Y.; Berry, P.M. Ecological impacts of climate change in Japan: The importance of integrating local and international publications. Biol. Conserv. 2013, 157, 361–371. [Google Scholar] [CrossRef]
  58. Colles, A.; Liow, L.H.; Prinzing, A. Are specialists at risk under environmental change? Neoecological, paleoecological and phylogenetic approaches. Ecol. Lett. 2009, 12, 849–863. [Google Scholar] [CrossRef] [PubMed]
  59. Kwon, T.-S.; Lee, C.-M.; Kim, E.-S.; Won, M.S.; Kim, S.-S. Analysis of Distribution Changes in Korean Butterfly; National Institute of Forest Science: Seoul, Republic of Korea, 2017. [Google Scholar]
Forests 16 01386 g001
Figure 1. The study area with the butterfly survey route as a white dotted line (approximately 4 km) in Gwangneung Forest in South Korea and land cover types in the surrounding area in 2020. Land cover data were obtained from the Environmental Geographic Information Service, EGIS (EGIS; http://egis.me.go.kr, accessed on 23 June 2024). The protected area is outlined with two different colored dotted lines: brown indicating the core area and green representing the buffer zone.
Figure 1. The study area with the butterfly survey route as a white dotted line (approximately 4 km) in Gwangneung Forest in South Korea and land cover types in the surrounding area in 2020. Land cover data were obtained from the Environmental Geographic Information Service, EGIS (EGIS; http://egis.me.go.kr, accessed on 23 June 2024). The protected area is outlined with two different colored dotted lines: brown indicating the core area and green representing the buffer zone.
Forests 16 01386 g001
Forests 16 01386 g002
Figure 2. (a) The relationship between the number of years observed and the abundance of species, (b) changes in abundance during the survey years, (c) changes in species richness during the study period, and (d) changes in low temperature in May measured at Pocheon weather station, operated by the Korea Meteorological Administration (http://www.kma.go.kr, accessed on 23 June 2024). Abundance was standardized by calculating the abundance in each year divided by the number of surveys in the corresponding year. The standardized abundance was log-transformed (ln(abundance + 1)).
Figure 2. (a) The relationship between the number of years observed and the abundance of species, (b) changes in abundance during the survey years, (c) changes in species richness during the study period, and (d) changes in low temperature in May measured at Pocheon weather station, operated by the Korea Meteorological Administration (http://www.kma.go.kr, accessed on 23 June 2024). Abundance was standardized by calculating the abundance in each year divided by the number of surveys in the corresponding year. The standardized abundance was log-transformed (ln(abundance + 1)).
Forests 16 01386 g002
Forests 16 01386 g003
Figure 3. Changes in butterfly abundance in each biological trait: (a) thermal adaptation types, (b) habitat types, and (c) diet breadth.
Figure 3. Changes in butterfly abundance in each biological trait: (a) thermal adaptation types, (b) habitat types, and (c) diet breadth.
Forests 16 01386 g003
Forests 16 01386 g004
Figure 4. Correlation coefficients between biological traits in the three categories. Blue scales indicate values surpassing the statistically critical value (p < 0.05).
Figure 4. Correlation coefficients between biological traits in the three categories. Blue scales indicate values surpassing the statistically critical value (p < 0.05).
Forests 16 01386 g004
Forests 16 01386 g005
Figure 5. The frequency of the slope values (i.e., slopes (a) of abundance (y) by year (x) regression by species, y = ax + b) of 67 abundant species in the linear regression analysis. Species observed for over 5 years were analyzed to evaluate the abundance changes through a linear regression analysis. The slope of each species was summarized as a frequency plot. The standardized abundance, calculated as the yearly abundance divided by the survey number, was log-transformed (ln(abundance + 1)). Abundance change was determined using the slope of a linear regression relative to years. The curve indicates a normal distribution, and the hatched vertical red line represents the zero slope.
Figure 5. The frequency of the slope values (i.e., slopes (a) of abundance (y) by year (x) regression by species, y = ax + b) of 67 abundant species in the linear regression analysis. Species observed for over 5 years were analyzed to evaluate the abundance changes through a linear regression analysis. The slope of each species was summarized as a frequency plot. The standardized abundance, calculated as the yearly abundance divided by the survey number, was log-transformed (ln(abundance + 1)). Abundance change was determined using the slope of a linear regression relative to years. The curve indicates a normal distribution, and the hatched vertical red line represents the zero slope.
Forests 16 01386 g005
Forests 16 01386 g006
Figure 6. Differences in (a) slope values in a linear regression analysis and (b) wingspans of species for three different thermal adaptation types. Different letters indicate significant differences among groups based on the Kruskal–Wallis test followed by the Dunn multiple comparison test (p < 0.05). Box plot: minimum, 25% percentile, median, 75% percentile, and maximum values. Open circle (o): outlier.
Figure 6. Differences in (a) slope values in a linear regression analysis and (b) wingspans of species for three different thermal adaptation types. Different letters indicate significant differences among groups based on the Kruskal–Wallis test followed by the Dunn multiple comparison test (p < 0.05). Box plot: minimum, 25% percentile, median, 75% percentile, and maximum values. Open circle (o): outlier.
Forests 16 01386 g006
Forests 16 01386 g007
Figure 7. The relationship between wingspan and slope values in a linear regression analysis. The three different thermal adaptation types are indicated with different colors.
Figure 7. The relationship between wingspan and slope values in a linear regression analysis. The three different thermal adaptation types are indicated with different colors.
Forests 16 01386 g007
Table 1. Numbers of species and dominant species (>2% of total abundance) recorded in the dataset observed in the survey area from 2009 to 2015.
Table 1. Numbers of species and dominant species (>2% of total abundance) recorded in the dataset observed in the survey area from 2009 to 2015.
Family Abundance (%)Number of species
Hesperiidae 729 (6.1)13
Lycaenidae 684 (5.8)25
Nymphalidae 7373 (62.1)48
Papilionidae 663 (5.6)8
Pieridae 2418 (20.4)6
Overall 11,867 (100.0)100
FamilyDominant speciesAbundance (%)Number of years observed
Hesperiidae Erynnis montanus377 (3.2)18
NymphalidaeLibythea lepita3540 (29.8)18
Argynnis paphia598 (5.1)17
Neptis philyroides401 (3.4)17
Sasakia charonda270 (2.3)16
Neptis philyra260 (2.2)16
Pieridae Pieris melete1989 (16.8)18
Pieris canidia332 (2.8)16
Table 2. Results of backward stepwise regression model. Significance: * p < 0.1, ** p < 0.05, ns p > 0.1. Diet breadth, habitat types, and thermal adaptation types are categorical variables, whereas others are continuous variables. (+): positive values of coefficients. AIC: Akaike information criterion.
Table 2. Results of backward stepwise regression model. Significance: * p < 0.1, ** p < 0.05, ns p > 0.1. Diet breadth, habitat types, and thermal adaptation types are categorical variables, whereas others are continuous variables. (+): positive values of coefficients. AIC: Akaike information criterion.
VariablesModel Performance
ModelAbundanceDiet BreadthHabitat TypeThermal AdaptationWingspanAdj. R2pAIC
Initialnsnsns** (+)* (+)0.072 0.13−434.2
Final ** (+)** (+)0.120 0.01−441.9
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Kwon, T.-S.; Kim, S.-S.; Yang, I.; Kim, A.R.; Park, Y.-S. Butterfly Community Responses to Urbanization and Climate Change: Thermal Adaptation and Wing Morphology Effects in a Conserved Forest, South Korea. Forests 2025, 16, 1386. https://doi.org/10.3390/f16091386

AMA Style

Kwon T-S, Kim S-S, Yang I, Kim AR, Park Y-S. Butterfly Community Responses to Urbanization and Climate Change: Thermal Adaptation and Wing Morphology Effects in a Conserved Forest, South Korea. Forests. 2025; 16(9):1386. https://doi.org/10.3390/f16091386

Chicago/Turabian Style

Kwon, Tae-Sung, Sung-Soo Kim, Ilju Yang, A Reum Kim, and Young-Seuk Park. 2025. "Butterfly Community Responses to Urbanization and Climate Change: Thermal Adaptation and Wing Morphology Effects in a Conserved Forest, South Korea" Forests 16, no. 9: 1386. https://doi.org/10.3390/f16091386

APA Style

Kwon, T.-S., Kim, S.-S., Yang, I., Kim, A. R., & Park, Y.-S. (2025). Butterfly Community Responses to Urbanization and Climate Change: Thermal Adaptation and Wing Morphology Effects in a Conserved Forest, South Korea. Forests, 16(9), 1386. https://doi.org/10.3390/f16091386

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop