Managing invasive alien plants in Gangwon State via alternative planting

1 Introduction

Biodiversity loss has emerged as a critical global environmental issue, and the introduction and spread of alien species are recognized as major drivers accelerating this phenomenon (Secretariat of the Convention on Biological Diversity, 2020; IPBES, 2023). In particular, some alien plant species negatively affect native species and ecosystem functions due to their high adaptability, rapid growth, and high reproductive capacity (Pyšek et al., 2020). Recent climate change and land-use change have further facilitated the introduction and expansion of alien plants, increasing the need for effective management of invasive alien plant species (Pyšek and Richardson, 2010; Bellard et al., 2012; Chen et al., 2024). These global patterns are further supported by foundational work on invasion biology (Blackburn et al., 2011; Mack et al., 2000; Simberloff et al., 2013).

Invasive alien plants are defined as introduced species that adversely affect, or are likely to affect, ecosystem structure and function (National Institute of Ecology, 2023; CBD, 2002; Pyšek et al., 2020). Through rapid growth and prolific reproduction, they often outcompete native vegetation, resulting in reduced biodiversity, altered vegetation structure, and ecosystem degradation (Pyšek et al., 2020). In the Republic of Korea, invasive alien species are designated and managed under the Act on the Conservation and Use of Biological Diversity (Ministry of Environment, 2024). As of 2026, 40 species have been designated as ecosystem-disturbing organisms, including 18 plant species such as common ragweed (Ambrosia artemisiifolia), giant ragweed (Ambrosia trifida), bur cucumber (Sicyos angulatus), Japanese hop (Humulus japonicus), and sheep sorrel (Rumex acetosella). These species frequently dominate natural ecosystems, agricultural lands, riverbanks, and roadsides, suppressing the growth of native vegetation (National Institute of Ecology, 2023). Their capacity to alter ecosystem structure and outcompete native species is well documented in the broader invasion-ecology literature (Richardson et al., 2000; Pyšek et al., 2004; Levine et al., 2003; Vilà et al., 2011).

Among them, bur cucumber is considered one of the most problematic invasive vines in riparian ecosystems, and landscape-genetic evidence indicates that multiple introductions and human-mediated dispersal, together with its close association with riverbank and floodplain habitats, have driven its rapid spread across Korea (Lee and Son, 2022). It covers surrounding vegetation, reduces light availability, and simplifies plant community structure (Lee et al., 2015). Similarly, giant ragweed and common ragweed expand rapidly in disturbed habitats, roadsides, and agricultural fields through prolific seed production and strong competitive ability (Kong et al., 2007; Choi et al., 2010; Zhao et al., 2021). Recent studies have also shown that the potential distribution of giant ragweed is expected to expand under future climate change scenarios (Chen et al., 2024), while its community structure and dominance vary depending on habitat conditions (Yu et al., 2026). Consequently, invasive alien plants pose a serious threat not only to individual species but also to ecosystem integrity and biodiversity conservation. Similar concerns have been raised for other high-impact invaders and for exotic-plant management in Korea more broadly (Rejmánek and Richardson, 1996; Lake and Leishman, 2004; Byun et al., 2024; Lim et al., 2023).

Gangwon State contains diverse ecological environments, including extensive forest ecosystems, rivers, agricultural landscapes, and the Demilitarized Zone (DMZ) and Civilian Control Zone (CCZ). Although these areas support high biodiversity, they also provide favorable conditions for the establishment and spread of invasive alien plants. In particular, giant ragweed, common ragweed, bur cucumber, and Japanese hop have continuously expanded across the region, especially in riparian habitats and the Civilian Control Zone (Choi et al., 2007; Gangwon State, 2025).

Since 2009, Gangwon State has implemented invasive plant removal programs in cooperation with the Ministry of Environment and local governments. Despite these efforts, invasive alien plants remain widely distributed, and reinvasion is frequently observed in previously managed sites. This suggests that current management practices focus primarily on removal activities, while relatively little attention has been paid to post-removal vegetation recovery and reinvasion prevention. This gap mirrors a recurring shortcoming identified in invasive-species management programs worldwide (Hulme, 2006; Kettenring and Adams, 2011; Dana et al., 2019; Pearson et al., 2016).

Recent studies have emphasized the importance of ecological restoration following invasive plant removal. Lee et al. (2015) reported that reinvasion is likely when vegetation restoration is not implemented after removal, while Jo and Park (2016) demonstrated that early vegetation establishment can suppress undesirable plant species in riparian restoration sites. The National Institute of Ecology (2023) also highlighted the need for post-removal restoration and long-term monitoring. Therefore, invasive plant management should extend beyond species removal to include vegetation recovery and reinvasion prevention. This principle underpins current best practice in restoration ecology (D’Antonio and Meyerson, 2002; Funk et al., 2008; Gaertner et al., 2012; Singh and Byun, 2023; Weidlich et al., 2020).

Accordingly, this study aimed to (1) analyze the distribution characteristics and dominant invasive alien plant species across 18 cities and counties of Gangwon State and (2) evaluate vegetation changes following alternative planting treatments in field-scale test-beds (TBs) established in Hwacheon-gun and Yanggu-gun. By assessing the effectiveness of alternative planting in promoting vegetation recovery and suppressing reinvasion, this study seeks to provide a scientific basis for transitioning from removal-oriented management to restoration-oriented management and to support the development of region-specific invasive alien plant management strategies for Gangwon State.

2 Materials and methods

2.1 Analysis of the distribution and management status of invasive alien plants

To analyze the distribution characteristics of invasive alien plants in Gangwon State, survey and management data collected by Gangwon State from 2009 to 2024 were utilized. The dataset included annual records of distribution and removal areas for invasive alien plants across all 18 cities and counties.

These distribution and removal area data, including those presented in Table 1 was compiled from field-based monitoring records maintained by the invasive alien plant control officers of each of the 18 cities and counties of Gangwon State, who are directly responsible for implementing ecosystem-disturbing species removal programs in their respective jurisdictions. Each year, the responsible officer conducted on-site field surveys and applied a direct visual assessment method to identify invaded areas and to confirm the areas from which invasive alien plants had been removed. Distribution area was recorded as the area over which a given species was visually confirmed to occur, whereas removal area was recorded as the area over which removal (mowing, uprooting) had been directly carried out and confirmed on site.

Table 1. Distribution and removal areas of invasive alien plants across 18 cities and counties of Gangwon State (km²).

The primary target species were giant ragweed (Ambrosia trifida), common ragweed (Ambrosia artemisiifolia), and bur cucumber (Sicyos angulatus), which are widely distributed throughout Gangwon State. Additional species, including sheep sorrel (Rumex acetosella), Japanese hop (Humulus japonicus), and prickly lettuce (Lactuca serriola), were also examined. Analyses focused on both spatial and temporal patterns. Distribution area and occurrence ratios were calculated to identify dominant species and distribution characteristics. Differences between the CCZ and non-CCZ areas were examined, and distribution and removal areas were compared among the 18 cities and counties. Long-term changes from 2009 to 2024 were also analyzed to evaluate expansion trends and management effectiveness.

2.2 Vegetation monitoring and impact assessment

2.2.1 Selection of test-bed sites

TB sites were selected from areas with continuous invasive alien plant management, feasibility for long-term monitoring, and administrative cooperation. After evaluating candidate sites in Chuncheon-si, Hwacheon-gun, and Yanggu-gun, Hwacheon-gun and Yanggu-gun were selected (Figure 1). The Hwacheon-gun site was located in a riparian area adjacent to the Papo Stream, while the Yanggu-gun site was located near agricultural lands along the Seocheon Stream. Both sites have experienced continuous occurrences of invasive alien plants, particularly bur cucumber and giant ragweed.

2.2.2 Experimental plot design and establishment

TBs were established in Yanggu-gun (May 12, 2025) and Hwacheon-gun (June 2, 2025) to evaluate the effects of alternative planting on vegetation recovery and reinvasion suppression following invasive plant removal (Figure 2). Monitoring plots consisted of 3 m × 3 m quadrats and included three treatments: bare ground (T1), common reed (Phragmites australis) planting (T2), and Korean forsythia (Forsythia koreana) planting (T3). Each treatment was replicated three times, resulting in nine quadrats with a total treatment area of 0.000027 km². Within each site, the nine quadrats were established along a single contiguous line following a fixed, repeating treatment sequence—bare ground (T1), reed (T2), and forsythia (T3)—repeated three times (i.e., T1, T2, T3, T1, T2, T3, T1, T2, T3; Figure 2). This arrangement kept quadrats of different treatments interspersed and in close proximity to reduce potential differences in ambient environmental conditions among treatments.

The bare-ground treatment served as a control to observe natural vegetation recovery and reinvasion. Reed was selected as a native riparian herbaceous species capable of rapidly establishing vegetation cover, while forsythia was selected as a native shrub with high establishment success and potential to suppress herbaceous invasion through canopy development. Reed seedlings grown in 3-inch pots and approximately 60-cm-tall forsythia seedlings were planted after the removal of existing vegetation. All treatments were maintained under identical management conditions. Forsythia was planted at a center-to-center spacing of ~30 cm in a square-grid arrangement (~11 seedlings/m²), within the 0.25–0.75 m range specified for hedge-type shrub planting in Korea’s Landscape Design Standards (MOLIT, 2016); this density followed the applicable planting standard rather than being optimized to maximize suppression, and density-dependent effects were not systematically assessed in this study.

2.2.3 Vegetation monitoring

Vegetation monitoring was conducted from May to October 2025, with field surveys performed one to two times per month. Survey variables included herbaceous plant invasion, planted vegetation performance, and environmental factors. Herbaceous plant invasion was assessed based on vegetation cover within each quadrat. Vegetation cover (%) was determined by visually identifying the area within each quadrat occupied by herbaceous plants, measuring that area directly with a tape measure, and calculating the proportion of the total quadrat area covered by vegetation, as follows:

Vegetation cover (%) = (area covered by herbaceous vegetation/total quadrat area) × 100.

where the total area of each 3 m × 3 m quadrat was 9 m². Because cover was recorded separately for each co-occurring species or vegetation layer (e.g., climbing vines such as bur cucumber growing over other plants) and then summed, total vegetation cover could exceed 100% where multiple overlapping layers occupied the same ground area, as observed in Table 2. The survival and growth of planted species were monitored throughout the study period. Environmental variables measured included soil moisture, soil temperature, light intensity, and soil pH. Soil moisture and temperature were measured six times per quadrat using portable instruments, while light intensity and pH were measured using a lux meter and a portable pH meter, respectively. Fixed-point photographs were taken during each survey to document seasonal vegetation changes. Collected data were summarized using mean values and standard deviations. Temporal changes in herbaceous plant invasion and differences among treatments were evaluated.

Table 2. Monthly herbaceous plant invasion (vegetation cover, %) in the Hwacheon-gun test-bed (mean ± SD).

2.2.4 Statistical analysis

Linear mixed-effects models (LMMs) were fitted to the Hwacheon-gun herbaceous cover data (n = 51 quadrat-survey observations from 9 quadrats), with vegetation cover (%) modeled as a function of treatment, days since establishment, and their interaction. Quadrat was included as a random intercept to account for repeated measurements. Models were compared using AIC and likelihood-ratio tests, and final coefficients were estimated by restricted maximum likelihood (REML). Planned contrasts were used to compare treatment-specific slopes and predicted cover at the end of monitoring.

Time to reach 50% cover (t50), estimated by linear interpolation, and cumulative species richness per quadrat were compared among treatments using one-way ANOVA and the Kruskal–Wallis test. Mean soil moisture, temperature, light intensity, and pH at the Hwacheon-gun TB (Table 3) were likewise calculated per quadrat and compared among treatments using one-way ANOVA and the Kruskal–Wallis test, and Pearson and Spearman correlations were used to examine associations between these environmental variables and mean vegetation cover.

Table 3. Environmental factors by treatment in the Hwacheon-gun test-bed (mean ± SD) (T1: bare ground; T2: reed planting; T3: forsythia planting.

Individual-level species records at the Hwacheon-gun TB, including species identity and stem counts, were also analyzed to determine which species contributed to the aggregate cover values in Table 2. The proportions of individuals belonging to the two target species (giant ragweed and bur cucumber) and other species were compared among treatments using a chi-square test of independence, assessing whether alternative planting suppressed the target species specifically or herbaceous vegetation more broadly. All statistical analyses were performed in Python using statsmodels, with statistical significance set at p < 0.05.

3 Results

3.1 Distribution and management status of invasive alien plants in Gangwon State

3.1.1 Distribution status of invasive alien plants

As of 2024, the total distribution area of invasive alien plants in Gangwon State was estimated at 12.280 km² (Table 4). Giant ragweed (Ambrosia trifida) occupied the largest area, covering 6.388 km² (52.0%), followed by bur cucumber (Sicyos angulatus) at 3.641 km² (29.6%) and common ragweed (Ambrosia artemisiifolia) at 1.896 km² (15.4%). Other invasive alien plant species collectively accounted for 355 thousand m² (2.9%). These results indicate that management efforts in Gangwon State should primarily focus on giant ragweed, bur cucumber, and common ragweed.

Table 4. Distribution of invasive alien plants in Gangwon State (km²).

*Other species: Rumex acetosella, Lactuca scariola, Aster pilosus, and Humulus japonicus.

Ambrosia trifida (giant ragweed) is an annual forb that germinates in April–May, reaches up to 3 m in height, and bears broadly ovate, deeply three- to five-lobed leaves on a coarse-haired, branching stem; it flowers from around August and sets 5–7-ridged achenes from September to October, typically colonizing disturbed, open ground such as roadsides and riverbanks. Sicyos angulatus (bur cucumber) is an annual climbing vine reaching 4–8 m via branched tendrils, with broadly pentagonal, five-lobed leaves and hairy, sticky stems; its small white flowers appear from around August and its spiny fruits mature in October, and its rapid vertical and lateral growth allows it to overtop and smother co-occurring vegetation. Ambrosia artemisiifolia (common ragweed) is smaller (30–100 cm), with erect, branching stems and fern-like, deeply pinnate leaves; it flowers from August to September, releasing abundant wind-dispersed pollen, and its achenes remain viable in the soil seed bank for several years.

Species composition differed markedly between CCZ and non-CCZ areas. Within the CCZ, giant ragweed occupied 4.659 km² of the total 6.042 km² distribution area, accounting for 77.1% of the total. In contrast, bur cucumber occupied 3.172 km² of the 6.238 km² recorded outside the CCZ, representing 50.8% of the total distribution area in non-CCZ regions. The dominance of giant ragweed in the CCZ is likely related to the extensive military facilities, roads, and disturbed habitats characteristic of northern border regions. In contrast, bur cucumber was primarily distributed along riverine and riparian habitats outside the CCZ. This composition reflects the most recent (2024) survey only; a species-resolved time series is not available for 2009–2024, so long-term shifts in species dominance or spatial configuration could not be assessed.

3.1.2 Distribution and management status by city and county

Analysis of invasive alien plant distribution across the 18 cities and counties of Gangwon State revealed that Cheorwon-gun had the largest distribution area (2.6445 km²), followed by Inje-gun (2.2937 km²), Chuncheon-si (1.6000 km²), and Hongcheon-gun (1.4290 km²) (Table 1). These regions represent major hotspots of invasive alien plant occurrence and should be considered priority management areas.

Cheorwon-gun and Inje-gun, which contain extensive border and CCZ areas, were characterized by high proportions of giant ragweed. In contrast, Chuncheon-si and Hongcheon-gun exhibited relatively high occurrences of bur cucumber, reflecting the influence of major river systems and riparian habitats. Mountainous areas with high forest cover, such as Taebaek-si and Jeongseon-gun, showed relatively low distribution areas.

Comparison of distribution and removal areas indicated that, in some cities and counties, annual removal areas were comparable to or greater than recorded distribution areas. This pattern suggests repeated re-establishment of invasive alien plants following removal, likely associated with persistent soil seed banks, dispersal from adjacent habitats, and bare-ground conditions created after removal activities.

3.1.3 Temporal changes in distribution and removal areas

Analysis of invasive alien plant distribution and removal areas from 2009 to 2024 (Figure 3) showed that management effort (removal area) expanded substantially over the study period, whereas the distribution area itself did not show a corresponding long-term decline. Despite management intensification between 2014 and 2017, no clear decline in invasive alien plant distribution area was observed after 2018, indicating that removal activities alone have been insufficient to reduce the long-term spread of invasive alien plants.

Distribution area and removal area were tested separately rather than as a combined total, because removal area reflects management activity carried out within already-distributed areas and can overlap with, rather than add independently to, distribution area (see Table 1 legend). For each series, we tested for a linear trend across the seven years from 2018 to 2024 and across the full 2009–2024 period, using both ordinary least-squares (OLS) regression and the non-parametric Mann–Kendall trend test. Over 2018–2024, neither series showed a statistically significant trend (distribution area: OLS slope = -0.43 km² per year, p = 0.161, Mann–Kendall p = 0.072; removal area: OLS slope = -0.04 km² per year, p = 0.649, Mann–Kendall p = 0.764), consistent with the absence of a clear decline after 2018. Over the full 2009–2024 period, removal area showed a clear increasing trend (OLS slope = +0.49 km² per year, p < 0.001; Mann–Kendall p = 0.010), whereas distribution area showed only a borderline trend that was significant by OLS (slope = +0.18 km² per year, p = 0.048) but not by the non-parametric Mann–Kendall test (p = 0.096). This indicates that the province-wide increase in invasive-plant management activity over the study period reflects predominantly expanding removal effort, rather than a statistically robust increase in the distribution area itself. This plateau likely reflects rapid recolonization of bare ground left after removal, persistent soil seed banks, and continued propagule input from adjacent unmanaged habitats (see Discussion).

3.2 Vegetation monitoring and impact assessment

3.2.1 Hwacheon-gun test-bed

The Hwacheon-gun TB was established in Sinpung-ri, Sangseo-myeon, in a riparian area adjacent to the Papo Stream, where bur cucumber was the dominant invasive species. Environmental conditions were similar among treatments (soil moisture 28.6–32.0%, soil temperature 32.8–33.4 °C, light intensity 115.5–116.0 thousand lux, pH 7.6–7.7; Table 3), indicating that differences in vegetation development were unlikely to reflect environmental variation among treatments.

Herbaceous invasion increased over time in all treatments (Table 2; Figure 4), with cover and height trajectories also plotted in Figure 5. Cover rose from 10.3 ± 0.3% to 119.9 ± 0.8% in bare ground, from 7.2 ± 3.7% to 113.9 ± 0.0% in reed, and from 3.4 ± 1.6% to 104.1 ± 2.9% in forsythia, which consistently showed the lowest invasion throughout monitoring (e.g., 35.5 ± 4.5% vs. 70.4 ± 12.5% for bare ground and 61.7 ± 10.8% for reed on July 9). This suggests that shrub establishment partially suppressed herbaceous invasion through space occupation and shading.

A linear mixed-effects model confirmed that cover increased significantly with time across all treatments (1.07% per day, SE = 0.13, p < 0.001). The treatment × time interaction was not significant overall (likelihood ratio = 5.05, df = 2, p = 0.080), so the three treatments’ rates of increase were not jointly distinguishable. A planned contrast restricted to reed (T2) and forsythia (T3) showed significantly different slopes (p = 0.029), though the predicted difference in final cover between them was not significant (p = 0.139); with only three replicate quadrats per treatment, this apparent gap in final cover should be interpreted with caution.

Time to reach 50% cover (t50) also differed among treatments (bare ground: 34.5 days; reed: 22.9 days; forsythia: 49.1 days; one-way ANOVA F = 6.10, p = 0.046), though this was not confirmed by the Kruskal–Wallis test (H = 4.56, p = 0.103), likely reflecting the small sample size (n = 2–3 per treatment). This provides suggestive, rather than conclusive, statistical support for forsythia plots reaching intermediate cover more slowly, consistent with an early-stage suppressive effect.

Cumulative species richness (distinct herbaceous species per quadrat) did not differ significantly among treatments (bare ground: 4.3; reed: 4.0; forsythia: 3.3; one-way ANOVA F = 0.50, p = 0.63; Kruskal–Wallis H = 0.69, p = 0.71), suggesting that cover differences among treatments reflected growth and spread rate rather than the number of colonizing species.

Although Table 3 shows numerically higher soil moisture and temperature but lower light intensity in bare ground than in the planted treatments, none of the four variables differed significantly among treatments at the quadrat level (n = 3 per treatment; soil moisture F = 0.70, p = 0.534; temperature F = 1.00, p = 0.422; light F = 0.03, p = 0.970; pH F = 0.08, p = 0.925; Kruskal–Wallis all p > 0.35). Between-quadrat variability (e.g., soil moisture ranging from 20.7% to 37.5%) exceeded the difference between treatment means, indicating that this pattern reflects natural microsite variability rather than a treatment effect on the physical environment.

Across the nine Hwacheon-gun quadrats, light intensity was significantly negatively correlated with mean cover (Pearson r = -0.83, p = 0.005; Spearman ρ = -0.83, p = 0.005), while soil moisture, temperature, and pH were not (all p > 0.12). Because light was measured at ground level, this likely reflects shading by the herbaceous canopy itself rather than ambient light limiting establishment. Combined with the absence of treatment differences in physical variables, this indicates that the treatment itself — the presence and identity of the planted competitor — was the main biological factor limiting herbaceous cover at this site.

We also tested whether reduced cover in forsythia plots reflected selective suppression of the two target species (giant ragweed and bur cucumber) or broader suppression of herbaceous vegetation. Individual-level counts showed that the combined share of these two species was similar, if anything higher, across treatments (T1: 35.6%; T2: 38.0%; T3: 45.0%), whereas the non-target species goosefoot (Chenopodium album) made up a much smaller share across treatments (T1: 34.4%; T2: 27.1%; T3: 8.0%; χ² = 37.03, df = 2, p < 0.0001). This indicates that forsythia suppressed herbaceous establishment broadly rather than selectively targeting the region’s dominant invaders, and is therefore better characterized as a general early-colonization suppressor than a targeted control measure.

3.2.2 Yanggu-gun test-bed

The Yanggu-gun TB was located in Hwanggang-ri, Guktojeongjungang-myeon, adjacent to agricultural land and the Seocheon Stream. Bur cucumber, giant ragweed, and common ragweed were observed in the surrounding area. Environmental conditions were similar among treatments; soil moisture ranged from 9.0 to 12.5%, soil temperature from 28.2 to 29.2 °C, light intensity from 82.0 to 89.5 thousand lux, and pH from 7.7 to 8.0 (Table 5).

Table 5. Environmental factors by treatment in the Yanggu-gun test-bed (mean ± SD).

At the initial survey, herbaceous invasion was similar among treatments. Although the reed treatment showed slightly lower vegetation cover on June 12, all treatments rapidly increased in vegetation cover and reached approximately 100% by late June. Vegetation cover on June 27 reached 101.0 ± 0.7% in the bare ground treatment, 100.6 ± 0.3% in the reed treatment, and 100.7 ± 0.5% in the forsythia treatment. Unlike the Hwacheon-gun TB, treatment effects were not clearly distinguished in Yanggu-gun, suggesting that alternative planting effects were limited during the early establishment stage under local site conditions.

4 Discussion

The results of this study indicate that invasive alien plant management in Gangwon State should transition from a removal-centered approach to a restoration-centered framework. The finding that giant ragweed and common ragweed dominate border and CCZ areas while bur cucumber prevails in non-CCZ riparian habitats is consistent with previous research demonstrating habitat-specific invasion patterns (Lee and Son, 2022). The marked difference in species composition between CCZ and non-CCZ areas underscores the need for region-specific management strategies rather than uniform approaches across the entire province. A similar point has been made for disturbance-invasion dynamics and biological invasions generally (Hobbs and Huenneke, 1992; Blackburn et al., 2011).

The persistence of high invasive plant coverage despite more than 15 years of removal programs highlights a fundamental limitation of removal-only approaches. Rapid vegetation development in bare-ground control plots suggests that removal activities may inadvertently create favorable conditions for reinvasion by eliminating competition while leaving disturbed, open ground. This is consistent with findings from Jo and Park (2016) and Lee et al. (2015), who demonstrated that bare ground following removal promotes rapid recolonization by invasive species. This echoes wider evidence that removal alone often triggers secondary invasion (Pearson et al., 2016; Kettenring and Adams, 2011).

The differential response between the two TB sites offers important insights into the context-dependency of alternative planting effects. The suppression of herbaceous invasion observed in forsythia plots at Hwacheon-gun aligns with the expectation that woody shrubs can reduce light availability and occupy space more effectively than herbaceous competitors. However, the absence of a clear treatment effect at Yanggu-gun, where all treatments reached near-complete cover by late June, suggests that site-specific factors such as propagule pressure from adjacent seed sources, soil conditions, and initial vegetation establishment rates may override the benefits of planted species during early stages. This is consistent with broader evidence on how exotic invasions reshape resident plant communities and ecosystem processes (Ehrenfeld, 2003; Pyšek et al., 2012).

We emphasize that propagule pressure, soil conditions, and establishment speed were not directly measured at either site in this study. The explanation offered above should therefore be regarded as a plausible hypothesis for future testing—for example, by measuring soil seed-bank density at both sites—rather than a demonstrated cause of the contrasting treatment responses between Hwacheon-gun and Yanggu-gun.

This study contributes methodological advances relative to previous invasive-plant assessments in Korea. Combining a province-wide, multi-year distribution and removal dataset with a controlled test-bed experiment links large-scale management outcomes with mechanistic evidence on why removal-only management fails to reduce reinvasion. In addition, the linear mixed-effects models used to test treatment-by-time interactions (Bolker et al., 2009), together with the quadrat-level correlation analysis between environmental factors and cover, provide statistical evidence — rather than description alone — that treatment differences are attributable mainly to biological competition from the planted species rather than physicochemical heterogeneity among plots (Craine and Dybzinski, 2013).

These findings have several management implications. First, post-removal restoration measures should be site-specific and account for dominant invader characteristics and landscape context. Second, restoration effectiveness should be evaluated using ecological indicators such as reinvasion rates, vegetation recovery, and establishment success of planted species rather than removal area alone. Third, long-term monitoring is essential to assess whether early-stage suppression effects in forsythia plots persist through subsequent growing seasons and lead to durable reductions in invasive plant cover. These priorities align with broader recommendations for effective invasive-species management (Dana et al., 2019; Weidlich et al., 2020).

Where alternative planting follows removal, our results suggest that canopy-forming woody species such as Forsythia koreana suppress herbaceous competitors more effectively during early establishment than herbaceous alternatives such as reed (cf. Meiners, 2007), and should be prioritized where rapid suppression of reinvasion is the objective. At sites such as Yanggu-gun, however, where high propagule pressure allowed all treatments to approach saturation cover within weeks, resource-intensive planting may offer limited benefit over lower-cost passive revegetation, and resources may be better directed toward pre-emptive removal before peak invasive growth.

5 Conclusion

This study demonstrated that invasive alien plant management in Gangwon State remains challenging despite sustained removal efforts, with giant ragweed, bur cucumber, and common ragweed continuing to dominate the landscape. Alternative planting with Forsythia koreana showed potential for suppressing herbaceous reinvasion at riparian sites, though effects were site-dependent. Because these results derive from a single growing season and two test-bed sites, they should be treated as preliminary evidence warranting confirmation through multi-year, multi-site monitoring before informing province-wide management decisions. Future management strategies should integrate removal with active vegetation restoration, adopt region-specific approaches based on dominant species and habitat type, and incorporate long-term ecological monitoring to evaluate and improve management effectiveness.

Generative AI statement

The author(s) declared that generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Footnotes

Edited by: Riddhika Kalle, Salim Ali Center for Ornithology and Natural History, India

Reviewed by: Shenglin Yang, Beijing Normal University, China

Hugo López Rosas, El Colegio de Veracruz, Mexico