A Study of the Initial Mulching Effect in a Sapling Planting Area: Focusing on Sapling Growth Rate and Weed Occurrence Characteristics
Article information
Abstract
Background and objective
In ecological restoration sites, the use of saplings is essential for ensuring rapid adaptation and growth in the target environments. However, planting efforts have often focused on medium-to large-diameter trees due to challenges such as weed competition and management issues. To address these problems, this study aimed to investigate the effects of varying mulching thickness on weed emergence and cover, as well as survival and growth rates of planted trees in the experimental plots where saplings were planted. The findings are expected to serve as foundational data for proposing proper planting and management strategies utilizing saplings.
Methods
To examine the dynamics of tree growth and weed emergence, Quercus acutissima saplings with a root collar diameter (RCD) categorized as R2 were planted at a high density of two saplings per square meter and managed with conventional weeding twice a year. Woodchip mulch was applied at three different thicknesses: 0 cm, 5 cm, and 10 cm. Changes in tree growth and weed emergence were monitored over a 30-month period. The study was divided into two key aspects: tree growth assessment and weed vegetation analysis. Tree growth was evaluated based on survival rate, growth rate, and canopy projection. Weed vegetation analysis included the number of emergent herbaceous species, presence and proportion of naturalized plants, and vegetation cover.
Results
The survival rate in the mulched plots was 1.75 times higher than in the control plot with 0 cm mulch. As mulch thickness increased and the canopy projection expanded, the number of weed species decreased, effectively controlling the invasion of tall herbaceous plants competing with planted trees, as well as limiting the establishment of naturalized species. Once the canopy layer was established, all mulched plots exhibited significantly lower weed cover after weeding compared to the 0 cm mulch control. As canopy vegetation cover increased, the composition of weed species changed before and after canopy closure. In the early stages of the experiment, prior to canopy formation, high canopy openness created a light environment similar to that of agricultural fields. This condition favored the dominances of crop-associated weeds, tall herbaceous plants, and naturalized species. However, following canopy establishment, the number of emergent weed species decreased, with herbaceous vine species becoming dominant.
Conclusion
Applying mulch at a thickness of 5–10 cm combined with biannual weeding can support the growth and development of saplings compared to untreated conditions. Additionally, for successful vegetation restoration using saplings, intensive management of tall herbaceous plants that suppress sapling growth is necessary before canopy formation. After canopy closure, management strategies are required to address the emerging dominance of herbaceous vine species.
Introduction
In ecological restoration, the use of saplings—young, small-sized trees—is generally recommended over mature trees (Clewell et al., 2005; National Institute of Ecology, 2015; Cho, 2021). Saplings offer a more economical alternative to mature trees, as they are cheaper to purchase and transplant, facilitating affordable restoration forest development (Miyawaki and Golley, 1993). However, a survey of practitioners in South Korea revealed that trees with a root collar diameter categorized as R8 or larger are generally regarded as desirable and are most commonly used in planting designs for ecological restoration projects (Cho, 2022). This is largely due to the intensive management required during the early stages of restoration, particularly the frequent removal of invasive species and weeds (Lee et al., 2020). Saplings are highly susceptible to being outcompeted by weeds (Clewell et al., 2005; Ki and Kim, 2012), necessitating active maintenance efforts, including regular weeding, which can significantly increase restoration costs (Robin and Maria, 2016). Therefore, this study aimed to address the gap between theoretical recommendations and practical applications by empirically presenting an initial management strategy for sapling planting, based on a comparative analysis of mulching thickness and planting density.
Trees suitable for transplanting are typically those with a diameter at breast height (DBH) of 4 cm or less, or a height of 3 to 4 meters or less (Lee et al., 2020). Dense planting of container-grown seedlings of even smaller sizes is also recommended (Miyawaki, 1999). This approach, even in artificially established sites, has been shown to enable the development of forest stratification in a shorter period of time than primary natural succession, particularly when 2- to 3-year-old saplings are planted at high densities (Jung et al., 2007). Previous studies have further demonstrated that transplantation success rates decline as root collar diameter (RCD) increases: success rates ranged from 76.6% for trees with an RCD of 10 cm or less to just 20.0% for those with an RCD of 50 cm or greater (Lee et al., 2015). These findings underscore the advantages of transplanting smaller-diameter trees over larger ones.
However, the use of saplings presents some challenges in the early stages of planting management. Owing to the underdeveloped canopies of saplings, the vigorous growth of naturalized plant species, especially in areas with abundant sunlight (Kim and Oh, 2006), can hinder the successful establishment of target restoration species (Jeon et al., 2022). Young saplings with limited height can be overtopped by naturalized plants and tall weeds, thereby negatively affecting their growth (Cho, 2022). Vegetation cover and seedling growth have been shown to be inversely correlated (Kim and Kim, 2012), and dense weed cover can create favorable conditions for herbivores such as grasshoppers and voles, which can cause severe damage to young plants (McCreary and Tecklin, 1993). In other words, saplings occupy an ecological position in which they directly compete with weeds. Therefore, research is needed on weed management strategies during the sapling stage, as well as on the time required for target species to achieve a sufficient canopy cover to stabilize the vegetation.
However, in South Korea, Article 7-2 of the Landscape Standards issued by the Ministry of Land, Infrastructure and Transport (January 7, 2022) stipulates a minimum RCD of 6 cm for trees. Since trees with an RCD of R8 are typically planted as part of the tree stratum in ecological restoration projects (Cho, 2022), there has been no detailed, size-specific research conducted on the transplant success rates of trees with an RCD of R10 or less. As a result, planting design approaches for ecological restoration projects are relatively underdeveloped in terms of relevant laws, guidelines, or standards (Cho, 2022). Moreover, detailed guidance and empirical research on on-site construction and maintenance techniques necessary for actual projects remain severely lacking (Cho, 2021).
Against this backdrop, we aimed to identify the correlation between weed occurrence characteristics according to mulching thickness — which is noted as a method for minimizing management — and the growth rate of saplings. Saplings with an RCD of R2 were planted at high densities under varying mulching thicknesses, and their mortality and growth rates were monitored. The timing of canopy closure was determined, and weed occurrence characteristics were analyzed in the periods before and after this event. Based on these findings, we sought to provide foundational data to assist in resolving post-planting management challenges, which are among the key factors contributing to the reluctance to plant saplings in the field. This research holds significance as a foundational study analyzing weed occurrence patterns in sapling planting within ecological restoration projects, with the goal of proposing optimal planting and management practices.
Research Methods
Research site and experimental plot development
The experimental site for this study is located at 1186-2 Toerae-ri, Hallim-myeon, Gimhae-si, Gyeongsangnam-do (35°29′28.03″N, 128°79′00.34″E). To the north of the site lies Mt. Toeraemisan, and within a 500-meter radius are fields, rice paddies, orchards, and industrial complexes. With no tall mountains or buildings nearby, the environmental conditions at the site—including incident light—are uniform, providing a consistent setting for plant growth experiments. The site has an average slope of 2–7%, and drainage is good. The topsoil texture is loam, with an effective soil depth ranging from 50 to 100 cm. Previously, the site was used for cultivating black barley (Hordeum vulgare L.), and pesticides were applied within three days of sowing the black barley in October 2021.
To assess the physicochemical properties of the soil, samples were collected from a depth of 10–20 cm and analyzed by the Korea Society of Forest Environment Research, a specialized analytical institution. The soil texture was classified as silt loam (SiL). Total nitrogen (T-N), available phosphoric acid (Av. P2O3), and exchangeable calcium (Ca2+) were within the upper range, while pH, organic matter content (OM), cation exchange capacity (CEC), exchangeable potassium (K+), and exchangeable magnesium (Mg2+) fell within the intermediate range (Table 1).
Results of soil chemical analysis
Agricultural land use is a major driver of ecological degradation in the Mt. Jirisan section of the Baekdudaegan Mountain Range. Among the 14 restoration sites, which were classified into five categories based on the cause of degradation, farmland represents a significant type (Lee et al., 2020). Accordingly, in the context of ecological restoration research focused on areas degraded by agricultural land use, sites previously utilized for farming are regarded as suitable candidates for restoration studies. In particular, former farmland sites are rich in organic matter, nitrogen, and phosphorus due to past cultivation activities, along with buried seeds (Cho, 2017). These characteristics make such sites ideal for investigating patterns of weed germination.
Temperature and precipitation in Gimhae-si were analyzed using data from the Korea Meteorological Administration (KMA) for the period from April 2022 to September 2024. Notably, nationwide precipitation in May 2022—immediately following the planting period—was only 6.1% of the climatological average, with regional disparities leading to prolonged drought conditions in the southern region (MOIS, 2024; Fig. 1).
Climatic conditions (temperature and precipitation) in Gimhae-si.
The testing material consisted of a single tree species, Quercus acutissima (sawtooth oak), a member of the Fagaceae family. The most fundamental approach to selecting plant species for ecological restoration involves the use of native species, species from adjacent areas, or species from reference ecosystems (Clewell et al. 2005; Clewell and Aronson, 2013). Quercus acutissima is commonly employed in ecological restoration projects in Korea—such as forest restoration, urban ecological corridor restoration (ME, 2024), ecological preservation levy return projects (Cho, 2021), and purchased riparian land restoration—due to its rapid growth and high adaptability to poor soil conditions (Jo et al., 2014). The Quercus acutissima saplings used in this study were cultivated in Iksan, Jeollabuk-do Province, and transplanted on April 20, 2022. An organic soil conditioner (Saengmyeongjeong, a commercially available product in Korea) was mixed into the soil during the planting.
The experimental plots were square in shape, each measuring 2 meters in width and length (2 m × 2 m). A total of 24 trees were planted across three experimental plots, with eight trees per plot, corresponding to a planting density of 2 trees per square meter (2 trees/m2). Mulching treatments were applied at three depths: 0 cm, 5 cm, and 10 cm (Fig. 2). Furthermore, palm mats were installed in the pathways between plots to control external variables. The planting density of 2 trees/m2 followed the high-density planting method proposed by Miyawaki (1999). Compared to planting density standards used in Korea—such as 0.05 to 0.15 trees/m2 for urban forest types (KFS and NIFoS, 2024) and 1 tree/m2 according to the standards for mountain restoration and ecological restoration under the Enforcement Decree of the Special Act on the Management of Mountainous Districts North of the Civilian Control Line (MAFRA, 2025)—this represents a significantly higher planting density. Previous research has shown that high-density sapling planting enhances canopy cover and tree height growth, thereby reducing the influence of invasive plant species and serving as an effective method for establishing natural forests within a relatively short period (Kim et al., 2022). With regard to mulching, Lee (2021) suggested an optimal mulch thickness of 5 to 7 cm after planting to improve soil conditions. However, this contrasts with the findings of Jo and Park (2016), who reported that a 5 cm mulch layer was insufficient to suppress herbaceous plant invasion, indicating the need for further empirical validation. Furthermore, although numerous studies in Korea have investigated the effectiveness of mulching in waterside green spaces (Jo et al., 2013; Han and Park, 2021), research on its effectiveness in the ecological restoration of farmland remains limited. Wood chips—a type of biodegradable mulch—were used. Wood chips are widely recognized for their ability to retain soil moisture, regulate temperature, enhance nutrient content, prevent erosion, reduce salinity, and suppress weed growth (Chalker-Scott, 2007).
Experimental plots.
Survey and analysis methods
The study was conducted over a period of 30 months, from April 2022 to September 2024. During the early growth stage, Quercus acutissima trees were evaluated for survival rate, growth rate, and canopy projection. Weed species and cover were also assessed. Survival rates were measured four times, considering potential mortality due to transplant stress: in May and August 2022 (four months after planting), and subsequently at one-year intervals in August 2023 and August 2024. Trees were considered dead if more than two-thirds of their canopy branches had died (MOLIT, 2025). However, even if the main trunk had died, trees exhibiting root suckers (root shoots) were considered alive (Han and Park, 2022). Plant mortality was determined through visual inspection by the surveyors. The survival rate was calculated using the following formula: (Number of surviving trees/Number of planted trees) × 100.
Relative growth rates (RGRs) were assessed using data collected from August 2022 to August 2024. Tree height (cm) was measured vertically from the ground to the apex of each tree using a direct-reading height gauge (Senshin, Japan). Root collar diameter (RCD, cm) was consistently measured at the same point, which was marked with a white circular ring at the location where the stem contacted the soil surface. Measurements were taken using a diameter tape (MDS, China), which was pulled taut without stretching and held perpendicular to the stem. RGRs for both height and RCD were calculated using the formula: [ln(final value) - ln(initial value)]/period (days). To evaluate the statistical significance of the RGRs, non-parametric tests were conducted using IBM SPSS Statistics version 29.0.2.0. The Kruskal-Wallis test was used to analyze differences among the three experimental plots with varying mulching depths, followed by the Mann-Whitney U test for post-hoc pairwise comparisons.
Canopy projection was comparatively analyzed in May, August, and October of 2022, 2023, and 2024. To obtain the necessary data, aerial images were captured using a drone (DJI Mavic 3) at an altitude of 25 meters above sea level on days with favorable solar radiation conditions. Using the acquired digital images, tree canopies were visually identified, and their outlines were extracted using AutoCAD 2024 to calculate the area. Canopy projection was calculated as (canopy area within an experimental plot/area of the experimental plot) × 100.
Manual weeding was conducted twice annually, in July and September, following the method reported by Seo (2020) to be effective for promoting tree growth. Given the shorter life cycle of herbaceous plants compared to woody species, seasonal weed surveys were carried out in spring (May), after the summer rainy season (July), and in fall (September) to assess patterns of weed occurrence and cover. In total, eight surveys were conducted: July and September 2022; May, July, and September 2023; and May, July, and September 2024. The number of occurring species, the presence and proportion of naturalized species, and overall weed cover were surveyed and analyzed to identify trends in weed vegetation during the early establishment phase of saplings, particularly those of small to medium-sized woody plants. Naturalized plant species were identified by compiling a list of emerging species and cross-referencing it with the Information of Korean Alien Species database provided by the National Institute of Ecology (NIE). The naturalization rate was calculated following the method proposed by Yim and Jeon (1980): (number of naturalized plant species/total number of plant species) × 100. Weed cover, a factor related to the area under management, was calculated by recording field data in a field notebook and using AutoCAD 2024 to determine the proportion of the experimental plot covered by weeds.
Results and Discussion
Survival rate
Most tree mortality occurred within four months of planting, with the lowest survival rate observed in the experimental plot without woodchip mulching (Table 2). The plot without mulch (0 cm) exhibited a survival rate of 50%, which was significantly lower than the rates observed in the mulched plots. In contrast, both the 5 cm and 10 cm mulched plots showed a final survival rate of 87.5%, approximately 1.75 times higher than that of the unmulched plot. This suggests that mulching mitigated moisture stress caused by drought during the early planting period. Mulching is a fundamental soil management approach that reduces soil moisture evaporation and prevents excessive soil temperature increases during the summer months (Lee, 2021). Soil moisture content (%) was measured using the Soil Meter 1 (manufactured by Purum Bio Co., Ltd.) and analyzed using the Friedman test. Although the differences were not statistically significant (p = .196), the average soil moisture content was highest in the 0 cm mulched plot (36.64%), followed by the 5 cm (33.71%) and 10 cm (33.55%) mulched plots (Fig. 3). However, the standard deviation of soil moisture content was lowest in the 10 cm mulched plot, followed by the 5 cm and 0 cm plots, indicating that the 10 cm mulching provided the most stable soil moisture environment. Additionally, the 5 cm thick mulching delayed the invasion of herbaceous plants only during the early growth period of Quercus acutissima saplings (Jo and Park, 2016), suggesting an increase in sapling establishment rates in May and June—when competition for resources between weeds and introduced saplings is at its peak (Kim et al., 2022). The high density of sapling planting likely also contributed to the formation of a colony more resilient to environmental stress, thereby promoting stable establishment (Kim, 2011).
Tree survival rate by experimental plot
Soil moisture content by experimental plot.
Relative growth rate
The Kruskal-Wallis test for the relative growth rate (RGR) in tree height revealed no statistically significant differences among mulching depths (p = .054), However, the p-value was close to the significance threshold (p = .05), suggesting a potential trend of differences among mulching treatments. One tree in the 10 cm mulched plot had a missing value and was therefore excluded from the statistical analysis (Fig. 4, Table 3). The 10 cm mulched plot exhibited the highest RGR, with average values increasing as mulching depth increased (Fig. 4, Table 3). These results may be attributed to the high survival rate observed in the mulched plots. Under natural conditions—without supplemental light or water—higher planting densities tend to increase competitive exclusion pressure due to intensified intraspecific competition (Bormann, 1965). As a physiological response to avoid this exclusion—and given the general tendency for tree height growth to decline under greater light availability—height growth was likely enhanced in the mulched plots with higher planting densities compared to the control plot with 0 cm mulch. Furthermore, the 5 cm mulched plot exhibited a large standard deviation in RGRs, which may reflect intensified competition among individuals and resulting disparities in growth between dominant and suppressed trees. In contrast, the 10 cm mulched plot, despite showing a similar survival rate to the 5 cm plot, had a smaller standard deviation, suggesting that mulching helped mitigate environmental stress—particularly drought and transplant shock—during the initial establishment period (Fig. 4). These findings indicate that, in the context of increasing extreme weather events driven by climate change, appropriate mulching practices can contribute to enhancing tree survival. Although a pairwise Mann-Whitney U test indicated a significant difference between the 0 cm and 10 cm mulched plots (p = .038), this significance did not hold after applying the Bonferroni correction.
Relative growth rate by experimental plot.
Average tree growth and growth rate by experimental plot
The Kruskal-Wallis test for RGRs in root collar diameter (RCD) revealed no statistically significant differences among treatments (p = .976). However, RGRs were highest in the 10 cm mulch treatment, followed by the 0 cm and 5 cm mulch treatments (Fig. 4, Table 2). Diameter growth tends to decrease as stand density increases, whereas at lower densities, greater light interception enhances photosynthesis, thereby promoting stem tapering in the lower part of the trunk (Han and Park, 2021). Based on this, it was expected that the 0 cm mulch treatment—which had the lowest survival rate—would exhibit the highest RGR. Contrary to this expectation, the 10 cm mulch treatment showed the highest RGR. This suggests that mulching may alleviate initial transplant stress and environmental pressures, thereby supporting growth rates comparable to those observed in non-mulched conditions.
Canopy projection of planted trees
In May 2022, the canopy projection ratio (CPR) was 30.55% for the 0 cm mulching treatment, 57.79% for 5 cm mulching, and 49.82% for 10 cm mulching. By July 2022, many trees in the 0 cm mulched plot had died, reducing the CPR to 23.37%. In contrast, the CPR increased to 54.56% and 67.44% for the 5 cm and 10 cm mulching treatments, respectively, despite the death of one tree in the 10 cm plot. In September 2022, the CPR for the 0 cm treatment showed a slight increase to 25.71%, though it remained lower than the May value. CPRs continued to rise for the 5 cm and 10 cm treatments, reaching 60.58% and 84.14%, respectively. By May 2023, canopy overlap between experimental plots began to occur. The CPR was 44.29% for 0 cm mulching, 95.47% for 5 cm mulching, and 94.58% for 10 cm mulching. From July 2023, complete canopy overlap was observed in the 5 cm and 10 cm mulched plots, making it difficult to distinguish between them. At this time, the CPR was 60.32% for the 0 cm treatment, 97.06% for 5 cm, and 100% for 10 cm. From this point onward, tree branches in the 5 cm and 10 cm mulched plots began to extend significantly beyond the boundaries of their respective plots. By September 2023, the 0 cm mulched plot reached a CPR of 66.32%, while both the 5 cm and 10 cm plots achieved 100%, indicating complete canopy closure except in the 0 cm treatment (Fig. 5, Table 4). In the plots with mulching of 5 cm or more, high-density planting led to canopy closure within 14 months. Two months later, branches had extended significantly beyond the plot boundaries, limiting the incident light reaching the ground surface within the plots at all times—regardless of the sun’s position or altitude. Successful ecological restoration requires both rapid early-stage growth and prompt canopy closure, which typically demand intensive management for approximately the first two years after planting (Kuusipalo et al., 1996). In this study, the 5 cm and 10 cm mulched plots achieved canopy formation within about one year after planting. This indicates that mulching supported stable early growth under high-density conditions, facilitating faster canopy development. Therefore, mulching is likely to contribute to more efficient management by shortening the intensive management period compared to non-mulched conditions.
Changes in canopy projection area. *(a) represents May, July and September 2022, (b) represents May, July and September 2023. In the figure, gray areas indicate regions covered by the canopy.
Canopy projection change and rate of increase/decrease
Weed vegetation analysis
A total of 36 weed species belonging to 19 families were identified beneath the trees in the three experimental plots (Table 5). The distribution of weed species across the plots was as follows: 9 species from the Poaceae family (25.0%), 6 species from Asteraceae (16.6%), and 2 species each from Fabaceae, Polygonaceae, Chenopodiaceae, and Amaranthaceae (5.5%). Additionally, 1 species each was recorded from the following families: Araceae, Convolvulaceae, Phytolaccaceae, Equisetaceae, Malvaceae, Boraginaceae, Commelinaceae, Euphorbiaceae, Scrophulariaceae, Rubiaceae, Solanaceae, Lamiaceae, and Brassicaceae (2.7% each). Seven weed species were consistently observed across all three experimental plots: Galinsoga quadriradiata Ruiz and Pav., Pinellia ternata (Thunb.) Breitenb., Digitaria ciliaris (Retz.) Koeler, Persicaria lapathifolia (L.) Delarbre, Galium spurium L. var. echinospermum (Wallr.) Desp., Setaria faberi R.A.W. Herrm., and Setaria viridis (L.) P. Beauv. In the first year following plot establishment, Galinsoga quadriradiata Ruiz & Pav. was the dominant species across all plots, while Digitaria ciliaris (Retz.) Koeler was primarily found in drier areas. Regardless of mulching depth, Pinellia ternata (Thunb.) Breitenb., Echinochloa crus-galli (L.) P. Beauv., Vicia tetrasperma (L.) Schreb., and Chenopodium album L. var. centrorubrum Makino appeared frequently. In the September survey, naturalized species such as Amaranthus patulus Bertol. and Chenopodium album L. were dominant in the 0 cm and 5 cm mulched plots. In contrast, only small numbers of native species, including Digitaria ciliaris (Retz.) Koeler and Persicaria lapathifolia (L.) Delarbre, were observed on the margins of the 10 cm mulched plot. Following the formation of tree canopies in the second year, from spring until weeding in July, the 5 cm and 10 cm mulched plots became dominated by vines—primarily Galium spurium L. var. echinospermum (Wallr.) Desp. and Calystegia hederacea Wall. In addition, several herbaceous species such as shepherd’s purse (Thlaspi bursa-pastoris L.), purple deadnettle (Lamium purpureum L.), and dandelion (Taraxacum platycarpum H. Dahlst.) were also present. After weeding in July, populations of Galium spurium L. var. echinospermum declined, and species from the Poaceae family became dominant. These included species of Setaria (e.g., Setaria viridis (L.) P. Beauv., Setaria faberi R.A.W. Herrm., and Setaria pumila (Poir.) Roem. & Schult. subsp. pallide-fusca (Schumach.) B.K. Simon), Panicum (Panicum bisulcatum Thunb. and Panicum dichotomiflorum Michx.), and Digitaria (Digitaria ciliaris (Retz.) Koeler and Eleusine indica (L.) Gaertn.).
Occurrence status of herbaceous plants
The number of weed species identified in each experimental plot was 26 species from 15 families in the 0 cm mulched plot, 16 species from 12 families in the 5 cm mulched plot, and 11 species from 8 families in the 10 cm mulched plot. The 0 cm mulched plot exhibited the highest diversity of weed taxa, which is likely attributable to the continuous disturbance from weeding activities under conditions of high light intensity on the soil surface before the formation of a canopy layer. As the number of weed taxa increases, a greater diversity of herbaceous growth forms—such as erect, prostrate, and tufted types—can emerge concurrently. Therefore, the frequency and effort required for manual weeding can vary depending on whether mulching is applied. In other words, mulching in planted areas can reduce the need for frequent weeding, thereby lowering overall maintenance costs.
Most of the weeds that emerged in the experimental plots were similar to those typically found in fields across Gyeongnam Province (Chang et al., 1990; Seong et al., 2015; Lee et al., 2017). However, while barley fields in Gyeongnam are primarily cultivated as winter crops on drained paddy fields, where vine species show low dominance (Seong et al., 2015), the experimental plots differed in that herbaceous vines, including Galium spurium L. var. echinospermum (Wallr.) Desp., became dominant from the second year onward. This difference is likely attributable to the growth cycle of Galium spurium L. var. echinospermum (Wallr.) Desp., which germinate around mid-October and mature by late May. In this study, manual weeding was conducted only in July and September, allowing these vines to escape disturbance and establish successfully. Additionally, woodchip mulching and subsequent canopy formation appear to have suppressed the emergence of other photophilous (light-loving) weed species, thereby favoring the emergence and growth of herbaceous vines. This finding aligns with previous studies reporting that woodchip mulching reduces overall weed populations and species richness while enhancing the dominance of vine species, which tend to outcompete other plants in community dynamics (Kim and Oh, 2006). Notably, vine species were not dominant in the first year, likely because initial weeding was carried out during plot establishment in April 2022. Moreover, since the optimal germination depth for Galium spurium L. var. echinospermum (Wallr.) Desp. is approximately 2 to 5 cm ( Lee et al., 1994), it is presumed that mulching may have physically inhibited their germination during the first year.
A total of 13 species of naturalized plants emerged across the experimental plots, with 10 species found in the 0 cm mulched plot, 4 species in the 5 cm mulched plot, and 3 species in the 10 cm mulched plot (Table 5). The overall naturalization rate in the experimental plots was 36%, with rates of 38.4%, 25%, and 27.2% in the 0 cm, 5 cm, and 10 cm mulched plots, respectively. In the bare 0 cm mulched area, tall herbaceous species such as Erigeron annuus (L.) Desf., Erigeron canadensis L., Chenopodium album L., and Sonchus asper (L.) Hill were observed. However, these tall herbs were maintained at a short height due to regular weeding. Naturalization rates in two forest ecosystems—one illegally cultivated and the other barren—where ecological preservation levy return projects were implemented, were 36.3% and 32.4%, respectively (Park, 2018), closely aligning with the results of this experiment. However, the naturalization rates in mulched areas were 11.2–13.4% lower than those in unmulched areas, which is also lower than the rates reported by Park (2018). This finding supports previous research indicating that naturalization rates and urbanization indices, which assess plant environments, decrease following woodchip installation (Kim & Oh, 2006). Unlike our study, the prior research involved woodchip installation one year after the establishment of greening beds. In any case, woodchip application appears to enhance soil moisture retention and light obstruction, thereby creating conditions that inhibit the invasion of naturalized plants that prefer dry, well-lit environments (Kim et al., 2016).
Weed cover
The mulched plots exhibited a reduction in weed cover beginning in 2023, attributed to both mulching and the onset of canopy formation. In contrast, the bare soil plots (0 cm mulch) maintained high levels of weed cover through 2024 (Fig. 6). During the early establishment phase (2022), both the 0 cm and 5 cm mulched plots exhibited 100% weed cover. However, the 10 cm mulched plots showed substantially lower weed cover—75% in July and 65% in August—particularly along the plot perimeters. Notably, in July 2022, three months after planting, herbaceous weeds in the 5 cm mulched plots were shorter than those in the 0 cm mulched (bare soil) plots, suggesting that woodchip mulching had a growth-suppressing effect on herbaceous species.
Weed cover by year.
After 2023, when the canopy projection ratio exceeded 90% in the 5 cm and 10 cm mulched experimental plots, a clear difference in weed cover emerged between the bare and mulched plots. In May 2023, herbaceous vines such as Galium spurium L. var. echinospermum (Wallr.) Desp. and Vicia tetrasperma (L.) Schreb., which had spread from the passages surrounding the 5 cm and 10 cm mulched plots, covered the ground and resulted in high weed cover. In July, weeding was conducted two weeks prior to the vegetation survey. Two weeks after weeding, the 0 cm mulched plot exhibited a 95% weed cover due to the re-growth of various naturalized species, including Setaria faberi R.A.W. Herrm., Erigeron canadensis L., Solanum americanum Mill., and Persicaria lapathifolia (L.) Delarbre. In contrast, the 5 cm and 10 cm mulched plots showed significantly lower weed cover and naturalization rates, at 5% and 3%, respectively.
In May 2024, weed cover in the experimental plots reached 100% in the 0 cm mulch treatment, 90% in the 5 cm mulch treatment, and 70% in the 10 cm mulch treatment. In the 0 cm mulched plots, the high weed cover was primarily due to the initial emergence of species such as Trifolium repens L., which were subsequently overtopped by Galium spurium L. var. echinospermum (Wallr.) Desp., which had spread into the plots from surrounding areas. Similar to the previous year, the 5 cm and 10 cm mulched plots were predominantly colonized by the herbaceous vine Galium spurium L. var. echinospermum (Wallr.) Desp., followed by the emergence of Calystegia hederacea Wall., resulting in high overall weed cover. However, after these herbaceous vine species withered due to weeding, the mulched plots—unlike the bare-ground (unmulched) plots—exhibited significantly reduced weed cover. This finding aligns with previous studies indicating that high-density plantings can suppress the mass emergence of sunloving pioneer species within one to two years by achieving rapid canopy closure (Han and Park, 2022). In riparian green spaces established along the four major river systems in South Korea (the Hangang, Nakdonggang, Geumgang, Yeongsangang, and Seomjingang Rivers), the canopy projection ratio was reported to be as low as 28%. This insufficient canopy coverage allowed for the proliferation of heliophilous pioneer species such as herbaceous plants, which in turn hindered the growth of planted trees and contributed to increased maintenance costs (Jo et al., 2014).
Conclusion
In ecological restoration sites, the use of saplings is essential for enabling adaptation to target environments and ensuring stable growth. However, due to their vulnerability to competition from weeds, managing saplings imposes a significant burden on practitioners. As a result, restoration efforts often focus on planting medium-to large-diameter trees instead. To address this challenge, this study aimed to determine optimal planting densities, mulching treatments, and weed management practices to support stable sapling growth. Specifically, the study evaluated survival rates, growth rates, and weed dynamics following planting. In April 2022, Quercus acutissima saplings (with a root collar diameter classified as R2) were planted at a high density of two trees per square meter in Gimhae-si, Gyeongsangnam-do. In addition to the conventional practice of biannual weeding, woodchip mulch was applied at three thickness levels: 0 cm, 5 cm, and 10 cm. Changes in sapling growth and weed occurrence were monitored over a 30-month period. The study analyzed two key aspects: tree growth and weed vegetation. Tree growth was assessed through measurements of survival rates, growth rates, and canopy projection. Weed vegetation was evaluated by surveying the number of weed species present in each plot, identifying naturalized species and their naturalization rates, and estimating overall weed cover.
The results of the study showed that survival rates in plots treated with woodchip mulch were 1.75 times higher than those in bare-ground plots without mulch. Despite frequent droughts caused by abnormal weather conditions during the early establishment period, woodchip mulching effectively suppressed early weed emergence and mitigated transplant shock, contributing to high survival rates. This outcome may also be attributed to the higher planting density facilitating early canopy closure, which alleviated environmental stress and promoted height growth through increased intraspecific competition.
Patterns in weed species occurrence varied significantly before and after canopy closure. In the early stages of planting, when canopy openness was high, field weeds were dominant. Tall herbaceous species such as Galinsoga quadriradiata Ruiz & Pav., Amaranthus patulus Bertol., and Chenopodium album L., along with various naturalized plants, were prevalent. After canopy closure, species richness declined; however, both winter and summer southern field weeds emerged, and herbaceous vine species became dominant. Notably, in this study, Galium spurium L. var. echinospermum (Wallr.) Desp. expanded its presence significantly from spring until just before weeding in July, becoming the dominant species during that period.
As mulching thickness and canopy projection increased, the number of herbaceous species decreased. Mulching was effective in limiting the establishment of tall herbaceous species and naturalized plants. In the early stages of planting, only the 10 cm mulch treatment effectively suppressed weed growth, whereas the 5 cm treatment failed to reduce initial weed cover. This suggests that thinner mulch was insufficient to prevent the germination of residual seeds already present in the soil and the influx of herbaceous species. However, once the crown layer was established, even the 5 cm mulch effectively suppressed herbaceous plant establishment. Therefore, in areas mulched to a depth of 10 cm, even if some of the mulch is lost due to environmental factors such as heavy rainfall or sloped terrain, the remaining 5 cm—comparable to the average organic layer thickness in Korean forest soils (Lee, 2021) —may help maintain the suppressive effect until the canopy closes, thereby creating a forest floor environment similar to that of natural forests.
In other words, the combination of high-density planting and 10 cm-deep wood chip mulching was found to be the most effective approach for promoting sapling growth and controlling weeds. This method significantly reduced the need for intensive, repetitive manual weeding, thereby enabling more efficient long-term maintenance and management. However, after the canopy has formed, management strategies need to be adjusted to address the dominance of herbaceous vines. Therefore, prior to canopy formation, it is advisable to maintain the existing weeding regime, with targeted removal of tall herbaceous species that compete with saplings, particularly in July and September. Following canopy closure, the weeding frequency can be reduced to once annually, with management efforts focused on controlling herbaceous vine species.
This study is based on the results of a 30-month experiment. However, the limited number of replicates within the experimental plots constrained statistical testing. The analysis was also confined to basic comparative assessments of canopy projection, weed species occurrence, and ground cover, which restricts the ability to draw broadly generalizable conclusions. Furthermore, the absence of a baseline survey of weed conditions prior to plot establishment hindered the analysis of weed species dynamics and intrusion routes. These limitations highlight the need for methodological refinement and further research. In the future, ongoing monitoring will be necessary to establish appropriate management strategies and modeling applications that align with ecological succession dynamics.