Planting Structures and Carbon Reduction Effects of Urban Greenspaces in Reclaimed Coastal Areas
Article information
Abstract
Background and objective
In order to enhance the contribution of urban greenspaces to carbon uptake, it is essential to assess the carbon reduction effects of these areas. Accordingly, this study examined the planting structures and carbon reduction effects of urban greenspaces in reclaimed coastal areas (RCAs), which are characterized by unique growth conditions, such as high salinity, that distinguish them from inland areas.
Methods
Field surveys were conducted across urban greenspaces in RCAs along the west and south coasts of Korea to document their planting structures and soil characteristics. Based on the obtained data, a quantitative analysis of the carbon reduction effect of trees and soil within these greenspaces was conducted.
Results
It was found that the average tree cover in the surveyed greenspaces was 37.2%, while soil salinity and organic matter content were measured at 0.03% and 4.2%, respectively. The carbon storage per unit area of the greenspace, attributable to both trees and soil, reached 59.2 t/ha, with an estimated total annual carbon uptake of approximately 5.7 kt/year. This uptake is equivalent to offsetting the carbon emissions of about 153,000 people based on the energy consumption of residential buildings. Drawing on these results, this study proposed strategies to further enhance carbon reduction in reclaimed coastal greenspaces. The recommended strategies include optimizing soil conditions for plant growth, implementing multi-layered and clustered planting with medium and large trees, and selecting tree species that tolerate high salinity.
Conclusion
Overall, this study makes a valuable contribution toward establishing a carbon uptake indicator for trees and soil in reclaimed coastal greenspaces, where relevant data are currently scarce. Future research should aim to include diverse types of greenspaces and larger sample sizes to more accurately capture annual soil carbon fluxes.
Introduction
Coastal reclamation is the process of constructing embankments along the coast and filling the enclosed area to create new land. It has been considered a way to efficiently expand and utilize limited land resources (Kim et al., 2000; Wang et al., 2014). Countries with coastlines, including the Netherlands, Korea, Japan, China, and Singapore, have reclaimed land from the sea, transforming these areas into farmland, industrial zones, airports, ports, and even new cities (Glaser et al., 1991; Suzuki, 2003; Hoeksema, 2007; Jun, 2011; Wang et al., 2014). In particular, Korea, which is surrounded by the sea on three sides and has more than 70% of its land area covered by mountains, has been reclaiming its relatively shallow western and southern coasts since the 1960s to create large-scale residential areas and improve land use efficiency. According to the Ministry of Agriculture, Food, and Rural Affairs (2015), Korea has developed approximately 197,158 hectares of reclaimed coastal areas (RCAs) to date, and plans to continue such projects in the future.
Meanwhile, greenspaces such as parks, street trees, and buffer zones have been created in RCAs as part of land development projects, alongside infrastructure like buildings and roads. However, the trees planted in these greenspaces are more susceptible to damage from salinity and onshore winds than those planted inland, making it harder for them to establish roots and resulting in poor growth (Bernstein and Ayers, 1971). Specifically, in areas with high soil salinity, such as RCAs, osmotic pressure causes water to move from plant roots into the soil, resulting in water and nutrient deficiencies for plants (Loveday, 1976; Samson et al., 2017). Moreover, salinity degrades the aggregate structure of the soil, impairing aeration, drainage, and water permeability (Kim et al., 2006). In fact, according to a previous study (Korea Land and Housing Corporation, 1995), the defect rate of trees in RCAs was 10–25% higher than that in general inland areas.
Due to these environmental conditions, most previous studies on greenspaces in RCAs have focused on analyzing the growth status of trees after planting (Kim et al., 2000; 2002; Choi et al., 2002; Park et al., 2003) and soil characteristics (Koo et al., 1999; 2000; Kim et al., 2011; Jun, 2011; Lyu and Chen, 2016; Beaven et al., 2018), or identifying species with high salt tolerance (Bouma et al., 2001; Paludan-Müller et al., 2002). Although not conducted in RCAs, a few studies have examined the damage caused by salinity in coastal greenspaces (Hallett et al., 2018). According to previous studies, 17–30% of all trees planted in RCAs or along the coast died due to salinity (Park et al., 2003; Hallett et al., 2018), and the photosynthetic rate of trees affected by salinity decreased by up to 50% (Paludan-Müller et al., 2002). On the other hand, tree species with high salt tolerance, such as Pinus thunbergii and Celtis sinensis, experienced relatively low mortality. Soil characteristics of reclaimed coastal greenspaces varied depending on the type of fill material, the depth of the planting soil, and the management method used.
Meanwhile, urban greenspaces (UGSs) have recently been considered as one of major carbon sinks in the fight against climate change, sparking growing global interest. Accordingly, many studies have been conducted to explore the carbon reduction role of greenspaces (Nowak and Crane, 2002; Nowak et al., 2013; McGovern and Pasher, 2016; Jo et al., 2019a; 2023), but there is a lack of research specifically focusing on greenspaces in RCAs worldwide. Some studies of carbon reduction in RCAs have been limited to determining changes in soil organic carbon (SOC) storage after reclamation. Deng et al. (2016) found that the SOC storage in RCAs in Zhejiang Province, China, was 3.18 kg/m2. Li et al. (2018) reported that the organic carbon storage of agricultural soil in RCAs in Jiangsu Province, China, ranged from 2.83 to 2.91 kg/m2. Although not RCAs, some coastal urban parks in New Zealand accumulated about 45 t/ha of carbon per unit area (Schwendenmann and Mitchell, 2014), while some coastal windbreak greenspaces in China sequestered between 26.6 and 79.8 t/ha of carbon (Wang et al., 2013).
To improve the carbon sink role of coastal windbreaks/greenspaces, it is essential to have baseline data on current carbon storage and uptake. However, as mentioned above, research on this topic is still lacking worldwide. It is well known that the carbon reduction effect of UGSs is affected by factors such as planting structure (e.g., species, density, and size), growth status, and soil environment. Applying the carbon storage and uptake data from inland areas with different growth environments to RCAs may lead to errors. Therefore, this study aimed to survey the planting structure of UGSs in the RCAs along the west and south coasts of Korea and to quantify the carbon reduction effects of trees and soil. This study is expected to contribute to the establishment of baseline data for determining the carbon storage and uptake of UGSs in RCAs, an area where data is currently lacking globally, and to enhance their effectiveness.
Research Methods
Selection of study greenspaces
In this study, a total of 20 UGSs in RCAs were selected as study sites, considering factors such as the size of the coastal reclamation project, location, year of construction, and area. Of these, 14 are located on the west coast, and 6 are on the south coast (Fig. 1). These UGSs, as defined by the Act on Urban Parks and Green Areas, consist of urban parks and green areas. They were selected as study sites since they are public goods that allow for field surveys and soil sampling. Meanwhile, the 20 study sites were selected as follows. First, UGSs in the RCAs along the west and south coasts, which had been planted for at least 5 years and could be accommodated at least three quadrats, each at least 10 meters wide, were identified. The final sites were then selected using stratified random sampling, considering both regional distribution and project size. According to the Ministry of Oceans and Fisheries (2022), about 70% of the RCAs designated for urban development, public facilities, and housing are on the west coast, and the remaining 30% are on the south coast. In this study, the selection of study sites was based on this ratio and the main areas of coastal reclamation projects. The greenspaces finally selected for this study were developed after 2000 on the west coast and between 1990 and 2000 on the south coast, indicating that development occurred earlier on the south coast than on the west coast.
Field survey of planting structures and soil environments
Field surveys were conducted in the late summer and early fall of 2022 and 2023 to assess the soil environment, planting structure, and tree growth status at each study site. Three quadrats, each at least 10 meters wide, were established at each site. After removing the ground cover from the center of each quadrat, 1 kg of topsoil was collected to a depth of approximately 30 cm and dried in the shade of the laboratory. The physicochemical properties of the topsoil were analyzed according to the soil analysis method suggested by the National Institute of Agricultural Sciences (NAS; 2000). Specifically, soil pH was measured using a pH meter in a 1:5 ratio of soil to distilled water, and organic matter (OM) content was estimated by a wet oxidation method. In addition, the physicochemical properties of available phosphate, exchangeable cations, and cation exchange capacity (CEC) were analyzed using the Lancaster soil test, atomic absorption spectrophotometry (AAS), and the 1N-ammonium acetate method, respectively.
The planting structure was measured for all trees within each quadrat, including species, stem diameter, crown width, and tree height. These data were analyzed quantitatively to determine species composition, density, size, cover, and similarity index. For the similarity index, we referenced data from previous studies on carbon reduction effects in urban parks in Seoul, Daejeon, and Daegu (Jo et al., 2019a; 2023) to compare the species of the planted trees with those of UGSs in RCAs in this study. In addition, the single-layer and multi-layer planting techniques that influence the height of plantings and their functions, including carbon reduction capacity per unit area, were also investigated and analyzed as part of the vertical planting structure.
The growth status of the planted trees was assessed by examining the mortality rate, the proportion of growth-deficient individuals, new shoot length, and leaf length for the major dominant tree species within the quadrat. Criteria for identifying growth-deficient individuals were specified for each tree part: crown imbalance (such as wilting or loss of main branches) for crowns, stem decay or trunk cavities for stems, dead spots on branches or poor new shoot growth for branches, and leaf dwarfism or abnormal leaf color for leaves.
Calculation of carbon storage and uptake
The carbon reduction effect of the trees at the study sites was quantified by applying quantitative models for carbon storage and uptake developed for urban landscape trees to each tree (Jo, 2002; 2020; Jo and Ahn, 2012; Jo and Cho, 1998; Jo et al., 2013; 2014; 2019b; 2019c; Greenhouse Gas Inventory and Research Center of Korea, GIR, 2023). This model estimates carbon storage and uptake based on factors such as diameter at breast height (DBH), tree height, and crown width. It was developed using seasonal CO2 exchange rate measurements or through direct harvesting including roots. Based on data from a field survey of the quadrats installed in the study greenspaces, the process for estimating the carbon reduction effect of trees is illustrated in Fig. 2. First, using the tree inventory from the field survey, a quantitative model for carbon storage and uptake was selected for each tree species with reference to Table 1 and Appendix 1–2 and applied to each tree.
For tree species without an applicable quantitative model, models for related species within the same genus or group were used as substitutes, or their averages were applied. For example, since there was no appropriate model for Cedrus deodara in Fig. 2, the model for Pinus densiflora, which belongs to the same family, was used. The carbon storage and uptake of the trees was then estimated by applying the DBH, the root collar diameter (RCD), the number of trees planted, and more to quantitative models of carbon storage and uptake corresponding to each tree species. For some tree species, both a model based on the CO2 exchange rate and one using the direct harvesting method were available. In these cases, the carbon reduction effect was estimated by averaging the results from both models. For tree species with a DBH larger than the range of the available quantitative models, a model for a related species within the same genus or group that could accommodate the larger DBH was used as a substitute. For example, a broadleaf tree model was applied for Chionanthus retusus.
Through this process, the carbon reduction effect of each tree was estimated and summed to derive the total carbon reduction effect. Soil carbon storage was analyzed for organic carbon content in 60 samples and then converted to area units using bulk density, gravel contents, and other factors based on the estimation method outlined by the National Institute of Forest Science (NIFS; 2007). The equation below suggests a method for converting carbon storage by organic matter into area units (NIFS, 2007).
Where: Y = carbon storage (t/ha); T = soil thickness (cm); BD = soil bulk density (g/cm3); C= organic carbon content; CF= gravel content ratio; and MFp= moisture content
However, annual soil carbon fluxes were not considered in this study due to a lack of established methods and baseline data. In addition, the carbon reduction effects of both trees and soil at the study sites were converted to area units by linking them to the area of each site after summing the individual values.
Statistical analyses
In this study, Microsoft Office Excel 2016 and SPSS Statistics 26.0 for Windows were used to estimate the mean carbon storage and uptake, tree density, cover, DBH, and other factors per unit area in the study sites, and then the mean error was statistically analyzed. Moreover, a t-test was performed to determine whether there were statistically significant differences in the greenspace structure, soil OM, and carbon storage and uptake between the west and south coasts. To identify factors affecting the carbon reduction effect, Pearson’s correlation coefficient was used to analyze the relationship between greenspace structure and soil OM and the carbon reduction effect.
Results and Discussion
Soil environments
The soil texture of the UGSs studied was mainly sandy soil or loamy sand, with the pH levels ranging from 5.4 to 7.9%, and an average of 6.4 ± 0.2% (Table 2). According to the soil classification system based on measurable characteristics proposed by the Korean Institute of Landscape Architecture (KILA, 2016), the pH of the UGSs ranged from medium to high levels, depending on the sites. Electrical conductivity (EC) and salinity are key indicators for diagnosing salinity damage to plants. It is known that values above 4.0 dS/m for EC and 0.05% for salinity can negatively affect plant growth (Marucum and Murdoty, 1990; Miller and Donahue, 1990; Tanhi, 1990). Previous studies found that the EC of greenspaces in Incheon’s RCAs ranged from 0.1 to 4.8 dS/m (Kim et al., 2011), while the salinity of greenspaces on the west and south coasts ranged from 0.02% to 0.18% (Choi et al., 2002). In comparison, the EC and salinity of the greenspaces studied were lower, with averages of 0.5 ± 0.1 dS/m and 0.03 ± 0.00%, respectively. However, in some of the study greenspaces, the EC ranged from 0.05 to 0.13 dS/m and the salinity ranged from 1.3% to 2. 0%, both of which were low levels or below the low threshold.
The OM, available phosphate, and cation exchange capacity of the study sites were 4.2 ± 0.2%, 115.5 ± 14.7 mg/kg, and 9.9 ± 0.7 cmol+/kg, respectively, with most of these values falling within the medium range. It has been recommended that RCAs maintain at least 3.0% OM to support the normal growth of planted trees (KILA, 1999). In comparison, the majority of the greenspaces in the RCAs studied met the minimum OM standard. Except for Ca2+, all exchangeable cations were at medium or higher levels. Soils in natural riparian forests in Korea were reported to have a pH of 4.9, OM content of 6.3%, total nitrogen of 0.32%, and cation exchange capacity of 14.7 cmol+/kg (Center for Aquatic Ecosystem Restoration, 2014). In comparison, the soils in the greenspaces in this study generally had lower values.
Meanwhile, statistically significant differences (p < . 05; Table 3) were found in the pH, EC, salinity, available phosphorus, and K+ levels between the soils of the west and south coasts. In general, the soils on the west coast had higher EC, salinity, and available phosphorus, while the K+ levels were higher on the south coast. Previous studies (Ismayilov et al., 2021; Mazur et al., 2022; Yan et al., 2023) have shown that EC and salinity are typically proportional to the levels of K+, Mg2+, Ca2+, Na+, and available phosphorus. However, the soil characteristics on the west coast mentioned above did not exhibit a significantly higher tendency compared to those on the south coast. Therefore, the higher salinity and EC found on the west coast are likely due to differences in salt leaching, which may be affected by the time elapsed since planting, in addition to the effects of the soil characteristics mentioned above. In other words, since the greenspaces along the south coast were developed about 10 years earlier than those along the west coast, salt leaching occurred over a longer period of time, which appears to have lowered the salinity and EC.
Planting structures
The tree density across the study sites ranged from a minimum of 1. 0 to a maximum of 15. 6 trees/100 m2, with an average density of 5.9 ± 0.9 trees/100 m2 (Table 4). When categorized into urban parks and buffer/landscape green areas, the tree density per unit area was 4. 5 trees/100 m2 in urban parks and 7.9 trees/100 m2 in buffer/landscape green areas. This indicates that buffer/landscape green areas have a tree density 1.8 times higher than that of urban parks. The tree cover of the study sites was an average of 37.2 ± 6.7%, and similar to tree density, tree cover tended to be higher in buffer/landscape green areas than in urban parks. This difference seems to be attributed to the nature of urban parks, which include various facilities for people’s recreation and leisure activities in addition to tree planting. Meanwhile, the tree density and cover in urban parks of major inland cities in Korea, such as Seoul, Daejeon, and Daegu, were reported to be in the range of 3.0–4.1 trees/100 m2 and 42.9–50.9%, respectively (Jo et al., 2019a; 2023).
Tree density in the study UGSs was generally higher than above, while tree cover was 5.7% to 13.7% lower. Compared to previous studies, these greenspaces had higher tree density but lower tree cover, apparently due to the size of the trees planted. The average DBH of the trees in the study greenspaces was 13.2 ± 0.9 cm, which is 70% of the DBH found in urban parks in Seoul, Daejeon, and Daegu. The DBH distribution in these greenspaces was as follows: 50.8% for trees with a DBH of 10–20 cm, 39.5% for those below 10 cm, 8.4% for 20–30 cm, 1.1% for 30–40 cm, and 0.2% for 40 cm and above. Trees in the growth stage were particularly dominant.
The average number of tree species planted per site in the study sites was 7.6 ± 1.4, with a total of 87 species. The similarity index of tree species planted between the UGSs in this study and the urban parks in Seoul, Daejeon, and Daegu ranged from 0.47 to 0.50. According to a previous study (Jo, 2020), the similarity index between UGSs in Seoul, Daejeon, Daegu, Chuncheon, and Suncheon was reported to range from 0.51 to 0.58. The similarity index between UGSs in these RCAs and urban parks in inland areas was lower, probably due to the fact that the tree species planted were mainly tolerant to salinity and onshore winds, which are characteristic of the RCAs.
In fact, the species with the highest relative dominance in the study greenspaces was P. thunbergii (23.7%), which has a high tolerance to both soil salinity and salt spray drift. It was followed by Zelkova serrata (10.7%), Chionanthus retusus (9.6%), Acer buergerianum (5.7%), Prunus yedoensis (5.0%), Rhododendron yedoense (3.8%), Cornus officinalis (2.7%), Quercus rubra (2.6%), Metasequoia glyptostroboides (2.6%), and C. sinensis (2.3%). Notably, after P. thunbergii, C. sinensis is also recognized for its high tolerance to soil salinity and salt spray drift (Oh et al., 2010). The vertical structure of tree planting in the UGSs studied was single-layer structure, in which only trees, shrubs or grass were planted, accounting for 90% of the total planted area, while multi-layer structure, in which trees, shrubs and herbs were planted together, accounted for only 10%. Considering that the multi-layer structure, in urban parks in Seoul, Daejeon, and Daegu ranged from 41.8 to 48.5% of the total area, these UGSs in the RCAs tended to be relatively dominated by single-layer structure.
Meanwhile, a comparative analysis of the planting structures in the greenspaces on each coast found statistically significant differences in tree cover and mean DBH between the west and south coasts (p < . 01; Table 5). Specifically, tree cover and mean DBH on the south coast were found to be 2.8 and 1.6 times greater, respectively, than those on the west coast. This seems to be because trees on the south coast grew for a longer period of time than those on the west coast, which resulted from variations in planting timelines. Although not statistically significant, tree density tended to be higher on the south coast compared to the west coast. This difference appears to be attributed to the fact that the study greenspaces on the south coast are adjacent to industrial complexes, while some greenspaces on the west coast are situated in residential areas of new cities, where various facilities have been integrated with tree planting to meet the needs of residents. The similarity index of planted tree species between UGSs on the west and south coasts was quite low, approximately 0.22. This is explained by the fact that, in addition to P. thunbergii, salt-tolerant tree species native to the southern region, including Quercus myrsinifolia, Machilus thunbergii, and Ilex rotunda, were planted on the south coast.
Growth conditions of planted trees
An analysis of tree growth in the study sites showed that approximately 0.7% of the total number of trees planted had died. This is relatively low compared to previous studies that reported tree mortality rates of 17–30% in reclaimed or coastal areas (Park et al., 2003; Hallett et al., 2018). The difference seems to be attributed to variations in the scale and criteria used to define dead trees in each study, as well as the medium or higher salinity resistant soil environment and tree species in the greenspaces in this study. In fact, the mortality rate in this study refers to the percentage of dead trees among trees surveyed in the field, whereas the previous studies estimated the percentage of dead trees out of all planted trees, resulting in different estimation bases for the mortality rate. Moreover, since the measurement periods were different, which may limit the direct comparison of the results. Meanwhile, the main species of dead trees in the studied UGSs included C. retusus, C. officinalis, and C. kousa.
An analysis of the proportion of growth-deficient individuals, categorized by tree part such as crown, stem, branches, and leaves, showed that 3.4% of all trees had low growth vitality. The proportion of growth-deficient individuals by tree part was 2.6% for the crown, 0.3% for the stem, 0.5% for the branches, and 1.4% for the leaves. The species with the highest proportion of growth-deficient individuals were C. retusus, P. thunbergii, M. glyptostroboides, Ulmus parvifolia, and Cornus officinalis, which were similar to the species identified as having the highest mortality. Among these species, all except P. thunbergii were classified as having medium or low salinity tolerance. Most of the P. thunbergii trees with poor growth were concentrated in areas with high salinity and EC. This suggests that even highly salt-tolerant tree species may exhibit poor growth if soil salinity control measures are inadequate.
Carbon reduction effects
Carbon storage by trees (i.e., tree carbon sequestration) per unit area in the study UGSs ranged from a minimum of 3.6 to a maximum of 85.6 t/ha, with an average of 25.7 ± 6.2 t/ha. Carbon uptake ranged from 0.4 to 8.4 t/ha/year, with an average of 2.9 ± 0.6 t/ha/year (Table 6). The carbon reduction effect per unit area varied by up to 24 times across the UGSs, likely due to variations in planting structure. Specifically, the greenspace with the highest carbon uptake per unit area had tree density, cover, and mean DBH that were 14.4, 11.9, and 2.4 times greater, respectively, than the greenspace with the lowest carbon uptake.
Meanwhile, carbon storage per unit area in coastal urban parks in New Zealand was reported to be 45 t/ha, while coastal forests in China showed a range of 26.6–79.8 t/ha (Wang et al., 2013; Hallett et al., 2018). Moreover the annual carbon uptake in urban parks in Seoul and Daejeon was 3.5 t/ha/year and 2.6 t/ha/year, respectively (Jo et al., 2019a; 2023). The carbon reduction effect in the study greenspaces was generally lower than in the above cases, likely due to differences in tree density, cover, size, and vertical structure. In fact, the tree cover and the ratio of multi-layered structure in urban parks in Seoul were higher than those in the study sites.
Soil carbon storage per unit area in the study greenspaces ranged from a minimum of 12.7 to a maximum of 62.1 t/ha, with an average of 33.0 ± 2.4 t/ha. Carbon storage across the greenspaces varied by a factor of up to 4.9 between the greenspaces, likely due to differences in OM content. Notably, the OM content of the site with the highest carbon storage was 1.7 times higher than that of the site with the lowest carbon storage. Previous studies have reported that carbon storage in forest soils typically ranges from 49 to 124 t/ha (Lee, 2011; 2012, Jo and Ahn, 2013; Lee et al., 2015). For urban parks, carbon storage was reported to be 22.4 t/ha in Seoul, 23.4 t/ha in Daegu, 12.8 t/ha in Daejeon (Yoon et al., 2016), 15.6–28.1 t/ha in Jinju (An et al., 2022), and 24.8 t/ha in Chuncheon (Jo and Han, 1999), while for riparian greenspaces, it was 26.4 t/ha (Jo and Park, 2015). In comparison, the carbon storage in the soil of the study greenspaces was lower than that of forest soils but higher than that of urban parks. This difference appears to be primarily attributed to differences in OM content. Specifically, the OM content in the study greenspaces ranged from 2.9% to 6.5%, while in forest soils, it ranged from 6.2% to 6.6%, which was up to 2.8 times higher. On the other hand, the OM content of the riparian greenspaces was 1.4%, which was lower than that of the study greenspaces. The higher OM content in the study greenspaces, compared to the existing riparian greenspaces, is attributed to the recommendation to maintain at least 3% OM content to improve the poor soil conditions caused by salinity (KILA, 1999).
As mentioned above, the main factors affecting carbon storage and uptake by trees, as well as soil carbon storage, included tree density, cover, size, and soil OM content. In this study, to identify the specific causal relationships between the carbon reduction effect and these factors, we analyzed the correlations between the carbon reduction effect and the density, cover, and average DBH of the planted trees, as well as the soil OM content. The results showed that tree density, cover, and DBH were all significantly positively correlated with tree carbon reduction (r ≥ 0.7), while soil OM content also had a significant positive correlation (r ≥ 0.7) with soil carbon storage (p < . 01; Table 7).
Meanwhile, a comparative analysis of the carbon reduction effects between UGSs on the west and south coasts revealed that the carbon storage and uptake per unit area in UGSs on the west coast were 4.3 times and 3.1 times higher, respectively, than those on the south coast (p < . 05; Table 8). As discussed in the greenspace structure analysis above, the higher carbon reduction effect of UGSs on the south coast compared to the west coast appears to be due to the greater tree cover and DBH on the south coast. This is likely the result of differences in planting periods, allowing the trees on the south coast to absorb and store carbon for a longer period of time. As shown in Table 5, tree cover and DBH are positively correlated with the carbon reduction effect, with higher values generally associated with a greater carbon reduction effect. Meanwhile, soil carbon storage did not show statistically significant differences between the coastal regions (p > .05). This seems to be due to the fact that OM content, which is related to soil carbon storage, also showed no differences between the coastal regions.
Based on the estimated carbon storage and uptake per unit area in the study greenspaces, the carbon reduction effects of trees and soil in the UGSs of Korea’s RCAs in Korea were estimated. The estimation was conducted under two scenarios: the first scenario used separate estimates for the west and south coasts, while the second scenario applied the average estimates for both coastal regions. In the first scenario, which applied separate estimates for the west and south coasts, the carbon storage and uptake in the UGSs of Korea’s RCAs were 115 kt and 5. 0 kt/yr, respectively. In the second scenario, which applied the mean estimates for both coastal regions, the carbon storage and uptake were 118 kt and 5.7 kt/yr, respectively, representing 1.02 and 1.14 times higher values compared to the first scenario. The estimated carbon uptake considered only the uptake by trees and did not include the annual soil carbon fluxes. Meanwhile, the annual carbon emissions per person based on the energy consumption of residential buildings in Korea in 2022 were reported to be approximately 0.04 t/yr (Korea Energy Economics Institute, 2023). Compared to the second scenario, which applied the average estimates for the two coastal regions, the current trees planted in the UGSs of Korea’s RCAs play a significant role in offsetting the annual carbon emissions of approximately 153,000 people based on the energy consumption of residential buildings. In the future, the carbon reduction effect of the UGSs in the RCAs is expected to increase when the annual net soil carbon influx is incorporated.
Strategies to enhance carbon reduction effects
Carbon storage by plants is proportional to the biomass of each part, such as stems, branches, leaves, and roots, while carbon uptake is proportional to the annual increase in biomass. This suggests that ensuring the normal growth of plants is a fundamental and essential condition for enhancing their carbon uptake capacity. Since trees planted in RCAs are relatively more vulnerable to salinity damage than those in inland areas, it is crucial to develop strategies that promote normal tree growth and mitigate salinity stress to enhance their carbon reduction effect. Based on the physiological characteristics of plants discussed above and the results of this study, the following strategies are proposed to improve the carbon reduction effect of the UGSs in the RCAs.
First, it is essential to control the salinity of the soil while enhancing its OM content. Although the soil environment in the study greenspaces was relatively favorable compared to previous research, some areas still had high salinity and EC. The proportion of growth-deficient individuals in these greenspaces ranged from 10.7% to 15.0%, which is 3.1 to 4.4 times higher than the overall average of 3.4%. Since poor soil environment is one of the main causes of tree growth inhibition, it is necessary to create a soil environment suitable for plant growth and enhance the carbon reduction effect based on normal tree growth. A representative method is to improve drainage and cation exchange capacity (CEC) by increasing OM content, which helps prevent salt accumulation in the soil (Kim et al., 2011). Increasing OM content not only supports normal plant growth but also contributes to enhanced carbon storage in the soil.
Second, highly salt-tolerant tree species should be prioritized for planting. Species with low salinity tolerance, such as C. officinalis, C. kousa, and M. glyptostroboides, showed poor growth and high mortality rates in the UGS studied (Oh et al., 2010). Even with the creation of salt barriers in the RCAs, soil salinity can still increase due to factors such as onshore winds, groundwater and capillary action, making it difficult for low salt-tolerant species to maintain normal growth. As trees with poor growth will eventually have a reduced carbon reduction capacity, it is desirable to plant highly salt-tolerant species that can maintain normal growth. According to a previous study (Cuiyu et al., 2019), plant biomass decreased by approximately 6–34% due to salt stress, suggesting that the carbon reduction capacity of the plants was consequently reduced by the same proportion.
Third, the carbon reduction effect per unit area should be enhanced by adopting multi-layer planting strategies rather than relying on the existing low-density, single-layer planting. The results of this study indicate that the low carbon uptake capacity of some greenspaces was due to simple-layer structures and low-density planting. Density, cover, and size of planted trees were all positively correlated with carbon reduction effect. This suggests that the greater the density, cover, and size of the planted trees, the greater the carbon reduction effect. To meet the above conditions in a limited planting area, multi-layer planting, in where trees, shrubs, and herbs are planted together, offers distinct advantages over single-layer planting. A previous study (Jo and Park, 2015) found that the tree density and carbon uptake per unit area in multi-layer planting, consisting of upper, middle, and lower tree layers, were 16 trees/100 m2 and 6.9 t/ha/year, respectively. These figures are 2.7 times higher for tree density and 2.4 times higher for carbon uptake compared to the greenspaces in this study, indicating that multi-layer planting can increase the carbon uptake of greenspaces, even within the same spatial area.
Conclusion
Coastal reclamation projects have been promoted as a means of providing new land in land-scarce regions around the world. In addition to infrastructure such as buildings and roads, RCAs include greenspaces for recreational use by local residents. However, trees planted in reclaimed coastal areas (RCAs) are vulnerable to damage from salinity and onshore winds, resulting in higher mortality rates than inland trees. Recently, urban greenspaces (UGSs) have been considered as a major carbon sink for carbon neutrality, in addition to their recreational function for local residents. However, there is a lack of research on the carbon reduction effect of UGS in RCAs worldwide. Therefore, this study conducted a field survey of UGSs in RCAs in Korea to determine their planting structure and carbon reduction effect.
The tree cover in the study RCAs was approximately 37.2%, with a higher frequency of salt-resistant species compared to inland areas. The soil organic matter (OM) content was approximately 4.2%, which is relatively good, but some areas still exhibited high salinity and EC, which hindered tree growth. The carbon storage per unit area by trees and soil in the study greenspaces was 59.2 t/ha, and the annual carbon uptake by trees was 2.9 t/ha/yr. The carbon reduction effect of trees and soil varied by a factor of up to 24 across the study sites, and had a significant positive correlation with tree density, cover, and size.
A comparative analysis of the structure of the greenspaces and their carbon reduction effects between the coastal regions showed that the greenspaces on the south coast generally had greater tree cover, size, and carbon storage and uptake than those on the west coast (p < . 05). This difference is likely due to the fact that the greenspaces on the south coast were established earlier, allowing the trees to grow for a longer period of time. Considering both the south and west coasts, the annual carbon uptake of UGSs in Korea’s RCAs was approximately 5.7 kt/yr, which is equivalent to the carbon emissions of about 153,000 people based on the energy consumption of residential buildings.
Based on the above findings, strategies were proposed to enhance the carbon reduction effect of UGSs in the RCAs. These include controlling soil salinity, increasing soil OM content, planting highly salt-tolerant tree species, and implementing multi-layer planting with medium and large trees. This study is significant in that it provides an index of carbon reduction by trees and soil in UGSs in the RCAs, for which there is little related data. However, this study has some limitations in that it does not consider other types of greenspaces, such as street trees and gardens, nor does it consider annual soil carbon fluxes. It is necessary to conduct further research that reflects the annual soil carbon fluxes by incorporating other types of greenspaces and expanding the sample size in the future.
Supplementary Information
Notes
This study was carried out with the support of the ‘R&D Program (Project No. C5202319792)’ provided by K-water. This study reflects part of Oh's master's thesis (2024).