INTRODUCTION

Urban parks have various functions ranging from improving the health of people, lowering temperature, storing and absorbing carbon, and contributing to the conservation of biodiversity (Ki et al., 2012; Han et al., 2014a; Han et al., 2014b; Kim, 2014). Until now, urban parks were mostly composed of installations for play and exercise for the leisure and rest of residents, and had simple plant and tree planting with the goal of providing green areas. However, in order to improve the quality of urban environments, discussions have recently become very active on quantitatively increasing urban parks, as well as the methods for the qualitative improvement of planted trees. Recently in Korea, efforts to establish ecological forests from a new perspective have been active for the qualitative improvement of green areas within urban parks (Heo et al., 2010). Lee et al. (2011) have recommended switching landscaping trees, usually comprised of foreign tree species, to be switched to natural wild tree species that can maximize the regional characteristics and Han et al. (2014b) have proposed multi-layer structures and a high tree planting concentration instead of using constant and repetitive tree planting.

From a management perspective of urban park greens, Song (2014) developed a quantitative evaluation index of tree vitality for planted trees in order to assess urban parks that were established in the past, and there were also studies made to review the appropriateness of tree planting concepts by spatial functions and to analyze the vegetation structure statuses (Han et al., 2014b). There are also studies conducted on the problems on ecological planting in urban parks, as well as improvement measures for them (Seong et al., 2011; Kim et al., 2012b; Oh, 2010).

However, studies up until now mostly dealt with existing methodologies, and there are only a few studies on areas established from a new perspective. There are a growing number of areas that have recently introduced new planting methods such as multi-layer forests in building green areas in urban parks, resulting in the current necessary to have appropriate methods for newly introduced tree planting methods. In particular among urban parks in Korea, the case of Seoul Forest Park, the creation of a large-scale multi-layer forest modeled after a natural forest, the limitations with managing park trees began to appear and it has become necessary to manage planted trees using a new paradigm.

Seoul Forest Park was built with the intent of strengthening functions as an urban forest, and it is different from ordinary park landscape tree planting methods. In particular, in the case of the Eco Forest regions and the Citizen’s Forest, taking into consideration the forest layer structure that can be seen in normal natural forests, they aimed at showing the appearance of a natural forest by planting trees, aborescents and shrubs centering on the trees of the Pinaceae and Quercus families. But most research cases were conducted on trees in parks made with existing tree planting methodologies, so it is difficult to use this for managing the multi-layer planted trees of Seoul Forest Park. The purpose of this study is to identify the growth status of trees based on long-term monitoring results for planted trees with the goal of constructing a forest within an urban park to provide preliminary data not only for the vegetation management of Seoul Forest Park, but also for the management of established forests through multi-layered planting in urban areas in the future.

MATERIALS AND METHODS

1.Research Target

1)Research Target

Seoul Forest Park is an area covering the Ttukseom region that was used as a water source park, golf course, horse race track, and sports park at an area where the Jungnang River and Han River of Seoul meets. The establishment of the Seoul Forest Park was made based on design plans selected through design contests, and after the design contest was completed, it was officially decided to name the park, the Seoul Forest on April 4, 2003. Afterwards, through a preparation period, construction of the park began in April 2004 and it opened on June 18, 2005. The core content of the design contest guide were, the “Consideration of regional features that are near the Han River and Jungnang River,” “Design in the concept of an environment-friendly large-scale forest,” “Forest focusing more on tree planting than facilities,” etc. The key texts of the design guide such as ‘large-scale forest’, ‘forest focusing on tree planting’, and ‘natural arrangement’ make clear that the intent was to establish as an urban forest. This is well reflected in the design plans and including the selected design of Dongshimwon, as well as most of the other awarded designs, they focus on the establishment of forests (Lee and Cho, 2004).

Likewise, Seoul Forest Park, which was established by placing important value on the establishment of the forest, was established in a different direction from the tree species and planting methods used in other urban parks. In particular, the Eco Forest and Citizen’s Forest area of Seoul Forest Park reflected a tree-planting concept that recreates the vegetation structure and vegetation succession that occur in natural forests. This study selected the Eco Forest and Citizen’s Forest, which well reflected such concepts in Seoul Forest Park, as the research target (Figure 1).

2.Survey and Analysis Method

1)Planted Tree Growth Monitoring

In order to monitor the growth status of the target area, the first survey was carried out in March 2005, right before the opening of the Seoul Forest Park, followed by a second survey in May 2008 and a third survey in May 2013. In the first survey, all trees with a diameter at chest height of 2cm or greater higher were examined excluding shrubs among the planted trees in the Eco Forest and Citizen’s Forest areas of Seoul Forest Park.. The survey was conducted by attaching serial number labels on the target trees and examining the tree species, tree height, diameter at breast height, timber height, and special features. In the second and third monitoring sessions, excluding the Eco Forest, which is inaccessible as it is used as a deer grazing area, and the area surrounding the Eco Forest, surveys were conducted on all trees remaining with labels attached in the first survey within the Citizen’s Forest area.

2)Growth Analysis of Planted Trees

In this study, instead of the method of installing fixed quadrates commonly used for tree growth analysis in other studies to accurately analyze the growth features of planted trees, growth was analyzed by assigning serial numbers to all trees in the research target area for monitoring. Growth was found through a change of diameter at breast height for each species over nine years, and growth features were studied by comparing with the existing research results of the same species. The existing research results used include the research data of natural forest areas with the same species and when the same tree species were planted as those of urban landscape trees. However, because the annual growth of trees was high in the beginning of growth and became lower later, the research data of trees with similar ages were collected, and only the parts required from existing research data were reanalyzed and utilized.

3)Growth Analysis by Region

The Eco Forest and Citizen’s Forest areas of Seoul Forest Park were artificially constructed based on a park design concept, and therefore, it is judged that numerous factors including construction methods, quality of soil used, planting patterns, management after planting, etc would impact the growth features. Such factors can form various growth features by region in the Citizen’s Forest as well, and therefore, this study aimed at seeking environmental factors that made an impact through planted tree growth status analysis by region. To accomplish this, an accurate map of planted trees was made and used for analysis. The map was drafted by installing a tape ruler in the form of 5 m by 5 m plots throughout the entire target area, and a GPS receiver with a ±1m margin of error was used to identify planting location. The ArcGIS 9.1 program was then used to draft the map of the location of trees. Based on this map, the total growth of each tree was indicated and the growth deviation by region was checked. Because the growth features are different for each tree species, it may be erroneous to make direct comparisons using the total growth of trees, thus, it was used after rating based on the total growth per tree species. The rating criteria was used by setting the rating criteria of a relative evaluation format based on the number of trees considering that it has a relative value evaluation among the same trees, and from the highest-most, it was distinguished from grades 1 to 5 in 20 % intervals. Areas with many trees with low grades were selected and field investigations were conducted to study the causes.

4)Soil Health Analysis

Soil is the physical support of plants and is a main supplier of moisture and nutrients, and therefore a very important factor in analyzing the growth environment of trees. The research target area is located near a large highway and is an urban park with many users. Therefore, it is expected that there would be a high quantity of various types of contaminants. Thus, as it is judged that such features will have a huge effect on the growth of trees, soil analysis was conducted focusing on the health of the soil. In order to analyze the health of the soil, physical and chemical analysis were conducted such as soil moisture content, soil organic substance content, and soil pH. In particular, based on the fact that there are different extracellular activities of organisms according to the contamination of the soil, the extracellular enzyme activity analysis of microorganisms was carried out to examine soil health (Sparling, 1997). In order to identify the level of the status of Seoul Forest Park, it was compared with other existing research results.

The collection of soil specimens was carried out in five areas considering the environment per vegetation type focusing on the Citizen’s Forest, such as the Quercus acutissima dominant area, the Quercus mongolica dominant area, the Pinus densiflora dominant area, the Pueraria lobata suppressed area, and the herbal area for analysis. The measurement of soil moisture content was made by heating the collecting soil for 24 hours at 103 °C in the lab and drying to calculate using the difference in weight. Soil organic substance content was calculated by heating the soil that was measured for moisture content for 24 hours at 600 °C and using the difference in weight after heating. The pH was measured by mixing soil and water at a rate of 1:2 and measured the supernatant using a pH measuring device (Orion, Model 250A, Berkeley, USA).

Microorganism activation was measured using the MUF-Substrate. In particular, in order to examine the activation of microorganisms regarding oxygen and nitrogen decomposition, the activation of β-glucosidase and N-acetly-glucosaminidase (NAG) were measured. ※

RESULTS AND DISCUSSION

1.Planted Tree Survey Results

For nine years from 2005 to 2013, the trees of the Eco Forest and Citizen’s Forest areas of the Seoul Forest Park were monitored on three separate occasions. The results of tree monitoring are as shown in Figure 2 and Table 1. After the Seoul Forest Park opened, the Eco Forest and its vicinity was used as a deer grazing area, and thus surveys could not be conducted as access was prohibited. Excluding trees that could not be monitored due to damaged labels, studies were conducted focusing on the Citizen’s Forest starting from the second survey.

In the first survey, 4,478 trees were selected for monitoring and surveys were conducted. According to the first survey results, commonly planted trees were Pinus densiflora (687 trees), Prunus yedoensis (672 trees), Quercus acutissima (432 trees), Quercus mongolica (304 trees), Acer palmatum (275 trees), Elaeagnus umbellata (212 trees), Acer buergerianum (194 trees), and Quercus serrata (185 trees). Numerous Pinus densiflora were used from medium to large treees to establish a Pinus densiflora multi-layered forest in the southwestern part of the Eco Park, and in particular, a broad-leaf forest using Quercus trees such as Quercus acutissima, Quercus mongolica, Quercus serrata, etc, which are not commonly planted in other urban parks was established. Quercus trees accounted for 24% of the total number of trees surveyed.

The second survey, conducted three years later, only targeted trees in the Citizen’s Park area, and it was conducted on a total of 1,520 trees. The most planted tree in the Citizen’s Park area was Prunus yedoensis, and a Pinus densiflora multi-layered forest of medium size or higher was established. Furthermore, focusing mainly on Quercus acutissima, Quercus mongolica, and Quercus serrata, a variety of tree species such as Quercus variabilis, Quercus aliena, and Quercus dentata were planted to establish a Quercus forest with a variety of tree species. In the third survey, five years after the second survey, trees, excluding those with damaged labels were investigated. There were a total of 370 trees surveyed. Most of the planted tree species were included and the surveyed trees were evenly distributed throughout the Citizen’s Forest, and thus it was judged that it was possible to analyze growth for nine years and to analyze the growth features by region.

2.Results of Growth Analysis by Tree Species

Growth was analyzed for 370 trees for which the data on the change in tree height and diameter at breast height for nine years (2005-2013) could be checked, and in order to examine the meaning of the change in growth, it was compared with existing research results in which the same tree species were analyzed for growth in natural forests or when planted as landscape trees (Table 2). In order to analyze growth, the annual mean growth per tree species and the mean total growth per tree species were deduced based on the change of diameter at breast height for use. The growth speed of trees differed by period from young stands and old stands, and in the case of diameter at breast height, it had a close relationship with planted density, and therefore, absolute comparisons with existing research results were impossible. Accordingly, as most of the trees being analyzed were between 20 and 30 years old, it was examined by comparing with existing studies taking into consideration the fact that growth would be relatively high. Furthermore, the total growth distribution status of each tree was taken into account to interpret the results (Figure 3).

Upon analyzing 71 Quercus acutissima, the diameter at breast height grew at an average of 3.2 cm, and grew at an annual mean of 3.6 mm. This is very high growth when compared to Quercus acutissima being planted as apartment landscaping trees, and it is also high growth compared to Quercus acutissima in natural forests. There were only a few trees with the total growth of Quercus acutissima being extremely high or low, so the average growth used has high reliability. When comparing with existing research results, even when taking into account the fact that Quercus acutissima in the Citizen’s Forest was in a period of high growth, it is still judged to display good growth.

Upon analyzing 54 Acer buergerianum, the diameter at breast height grew at an average of 3.8 cm, and grew at an annual mean of 4.3 mm. While it is impossible to compare as there are no existing research results on Acer buergerianum, when compared with other species planted in the target area, it displayed good growth. However, when examining the total growth of each tree, as there are many trees with higher or lower growth than average total growth, it is judged that there is a large difference in growth depending on the growth environment.

Upon analyzing 47 Pinus densiflora, the diameter at breast height grew at an average of 2.6 cm, and grew at an annual mean of 2.9 mm. It showed superior growth compared to the Pinus densiflora planted as landscape trees in other parks, and it is judged that it will be similar to Pinus densiflora in natural forests. Such results make it possible to judge Pinus densiflora as growing well in the target area, even when taking into consideration the fact that a few Pinus densiflora with extremely high growth raised the average growth.

Upon analyzing 36 Quercus mongolica, the diameter at breast height grew at an average of 2.7 cm, and grew at an annual mean of 3.1 mm. When compared with Quercus mongolica of natural forests, it was found to be similar, or slightly higher. When examining the total growth of each tree, there were few extremely high or low trees, so the reliability on average growth is high. It is judged that Quercus mongolica grows well in the target area.

Upon analyzing 34 Prunus yedoensis, the diameter at breast height grew at an average of 4.0 cm, and grew at an annual mean of 4.4 mm. The growth is higher than Prunus yedoensis research results of natural forests, but lower than cases in which it was used as landscape trees in parks. When taking into consideration the fact that Prunus yedoensis is a rapid growth tree and that the diameter at breast height grows quickly when planted independently, it is judged that compared to when planted as independent trees in other urban parks, multi-layered planting modeled after forests can lower the growth due to the effects of dense planting. In particular, when examining total growth among individual trees, it varies widely despite being the same tree species, and it is thus judged that there will be a large difference in growth depending on the conditions of the surrounding environment.

Upon analyzing 34 Pinus strobus, the diameter at breast height grew at an average of 2.4 cm, and grew at an annual mean of 2.7 mm. It shows high growth when compared to the research results of those planted in apartment complexes. However, most Pinus strobus planted in apartment complexes are planted densely, so they tend to have poor growth. In the case of Pinus strobus in the research target area, it is judged to be the result of sufficient growth space being available.

Upon analyzing 22 Quercus serrata, the diameter at breast height grew at an average of 3.2 cm, and grew at an annual mean of 3.6 mm. Compared to the research cases of trees planted in urban parks, it has high growth and it displays similar growth with natural forests. However, the total growth distribution status of individual trees varies, so it is judged that there is a big difference in growth depending on environmental conditions.

Upon analyzing 21 Carpinus laxiflora, the diameter at breast height grew at an average of 1.2 cm, and grew at an annual mean of 1.4 mm. This appears to be similar to the growth of the Carpinus laxiflora of natural forests. The total growth of individual trees was found to vary, so it is judged that there is a high difference in growth depending on the surrounding environment.

3.Results of Growth Analysis by Region

There are many results with clear differences in growth even for the same tree species. Therefore, it is judged that the growth environment by region within the research target area would have affected the growth of trees. Accordingly, based on the tree location map based on spatial information, growth by region was analyzed, and areas having problems with growth were found to check on site. Since growth features by tree species are different, there can be errors in making direct comparisons using total growth, and thus, rated results using total growth per tree species were used (Figure 4). Upon analyzing growth by regions, it is evident that there are regions with clear differences in growth. Areas in which trees with relatively low growth of grades 4 and 5 are judged to be the areas A, B and C.

Results from the field surveys of A area showed that Pinus densiflora, Quercus acutissima, Quercus serrata, and Acer buergerianum were planted. It is judged that this area has a very high planting density compared to other areas. The multi-layered forest was made by establishing a mixed forest of soft and hardwood with Pinus densiflora and Quercus acutissima in the tree statum, and in the aborescents layer, Acer buergerianum was planted in large scale. However, Acer buergerianum was planted too excessively, resulting in a higher overall planting density. As a result, the branches of most trees overlap and excluding the upper part of the tree crowns, most branches have poor growth, and many dead branches could also be seen. Accordingly, it is judged that there is a drop in growth due to dense planting in area A.

The results from the field surveys of B area showed that Zelkova serrata, Quercus acutissima, and Quercus mongolica were planted, and this area is made up of a part where the soil slope is soft and there is a flat area towards the bottom. This area has many traces of water pooling and water ditches, and taking into consideration, unlike other surrounding areas, marshy herbs are dominant, it is judged to be an area with poor drainage. If there is excessive moisture in the soil due to drainage problems, the young branches of trees can die and it can also cause a drop in vitality due to irregular breathing from the roots. In most trees in this area, the dried leaves of trees, the death of young branches in the lower parts, and increased number of dead branches in the upper parts were spotted, and it is judged that there was a drop in growth due to problems with excessively moist soil.

Area C is where Quercus acutissima are planted and Pueraria lobata is flourishing. Most young trees died and in the case of adult trees, it was found that many were damaged by Pueraria lobata. Pueraria lobata has very high growth rate and it disrupts tree growth and can easily cause the death of trees. It is thus judged that a drop in growth was caused by the suppression of Pueraria lobata.

4.Results of Soil Health Analysis

1.Physical and Chemical Features of Soil

Soil organic substance content is highly related to the fertility of soil, and the higher it is, the more fertile the soil. Upon analyzing the physical and chemical properties of the soil, the soil organic substance was analyzed to be between 2 and 26 % depending on where the specimens were collected (Figure 5a). Study results by comparing this with past studies are as shown in Figure 6. Upon analyzing the organic substance content of soil in street trees around Seoul and Hongneung, it was found that they had 1-7 % soil organic substances. Considering that the standard for organic substance for planting landscaping trees by Craul (1999) is 1-5 %, it was found that the soil of Seoul Forest Park satisfied the standards for planting landscape trees. However, referring to the report by Jang and Kang (2010), which said that the soil organic substance was 18-23 % when measured in Gangwon-do at Jeongbongsan, which has the best developed forest in Korea, it is judged that soil fertility is much lower than in natural forests.

In the case of soil pH, soil in most of the target areas was close to 7, and some soil showed high values at 7.3 (figure 5b). Jang & Kang (2010) measured the pH of the soil at Mt. Jumbongsan for two years and reported acidic soil of 4.8-5.5. Furthermore, many studies commonly report that forest soil manifests acidity due to the precipitation of acidic substances from the atmosphere and the effects of plant exudates. However, according to results from analyzing street tree soil in the city, the pH was found to be high at 5.7-7.8. There are many cases in which soil in the city display higher alkalinity compared to other soils because of various contaminants such as deicing materials used in the winter and the corrosion of concrete structures. When comparing with existing studies, the soil of Seoul Forest Park has similar alkalinity with urban street tree soil, thus being highly different with the soil pH of natural forests.

2.Results of Soil Microorganism Activity Analysis

The results of the soil microorganism activity analysis are as shown in Figure 7. β-glucosidase, which shows the activity of microorganisms that break down carbon from 0.55 to 11.55 nmol g^-1 dry soil min-1. N-acetyl-glucosaminidase, which shows the activity of microorganisms involved with decomposing nitrogen that showed values of 1.06 - 12.10 nmol g^-1 dry soil min-1.

Microorganism extracellular activity is an index that shows the quantity and activity of microorganisms within the soil. According to Sparling (1997), because there is a difference in extracellular activity according to the pollution levels of the soil, it is an indirect index that shows the health of soils. When examining microorganism activity measured at Mt. Jumbongsan in Gangwon-do using the same method, β-glucosidase was found to be 16.20 - 84.49 nmol g^-1 dry soil min-1, and NAG to be 3.25 - 36.37 nmol g^-1 dry soil min^-1. When compared to natural forests, it is evident that the microorganism activity of the target area is very low. In an experiment conducted with the same method for soil for street trees in downtown Seoul, β-glucosidase was 1.63 - 4.87 nmol g^-1 dry soil min^-1, showing that it is in a similar range with that of the Seoul Forest Park.

When examining the correlation analysis between microorganism activity and environmental factors, both b-glucosidase and NAG had a positive correlation (P=0.005, r=0.679, n=15 for β-glucosidase, and P=0.002, r=0.725, n=15 for NAG) with soil organic substance contents (Figure 8). Through this, it was judged that the microorganism activity of soil in Seoul Forest Park was mostly affected by organic substance content.Fig .9

5.General Discussion

It was found that the growth of most trees was higher than the landscape trees in urban parks, and also showed similar or higher growth than that of natural forests. Even when considering the fact that most of the trees in the research target area are young stands, in this study it cannot be said that the growth of trees in the target area is lower than that of natural forests, and the results show that they are at similar levels. This shows that when using planting methods similar to multi-layered forests modeled after natural forests when establishing urban parks, trees can grow well. However, the difference in growth between regions was checked through growth analysis by region, and the cause for this was analyzed. As a result, problems such as dense planting, soil issues, and Pueraria lobata suppression were found. It is believed that areas with high planting, which is not seen in natural forests, will require tending such as the thinning or removal of dead branches. Areas that had problems with excessively moist soil will require the installation of drains and soil enhancement, while areas with Pueraria lobata suppression requires Pueraria lobata control. These operations are used as the preliminary work of forest management, but they are not commonly conducted for the management of urban parks. However, as the research target area was planted modeled after a forest, it is necessary to go beyond the frame of vegetation management of urban parks, and to take an approach from a forest management perspective.

Upon analyzing the soil, when comparing it with the soil of natural forests, the Citizen’s Forest area has a very low organic substance content, and it was also found that the alkalization of the soil was in progress. The results of the microorganism activity analysis, which can show the health of the soil, was also much lower than that of natural forests, thus displaying very poor health at levels similar to urban street tree soil. In the long run, these will be factors that interfere with the growth of trees, and thus it is necessary to find the exact reasons and manage them. Therefore, it is essential to conduct ongoing monitoring and detailed soil surveys to resolve these problems.