Korean white pine (Pinus koraiensis Siebold & Zucc.) is a major economic coniferous tree species used for timber and seed production; it accounts for 151,946 ha in Korea, which is 7.3% of the total coniferous forest area of 2,319,832 ha; most individuals of this species grow in plantation (Korea Forest Research Institute 2012b; Korea Forest Service 2021a). In addition, the management of the Korean white pine plantation is urgently required, as it accounts for an average annual planting area of 327 ha for the past five years (Korea Forest Service 2021a). Such plantation refers to a forest cultivated by artificial and intensive forestation process, as a stand created by planned planting or direct seeding (Lee et al. 2010); in particular, it is widely used as a timber-producing forest among functions of the forest, and has contributed to wood production, based on the targeted products such as large-, medium-, small-diameter trees (Korea Forest Research Institute 2005).

Meanwhile, the natural deciduous forest in Korea occupies a large area of 3,664,691 ha (61.2%), which is the sum of deciduous forests (2,002,150 ha, 33.5%), and mixed forests (1,662,541 ha, 27.8%), out of the total stocked forest area of 5,984,523 ha (Korea Forest Service 2021a). In particular, among the total number of 445,264 deciduous tree species measured by the National Forest Inventory, oak was found to show a high frequency of occurrence, a total of 222,345 observations (31.5%), including Quercus mongolica Fisch. ex Ledeb. (105,946 observations, 15.3%), Q. variabilis Blume (61,420 observations, 8.8%), Q. serrata Murray (35,195 observations, 5.1%), and other Quercus spp. (19,784 observations, 2.3%) (Korea Forestry Promotion Institute 2017). Such natural deciduous forests were treated as stands with useless trees and were neglected without much attention since 1980s. The importance has been recently emphasized in the environmental, ecological, and public value evaluation of forests (Chung et al. 2018; Korea Forest Research Institute 2012a; Lee et al. 2018b; Noh et al. 2020; Park et al. 2020).

In the silvicultural operation, thinning is a representative method of tending operation, and in Korea, it is classified as pre-commercial thinning or commercial thinning in plantation and natural forest improvement or natural forest tending to natural forests (Korea Forest Research Institute 2005; Korea Forest Service 2021a). In general, thinning is known to promote the diameter at breast height (DBH) growth of residual trees and improve stand health by preventing natural mortality (Choi et al. 2014a). According to the Statistical Yearbook of Forestry, the type of thinning classified as mature forest management was performed on an average annual forest area of 78,446 ha for the past five years (Korea Forest Service 2021a). Furthermore, the Korea Forest Service set a budget equivalent to South Korean won (KRW) 188.4 billion, which is 7.5% of the total annual expenditure budget of KRW 2,528.2 billion in 2021, and the largest budget has been continuously allocated to this single project of the Korea Forest Service (Korea Forest Service 2021b). Despite the large area for thinning, and expenditure budget, the quantitative effect on thinning has not yet been sufficiently and academically proven, and the model or silvicultural operation guidelines for dynamic growth are still insufficient.

Silvicultural guidelines by such thinning should be conducted based on scientific and empirical research results through monitoring data on experiment forests. In advanced countries in forestry, various silvicultural treatments such as thinning from below, crown thinning, and row thinning are carried out in order to estimate changes in stand structure and growth of standing trees based on thinning intensity; there have been active long-term monitoring studies through constantly repeated measurement in the field by installing permanent plots (Amateis 2000; Amateis and Burkhart 2005; Cao et al. 2000; Huuskonen and Hynynen 2006; Knoebel et al. 1986; Mäkinen and Isomäki 2004a; Mäkinen and Isomäki 2004b; Nishizono et al. 2008; Pelletier and Pitt 2008; Pfister et al. 2007; Wallentin and Nilsson 2011).

However, in Korea, the growth investigation through monitoring studies based on thinning intensity of the Korean white pine plantation has been partially reported by Bae et al. (2010), Choi et al. (2014a), and the National Institute of Forest Science (2021) as quantitative research results, but it is still insufficient. Regarding natural deciduous forest studies, there is academic research on vegetation distribution, cluster classification, successional tendency, and ecological status assessment (Chung et al. 2014; Hwang et al. 2012; Hwang et al. 2016; Kim and Kim 2001; Um and Lee 2006); there have been some studies on stem analysis using felled trees and repeated measurement of plots to analyze the growth (Choi et al. 2007; Choi and Yoo 2006; Lee et al. 2000a; Park et al. 1996). However, long-term monitoring studies of natural deciduous forests in which oaks (Quercus spp.) are dominant species, and studies on natural forest tending are highly insufficient.

In this sense, this study was carried out to provide the basic information for developing the silvicultural guideline for the Korean white pine plantation and oak-dominated natural deciduous forest, by conducting a comparative analysis of forest characteristics, in terms of growth of standing trees and stand structure based on thinning intensity. In particular, the growth patterns of DBH and volume were analyzed using the repeatedly measured monitoring data from the permanent plots. Furthermore, by comparing the status of mortality and ingrowth trees, we attempted to figure out the impacts of thinning on changes in each stand structure and growth.

Study site

The study site is located in the research forest of College of Forest and Environmental Sciences, Kangwon National University in Wonchang-ri, Dongsan-myeon, Chuncheon-si, Gangwon Province, Republic of Korea (Fig. 1). It has a continental temperate climate, and is hot and humid in summer and cold and dry in winter. According to the statistics of the average annual temperature in the Chuncheon area for the last three decades, from 1991 to 2020, revealed by the Korea Meteorological Administration, the average temperature was 11.4°C, the highest of 17.4°C, the lowest of 6.3°C, and the precipitation of 1,341.5 mm (Korea Meteorological Administration 2021). Most of the soil consists of loam and sandy loam, along with rich organic matters and suitability for plant growth due to a mixture of decomposed fallen leaves and minerals (Choi 2003).

Figure 1. Map of the study area. (A)The location of the research forest of Kangwon National University, Chuncheon-si, Gangwon-do, South Korea. (B) The location of the experimental sites conducted within the research forest for the Korean white pine plantation and oak-dominated natural deciduous forest.

As for the status of the research forest (3,140 ha in total) of College of Forest and Environmental Sciences, Kangwon National University, the plantation of 750 ha mostly consists of coniferous forest, mainly the Korean white pine (521 ha, 69%) and the Japanese larch (Larix kaempferi) (164 ha, 22%). The most natural forest of 2,389 ha is deciduous, along with the forest area (2,358 ha, 99%) with such dominant tree species as Quercus spp. including Q. mongolica Fisch. ex Ledeb., Q. variabilis Blume, Q. dentata Thunb., and Q. aliena Blume (Kangwon National University 2020).

Experiment design and thinning treatment

The thinning experiments were designed to study quality improvement of the standing trees and production of large-diameter trees, and installed with four plots of different thinning intensities (i.e., heavy thinning treatment, moderate thinning treatment, light thinning treatment, and unthinned treatment), in line with each stand condition (Table 1). Two thinning experimental sites of each stand type were established as replicates. As for the first experimental site with the Korean white pine plantation, four plots of 20 m × 20 m were installed in 1996 (age 19) in two adjacent areas (Choi et al. 1996b); as for the second experimental site with the Korean white pine plantation, four plots of 30 m × 30 m were installed in 1999 (age 21) in one area on the same contour (Choi et al. 1999). Regarding the first and second experimental sites with the oak-dominated natural deciduous forest, four adjacent plots of 20 m × 20 m were installed in 1996, in each area; site 1 with the natural deciduous forest was dominated by Q. mongolica Fisch. ex Ledeb., whereas Q. variabilis Blume was the most dominant in the site 2 (Choi et al. 1996a).

Table 1 . Summary of experimental site characteristics for Korean white pine plantation and oak-dominated natural deciduous forest.

Stand type	Major species	Experimental site	Treatment area (m2)	No. of plots	Plot area (m2)	Year of plot establishment	Age at treatment start*	Year of last measurement	No. of measurements	Site index† (m)
Korean white pine planation	Pinus koraiensis	Site 1	3,200	8	400	1996	19	2014	4	14
		Site 2	3,600	4	900	1999	21	2016	3	16
Oak-dominated natural deciduous forest	Quercus mongolica and Q. variabilis	Site 1	1,600	4	400	1996	23 (15–28)	2017	4	-
		Site 2	1,600	4	400	1996	42 (35–49)	2017	4	-

*Age of natural deciduous forest is mean with min–max in parenthesis. †Site index of Korean white pine planation was calculated using base age and dominant height according to the site index table provided by the Korea Forest Research Institute (2012c).

The thinning was carried out only one time during the whole measurement period and the thinning was performed immediately after the plot installation (site 1 of pine stand in 1996, site 2 of pine stand in 1999, and site 1 and 2 of oak stand in 1996). The thinning from below was performed with diverse intensities, while considering the qualitative and quantitative factors to control the growth space for the stand, and induce the production of high-quality large-diameter trees (Choi et al. 1996a; Choi et al. 1996b; Choi et al. 1999). Based on the number of trees, the thinning rate of the Korean white pine plantation was between 42% and 62% in the experimental site 1 and between 18% and 61% in the experimental site 2; the thinning rate of the natural deciduous forest was between 64% and 76% in the experimental site 1, and between 64% and 67% in the experimental site 2 (Table 2). The thinning quotient, the quotient between the mean diameter of removed trees and the mean diameter of all trees before thinning, was varied by thinning intensity and it indicated the low thinning (thinning from below) as the calculated values were less than 1.0. The site index of the Korean white pine plantation was 14 m in the site 1, and 16 m in the site 2; the natural deciduous forest was confirmed to have diverse ages of trees during the first measurement, indicating that it was impossible to calculate the standard age and dominant tree height, and then the forests were excluded in the calculation of site index (Table 1).

Table 2 . Thinning intensity and number of stems before and after thinning in two stand types.

Stand type (major species)	Thinning experimental site	Plot	No. of stems before thinning (trees/ha)	Thinning intensity (%)		No. of stems after thinning (trees/ha)	Thinning quotient
				Stem number basis	Basal area basis
Korean white pine plantation (Pinus koraiensis)	Site 1	A	2,874	62	44	1,090	0.85
		B	2,933	50	32	1,468	0.80
		C	3,134	42	25	1,823	0.77
		D	2,891	0	0	2,891	0
	Site 2	A	1,595	61	48	618	0.88
		B	1,879	43	23	1,063	0.73
		C	1,683	18	8	1,471	0.66
		D	1,787	0	0	1,787	0
Oak-dominated natural deciduous forest (Quercus mongolica and Q. variabilis)	Site 1	A	3,722	76	63	876	0.91
		B	3,699	67	57	1,224	0.92
		C	4,321	64	42	1,571	0.81
		D	4,272	0	0	4,272	0
	Site 2	A	1,304	67	55	435	0.89
		B	1,515	62	53	572	0.87
		C	1,925	64	42	700	0.78
		D	2,593	0	0	2,593	0

For the Korean white pine plantation, thinning was undertaken in 1996 at age 19 for site 1 and in 1999 at age 21 for site 2. For the oak-dominated natural deciduous forest, thinning was undertaken in 1996 for site 1 and 2. The thinning quotient is the quotient between the mean diameter of removed trees and the mean diameter of all trees before thinning. The number of trees was calculated using the whole plot size (e.g., 20 m × 20 m or 30 m × 30 m) by plot of each experimental site.

Inventory method and measurements

The field investigation analyzed in this study was conducted 3 to 4 times, and the inventoried year and measurement intervals were different for each site. In specific, there were four measurements recorded in the site 1 of the Korean white pine plantation: 1996 (age 19), 2001 (age 24), 2008 (age 31), and 2014 (age 37). There were three measurements recorded in the site 2 of the Korean white pine plantation: 1999 (age 21), 2010 (age 32), and 2016 (age 38) (Choi et al. 1996b; Choi et al. 1999; Choi et al. 2001a; Choi et al. 2010b; Choi et al. 2014b; Choi et al. 2016). There were four measurements recorded in the site 1 and 2 of the oak-dominated natural deciduous forest: 1996, 2001, 2008, and 2017 (Choi et al. 1996a; Choi et al. 2001b; Lee and Choi 2018a).

As these experimental sites aim to analyze the growth monitoring per main factor in line with thinning intensity after thinning, we targeted all residual trees which were higher than 1.2 m in tree height, and investigated tree species, DBH, tree height, location of standing trees, and status of mortality (Choi et al. 2010a; Choi et al. 2010b; Choi et al. 2014b; Choi et al. 2015). The location of standing trees was measured in meters (m) by taking one vertex of a plot as the origin and measuring the distances in the X and Y directions. We gave identifying numbers to all measured standing trees within the field and recorded them on the field note. We measured ingrowth trees above 3 cm DBH which had grown since the second inventory. Additionally, a tree was counted as multiple stems and measured separately if the stems were forked below DBH at 1.2 height above the ground (Burkhart et al. 2019). The current status of the permanent plots is shown in Figure 2.

Figure 2. Photos of the Korean white pine plantation and oak-dominated natural deciduous forest in site 1 by plot, January 2022. The plots were thinned with different intensities at the first measurement period: Plot A with heavy thinning treatment, Plot B with moderate thinning treatment, Plot C with light thinning treatment, and Plot D with unthinned treatment. Specific thinning intensities are provided in Table 2.

Forest growth monitoring analysis

The samples targeted in the analysis are basically all the standing trees in the experiment, but the first experiment with the pine plantation, and the first and second experiments with the oak-dominated natural deciduous forest were initially installed adjacent to each other without a buffer zone between thinned plots (Choi et al. 1996a; Choi et al. 1996b). Accordingly, a buffer zone of 2.0 m, that is, up, down, left, and right, was set from the boundary of each plot, in consideration of the number of standing sample trees for the analysis, and then, analysis was performed only for the trees within the area of 16 m × 16 m. As for the second experiment with the Korean white pine plantation, since there was a buffer zone between the thinned plots, we targeted and analyzed all standing trees grown within the plot of 30 m × 30 m (Choi et al. 1999).

In this study, we analyzed both tree- and stand-level results focusing on DBH and volume, which are the main variables based on thinning; we figured out change patterns of stand structure by identifying the periodical occurrences of mortality and ingrowth trees. Regarding the calculation of the site index, we targeted the experimental sites on Korean white pine stands which were the plantations with the same age among trees, and applied the site index table of the National Institute of Forest Science for Korean white pine to the dominant height and corresponding age of trees within the plot (Korea Forest Research Institute 2012c). Additional variables were calculated based on the factors measured in the field and used for the analysis of the results. Basal area (BA) was calculated, as shown in Equation 1, based on DBH at 1.2 m above the ground; the volume of standing trees (V) was calculated, as indicated in Equation 2, based on Model 4 for Korean white pine among the bivariate volume equations reported in the Lee et al. (2017). The volume of standing trees within the oak-dominated natural deciduous forest was calculated, based on the calculated values of Q. mongolica Fisch. ex Ledeb. and Q. variabilis Blume, which were provided by the forest resource assessment program of the National Institute of Forest Science (Korea Forest Research Institute 2012c).

BA: Tree basal area (m²), D: Tree diameter at breast height (cm),

π: mathematical constant approximately equal to 3.14159,

V: Tree volume (m³), H: Tree height (m).

In this paper, all figures were presented based on the tree- and stand- characteristics after the thinning was operated. The results of the Korean white pine plantation and oak-dominated natural deciduous forest were derived from each of the experimental sites 1 and 2, while considering forest type, repeated measurement period, site index, and stand structure. As the site 1 for the Korean white pine plantation has two blocks with four plots each, average values of plots by thinning intensity were applied to analyze. We analyzed the targeted standing trees which lived during the whole measurement period in terms of the DBH and volume of individual tree; as for the analysis of stand characteristics and ingrowth trees, we targeted the standing trees above 3 cm DBH. Regarding the main analysis results between the thinned plots, we performed one-way analysis of variance (ANOVA) and Duncan’s multiple range test. In terms of the growth model of plot-level DBH periodic annual increment (PAI) in line with stand density, a linear regression model was applied in consideration of the number and pattern of samples by targeting the trees with DBH above 6 cm in each plot, and R statistical software was used for statistical analysis in this study (R Core Team 2019).

Stand characteristics and DBH distribution

The thinning applied during the first inventory (in 1996 for site 1 of pine stand and site 1 and 2 for oak stand, and in 1999 for site 2 of pine stand) in these sites was carried out in the range between 8% and 48% for the Korean white pine plantation, and between 42% and 63% for the oak- dominated natural deciduous forest, based on basal area by plot. Accordingly, the numbers of residual trees were different as follows depending on thinning intensity: between 990 and 2,283 trees/ha in the site 1 and between 618 and 1,787 trees/ha in the site 2 for Korean white pine plantation; between 948 and 2,672 trees/ha in site 1, and between 491 and 2,144 trees/ha in site 2 for oak-dominated natural deciduous forest (Table 3). In the natural deciduous forest, compared to the Korean white pine plantation, there were diverse tree species, and stand structure and growth status at the final measurement were different by plot (Table S1).

Table 3 . Tree and stand characteristics of a Korean white pine plantation and oak-dominated natural deciduous forest.

Stand type	Experimental site	Plot	At first measurement after thinning							At last measurement
			Average tree DBH (cm)	Average tree height (m)	Average tree volume (m3)	No. of stems (trees/ha)	Stand basal area (m2/ha)	Stand volume (m3/ha)		Average tree DBH (cm)	Average tree height (m)	Average tree volume (m3)	No. of stems (trees/ha)	Stand basal area (m2/ha)	Stand volume (m3/ha)
Korean white pine plantation (Pinus koraiensis)	Site 1	A	13.2	9.5	0.076	990	13.8	75.0		22.2	15.8	0.324	970	39.5	315.2
		B	13.6	9.4	0.081	1,511	22.2	119.0		21.2	15.7	0.294	1,406	51.9	413.2
		C	13.5	9.3	0.079	1,774	25.6	136.2		20.3	15.2	0.264	1,668	56.9	441.4
		D	13.2	9.2	0.076	2,283	30.3	161.1		19.2	14.6	0.233	1,844	57.4	426.9
	Site 2	A	19.2	12.0	0.190	618	18.6	117.5		28.8	16.9	0.564	618	41.4	348.4
		B	16.8	10.3	0.129	1,063	24.1	134.9		24.5	15.7	0.385	1,026	50.4	395.5
		C	16.6	13.1	0.172	1,471	33.4	248.3		24.2	17.8	0.439	1,185	58.1	515.1
		D	18.0	14.2	0.203	1,787	38.2	285.1		23.0	17.0	0.376	1,170	51.4	440.3
Oak-dominated natural deciduous forest (Quercus mongolica and Q. variabilis)	Site 1	A	10.1	8.7	0.046	948	8.3	43.4		18.2	12.1	0.193	905	20.5	131.6
		B	8.8	8.8	0.037	1,250	8.6	45.6		15.5	11.9	0.140	1,595	26.2	174.1
		C	10.7	9.3	0.052	1,810	16.5	89.2		15.6	11.6	0.140	2,069	34.8	216.2
		D	9.5	9.8	0.048	2,672	21.0	122.9		14.7	11.4	0.133	1,940	33.9	222.1
	Site 2	A	20.1	14.0	0.250	491	16.2	123.0		27.6	19.6	0.664	1,474	38.6	340.5
		B	21.3	15.5	0.303	715	25.7	214.1		28.0	16.4	0.534	1,429	45.1	364.9
		C	21.3	15.3	0.309	983	35.6	294.2		26.7	16.1	0.519	2,054	62.2	503.6
		D	16.7	13.1	0.218	2,144	53.8	430.7		22.1	13.6	0.349	2,144	72.5	567.5

The average values at tree level statistics were provided based on the living trees through all measurement periods by excluding any mortality and ingrowth trees. The inventory of Korean white pine plantation was carried out firstly in 1996 and lastly in 2014 for the site 1 and firstly in 1999 and lastly in 2016 for the site 2. The inventory of oak-dominated natural deciduous forest was carried out firstly in 1996 and lastly in 2017 for the site 1 and 2. The statistics and no. of trees per ha were calculated based on the plot size applied with the buffer zone (e.g., 16 m × 16 m or 30 m × 30 m) by plot of each site.

DBH: diameter at breast height.

As a result of examining the tree- and stand-level status during the first and last inventory periods, DBH and volume of individual trees in the Korean white pine plantation tended to increase in line with thinning intensity, whereas there was no clear tendency in the oak-dominated natural deciduous forest. Furthermore, the number of trees in the Korean white pine plantation was observed to decrease in all plots depending on the occurrence of dead trees. On the other hand, in the oak-dominated natural deciduous forest, a higher number of trees was observed due to the newly occurred ingrowth trees.

As for the DBH distribution pattern of standing trees per measurement, DBHs at the first measurement after thinning in all sites showed a similar average per plot, and they were distributed in a form of a single bell-shape curve; as the growth period passes, the shape of DBH distribution per plot was observed to be different (Fig. 3). In particular, as the thinning intensity is stronger in the Korean white pine plantation, the DBH distribution was skewed to the left; this tendency was more evident during the last measurement period, e.g., plot A of site 2 in pine stand. That is, as the thinning intensity was stronger, there were more trees with larger DBHs, and as the thinning intensity is weaker, there were more trees with smaller DBHs in Korean white pine plantation.

Figure 3. DBH distribution by thinning intensity over measurement instances for each site in the Korean white pine plantation and oak- dominated natural deciduous forest. Only the trees that survived the entire measurement period were analyzed. DBH class was 2 cm of bandwidth. Each site and plot were thinned with different thinning intensities from heavy thinning treatment to light thinning treatment according to Table 2. DBH: diameter at breast height.

On the other hand, the DBH distribution in the oak- dominated natural deciduous forest did not show the same pattern as that in the Korean white pine plantation, and there was no significant difference in line with the thinning intensity. In particular, the overall DBH distribution over time did not move to the right for the oak-dominated natural deciduous forest as much as in the Korean white pine plantation. Taking these results into account together, the implementation of an appropriate thinning regime for the production of large-diameter trees is considered more effective in Korean white pine plantations than in oak- dominated natural deciduous forests.

DBH and volume trend by thinning over time

As a result of comparing the changes in the average DBH by measurement period, a larger DBH was found in the Korean white pine plantations compared to the oak-dominated natural deciduous forests. As for the Korean white pine plantation, Seo et al. (2018b) reported that the average DBH was 15.7 cm at age 24 and 21.3 cm at age 37, and Seo et al. (2018a) analyzed a similar DBH trend over time. The results of the previous studies were consistent with the observations of our study. In addition, both the sites 1 and 2 indicated the tendency of higher average DBH, as thinning intensities were higher (Fig. 4). In particular, the difference between the heavy thinning plot and the unthinned plot was the most distinct; at the last measurement period, the DBH in the heavy thinning plot of the site 1 was 22.2 cm, which was 116% higher than the DBH of 19.2 cm in the unthinned plot. In the case of the site 2, at the last measurement period, DBH in the heavy thinning plot was 28.8 cm, which was 125% higher than the DBH of 23.0 cm in the unthinned plot, indicating that the growth differences between thinning intensities were more evident over time.

Figure 4. Average DBH change over time in the Korean white pine planation and oak-dominated natural deciduous forest. Only the trees that survived the entire measurement period were analyzed. The plots were thinned with different intensities at the first measurement period: Plot A with heavy thinning treatment, Plot B with moderate thinning treatment, Plot C with light thinning treatment, and Plot D with unthinned treatment. Specific thinning intensities are provided in Table 2. DBH: diameter at breast height.

Meanwhile, in the oak-dominated natural deciduous forest, the average DBH was also the smallest in the unthinned plot; the average DBH tended to be in the order of thinning intensity similar to the Korean white pine plantation. In general, the thinning effect on the increase in diameter in the natural deciduous forest was analyzed to be relatively lower. Park et al. (1996) reported the average DBH of sample trees ranged between 8 cm and 16 cm from age 20 to age 35, and this result was similar to the site 1 of the natural deciduous forest in the present study.

As a result of analyzing the average volume of individual trees over time, a higher average volume in the Korean white pine plantation was clearly found at higher thinning intensities (Fig. 5). In particular, during the final measurement period, the average tree volume in the heavy thinning plot of the site 1 was 0.324 m³, which was 139% higher than the volume of 0.233 m³ in the unthinned plot; the average volume in the heavy thinning plot of the site 2 was 0.564 m³, which was 150% higher than the volume of 0.376 m³ in the unthinned plot. This result was consistent with the result of a previous study in which the average tree volume of Korean white pine was 0.314 m³ at age 30 (Seo et al. 2018a).

Figure 5. Average tree volume change over time in the Korean white pine planation and oak-dominated natural deciduous forest. Only the trees that survived the entire measurement period were analyzed. The plots were thinned with different intensities at the first measurement period: Plot A with heavy thinning treatment, Plot B with moderate thinning treatment, Plot C with light thinning treatment, and Plot D with unthinned treatment. Specific thinning intensities are provided in Table 2.

Similar to the observed significant difference in the Korean white pine plantation, the average volume of individual trees in the natural deciduous forest indicate distinct tendency in line with thinning intensity. Considering the average volume of the natural deciduous forest at the final measurement taken in 2017, the average tree volume in the heavy thinning plot of the site 1 was 0.193 m³, which was 145% higher than the volume of 0.133 m³ in the unthinned plot; the volume in the heavy thinning plot of the site 2 was 0.183 m³, which was 190% higher than the volume of 0.349 m³ in the unthinned plot.

On the other hand, the stand volume per ha was contrary to the tendency of the average tree volume; the lower thinning rate and the higher stand density indicated the higher stand volume (Fig. 6). This trend continued over time, but at the time of the final measurement, the highest stand volume was observed in the light thinning plots (Plot C) for the Korean white pine planation. Meanwhile in the oak-dominated natural deciduous forest, the unthinned plots had the highest stand volume, but the difference from the light thinning plot was smaller at the final measurement than at the first measurement. These results may imply that taking self-thinning into account, light thinning treatment can be considered even for the maximum stand volume (Mäkinen and Isomäki 2004b).

Figure 6. Stand volume change over time in the Korean white pine planation and oak-dominated natural deciduous forest. The plots were thinned with different intensities at the first measurement period: Plot A with heavy thinning treatment, Plot B with moderate thinning treatment, Plot C with light thinning treatment, and Plot D with unthinned treatment. Specific thinning intensities are provided in Table 2.

Periodic annual increment of DBH and volume

As a result of analyzing the DBH PAI per growth period, a clear difference was observed in the Korean white pine plantation based on thinning intensity (Fig. 7). The ranges of average growth per plot in the site 1 were as follows: 0.61–0.84 cm/year in the period between 1996 and 2001 (age 19–24), which was just after thinning, 0.33–0.52 cm/year in the period between 2001 and 2008 (age 24–31), and 0.22–0.26 cm/year in the period between 2008 and 2014 (age 31–37). These results are supported by a previous study in which the 3-year DBH increment at age 32 ranged from 1.6 cm/3 year to 2.1 cm/3 year by thinning intensity (Seo et al. 2019). In the site 2, there was the growth difference by thinning intensity: 0.34–0.66 cm/year in the period between 1999 and 2010 (age 21–32), just after thinning, and 0.28–0.43 cm/year in the period between 2010 and 2016 (age 32–38). Overall, the DBH PAI was higher in younger stands and shorter time after thinning, which is similar to the trend observed in previous research where the DBH PAI of Korean white pine peaked at 1.07 cm/year at age 15 (Seo et al. 2018a). Based on our results of DBH PAI, the second thinning for Korean white pine plantation can be suggested to perform 10–15 years after the first thinning. This is consistent with the previous literatures where the second thinning is recommended 10 years after the first thinning (Choi et al. 2014a; Korea Forest Research Institute 2005).

Figure 7. Periodic annual increment of DBH by thinning intensity during each growth period in the Korean white pine planation and oak-dominated natural deciduous forest. Only the trees that survived the entire measurement period were analyzed. The plots were thinned with different intensities at the first measurement period: Plot A with heavy thinning treatment, Plot B with moderate thinning treatment, Plot C with light thinning treatment, and Plot D with unthinned treatment. Specific thinning intensities are provided in Table 2. Significant differences between thinning treatments at each measurement are indicated by different letters based on Duncan’s multiple range test (alpha = 0.05). DBH: diameter at breast height.

On the other hand, the ranges of average growth per plot in the site 1 of the natural deciduous forest were as follows: between 0.27 and 0.52 cm/year in the period between 1996 and 2001, which was just after thinning, between 0.28 and 0.44 cm/year in the period between 2001 and 2008, and between 0.21 and 0.27 cm/year in the period between 2008 and 2017 (Fig. 7). As for the site 2, the ranges were as follows: between 0.23 and 0.35 cm/year in the period between 1996 and 2001, between 0.30 and 0.49 cm/year in the period between 2001 and 2008, and between 0.22 and 0.32 cm/year in the period between 2008 and 2017, indicating the lower growth than the Korean white pine plantation in general. Furthermore, the DBH PAI in the oak-dominated natural deciduous forest showed significant differences between thinning treatments during early period, but less amount of growth was observed for these forests, which is in contrast to those observed for the Korean white pine plantation. The DBH PAI of oak stands in this study were consistent with those observed in previous studies: 0.64 cm/year for Q. mongolica by Park et al. (1996), 0.26 cm/year for Q. mongolica by Song and Lee (1996), 0.39 cm/year for Q. variabilis by Chung and Lee (1999), and 0.28 cm/year for Q. mongolica and 0.27 cm/year for Q. variabilis by Choi and Yoo (2006). Based on our results of insignificant difference between thinning treatment at the last period, the second thinning for oak-dominated natural deciduous forest can be suggested to perform 10–15 years after the first thinning, as Korea Forest Research Institute (2005) recommended the second thinning 10 years after the first thinning for oak stand (Quercus acutissima Carruth.).

In the case of the Korean white pine plantation, the PAI of the average tree volume tended to be higher along with the higher thinning intensity (Fig. 8). The trend was similar in the natural deciduous forest, but the significant difference among thinning treatment was not as clear as in Korean white pine plantation. Compared to DBH growth, furthermore, the PAI of average tree volume of the thinning plots continued to be higher than the unthinned plots over time both in the Korean white pine plantations and the oak-dominated natural deciduous forest. Seo et al. (2018a) reported that the PAI of individual trees in Korean white pine plantations remained high and did not meet the mean annual increment in their study. They assumed that volume growth was vigorous until the rotation age of a stand was reached. The result of the present study is similar to the preceding research as there was no clear decrease in tree volume PAI over time.

Figure 8. Periodic annual increment of tree volume by thinning intensity during each growth period in the Korean white pine plantation and oak-dominated natural deciduous forest. Only the trees that survived the entire measurement period were analyzed. The plots were thinned with different intensities at the first measurement period: Plot A with heavy thinning treatment, Plot B with moderate thinning treatment, Plot C with light thinning treatment, and Plot D with unthinned treatment. Specific thinning intensities are provided in Table 2. Significant differences between thinning treatments at each measurement are indicated by different letters based on Duncan’s multiple range test (alpha = 0.05).

The DBH PAI was illustrated as scatter plots, in line with the stand basal area and the number of trees, and a linear regression analysis was performed; significant coefficients were calculated (Fig. 9). In both the Korean white pine plantation and the oak-dominated natural deciduous forest, a lower tendency of PAI was shown for higher stand densities; the PAI of the Korean white pine plantation was significantly higher than the PAI of the oak-dominated natural deciduous forest. Lee et al. (2004) compared red pine and oak species (P. densiflora and Q. variabilis) in Korea and reported that the growth of the red pine species was more vigorous than that of the oak species. The growth model in line with the stand basal area had higher adjusted coefficients of determination than the growth model based on the number of trees. It is expected in future studies that the more accurate models for both variables will be developed through the repeated monitoring of permanent plots.

Figure 9. Scatterplots of DBH periodic annual increment (PAI) over stem number and stand basal area by stand type. The lines represent the models based on linear regression. TPH is trees per ha, BA is basal area, and Pine and Oak are the dummy variables for stand type. All the coefficients were significant (p < 0.001) except for TPH (p = 0.066). R²_adj is adjusted coefficient of determination. S_y‧x is standard error of the estimate. DBH: diameter at breast height.

Mortality analysis

The mortality in each site decreased along with higher thinning intensity, and the cumulative total number of mortality was the highest in the unthinned plot (Fig. 10). In terms of the mortality of unthinned plots in the Korean white pine plantation, there were 460 trees/ha in the site 1, and 617 trees/ha in the site 2. As for the mortality in the oak-dominated natural deciduous forest, there were 1,164 trees/ha in the site 1, and 804 trees/ha in the site 2; there were more dead trees than in the Korean white pine plantation (Fig. 10). In particular, dead trees were mostly small-diameter trees with suppressed DBH and tree height; there was a tendency for a higher number of dead trees over time, compared to the initial investigation period. This is assumed that the diameter growth of trees and stand development intensified the competition between individual trees, leading to the limit of the maximum stem number per unit area (Lee et al. 2000b). In the future study, when such repeated measurements on a permanent plot are sufficiently secured, it is expected to be possible to develop the models of maximum stand density and natural mortality for the oak-dominated natural deciduous forest, similar to the model of maximum stand density developed by Lee and Choi (2019; 2020) for Korean red pine, Korean white pine, and Japanese larch.

Figure 10. Tree mortality by thinning intensity over measurement period for each site in the Korean white pine planation and oak- dominated natural deciduous forest. Plot A is heavy thinning treatment, plot B is moderate thinning treatment, plot C is light thinning treatment, and plot D is unthinned treatment.

Ingrowth analysis

During the entire monitoring period, there were no ingrowth trees in the Korean white pine plantation, but in contrast, there were many ingrowth trees in the oak-dominated natural deciduous forest (Table 4). As a result of analyzing ingrowth trees by species, Korean white pine (P. koraiensis Siebold & Zucc.) had the most ingrowth trees with 845 trees/ha, followed by other species in descending order: 252 trees/ha for Q. mongolica Fisch. ex Ledeb., 81 trees/ha for Prunus sargentii Rehder, 78 trees/ha for Fraxinus rhynchophylla Hance, and 55 trees/ha for Q. variabilis Blume. As such features underpin the result of Kim and Kang (2005), it is considered as an early stage of the transition of oak-dominated natural deciduous forest to a mixed forest with Korean white pine saplings.

Table 4 . Ingrowth tree species by experimental site and thinning intensity in oak-dominated natural deciduous forest.

Species name	Experimental site and plot id								Average (trees/ha)
	1A	1B	1C	1D	2A	2B	2C	2D
Pinus koraiensis Siebold & Zucc.	474	819	1,164	862	1,161	625	983	670	845
Quercus mongolica Fisch. ex Ledeb.	302	647	216		179	357	313		252
Prunus sargentii Rehder	129	388	86	43					81
Fraxinus rhynchophylla Hance			43	43	134	134	89	179	78
Quercus variabilis Blume	129						313		55
Castanea crenata Siebold & Zucc.	43								5
Quercus acutissima Carruth.	43								5
Quercus dentata Thunb.	43								5
Styrax obassis Siebold & Zucc.		43							5
Sum (trees/ha)	1,164	1,896	1,509	948	1,474	1,117	1,697	849	1,332

In addition, as a result of examining the number of ingrowth trees based on the stand density immediately after thinning per plot, there was no clear trend between the thinning plots, but in the sites 1 and 2, there were the lowest numbers of ingrowth trees in unthinned plots, that is, 849 trees/ha, and 948 trees/ha, respectively (Fig. 11). Regarding this result, Kim and Kang (2005) assessed growth characteristics of saplings of Korean white pine in the natural deciduous forest; it was reported that saplings of Korean white pine with shade tolerance can have inhibited germination and growth if the crown density of the upper story is too high. The results in the present study are consistent with the outcomes of previous studies, indicating that the open state of the upper canopy in the natural deciduous forest can sufficiently affect the development of ingrowth trees.

Figure 11. Ingrowth saplings by growth period based on the residual trees at the thinned year of 1996 in oak-dominated natural deciduous forest.

We analyzed the differences among stand structures and growth changes of Korean white pine plantation and the oak-dominated natural deciduous forest based on thinning intensity. The average DBH and volume showed a greater difference by thinning intensity over time, and the trend was clearer in the pine stand. The stand volume was lower as the thinning intensity was higher over time in both stand types. However, light thinning treatment can be recommended to retain the maximum stand volume based on our study. In the heavy thinning plot of the pine stand, there were more trees with larger DBH over time, indicating the obvious thinning effect. It is considered that a clearer growth effect is expected in the Korean white pine plantation if the production of large-diameter trees is targeted by performing appropriate thinning. Furthermore, considering forest growth, Korean white pine plantation can be more profitable than oak-dominated natural deciduous forest as it developed superior growth characteristics within a short period after appropriate thinning treatment. Especially, heavy thinning treatment is suggested for trees with high DBH and volume of individual trees.

The DBH PAI of individual trees was higher in the pine stand along with the heavier thinning intensity. Based on the statistical results, the second thinning can be suggested to be performed 10–15 years after the first thinning in both stand types. The PAI of the tree volume was higher in the pine stand than in the oak stand. The models of DBH PAI based on stand basal area presented the decreasing growth trend along with the higher stand density. The PAI of the pine stand was significantly higher than that of the oak stand. Dead trees were the most found in unthinned plots in all sites. Korean white pine was the most dominant for the ingrowth trees in the natural deciduous forest. The main results of this study are expected to provide useful basic data to prepare silvicultural guidelines for the Korean white pine plantation and the oak-dominated natural deciduous forest for practical forestry.

Supplementary information accompanies this paper at https://doi.org/10.1186/jee.22.013

Table S1. Summary statistics of tree and stand characteristics for the sites in oak-dominated natural deciduous forest at the last measurement in 2017.

The Research Forest and the Forest Resource Monitoring Center on Climate Change (FRMCCC) at Kangwon National University provided data maintenance support. In establishing permanent plots and collecting field data, our authors gratefully acknowledge the many contributions of Dr. Inhwa Choi, Emeritus Professor of the Department of Forest Management, the College of Forest and Environmental Sciences, Kangwon National University.

ANOVA: One-way analysis of variance

DBH: Diameter at breast height (cm)

PAI: Periodic annual increment

DL conceived the ideas, conducted field study, conducted the data collection and formal analysis, and wrote and reviewed the manuscript. JC conceived the ideas, conducted field study, reviewed the manuscript, and secured the research funding. All authors read and approved the final manuscript.

This study was funded by the Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education in 2022 (No. NRF-2016R1D1A1B02011648).

The datasets generated and/or analyzed during this study are available from the corresponding author upon reasonable request.

Not applicable.

Not applicable.

The authors declare that they have no competing interests.