Journal Search Engine
Search Advanced Search Adode Reader(link)
Download PDF Export Citaion korean bibliography PMC previewer
ISSN : 1598-5504(Print)
ISSN : 2383-8272(Online)
Journal of Agriculture & Life Science Vol.56 No.2 pp.25-34
DOI : https://doi.org/10.14397/jals.2022.56.2.25

Selection of Elite Tree of Evergreen Oaks and the Growth Characteristics of Selected Individuals

Seong-Hyeon Yong1, Do-Hyun Kim2, Kwan-Been Park3, Dong-Jin Park4, Hyun-Jin Song5, Hak-Gon Kim6, Myung-Suk Choi7*
1Researcher, Department of Environmental Forest Science& Institute of Agriculture and Life Science, Jinju, 52828, Korea
2Graduate student, Department of Environmental Forest Science& Institute of Agriculture and Life Science, Jinju, 52828, Korea
3Graduate student, Department of Environmental Forest Science& Institute of Agriculture and Life Science, Jinju, 52828, Korea
4Researcher, Department of Seed and Seedling Management, National Forest Seed and Cultivar Center, Chungju, 27495, Korea
5Researcher, Department of Variety Examination, National Forest Seed and Cultivar Center, Chungju, 27495, Korea
6Researcher, Forest Research Department, Gyeongsangnam-do Forest Environment Research Institute, 52615, Korea
7Professor, Department of Environmental Forest Science& Institute of Agriculture and Life Science, Jinju, 52828, Korea
* Corresponding author: Myung-Suk Choi (Tel.) +82-55-772-1856 (E-mail) mschoi@gnu.ac.kr
March 23, 2022 April 15, 2022 April 19, 2022

Abstract


Elite trees of evergreen oaks (Quercus glauca, Quercus acuta, Quercus salicina, and Quercus gilva) were selected from Jeju Island and Wando Island. Elite trees were carried out by modifying the tree selection criteria. Elite trees were selected by height, DBH, clear length, crown diameter, leaf length, and leaf depth. Regarding height, Q. acuta was the highest, and the other three tree species were similar. Clear length showed the same trend as height. In the case of Q. glauca, height showed a positive correlation with DBH, crown diameter and leaf depth. In the case of Q. acuta, positive correlations were shown with all characteristics of DBH, and correlation analysis between DBH and crown diameter, and leaf length and leaf depth also showed positive correlations. In the Pearson correlation coefficient of Q. salicina, height showed a positive correlation with DBH. In the case of Q. gilva, height showed a positive correlation with DBH (0.539). As a result of analyzing the principal components for each of the six growth characteristics, the four species were divided into two principal components with an eigenvalue of 1 or higher, and the cumulative explanatory power was 57% or more. Based on the principal component results, it was possible to confirm the relationship between growth and trait characteristics by species. Still, it was not easy to understand the relationship among each selection tree, so a cluster analysis was performed using the principal component score. Based on the distance levels 5.0 and 6.0 of the selection tree of each species, they were classified into 4-5 clusters. It is judged that the above results can be used as data for the selection of elite trees of evergreen oaks.



초록


    Introduction

    According to the climate change scenario, in the case of the Korean Peninsula, as the temperature rises, precipitation increases and the concentration of carbon dioxide in the atmosphere increases (Korea Meteorological Administration, 2020). Climate change affects various ecosystems and, in particular, causes various problems in the forest ecosystem, such as vegetation distribution, growth limit changes, and biodiversity (Cha, 1998;Chung et al., 2017). The establishment of countermeasures is urgently required.

    Global warming threatens healthy forests because it increases environmental stressors such as extreme heat, extreme cold, and drought (Lim et al., 2008). In addition to woody plants, freezing of crops, fruit trees, and landscape trees due to climate change have been continuously reported (Hwang & Kim, 2012;Jung et al., 2014).

    As the distribution of warm temperate tree species is moving north due to climate change, not only is the resource value increasing, but also in winter, leaves do not fall, so it is very valuable as a material for urban green spaces in terms of management (Jung et al., 2014).

    As a long-term countermeasure against changes in the ecosystem, it is necessary to secure breeding resources for warm temperate forest tree species and improve the secured warm temperate forest tree species so that they can be used as future timber resources (NIFOS, 2013). Therefore, it is necessary to closely investigate and analyze promising resources among forest genetic resources in temperate regions. Through this, promising tree species should be selected, and future industrial use of warm temperate tree species should be dealt with in advance.

    Selective breeding is one of the critical methods for cultivating trees (Hyun et al., 1985). The first step is to select many excellent individuals from within the native population and form a primary breeding group to obtain a high improvement effect. Among the selected individuals created in the primary breeding group, various crosses are carried out according to the breeding design method to investigate the growth status of the seedlings. It should be possible (Hyun et al., 1985). In particular, since the gene source of evergreen broadleaf forests in Korea is poor, it is necessary to set up a seed garden exclusively for temperate tree species by tree selection (Song et al., 2011).

    In addition, to establish a breeding plan for a tree group, it is first necessary to study the genetic characteristics or trait variation of the target tree species. Research on the scope is essential (Goodall-copestake, 2005).

    Six evergreen oaks are growing in Korea, distributed in the southern region. Q. glauca, Q. acuta, Q. gilva, Q. salicina are native species, and Q. phillyraeoides is known as an introduced species (Choi et al., 2021). Evergreen oaks have high utility value, such as producing excellent wood and acorn jelly and have applications as biomaterials such as cosmetics and pharmaceuticals (Shin et al., 2006). The demand for seedlings of temperate evergreen broadleaf species is increasing due to the restoration of temperate forests and new reforestation due to climate warming, landscaping trees around cities, street trees, park, and garden trees, and the planting area is also expanding (Park et al., 2010;Song et al. al., 2015). Q. myrsinaefolia, one of the evergreen oaks, was recently selected as the target plant for climate change adaptation on the Korean Peninsula and is a primary observational plant for climate warming effects (Oh et al., 2010). Evergreen oaks represent Korea's warm temperate evergreen broadleaf trees and are continuously planted for restoration and artificial reforestation in warm temperate regions.

    However, research on the selection of elite trees of evergreen oak is incomplete. Therefore, in this study, four species of an evergreen oak family were selected, and to classify them in a practical way, the growth characteristics of the selected trees were investigated. The selection characteristics were identified using principal component analysis and cluster analysis.

    Materials and Methods

    1. Materials and selection sites

    Elite tree selection was performed in the southern region of Korea, where evergreen oaks are widely distributed. The species collected were Quercus glauca Thunb., Quercus acuta Thunb., Quercus salicina Blume. and Quercus gilva Blume. The collection site is shown in Fig. 1.

    2. Selection of elite tree

    The elite tree selection was performed by modifying the method of Son et al. (2011) and the tree selection criteria of the Korea Forest Service. The selection criteria were trees with excellent stem straightness and crown development, trees with no signs of disease, and trees with high clear length. For elite tree selection, a total of 105 trees, including Q. glauca 28 trees, Q. acuta 31 trees, Q. salicina 31 trees, and Q. gilva 15 trees were selected over 4 years from 2017 to 2020, and their characteristics were investigated.

    3. Determination of growth characteristics of selected elite trees

    Height, Diameter at Breast Height (DBH), clear length, and crown diameter of each selected elite tree were investigated. The elite tree also examined leaf length and leaf depth by collecting leaves from branches above clear length.

    For the investigated data, the differences in growth characteristics were analyzed using the SPSS program (Statistical Package for Social Science, ver. 25.0) for each species. The Pearson correlation and the Spearman rank correlation method were used for analysis. In addition, the distance between regions was calculated from the correlation matrix between the morphological characteristics investigated through principal component analysis, and the contribution of each principal component to the eigenvalue and total variation was obtained. UPGMA (Unweighted pair-group method using arithmetic averages) cluster analysis was performed, and the calculated distances were expressed as dendrograms to ensure the selection diversity for each evergreen oaks species.

    Results and Discussion

    1. Describe the approximate age of the selected species.

    The investigation results of the morphological characteristics of 105 evergreen oaks are shown in Table 1. Significant differences were recognized when the results were confirmed by grouping by species. Regarding tree height, Q. acuta was the highest, and the other three tree species were similar. Clear length showed the same trend as height. DBH showed no significant difference between species. Q. glauca was the most widely measured for crown diameter, and Q. salicina was the narrowest. When comparing the leaf size, the Q. acuta had the longest and widest leaf, followed by Q. glauca and Q. salicina.

    2. Correlation analysis

    The results of the correlation analysis on the 6 characteristics used to secure the selection diversity of evergreen oaks are as follows (Table 2). In the case of Q. glauca, positive correlations were shown with DBH (0.560**), crown diameter (0.517**), and leaf depth (0.410*) for height in Pearson correlation. Also, DBH showed a positive correlation with the crown diameter (0.406*). In the Spearman rank correlation coefficient, as in the previous results, height showed a positive correlation with DBH, crown diameter, and leaf depth, but DBH and crown diameter were not significant (p>0.5).

    For Q. acuta, the Pearson correlation analysis and the Spearman rank correlation analysis showed the same results with Q. glauca (Table 3). DBH showed positive correlations with all characteristics, and correlation analysis between DBH and crown diameter, leaf length, and leaf depth also showed positive correlations.

    In the Pearson correlation coefficient of Q. salicina, height showed a positive correlation with DBH. A positive correlation was also found between DBH and crown diameter, leaf length, and leaf depth (Table 4). The Spearman correlation coefficient also showed the exact positive correlation as the previous result. Also showed a positive correlation between leaf length and leaf depth.

    In the case of Q. gilva, the Pearson correlation showed a positive correlation with DBH (0.539*) for height (Table 5). Also, Clear length showed a positive correlation with crown diameter (0.546*). In the Spearman rank correlation coefficient, as in the previous results, height showed a positive correlation with DBH, but the correlation between DBH and crown diameter was not significant (p>0.5).

    In general, trees have a close relationship between height and DBH (You et al., 2004). Calculating the coefficients of determination of the height-DBH growth models for 8 major tree species distributed in the Chungcheong-do region showed a high explanatory power of 94% or more (Seo et al., 2011). Also, as a result of estimating the relative growth equation for Q. acuta seeds, it was stated that the coefficient of determination was somewhat higher in the relative growth equation using DBH and height as independent variables than in the relative growth equation using only DBH as independent variables. In the results of this study, a positive relationship between height and height of chest was found in all Quercus spp., suggesting that we selected trees with good growth.

    3. Principal Component Analysis

    Principal component analysis was performed for each of the 6 growth characteristics, and the intrinsic value of each obtained characteristic was analyzed. As a result of Q. glauca's investigation, it was divided into two Principal components with an eigenvalue of 1 or higher, and the cumulative explanatory power was 72.27%. The eigenvalue of the first principal component was 3.05, which had an explanatory power of 50.85% of the total variance. The first principal component correlated leaf length, leaf depth, height, and clear length and ranged from 0.608 to 0.852. The eigenvalue of the second principal component was 1.285, which had an explanatory power of 21.42% of the total variance. Crown diameter (0.886) and DBH (0.865) correlated with the second principal component.

    As a result of the analysis of Q. acuta, it was divided into two principal components with an eigenvalue of 1 or higher, and the cumulative explanatory power was 57.31% (Table 7). The eigenvalue of the first principal component was 2.26, which had an explanatory power of 37.63% of the total variance. The first principal component had a high correlation in DBH, height, and crown diameter and ranged from 0.742 to 0.822. Also, it showed a negative correlation (-0.413) with clear length. The eigenvalue of the second principal component was 1.181, which had an explanatory power of 19.69% of the total variance. The second principal component had a high correlation in leaf depth, clear length, and leaf length and was located in the range of 0.457 ~ 0.703.

    As a result of the analysis of Q. salicina, it was divided into three principal components with an eigenvalue of 1 or higher, and the cumulative explanatory power was 78.65% (Table 8). The eigenvalue of the first principal component was 2.17, which had an explanatory power of 36.14% of the total variance. DBH and crown diameter had a high correlation in the first principal component. The eigenvalue of the second principal component was 1.29, which had an explanatory power of 21.66% of the total variance. In the second principal component, leaf length and leaf depth are correlated. The eigenvalue of the third principal component was 1.25, and it had an ex- planatory power of 20.85% of the total variance, and clear length and height had a high correlation.

    As a result, Q. gilva analysis, it was divided into two principal components with an eigenvalue of 1 or higher, and the cumulative explanatory power was 67.32% (Table 9). The eigenvalue of the first principal component was 2.11, which had an explanatory power of 35.11% of the total variance. The first principal component had a high correlation in the order of clear length crown diameter. The eigenvalue of the second principal component was 1.93, which had an explanatory power of 32.21% of the total variance. DBH and height showed a high correlation in order for the second principal component, and leaf depth and leaf length showed a negative correlation.

    In general, even in the same species, genetic differences appear due to geographic characteristics and various environmental factors, and these tendencies are also very diverse (White et al., 2007). Research on trait characteristics or genetic variation is essential to select and classify these genetic resources. Various original data must be accumulated, and it must be analyzed effectively. The principal component analysis is used to solve these problems. Principal component analysis selects characteristic variables from multivariate data and finds the linear combination showing the most significant variance. Results not shown in the original data can be interpreted. Accordingly, the main trait can be identified by checking the correlation coefficient between the principal component and the trait within the principal component showing high explanatory power (Kang, 2005).

    In this study, various characteristics for each component showed a high correlation coefficient, and a relatively large tendency between the derived characteristics was shown. It is expected that the interpretation of the data using the principal component will contribute to increasing the usability as a selection index.

    4. Cluster Analysis

    Based on the principal component results, it was possible to confirm the relationship among growth characteristics for each species, but it was not easy to understand the relationship between the group for each selected tree.

    For Q. glauca, 28 selection trees were classified into 5 clusters according to distance level 5.0. When the clusters were represented by arranging the selection trees used for the analysis in a two-dimensional space based on the principal component values, the I, II, and III groups were distributed in the negative region of the first principal component. Taxon II was also distributed in the negative region of the second principal component. In the case of Q. glauca, since the main collection area is located on Jeju Island, it was expected that the group affiliated with would be messed up when performing cluster analysis. It was confirmed that the groups were grouped. The cluster analysis results and the principal component analysis on the Q. glauca confirmed that group Ⅱ was not suitable as a selection tree. In the case of groups Ⅳ and V, the height and height below the ground tends to be high, which is confirmed to be a suitable individual as a lumber tree among the first trees (Fig. 2).

    Q. acuta 31 selection trees were classified into 4 clusters based on an average distance of 5.0. When clusters were represented by arranging the individuals used in the analysis in a two-dimensional space based on the principal component values, groups II and III were distributed in the negative region of the first principal component. When viewed together with the principal component analysis results, the Q. acuta individuals selected in Jeju Island were grouped into the taxon III, and their characteristics were measured to be low compared to the individuals in other regions. This means that it is not suitable as an Elite tree. As for the suitability as a elite tree, Ⅰ and Ⅳ groups were found to be suitable, and the groups with high value as a selection tree were group Ⅰ (Fig. 3).

    The selected 31 trees of Q. salicina were classified into 4 clusters based on distance level 6.0. When the clusters were represented by arranging the individuals used in the analysis in a two-dimensional space based on the principal component value, most individuals were located in the Ⅱ group. The difference in the third principal component while distributed in the negative region of the first principal component was not seen. For each taxon, clusters were classified into Jeju Island (Class I), Wando (Class II), Haenam (III), and Jindo (IV) regions, and it was found that there were differences in the selected characteristics for each region. Representatively, in the case of taxon II, the diameter and crown width at breast height was low, but the leaves showed significant morphological characteristics, whereas, in the taxon group IV, the diameter and crown width were high, but the leaves showed small morphological characteristics (Fig. 4).

    The selected 15 trees of Q. gilva were classified into 4 clusters according to street level 5.0. When clusters were represented by arranging the objects used for analysis in a two-dimensional space based on the principal component values, most of the objects were classified as positive (I) and negative (II) of the second principal component, and in the case of two objects, each grouping was confirmed. Among them, in the case of the second principal component, the height and height of the chest are the principal components, and it was confirmed that the value of group 1 as fertility water was higher when compared with other groups. As an endangered species, it was collected only within the Jeju-do region, but it is judged that the individual selection was effective due to the diverse characteristics of each species (Fig 5).

    However, when a cluster analysis for each species was performed, it was formed according to the same characteristics. It was judged that a clear distinction was not made.

    Key growth characteristics such as height and root growth are the basis for establishing a selective breeding plan (Han et al., 2007). Ahn et al. (1994) also reported that, as a result of examining the pedigree height and leaf bud characteristics of fir tree windmills, there was a significant height difference but no significant difference in leaf depth and stem length.

    It was possible to combine the above results into various groups in the selection tree only with growth characteristics such as height and DBH and a small number of growth characteristics. That is expected to result from the small number of features used in the experiment, so more morphological traits can be additionally measured, or genetic variation between groups by RAPD (Random amplified polymorphic DNA) or RFLP (Restinction fragment length polymorphism) methods will have to investigate. In addition, these selected trees need to be established as a progeny test forest and confirmed their value as a tree through in-depth observation for many years. It is judged that this study can be used as primary data for the selection of elite trees and seed orchard composition of evergreen oaks, which are in the spotlight as a species to cope with climate change or a species with high carbon absorption.

    Acknowledgements

    This study was carried out with the support of “Forest BioResource Collection, Conservation and Characteristic Evaluation” of the National Forest Seed and Variety Center.

    Figure

    JALS-56-2-25_F1.gif

    Collection area of evergreen oaks.

    a: Jinju, b: Hampyeong, c: Yeosu, d: Goheung, e: Haenam, f: Jindo, g: Wando, h: Jeju

    JALS-56-2-25_F2.gif

    Cluster analysis of Q. glauca selected trees.

    A: Cluster dendrogram, B: First and Second principal component axes. Collection area (refer to Fig. 1.) : a (Jinju), c (Yeosu), d (Goheung), h (Jeju)

    JALS-56-2-25_F3.gif

    Cluster analysis of Q. acuta selected trees.

    A: Cluster dendrogram, B: first and Second principal component axes. Collection area (refer to Fig. 1.) : b (Hampyeong), g (Wando), h (Jeju)

    JALS-56-2-25_F4.gif

    Cluster analysis of Q. salicina selected trees.

    A: Cluster dendrogram, B: first and Second principal component axes. Collection area (refer to Fig. 1.) : e (Haenam), f (Jindo), g (Wando), h (Jeju)

    JALS-56-2-25_F5.gif

    Cluster analysis of Q. gilva selected trees.

    A: Cluster dendrogram, B: First and Second principal component axes. Collection area (refer to Fig. 1.) : h (Jeju)

    JALS-56-2-25_F6.gif

    Selected Quercus spp. elite trees.

    Table

    Growth characteristics of 4 evergreen oaks selected trees

    Correlation among growth characteristics of Q. glauca selected trees

    Correlation among growth characteristics of Q. acuta selected trees

    Correlation among growth characteristics of Q. salicina selected trees

    Correlation among growth characteristics of Q. gilva selected trees

    Principal component analysis for growth characteristics of Q. glauca selected trees

    Principal component analysis for growth characteristics of Q. acuta elected trees

    Principal component analysis for growth characteristics of Q. salicina selected trees

    Principal component analysis for growth characteristics of Q. gilva selected trees

    Reference

    1. Ahn JK and Lee WY. 1994. Heritabilities of growth and wood quality characters in open-pollinated progenies of Abies holophylla Max. J. Korean For. Soc. 83(4): 480-485.
    2. Cha GS. 1998. Estimation of changes in potential forest area under climate change. J. Korean For. Soc. 87(3): 358-365.
    3. Choi EJ , Yong SH , Seol YW , Park DJ , Park KB , Kim DH , Jin EJ and Choi MS. 2021. Propagation by in vitro zygotic embryos cultures of the Quercus myrsinifolia. J. for. Environ. Sci. 37(4): 323-330.
    4. Chung JM , Kim HS , Kim MS and Chun YW. 2017. Correlation analysis between climatic factors and radial growth and growth prediction for Pinus densiflora and Larix kaempferi in South Korea. J. Korean For. Soc. 106(1): 77-86.
    5. Goodall-Copestake WP , Hollingsworth ML , Hollingsworth PM , Jenkins GL and Collin E. 2005. Molecular markers and ex situ conservation of the European elms (Ulmus spp.). Biol. Conserv. 122(4): 537-546.
    6. Han SU , Kang KS , Cheon BH and Kim CS. 2007. Realized genetic gains and heritabilities for height, DBH and volume growth in open-pollinated progenies of Pinus thunbergii. Korean J. Breed. Sci. 39(1): 15-19.
    7. Hwang JG and Kim YD. 2012. A survey low temperature damage of tea tree at South Korea in 2011. Korean J. Agric. For. Meteorol. 14(4): 246-253.
    8. Hyun JO , Lee SK and Lee KJ. 1985. Status and prospect of tree breeding research in Korea. Korean J. Breed. Sci. 17(3): 273-285.
    9. Jung SY , Lee KS , Yoo BO , Park YB , Ju NG , Kim H and Park JH. 2014. Freezing injury characteristics of evergreen broad-leaved trees in Southern Urban area, Korea. J. Korean For Soc. 103(4): 528-536.
    10. Kang BS. 2005. Multivariate statistics with SPSS. HanKyungsa.
    11. Korea Meteorological Administration2020. Korea Meteorological Administration. 2020. Korea climate change assessment report 2020.
    12. Lim JH , Chun JH , Woo SY and Kim YK. 2008. Increased declines of Korean fir forest caused by climate change in Mountain Halla, Korea. In: Oral Presentation At: International Conference Adaptation of Forests and Forest Management to Changing Climate with Emphasis on Forest Health: A Review of Science, Policies, and Practices. Umea, Sweden, FAO/IUFRO, 25-28.
    13. National Institute of Forest Science(NIFOS).2013. A study on the establishment of a foundation for nurturing promising species of warm temperate resources. Research Report No. 497.
    14. Oh BW , Jo DK , Ko SC , Choi BH , Baek WK , Kyu KY , Lee YM and Chang CK. 2010. Plants subject to adaptation to climate change on the Korean Peninsula 300. National Arboretum, Korea Forest Service. 492p. (In Korean)
    15. Park BB , Byun JK , Kim WS and Sung JH. 2010. Growth and tissue nutrient responses of Fraxinus rhynchophylla, Fraxinus mandshurica, Pinus koraiensis, and Abies holophylla seedlings fertilized with nitrogen, phosphorus, and potassium at a nursery culture. J. Korean For. Soc. 99(1): 85-95.
    16. Seo YO , Lee YJ , Rho DK , Kim SH , Choi JK and Lee WK. 2011. Height-DBH growth models of major tree species in Chungcheong Province. J. Korean For. Soc. 100(1): 62-69.
    17. Shin HC , Park NC and Hwang JH. 2006. Korea's warm temperate tree species. NIFOS Research Report. (In Korean)
    18. Son SG , Kim HJ , Kang YJ , Oh CJ , Kim CS and Byun KO. 2011. Establishment of breeding population for Quercus glauca and climatic factors. Korean J. Agricultural and Forest Meteorology 13(3): 109-114.
    19. Song KS , Choi KS , Sung HI , Jeon KS , An KJ and Kim JJ. 2015. Characteristics of seedling quality of Daphniphyllum macropodum 2-year-old container seedlings by shading level. J. Korean For. Soc. 104(3): 390-396.
    20. White TL , Adams WT and Neale DB. 2007. Forest genetics. Wallingford, UK: CAB International.
    21. You JH , Cho HW , Jung SG and Lee CH. 2004. Correlation analysis between growth and environmental characteristics in Abeliophyllum distichum habitats. Kor. J. Env. Eco. 18(2): 210-220.
    오늘하루 팝업창 안보기 닫기
    오늘하루 팝업창 안보기 닫기