Evaluation of early growth performance of 41 clones of teak(Tectona grandis Linn. f.) at four microsites in Purwakarta,Indonesia

Asep Mulyadiana · Trikoesoemaningtyas · Iskandar Z. Siregar

Abstract Teak(Tectona grandis Linn.f.),one of the mostvalued tree species in the world, is slow-growing with a long period until it can be harvested; therefore, ensuring that only high-quality seedlings or clones are selected for planting is critical. The main objective was to determine performance and repeatability of selected clones in terms of growth and survival rates in different micro-environments. A 2-year clonal trial using 41 clones and a local seedling of teak as a control were grown at 4 microsites differing in spacing, soil fertility and alley crops to assess tree height, diameter and survival rate that was evaluated in Purwakarta, West Java, Indonesia using a randomized complete block design with four replicates of each clone at each microsite. Teak growth was influenced by clone(p ＜0.01), microsite (p ＜0.05) and clone × microsite interaction (p ＜0.01). The interaction clone × microsite resulted in several potential superior clones that differed in terms of growth rates at each microsite. On the basis of diameter only, superior clones (nos. 14,18,24,30 and 37)were identified. Repeatability estimated was = 0.84 for diameter and = 0.77 for height.Growth performance of teak trees varied among microsites. Repeatability values for diameter and height characters were high.The effect of variable growth on each clone was influenced by genetic factors, environmental factors and the interaction of genetics × environment. Microsite significantly affected growth of teak clones. Clone × microsite interaction significantly affected growth of clones and led to the growth of superior clones at each microsite.

Keywords Clonal trial · Clones · Repeatability · Survival rate · Tectona grandis

Introduction

Teak (Tectona grandis Linn. f.), is planted widely in various places for its highly prized and valuable wood for various needs, from furnitures to buildings (Soerianegara and Lemmens 1994). Because the trees grow slowly, the time until long harvesting period; thus only superior teak seedlings should be planted. Superior teak seedlings are obtained from tree breeding programs by mass selection of some teak stands to be chosen as the plus tree (Wibowo 2005). The plus trees (ortet) are then vegetatively propagated to develope clones. Generally, clones derived from the same ortet (ramet) are rarely tested in the field trials.Clones, including ramets, were produced through standardized shoot cutting methods developed by a commercial company.

Clonal testing aims to obtain superior clones according to desired characteristics (Wright 1976), in particular tree growth characteristics(Zhang et al.2003)such as diameter,height, and survival rate (Monteuuis et al. 2011; Hidayati et al. 2013). Clonal tests are set up in a randomized complete block design with a variety of treatments at various locations (Williams et al. 2002).

Locations for clonal testing of teak are generally flat and inclined areas in a valley, close to river or other water source(about 0.8 m above the water surface)with trenches near the trial(Goh et al.2013).However,in Indonesia,it is difficult tofind a trial location that is free of community participation. According to Siswamartana (2005), the community around a teak forest has a role in planting,maintenance of main trees and the alley crops, and harvesting. An alley crop is planted and maintained for shortterm needs while the teak clones are maintained as the primary trees for long-term goals. At the clone site, cash crops may be planted as alley crops among the main trees,in cooperation with the community (Chundawat and Gautam 1993).

Considering the above matters, trials of teak clones should be done in a location involving communities or forest farmers. The present study addresses how clones perform in a cropping system that involves the community with objectives to (1) evaluate the performance of 41 clones from vegetative propagation in four microsites; (2)estimate genetic parameters for growth characters of 41 clones to include repeatability, genetic correlation and genetic gain;and(3)assess the effect of microsites on each clone related to plant spacing, fertilizers and alley crops and any interaction between clones and microsites to identify superior clones.

Materials and methods

Plant materials and trial characteristics

Teak stecklings (ramets) from vegetative propagation(clone)were used for this study and derived from 41 clones that have different mother trees. Ramets were sent to the trial site from the company nursery in Central Java and in topsoil-containing polybags (10 cm × 15 cm).

The research location was developed in January 2011 in Purwakarta, West Java. In January 2013, trees were measured and observed.In general,the study sites were 300 m above sea level with a slope from about 10° to 15°. Mean monthly temperature ranged from 22 to 28 °C, and mean annual rainfall was 3093 mm year-1.

The study was set up in a complete randomized block design consisting of 4 microsites. A microsite was a block with a series of contiguous plots and its own planting characteristics such as spacing, fertilizers and alley crops.Each microsite consisted of 4 replications that had a plot comprising 41 plants and 1 plant of local identity as a control,which each one having an individual clone(ramet)as a replication. Four trees (ramet) for each replication of each teak clone and control were set up (4 replications/microsite × 4 trees/clone = 16 replications/clone/microsite). Generally, each teak clone consisted of 4 individuals (ramet) × 4 replications × 4 microsites = 64 individuals, which should equal 2688 individual teak trees total on the four microsites. However, due to inadequate field situations, the total was only 2644 individuals. The characteristics of each microsite were as follows:

1. Microsite 1 had 636 individual trees with 3 m × 4 m spacing, 3 kg basic fertilizer and rice (Oryza sativa),pepper (Capsicum frutescens), and taro (Colocasia esculenta) as alley crops.

2. Microsite 2 had 664 individual trees with 3 m × 4 m spacing,5 kg basic fertilizer and rice(Oryza sativa)as the alley crop.

3. Microsite 3 had as many as 672 individual trees with 5 × 2 m spacing, 3 kg basic fertilizer and ginger(Zingiber officinale), corn (Zea mays), rice (Oryza sativa) and peanut (Arachis hypogaea) as alley crops.

4. Microsite 4 had as many as 672 individual trees with 5 m × 2 m spacing, 5 kg basic fertilizer and ginger(Zingiber officinale), peanuts (Arachis hypogaea) and soybean (Glycine max) as alley crops.

Measurement of growth variables

Plant height was measured using pole meters from the ground to the top of plants (apical), and diameter was measured using a caliper. Individual clones teak living or dead were assessed as living or dead to determine survival rate.

Data analysis

Analysis of variance

Analysis of variance (ANOVA) was used to test for differences among clones and the interaction between genotype and environment for growth characteristics. Analysis of variance was performed for each microsite to view genotype effects. The effect of genotype, microsite and genotype × microsite interaction were analyzed in a combined analysis of variance. Genotype and microsite factors were assumed to be random so that the constructed analysis model was a model of random linear equations and structural analysis of variance (SAS Institute Inc 2006).Additive model of linear equations for the combined analysis of variance was as follows (Zhang et al. 2003):

where Xijkis the value of the ith measurement for genotype in block j at microsite k; μ is the overall mean for trait measured; Ciis theith genotype (clone) effect; Ljis the effect of microsite (location) jth; CiLjis interaction between the ith clone and jth microsite (location); εijkis random error.

Repeatability

Repeatability is ideally broad sense heritability for a trait with a value ranging between 0 and 1 (Donaldson and Burdon 1995). Repeatability (） for each clone was calculated as follows (Zhang et al. 2003):

Genetic correlation

Genetic correlation between clone traits was calculated by as follows (Zhang et al. 2003):

where r（x，y）is the clonal correlation between traits x and y;σc（x，y）is the estimated covariance between x and y for the clone;is the estimated clonal component of variance for trait x;is estimated clonal component of variance for y.

Expected genetic gain

This study assumed that selection proportions will be made as much as 61% (25 clones of the 41 clones) with a selection intensity 0.629 (Becker 1992). Estimation of genetic gain(ΔG)at trait y based on clonal selection at trait x was calculated using the formula (Falconer 1981):

where i is intensity of selection;Rxis repeatability for trait x; σyis clonal standard deviation for trait y; rxyis the genetic correlation between traits x and y.

Results and discussion

Analysis of tree growth

Individual 2-year-old teak clones varied in mean growth characters among the microsites (Table 1). Microsite 3,which had the best environment, yielded the greatest plant height and stem diameter, which had a low coefficient of variance (CV). Of the microsites, microsite 3 is also the flattest and at the highest altitude.Tree density is high,and the farmers are very active in maintaining both alley crops and the teak trees as the main plant.

The CV for the diameter and height of the tallest tree was 22.77% and 27.01% at microsite 4, apparently due to its 5 m × 2 m tree spacing and varied conditions. In addition, some farmers worked actively and others not, so the maintenance of teak clones was less uniform than at microsite 3. The activities undertaken by farmers on microsite 4 were the maintenance of teak and tillage between teak trees planted with peanuts,ginger and peas as alley crops. The application of additional fertilizer on the alley crop also affected teak growth. Some parts of microsite 4 were not planted with an alley crop due to lessactive farmer use of the land. In general, the level of maintenance by active farmers and less-active farmers at microsite 4 caused the greater variability in teak growth compared to that at microsites 1, 2 and 3.

The mean survival rate at the four microsites was above 90%(97%,98%,99%and 96%;Table 1),a good indicator in trial plantings (Mahfuz et al. 2010). According to Monteuuis et al.(2011),the survival rate was stable at 84%in the growth of teak varies between 13 and 106 months old. Survival rates of teak plant at 2 years were influenced by various factors such as level of precipitation and its absorption during the rainy season and tree ability to adapt well in the dry season when water is limited. Hence, the high mean survival rates of the teak clones indicates a tendency for good adaptability at all microsites.

Homogenity test of variances

Levene’s(1960)test for homogenity of variances is used to test for similarities in some of the population variances. In the present study, the homogeneity test of variances was carried out to determine whether the microsites have homogeneous variances for growth variables such as diameter, height and survival rate. The result of Levene’s test indicated that the variable of height and diameter at the four microsites have homogeneities of variance and demonstrated that values did not differ significantly(Table 2). Conversely, the test for survival rate indicated that the variances of microsites were not homogeneous.

Table 1 Mean values for growth variables of teak clones at four microsites Microsite 4 Microsite 3 Microsite 2 Microsite 1 Variable CV (%22.77 27.01 11.10 Range 1.26-6.41 50-100 Mean 3.74 ± 0.85b 435 ± 1.17ab 133-821 96 ± 0.17 CV (%)19.47 21.70 6.24 Range 2.15-6.20 215-737 67-100 Mean 4.02 ± 0.78a 450 ± 0.98a 99 ± 0.10 CV (%)19.78 23.82 8.57 Range 1.94-6.39 168-742 50-100 Mean 3.85 ± 0.76b 428 ± 1.02b 98 ± 0.13 CV (%)20.42 23.33 10.14 Range 0.73-5.84 90-665 33-100 Mean 4.01 ± 0.82a 444 ± 1.04a 97 ± 0.15 Diameter (cm)Height (cm)Survival rate (%)CV coefficient of variance; the same letter on variance indicates not significantly different between variables and different letters indicates significantly different of variables

Furthermore, variable values of survival rate in four microsites could not be combined, indicating data can be separated for each microsite.

Variance for plant growth and repeatability

Repeatability values for diameter,height and survival rates were 0.84, 0.77 and 0.17, respectively (Table 3).Repeatability for both height and diameter was high, suggesting that genetic factors are very dominant. Survival rate,however,had a low value for repeatability,suggesting that environmental factors are more dominant than genetic factors.

Another study on teak clones showed that the repeatability values were influenced by genetic factors, for instance,repeatability values for diameter and height of 61 clones at the age of 3.5 years were 0.37 and 0.28(Callister and Collins 2008). Hidayati et al. (2013) reported that the repeatability of the diameter and height in 15 clones of 12-year-old teak trees planted was, respectively, 0.50 and 0.39 at one site and 0.44 and 0.76 at another. For a fastgrowing poplar (Populus spp.), the repeatability value of the diameter and height at 3.5 years were affected by genetic factors, respectively, 0.79 and 0.85 at one site and 0.74 and 0.72 at the another (Zhang et al. 2003).

Our variance analysis of 2-year-old teak clones showed that the variance of microsite for diameter and survival rate differed significantly at the 1% level and for height at the 5% level. Thus, microsite influenced the growth of teak clones.

Some clones differed significantly for diameter and height, but not for survival rate because each clone performs differently depending on the mother tree; thus their differences are sighted on survival rate. Interaction between microsite and clones of these growth characteristics was highly positively significant. Accordingly, growth characteristics were affected by both genetic and environmental factors.

In this study,the coefficient of variance (CV) was quite large due to the lack of uniformity in clonal growth.Another study showed that the source of error is also quite large for teak plants grown in Australia (Callister and Collins 2008). According to Zhang et al. (2003) in a study of 3-year-old hybrid Populus, the error variance components were ＞60% due to differences in the growth environment.

In Table 4, the estimate of repeatability values at each microsite show that repeatability differed at the four microsites. Microsite 4 showed the highest repeatability values for height and diameter. Thus, the performance for these characters is most influenced by genetic factors.

Table 2 Levene’s test for homogeneity of variances of growth variables

The genetic and phenotypic correlations

Table 5 presents the genetic and phenotypic correlations among teak growth characters measured. Genetic and phenotypic correlations between height and diameter showed a strong and positive correlation. These correlations indicated that each clone has a close relation (Monteuuis et al. 2011; Hidayati et al. 2013). For selection,diameter or height traits are good criteria in tree breeding program.

Implications for tree breeding

Table 6 presents the genetic gain and estimation of response between variables. Genetic gain is improvement of the mean value for a trait from one generation to next generation as a result of selection in a breeding program(Soekotjo 2009; Hidayati et al. 2013). Genetic gain is a quantitative value of a population response to selection in population. In addition, selection is necessary to establish correlation between genetic gains obtained from an early growth stage to a mature growth stage(Wahid et al.2012).Generally,genetic gain is related to the repeatability value of each character in which the higher repeatability values will increase the expected genetic gain.

In this study, 25 clones were selected from the original 41 teak clones according to the standard for Seed Clonal Orchard in regulation No. P.72/Forest Ministry-II/2009.When the selection is based on diameter, the response ranges from 10.40 to 19.97%. When selection is based on height, the response to diameters ranges from 13.28 to 18.70% (Table 6).

The implication of this study for tree breeding programs is to select teak clones in an early thorough evaluation of the clones for genetic correlation between growth characteristics in the nursery and a wide range of clonal trials(Monteuuis et al. 2011; Wahid et al. 2012). Furthermore,the best clones will have superior growth in various microsites and are assessed by ranking the growth of 2-year-old clones at the different microsites(Fig. 1).Based on diameter, clone 14 was selected as a superior clone in two microsites and is called consistently superior. Clones 37 and 24 are superior clones in a microsite or superior for a specific site. For height, clones 37, 26, 7 and 10 ranked first in microsites 1, 2, 3 and 4, respectively. In general,clones 37 and 24 were superior for both diameter and height. Superior clones may be affected by various factors including genetic factor, environmental factor and genetic × environment interactions.

Table 3 Analysis of variance and repeatability for growth variables

Table 4 Estimation of the value of repeatability at each microsite

Our study successfully demonstrated that the performance of teak clones is influenced by the interaction clone × microsite (genotype × environment), causing distinct performance of the clones in each environment.Yu and Pulkkinen (2003) showed a significant interaction for clone × environment for 24 clones of a hybrid aspen(Populus tremula × P. tremuloides) in Finland. Baltunis and Brawner(2010)reported that the stability of clones of Pinus radiata in two regions of New Zealand and Australia could be detected based on the interaction of genotype (clone) × environment. Similarly, a 4-year-old clone of Eucalyptus globus in Portugal showed an interaction between genotype and environment (Costa E Silva et al. 2004).

In addition, microsites with different microenvironments (plant spacing, fertilizer and alley crop) may be used to test for a clone × microsite interaction, such as done for Tectona grandis (Callister and Collins 2008;Monteuuis et al. 2011; Hidayati et al. 2013; Goh et al.2013) and hybrid poplar (Zhang et al. 2003).

In the present study, many clones were superior over other clones;they excelled and adapted at each microsite.Hence, these data on teak growth can be used to select clones in breeding programs.

Table 5 Coefficients of genetic correlation between characters of teak clones (below diagonal) and phenotypic correlations (above diagonal)

Table 6 Estimation of genetic gain and percentage response by selection criterion and microsite Height (cm)Microsite 4 Diameter (cm)Height (cm)Microsite 3 Diameter (cm)Height (cm)Microsite 2 Diameter (cm)Height (cm)Microsite 1 Diameter (cm)Criterion 19.10 (19.97%)38.37 (32.01%)0.21 (21.87%)20.18 (16.84%)12.49 (13.03%)21.23 (17.71%)0.20 (20.34%)15.93 (13.28%)10.40 (10.40%)18.45 (15.39%)0.19 (33.08%)16.92 (14.11%)18.60 (15.52%)26.84 (22.40%)0.27 (28.76%)22.41 (18.70%)Diameter (cm)Height (cm)Response estimation value is in parentheses

Fig. 1 a Diameter and b height of five superior clones at each microsite

Conclusion

Growth performance of 2-year-old teak trees varied among microsites. Furthermore, repeatability for diameter and height was high and influenced by genetic factors. In addition, genotype × environment interactions were important factors for determining preliminarily several clones with superior growth characters.

AcknowledgementsWe gratefully acknowledge the funding from KPWN(Wanabakti Nusantara Cooperative for Housing).In addition,we appreciate USAID through SHERA program-Centre for Development of Sustainable Region (CDSR) for facilitating this research. Authors thank also the farmers and field assistants who helped the research and the reviewers for comments that improved the manuscript.