Root architecture and its relation with the growth characteristics of three planted Shorea species (Dipterocarpaceae)

Root and its architecture are important for trees to support their growth and development. However, there have been still few reports on tree roots architecture of important tropical forest tree species such as Shorea species. In this study, the differences of root architecture formation and the relation of root architecture and phenotype (growth) characteristics, including Index of Roots Anchoring (IRA) and Index of Roots Binding (IRB) of S. stenoptera, S. palembanica, and S. leprosula in Gunung Walat Education Forest were examined through root growth patterns, root morphology, and root growth variables. The results showed that the root growth patterns and the root morphology of three species were similar, which were classified into Champagnat group model with monopodial typical growth and tended to form orthotrophic branching. The total of root length significantly influenced tree height and trunk volume, while the depth root significantly influenced tree height and the crown diameter. The IRA values of S. stenoptera dan S. palembanica were categorized into medium group, while S. leprosula belonged to high group, suggesting that S. leprosula has a strong root anchorage compared to other two species. The IRB values of three Shorea species were categorized into low group, indicating low root binding ability in soil.


Introduction
Shorea spp. (meranti) is an important timber tree species that has a high economic and ecological value [1]. As the largest members of the Dipterocarpaceae family, meranti dominates the lowland mixed dipterocarp forest of West Malesia [2,3]. However, Shorea population in Indonesia are decreasing due to the exploitation of the valuable timber and land conversion of the forest, where this species exists, from forests to other uses including forest plantation and shifting cultivation activity [1,4]. A total of 147 species of meranti are included in the International Union for Conservation of Nature (IUCN) Red List of Threatened Species with 94 species are clasified as critical endangered, 35 species are endangered, 9 species are vulnareble, 3 species are in low risk, 3 species are near threatened, 1 species is least concern and 1 species is extinct [5]. Based on these data, conservation efforts of these species either in-situ or ex-situ are needed. One of the successful effort of meranti exsitu genetic conservation areas in West Java, Indonesia, is located in Gunung Walat Educational Forest  (GWEF). In order to support the conservation of genetic resources of meranti in GWEF, an understanding of the biological aspects of the species such as root architecture is necessary [6].
Root and its architecture are important for trees to support their growth and development, such as for anchoring, air and nutrition absorption, and their distribution [7]. However, there have been still few reported research on tree roots architecture especially in forest tree species such as Shorea species. Therefore, the objectives of the study were (i) to determine the difference in the shape of root architecture on three meranti species (Shorea stenoptera, Shorea palembanica, and Shorea leprosula) in GWEF and (ii) to analyze the relationship of root architecture with growth characteristics (phenotypes), including Index of Roots Anchoring and Index of Roots Binding (IRB), of the three Meranti species (Shorea stenoptera, Shorea palembanica, and Shorea leprosula) in GWEF. This report provided a description of the root architecture of the three meranti species (Shorea stenoptera, Shorea palembanica, and Shorea leprosula) in the ex-situ conservation area of GWEF, as well as its relationship to plant growth characteristics.

Study site
The study was conducted from August to October 2013 in the 0,6 ha of meranti (Shorea spp) ex-situ conservation Demonstration Plot No. 1 in Gunung Walat Education Forest (GWEF) (figure 1). The demonstration plot was established in 2004 with an area of ± 1.2 ha and meranti species were planted under moderate density of Agathis loranthifolia stands [6]. The GWEF itself is a 359 ha mature plantation forest managed by the Faculty of Forestry, Bogor Agricultural University (IPB), Indonesia, since 1968 [6]. The forest is located at Cicantayan and Cibadak sub-district, Sukabumi district, West Java province, which is around 60 km away from IPB Darmaga Campus, Bogor (figure 1).  Selection of Shorea tree used for root sampling was based on purposive sampling method at meranti ex-situ conservation area. Root samples were taken from 9 individuals of 10-year-old live trees, which had the best phenotype (straight rod, stem diameter, and tree height) in each species. Furthermore, three Shorea species i.e. Shorea stenoptera, S. leprosula, and S. palembanica were selected for this study (figure 2).

Measurement of Shorea tree growth variables.
Tree growth of each Shorea was indicated by tree height, stem diameter, crown diameter and stem volume variables.
-Tree height. The total of tree height was measured from ground line to top of terminal bud using Haga Hypsometer. -Stem diameter. The data was obtained by initially measuring the tree's circumference using measuring tape. The tree circumference was measured at 1.3 m above the ground (diameter breast height/dbh) for the trees that have a height of more than 4 m, whereas the calculation was conducted at 30 cm above the ground for the trees that have a height of less than 4 m. Afterwards, stem diameter was calculated by dividing the circumference measurement by 3.14.
-Crown diameter. The data was obtained by measuring the canopy length from north to south, and canopy length from west to east was measured on each individual tree using measuring tape. Furthermore, the crown diameter was calculated by dividing the sum of the canopy length from north to south and canopy length from west to east by 2. -Stem volume. Calculation of stem volume for each individual tree was performed using the formula as in equation (1).
, π is 3.14, D is diameter (m), H is height (m) and f is form factor (0.7).

Roots excavation.
Root excavation was firsthand performed on the western side of the roots, then continued on the eastern side to prevent the fall of the tree. First, root excavation was carried out by digging the soil at a distance of 50 cm from the outermost canopy. Furthermore, the excavation was continued at a distance of every 10 cm using a small shovel to brush up to the open root ( figure 3).  Figure 3. Roots excavation process, where D is the distance of 50 cm from the outermost canopy and d is the distance of 10 cm.

Root architecture.
Roots that have been excavated were further divided into 4 quadrants and determined its north, east, south, and west direction. The division of the quadrant was undertaken by drawing a straight line with a rope from north to south, and from west to east. Furthermore, each root type (primary root, secondary root, and tertiary root) was classified using different colored ropes. Each root that has been roped was then marked by using black markers as measurement points. The measurement points were made at the place, where the root direction changes (figure 4). The more measurement points were made, the closer fabrication of 3D root diagrams to the original. In addition, root length (cm), root diameter (mm), root depth (cm), and root orientation [8] were also measured.

Index of Roots Anchoring (IRA) and Index of Roots Binding (IRB).
The diameter of the vertical root and horizontal root data on each root were required in order to calculate Index of Roots Anchoring (IRA) and Index of Roots Binding (IRB) through the equation (2).

Data analysis
Root data taken from the field were analysed using 3D Virtual Branch 1.0.3 software and Photoscape. While, root photos taken from the field were analysed using Ms. Office Picture Manager. Furthermore, the 3D root diagram and root photos of each Shorea species were analysed to determine root growth pattern, root morphology, and root characteristics. The relationship between root architecture and growth characteristics were analysed using correlation model, while the effects of relationship between root architecture and growth characteristics were analysed using regression model. Root data and growth characteristics were analysed using Ms. Excel 2007 and SPSS 20.0 software.
Analysis on soil samples was undertaken at the Forest Influence Laboratory of the Faculty of Forestry, Bogor Agriculture University (IPB). The analysis of soil chemical properties included pH, Corganic, CEC, and soil texture.

Tree growth characteristics of Shorea spp.
In this study, characteristics of Shorea species was revealed by measuring the tree growth variables (table 2). The results showed that the highest average value of tree height was in S. stenoptera (T=12.33 m), followed by S. leprosula (T=9.33 m) and S. palembanica (T=9.17 m) (table 2). The highest average value of stem diameter was in S. leprosula (θ=10.36 cm), followed by S. stenoptera (θ=9.24 cm) and S. palembanica (θ=8.81 cm). The highest average of crown diameter was in S. palembanica (θ j=651.67), followed by S. stenoptera (θ j=642.50) and S. leprosula (θ j=450.00). While, the highest average value of stem volume (V=0.06 m 3 ) were in S. stenoptera and S. leprosula and the lowest was in S. palembanica (V=0.04 m 3 ).
Shorea leprosula is a meranti species that undergoes the fastest growth when compared with other meranti in Borneo [10] such as S. platyclados, S. parvifolia, S. virescens and S. johorensis [4]. However, based on the measurement results, S. stenoptera has the highest average of tree height and S. leprosula has the highest stem diameter. These results occurred because the stand density in the plot of S. stenoptera was higher than the stand density in S. leprosula plot. The high density indicates the upper stratum of the forest canopy is closed and prevent most radiation [11,12], resulting in high competition in obtaining light between the tree communities. This light competition has a strong impact on the height growth over diameter growth in the tree community [13,14], so that individual trees under this pressure typically appear more slender than their dominant counterparts [15,16,17].

Shorea spp. root growth pattern
Root growth patterns of Shorea spp. were revealed by observing the typical root growth, axis branching pattern and orientation of the axis. The results showed that there was no difference on root growth pattern between the three Shorea spp. The typical root growths were monopodial, axis branching patterns were lateral and the orientations of the axis were orthotropic (table 3). This pattern suggested that the root architecture model of the three species was Champagnat's. The Champagnat's model is an architectural model that has pseudo-monopodial growth properties composed by a mixture of axes that are mostly oriented orthotropic. The branching occurs in the distal part, and some axes grow in plagiotropic [18]. Table 3. Shorea spp. root growth pattern.

Root morphology of Shorea spp.
Root morphology refers to the surface and branching of roots as an organ, including the characteristics of the root epidermis. The results showed that the root morphological characters of S. stenoptera were relatively hard and slippery surface, and less root hair (figure 5). Root color was 7.5 YR 8/2 (Munsell color standards), with clear-cut branching. The root system developed vertically, with the size of the primary proximal root diameter was quite large. The morphological characters of S. palembanica root were relatively hard and slippery surface, and have root hair on its surface (figure 6). Root color was 7,5 YR 7/2 (Munsell color standards), with more complicated branching. Root system developed vertically and the diameter of the root was the lowest when compared to other species. Furthermore, the morphological characters of the S. leprosula root were relatively hard and slippery surface and have a root hair on the distal part (figure 7). Root color was 7.5 YR 7/2 (Munsell color standards), with clear branching and the root system developed vertically.

Species
Typical This study investigated that the root growth pattern and root morphology of Shorea spp. were relatively the same. This pattern was supported by Tomlison [19], which reported that the shape characterizes the appearance of a biological group, meaning that the roots of the same biological group tend to have similarities in architectural form and pattern in this case.

Relationship between root architecture and growth characteristics
The relationship between root architecture and growth characteristics of Shorea spp. were analyzed using Pearson correlation model (table 4) and regression model at significance level of 5% ( figure 9). The results showed that the total length of primary root, the total length of the secondary 8 root, and the overall root length significantly affected tree height and volume. The highest correlation value was found in the relationship between the total length of primary root and tree height (77.8%), followed by the relationship between root depth and tree height (75.3%), and crown diameter (67.1%). Table 4 also evidenced that the total length of the root significantly affected tree height (67%), and tree volume (66.1%). Positive correlation values indicated that the increasing of total root length led to the increasing tree height as well as tree volume.  The coefficient of determination between total root length and tree height was 44.8%, while the coefficient of determination between the total root length and volume was 45.2%. Those coefficient of determination value showed that the relationship between each variable was low. However, the results of regression model at significant level of 5% (figure 8) denoted that there was a relationship between the total root length and height (r 2 = 44.8%) (figure 8a) and there was a relationship between the total root length and volume (r 2 = 45.2%) (figure 8b). The regression model at significant level of 5% (figure 9) also displayed that there was a relationship between the root depth and height (r 2 = 56.6%) (figure 9a) and there was a relationship between the root depth and volume (r 2 = 44.9%) (figure 9b).

Roots Anchoring (IRA) and Index of Roots Binding (IRB)
Index of Root Anchoring (IRA) is a comparison of the diameter of vertical roots and stem diameter, while Index of Root Binding (IRB) is the ratio between the diameter of horizontal roots with stem diameter [20]. The results showed that S. stenoptera and S. palembanica possessed a moderate level of IRA and low level of IRB pointed that the roots of the two species were quite capable of supporting the tree trunks to remain upright, but were not good enough to grab the soil (table 5). While, S. leprosula held a high level of IRA value and a low level of IRB value, which described that this species had strong root anchoring, but the roots were not good enough to grab the soil (table 5). This work revealed that the three Shorea species studied did not have roots that were strong enough, so the trees were prone to collapse when the tree faced an interference. The study of Abe and Ziemer [20] reported that the horizontal roots that spread in the surface layer of the soil would grip the soil and vertical roots as anchors would support the uptake of the tree so that the tree would not easily collaps by the mass movement of land.
The robustness of tree is one of the factors influencing the survival of a tree species. The level of robustness of the tree can be examined by dividing the tree height with tree diameter [21]. A high level of tree robustness value shows a low life capability because of the unbalanced ratio between height and diameter. The present study showed that S. stenoptera had the highest level of robustness (133.44), followed by S. palembanica (104.09) and S. leprosula (90.06). This pattern indicated that the trees, which had moderate level of IRA value and low level of IRB value would also have a low robustness level.

Conclusions
The root growth patterns on three meranti species (Shorea stenoptera, Shorea palembanica, and Shorea leprosula) were similar, which were classified into Champagnat group model with monopodial typical growth and tended to form orthotrophic branching. The root morphology was also similar, and the variance of root characteristic variables was not significant between species. The total of root length significantly influenced tree height and trunk volume, while the depth root significantly influenced tree height and the crown diameter. The IRA values of S. stenoptera dan S. palembanica were categorized into medium group, while S. leprosula was categorized into high group, suggesting that S. leprosula had a strong root anchorage compared to other two species. Furthermore, the IRB value of three Shorea species was categorized into low group, indicating low root binding ability in soil.