Tree Association with Pometia and its Structure in Logging Concession of South Papua Forest

Part of forests in Papua is still as logging concession. Pometia spp. are target species, but there is still a lack of information regarding the ecological condition of those species. Thus, the objectives of this research were to describe what tree species (small and large individuals) associated with Pometia, how logging and soil properties influence the association and to analyze the structure of Pometia in term of diameter distribution. Canonical correspondence analysis (CCA) was applied to describe the association and its relationship with environmental factors (soil and litterfall). The results showed that association of small and large individuals of trees with both Pometia showed a different pattern in which the small individuals had a positive association and had certain tree species as a community. This association resulted from logging activity leading to the change in ecological conditions. Conversely, the association between large tree species with Pometia acuminata Radlk. and Pometia pinnata J. R. Forst. & G.Forst. showed negative pattern and tree species correlated with both Pometia were different. C content of litterfall had a positive correlation with large Pometia acuminata and its community from environmental factors. Furthermore, the small individuals of Pometia were dynamic as a response to logging in which a number of the small individuals of Pometia tended to increase after logging.


Introduction
Forest of south Papua is characterized as lowland area and some parts of them are still intended as logging concession (Murdjoko 2013;Kuswandi 2014;Kuswandi & Murdjoko 2015). During logging activities, ecological conditions change in term of understory species composition, tree structures, soil conditions, and microclimatic circumstances (Arbainsyah et al. 2014). In this area, tree regeneration is natural since there is no plantation program in this logged forest because the plantation is only done in ex-skid trail and ex-log yard. Therefore, understory species especially seedlings resulted from natural processes as seedling establishment (Murdjoko 2013).
Trees as target species are selectively logged in this concession area and the rest are left as remaining trees (Sandor & Chazdon 2014;. One of the target species is from genus Two species that Pometia. have been taxonomically identified are Pometia acuminata Radlk. and J.R.Forst. & G.Forst. (Kuswandi Pometia pinnata et al. 2015;Murdjoko et al. 2016). Both species are less studied concerning population and distribution. Hence, we singled out both species as the focus of this study. Furthermore, the population dynamics of in forests Pometia is generally as an impact of abiotic and biotic factors where the logging leads to alteration of conditions. Thus, the pattern of understory establishment would possibly differ from condition in the primary forest where the condition remains stable (Win et al. 2012). Besides that, edaphic factors like soil properties are responsible for providing place for growing. Parent materials have contributed to soil characteristic during the forming of soil through decay process since long time ago. Decomposition of organic matter also takes place in the soil. Furthermore, nutrients and water are stored in soil (Khairil et al. 2014;Chiti et al. 2015) Those processes can be facilitated when climatic factors supported by creating suitable circumstances. Climatic factors in the tropical rainforest can be described as microclimate and macro climate. The microclimate is mainly as a result of the condition of tropical rainforest such as moisture understory (Cicuzza et al. 2013;Sawada et al. 2015).
Biotic factors can also influence dynamics of the stand where flora and fauna have functioned as either facilitation or competition (Velho et al. 2012). In the tropical rainforest, many trees grow in the same area resulting in competition to get a place. Hence, density and basal area of trees can affect growth and even mortality of trees. The presence of other trees can also be seen as a factor that affects trees (Ruslandi et al. 2012). Even though fauna also plays important role in trees, the contribution of fauna will not be taken into account during this study.
As described above, the presence of other trees in selectively logged over forest play a crucial role in the dynamic of trees. Hence, this study described trees Pometia that had an association with in unlogged and logged Pometia forest. Besides that, the structure of based on its Pometia density between unlogged and logged forest was compared. In this study, primary forest is seen as the approach of trees before logging in which tree descriptions in both conditions were analyzed whether the tree description based on the density was similar or not. Moreover, hypothetically, after several years logged over forest condition in term of stem density, species composition, and abiotic conditions will be close to primary forest (Ding et al. 2012;Rutten et al. 2015).
To analyze that condition, canonical correspondence analysis (CCA) will be applied to figure out tree communities in both conditions The CCA is useful to take into account . tree species, plot distribution and environmental factors as integrated calculation (Ter Braak 1986;Ter Braak 1987). Over the time, alteration of tropical rainforest can take place resulting from natural and anthropogenic causes (Huth et al. 2004). In this study, the focus of alteration is as a result of selective logging as the anthropogenic cause which can change the condition of tropical rainforest. The biotic and abiotic situation of this forest before selective logging will differ from the condition after selective logging. In brief, Pometia trees in both situations will be certainly affected.
For that reason, to what extent the trees Pometia influenced by selective logging is, therefore, interesting to be investigated. The objectives of this research were: to describe what tree species (small and large individuals) associated with , how logging and soil properties Pometia influence the association, and to analyze the structure of Pometia trees in the stands in term of its diameter distribution. S has an annual rainfall ranging from about 3000-4000 mm and daily moisture was on the average between 75 and 85% % (Petocz 1989). Families of , Dipterocarpaceae Lauraceae and dominate this area . This study area is a Myrtaceae logging concession of PT Tunas Timber Lestari where the forest is isolated by Muyu and Uwim Merah River in the west and Fly River in the east, while in the northern part is mountainous area and in the southern part is an ex-timber concession. Data were collected in the unlogged forest as primary forest and logged forest consisting of one-year, fiveyear, ten-year, and fifteen-year logged forest.

Methods
Data collection and sampling Tree species in this forest were collected that were divided into four phases as seedlings, saplings, poles, and trees. Seedlings are typified to have a height less than 1.5 m. Saplings were characterized with a height greater than 1.5 m and diameter of less than 10 cm. Poles were characterized to have a diameter between 10 and 20 cm. Trees were typified to have a diameter greater than 20 cm (Forestry Department 1989). Seedlings and saplings were then grouped as small individuals while poles and trees were classified as large individuals. Plots were placed systematically using nested sampling where seedling was 2 m × 2 m, sampling was 5 m × 5 m, the pole was 10 m × 10 m and tree was 20 m × 20 m. In the primary forest, 46 plots were placed while in logged forest 120 plots were established. Data in each plot consisted of species of individuals -each was identified according to scientific name; -Number of individuals in each Number of individuals species per plot were documented; Diameter at Diameterbreast height (DBH) or 20 cm above the buttress was measured for trees or individuals > 5 cm in diameter. Edaphic factors-Soil property was soil organic matter (SOM) while litterfall was also collected in a plot with 1 m × 1 m in size for each plot to analyze C content and dried weight. Analysis of soil and litterfall to obtain the estimates of C content and dried weight was done in Laboratorium Balai Pengkajian Teknologi Pertanian Yogyakarta.

Data analysis
To analyze tree association with , Pometia canonical correspondence analysis (CCA) was applied as multivariate analysis (MVA) to see the distribution of tree species corresponding to locations of both logged forest and Scientific Article ISSN: 2087-0469 Figure 1 Location of research in South Papua. JMHT Vol. 22, (3): 180-191, December 2016EISSN: 2089-2063 primary forest. In this analysis, the variable of importance value index of tree species is the value of species as a row (m) while the column is as plot (n). Then, those were expressed as matrix m × n. Environmental factors used in this analysis were abiotic factors namely organic matter, litterfall and time after logging. The environmental factor is defined as matrix m × q. The computation was performed using R statistical program with VEGAN package version 3.3.1.
(R Development Core Team 2005; Oksanen 2013). After et al. getting CCA graph, tree species positively associated with either or were obtained from P. acuminata P. pinnata Euclidean distance between tree species in the same quadrant and either or The Euclidean distance P. acuminata P. pinnata. between them was calculated as average and confidence interval of 95% was applied to decide the positive association.
The structure of Pometia was analyzed by means of plotting density against the class diameter of DBH where the density was the number of individuals of Pometia species per -1 hectare (trees ha ). The relationship can be described mathematically as where y = f(DBH) where y is a number of -1 individuals of Pometia per hectare (tree ha ), DBH is diameter at breast height (m) and f is a function of the relationship. The function was determined using either linear or nonlinear equations. Furthermore, bias (E) and the 2 adjusted coefficient of determination (R adj) were used to be criteria to choose the best equation.

Results and Discussion
Association between pometia and tree species in unlogged and logged forest The tree species in both unlogged and logged forests showed a pattern of species association. This research grouped tree species association as seedlings and saplings as small individuals while poles and trees as large individuals. In this research, 176 tree species were recorded in both unlogged and logged forest. Then, 159 tree species were categorized as small individuals and 127 tree species were classified as large individuals (Appendix 1). The species of were and . By Pometia P. acuminata P. pinnata means of CCA, tree species and edaphic factors were plotted in Figure 2. At that time, the associations were based on the distance between other individuals of other species and both Pometia acuminata Pometia pinnata and . In small individuals (Figure 2 A), a total of 24.5% was explained as the variance of both axes in which the first axis (CCA1) was 13.4% and a second axis (CCA2) was 11.1%. The tree species were distributed in four directions of the quadrant. The both small individuals of Pometia species were on the lower left quadrant. The tree species that had a positive association with P. acuminata (the blue boxes with dashed line) were 29 tree species and the tree species that had a positive association with P. pinnata (the blue boxes with solid line) were 7 species (Table1). The Euclidean distances of those species were below 2.18 with P. pinnata. The large individuals of tree species positively associated with and (Figure 2. B.) were P. acuminata P. pinnata distributed in opposite directions, which were on upper left quadrant and lower right quadrant, respectively. The both axes of CCA explained 34.8 % of the variation in which the first axis showed 17.6 variation and second axis showed % 17.2 variation. The tree species close to (the % P. acuminata blue boxes with dashed line) were 13 species (Table 2). Those tree species had Euclidean distance under 0.99 with P. acuminata . Moreover, 47 tree species (the black boxes with solid line) had a positive association with ( Table  P. pinnata 2). The Euclidean distances of those species were less than 1.33 compared with . P. pinnata Perturbation of edaphic condition after logging Soil and litterfall conditions after logging tended to have opposite directions with a period of logged forest (Figure 2A and Figure 2B.). SOM which symbolizes soil organic matter (%) and LF_DW which denotes dried weight (g) went to upper right quadrant while time that is a period of logged forest move to lower left quadrant. On the other hand, LF_C which is C content in litter fall (%) went to upper left quadrant. The longer period of logged forest led to the gradual decrease of SOM and dried weight of litterfall. In contrast, the C content of litterfall tended to increase steadily in that period.

Structure of Pometia in unlogged and logged forest
The individuals of Pometia in the unlogged forest were distributed from small class diameter to large class diameter (green bar). In logged forest, some individuals were absent in certain class diameter. In general, the number of individuals of Pometia in unlogged forest differed from in logged forest.
In one year logged the forest, the number of individuals of Pometia declined especially the small individuals. Afterward, the number of small individuals in 5 and 10 years logged forest rose about fourfold compared to the unlogged forest. Then, in 15 years logged the forest, the number of individuals with a diameter below 10 cm were absent while the number of individuals with a diameter between 10 cm and -1 -1 20 cm appeared by about 60 ind ha and 20 ind ha , respectively. From Table 3, only equations of 1 year logged forest was not significant (P > 0.05), but the other four equations were significant (P < 0.05) for power equations. Therefore, the equations were used to obtain patterns of the structure of  Table 1 Small individuals of tree species had a positive association with both Pometia acuminata Radlk. and Pometia pinnata J.
R. Forst. & G. Forst. in which there were 29 tree species associated with P.acuminata and 7 tree species with P. pinnata.
In general, the structure of followed reverse J-Pometia shaped distribution from small individuals with a diameter below 5 cm (unlogged forest, five years logged forest and ten years logged forest) and a diameter between 10 cm up to 20 cm. The highest number of small individuals of was Pometia in ten years after logging, then followed by the number of individuals of in five years logged forest and Pometia unlogged forest. Afterward, the number of individuals of Pometia was seemingly similar for all forest types starting from class diameter 20−24 cm to 45−49 cm, only in fifteen years after logging that the individuals were present in class diameter up to 55−59 cm.
Environmental alteration, tree communities and pometia structure during post-selective logging The pattern of association was different between small and large individuals. The small individuals of P. acuminata and P. pinnata tended to be close each other along with other tree species, which have positive association (Figure 2 A). In contrast, large individuals of Pometia itself were not distributed in the same quadrant (Figure 2 B). It suggests that those large individuals of both P. acuminata and P. pinnata showed negative association. In this forest, the both P. acuminata and P. pinnata were not close together to grow (Murdjoko et al. 2016) as they had negatively conspecific   JMHT Vol. 22, (3): 180-191, December 2016EISSN: 2089-2063 association. Therefore, the tree species associated with both Pometia were different. In the tropical rainforest, individuals of same species showed nonspecific associations whether positive or negative (Nichols et al. 1999;Bagchi et al. 2010;Howe 2014;Sawada et al. 2015). The Pometia had a negative association between small and large individuals in this forest (Murdjoko et al. 2016). In logged forest, there is no plantation and enrichment program. Thus, regeneration of the logged forest is natural processes (Murdjoko 2013;Kuswandi & Murdjoko 2015). Furthermore, seeds produced by mature trees were spread out to a particular area. That process was the beginning of association development in this forest. Therefore, seedling establishment after germination grew in the different tree composition. Some seedlings survived and then benefit from certain ecological circumstances, resulting in the positive association. In contrast, some seedlings were suppressed to the environment, leading to the negative association (Zambrano et al. 2014). Thus, small individuals of both Pometia showed a positive association, suggesting that both seedlings were able to share area to grow. On the other hand, the other tree species located in upper right quadrant were a negative association with both Pometia as they have opposite direction (Figure 2 A). The positive association of seedlings was presumably established dynamically in which after logging seedling composition changed as environmental factors altered such as light availability and nutrients in the soil (Corrià-Ainslie et al. 2015;Toriyama et al. 2015;Shen et al. 2016). That can be seen in the structure of Pometia altered as respond to the circumstance change. Thus, the number of small individuals of Pometia increased in five and ten years after logging (Figure 3). Later on, the small individuals grew in fifteen years after logging. Therefore, the number of small individuals of Pometia increased during post-logging.
In large individuals, both Pometia has negative association, bringing about certain tree species associated with either P. acuminata or P. pinnata. The association of large individuals has been established before logging period. This can be said that the association of tree species with large individuals of either P. acuminata or P. pinnata was original association in this tropical forest. The tree species on the upper right and lower left quadrant (Figure 2 B) were not associated with large individuals of either P. acuminata or P. pinnata. The change of edaphic factors as a result of logging did not affect the association since the soil organic matter and amount of litterfall have upper right quadrant as direction (Figure 2 B). It is a presumption that the edaphic change in this logged forest probably affected the growth of both Pometia and other tree species that had a positive association with one of them. Therefore, research on dynamics of individuals after logging would be necessary to find out the effect of logging on remnant trees, especially Pometia.
In general, distribution of both had a similar Pometia pattern of distribution of individuals in tropical rain forest in which a number of small individuals were more abundant than a number of large individuals (Murdjoko 2013;Kuswandi & Murdjoko 2015). As a result, the natural Scientific Article ISSN: 2087-0469  Scientific Article ISSN: 2087-0469 regeneration in this forest occurred continuously. However, there is no positive conspecific association between small and large individuals of (Murdjoko et al. 2016). As Pometia a consequence, this can be a consideration in term of enrichment planting program that artificial plantation of Pometia should be close the tree species that have a positive association with .

Pometia
The structure of individuals showed a different Pometia number between a condition in unlogged and logged forest (Figure 3). The number of small individuals (diameter less than 20 cm) were higher after logging especially in five and ten years after logging. At that time, seedling establishment of benefited from the opening of canopy gap Pometia resulting from logging activities where irradiance could reach understory. Most of the early seedling establishment in tropical rainforests require the irradiance to grow (Duah-Gyamfi et al. 2014;Goodale et al. 2014;Whitfeld et al. 2014). On the other hand, the absence of larger individuals of Pometia in logged forest indicated that the logged forests are still recovering from logging impact. Some studies in tropical rainforest addressed that to recover from logging affect; it would take more than 40 years (Gourlet-Fleury et al. 2013;. Hence, based on the structure of fifteen years after logging the condition Pometia, of logged forests have not recuperated.

Conclusion
Association of small and large individuals of trees with both P. acuminata and P. pinnata showed a different pattern in which the small individuals had a positive association. The small individuals of P. acuminata and P. pinnata tended to grow closely. Hence, small tree species positively associated with both Pometia were similar. In contrast, the association between large tree species with P. acuminata and P. pinnata was different where the association showed a negative pattern. Thus, tree species correlated with both Pometia were different. The different pattern of small individuals was a result of logging impact in which ecological circumstance changed resulting in alteration of microclimate. Of environmental factors, only C content of litterfall had a positive correlation with large P. acuminata and its community. Based on the distribution of individuals of Pometia, the small individuals of Pometia were dynamic as a response to logging in which a number of the small individuals of Pometia tended to increase after logging.