Effect of age and spacing on biomass... EFFECT OF AGE AND SPACING ON BIOMASS PRODUCTION IN FOREST PLANTATIONS

1 Received on 23.11.2017 accepted for publication on 19.06.2018. 2 Universidade Federal de Santa Maria, Departamento de Engenharia Florestal, Frederico Westphalen, RS-Brasil. E-mail: <eloyelder@yahoo.com.br> and <otomarcaron@yahoo.com.br>. 3 Universidade Federal do Paraná, Departamento de Engenharia e Tecnologia Florestal, Curitiba, PR-Brasil. E-mail: <dimas.agostinho.silva@gmail.com>. 4 Universidade de São Paulo, Programa de Pós-Graduação em Engenharia de Sistemas Agrícolas, SP-Brasil. E-mail: <elvisfelipeelli@yahoo.com>. 5 Universidade de São Paulo, Programa de Pós-Graduação em Fitotecnia, SP-Brasil. E-mail: <felipe_schwerz@hotmail.com>. *Corresponding author.


INTRODUCTION
Over the last decades, the demand for energy has mostly been met by non-renewable sources, which results in a variety of difficulties regarding energy supply and the maintenance of the balance between economy and environment.Therefore, many countries are looking for alternatives that may potentially mitigate these problems, particularly by intensification in the production of renewable sources, including forest biomass (Carneiro et al., 2014).
Forest biomass here is referred to as the quantity of total plant mass occurring in a forest, and it can be determined by two different approaches, destructive and non-destructive.Destructive methods (also referred to as 'direct' methods) use measurements on the plant material directly in the forest, upon harvesting.Nondestructive methods ('indirect') employ models that relate the variable 'biomass' with other variables commonly measured, e.g. during forest inventories (Sanquetta et al., 2014a).According to Schikowiski et al. (2013) the direct method of weighing all materials is, however, the most accurate approach.
Studies on forest biomass have several purposes; among them is the assessment of forest growth to investigate energy production potential.Climate changerelated issues caused by increased concentrations of greenhouse gases, particular ly carbon dioxide (CO 2 ), have also attracted interest owing to biological carbon sequestration and CO 2 fixation by forest plants (Sanquetta et al., 2014b).
Currently, 8.2% of the energy produced originates from wood and charcoal, according to official sources of energy connecting area (Brasil, 2016).Biomass as a source of energy is typically used by developed countries, which employ advanced and highly efficient technologies with low emissions.In contrast, underdeveloped countries primarily rely on nonrenewable energy sources, and are thus responsible for a substantial amount of pollution, which is also because of the inefficiency of their energy generation methods.
The advances achieved in Brazilian silviculture regarding the use of forest biomass for energy generation are promoted by the beneficial edaphoclimatic conditions.In this context, the role of short-rotation plantations as a means of producing biomass should be emphasized, taking economic, social, and environmental dynamics into account.
Decisions on silvicultural treatments and forest management aiming at the production of biomass for energy depend on the final wood processing method, the choice of suitable genetic material, planting density, and the rotation period.One of the most important aspects to be considered in the formation of forest stands for energy generation purposes is the spacing of individual plants.
Plant spacing is a complex aspect from a silvicultural, technological, and economic point of view, as it affects growth rates, cutting age, wood quality, silvicultural practices, and, consequently, production costs (Caron et al., 2015).Therefore, defining the spacing for forest planting is very important, considering its influence on the growth rate and the amount and quality of raw material, and thus on production costs.Optimal spacing is therefore defined by the capability of producing the highest product amount in the desired size, shape, and quality; however, optimal spacing depends on the plant species, site characteristics, and genetic potential of the reproductive material used (Eloy et al., 2015).
When targeting timber production for energy purposes, typically, dense spacing is recommended to produce the largest quantity of biomass per unit area in the shortest amount of time possible (Eloy et al., 2017a).Therefore, forestry management strives to identify species with sufficient environmental plasticity, high productivity, and superior energy-generation properties.
In this context, the objective of this work was to determine the effect of age and plant spacing on biomass production capacity in four forest species, distributed in four spacings, at three different ages after planting.

Study area
This study work was carried out in an area belonging to the Federal University of Santa Maria (UFSM/FW; 27°22' S; 53°25' W), at 480 m altitude, in the city of Frederico Westphalen, Rio Grande do Sul.
The climate in this region is rated as 'Cfa,' according to the Köppen climate classification system: a humid subtropical climate with an average annual temperature of 19.1 °C (ranging from 0 °C to 38 °C) and an average annual rainfall of 1606 mm.The experimental area is approximately 30 km from the city of Iraí, to which this climate data refers to.Iraí has an average annual temperature of 18.8 °C, and average temperature of 13.3 °C in the coldest month (Maluf, 2000).
The experiment was set up in a randomized complete block design, arranged in a 4 × 4 × 3 factorial experiment, i.e., four forest species (Acacia mearnsii De Wild, Eucalyptus grandis W. Hill ex Maiden, Mimosa scabrella Benth, and Ateleia glazioviana Baill), four spacings (2.0 × 1.0 m, 2.0 × 1.5 m, 3.0 × 1.0 m and 3.0 × 1.5 m), and three periods after planting (1 st , 3 rd , and 5 th year).The experiment were performed in three replicates in subdivided plots, where each plot was represented by the spacing and the species, and the subplot by the age of the plants.The block contained 16 experimental units, with 45 plants distributed in 5 rows in each unit.The experimental units were divided into three sections, with one being evaluated at each age.
The soil of the area (located within the mapping unit of Passo Fundo) is classified as typical dystrophic red latosol, with a well-drained clayey texture (Embrapa, 2006).The experimental area was in a lowland region with good soil deposition.In preparation for planting, subsoiling and harrowing operations were carried out.Planting was done manually in September 2008, and each seedling was fertilized with 150 g of nitrogen, phosphorus, and potassium, in formulation 8-24-12.

Sampling
The destructive evaluations were carried out in the 1 st year (2009), 3 rd year (2011), and 5 th year (2013) after planting.A total of 144 trees was selected in each year of evaluation, corresponding to 36 trees per spacing.From these trees, six discs with approximately 2 cm thickness were sampled at the following positions along the stem: 0% (base), 1.30 m (diameter at breast height -DBH), 25%, 50%, 75%, and 100% of the total height of the tree.The discs were numbered according to their relative trunk position and experimental location.After this, they were packed in plastic bags and transported to the Laboratory of Agroclimatology of UFSM/FW, where they were marked, the bark and wood were separated, and two opposite symmetric wedges of each disc were cut.
Branch and leaf samples were collected in a stratified way, i.e., in the lower, middle, and upper stratum of the tree canopy, in order to obtain homogeneous samples representing the whole extension of the crown.The samples were labeled and dried in greenhouses under circulating air exchange to produce dry matter.Samples of wood, bark, leaves, and branches were dried at 103 ± 2 °C to constant mass.
Advanced leaf senescence during the evaluation periods (i.e., September) did not permit leaf sampling of Ateleia glazioviana.

Determination of biomass
For determination of the different biomass components, viz., wood biomass, bark biomass, branch biomass, and leaf biomass of different forest tree species distributed in the different spacings at different age classes, a direct method was employed, which consisted of harvesting and weighing of the different components (Sanquetta, 2002).The total fresh mass of each sampled tree was determined in the field, and moisture of each component was measured in the laboratory.
The dry biomass of the aerial part, in t ha -1 , was calculated considering the population density of each spacing, with a survival rate of 100% in the 1 st and 3 rd year, and 94% in the 5 th year after planting, as confirmed by the experimental conditions.

Data analysis
Statistical analyses were performed using the Statistical Analysis System (SAS, 2003) software.We performed an analysis of variance, and an F-test; the assumption of normal distribution was tested by Shapiro-Wilk's test, heteroskedasticity was tested using Bartlet test; a regression analysis and Tukey's test were performed, all at 5% error probability.

RESULTS
In the analysis of variance, we observed a significant difference in the biomass production by wood, bark, branch, leaf, and total biomass, depending on the three analyzed factors: year, species, and spacing.In the same way, this characteristic was confirmed for all tested interactions, including species × spacing; year × species; year × spacing; and year × species × spacing, for all variables.Taking into account the simple effect analysis and evaluating the influence between the three factors, a significant difference was found in all observations, except from bark biomass in the first year (Table 1).

ELOY E et al.
Eucalyptus grandis produced a larger quantity of leaf and total biomass in all spacings in the first year, compared with that in the other species.Also, this species produced larger quantities of biomass in terms of wood, bark, and branch; however, this was not statistically different from those produced by Acacia mearnsii and Mimosa scabrella, except for in the 2.0 × 1.0 m spacing of Mimosa scabrella (Figure 1).
In the 3 rd year after planting, Acacia mearnsii produced the largest quantity of branch biomass in all spacings.Furthermore, Eucalyptus grandis produced the largest quantity of biomass in terms of wood (91.0 t ha -1 ), bark (16.3 t ha -1 ), leaf (17.4 t ha -1 ), and total (142.9t ha -1 ) in the densest spacing (2.0 × 1.0 m), compared with those in the other species.The species Ateleia glazioviana showed the lowest values regarding wood, bark, and total biomass for all spacings, which was not significantly different from Mimosa scabrella in the 3.0 × 1.0 m and 3.0 × 1.5 m spacings.Branch biomass, however, was superior to Mimosa scabrella in all spacings (Figure 2).
Where: (*) Indicate statistical significance, and ( ns ) indicates non-significance at a 5% probability level of error, according to Fisher's distribution, (-) indicates no evaluation was performed.

Effect of age and spacing on biomass...
A direct relationship of planting density and biomass distribution in the different components was found in the analyses of the means of total aerial biomass of the 1 st year (Figure 1), 3 rd year (Figure 2), and 5 th year (Figure 3) after planting: higher density treatments showed the highest values of biomass, compared with that in the low density treatments.
The significant models for biomass production of the four forest species along the three evaluated periods and biomass proportion in relation to tree age is presented in figure 4. In general, an increasing trend of biomass production over the years was found, mainly in terms of wood and total biomass.
Eucalyptus grandis showed the greatest development of wood biomass, compared with that in other species and other biomass components, varying from 25.8% to 83.7% from the 1 st to 5 th year, respectively.In contrast, leaf ratio decreased with stem growth, from 46.3% to 5.6%, respectively, over the study period (Figure 4a).
Where: Means followed by the same lowercase letters do not differ between spacings, in the respective species; Means followed by upper case letters do not differ between species, in the respective spacings, at 5% of error probability according to Tukey's test.
Acacia mearnsii (Figure 4b) and Mimosa scabrella (Figure 4d) showed similar characteristics of proportions over time; however, the differences were less pronounced than that in Eucalyptus grandis.Regarding wood production, a variation of 35.9% to 61.1%, and 30.4% to 69.6%, respectively, was found from the 1 st to 5 th year.In contrast, Ateleia glazioviana (Figure 4c) showed a decrease in the proportion of wood biomass over time, which was low until the 3 rd year, with a variation from 79.9% to 46.8%, and increased until the 5 th year (71.1%).

4.DISCUSSION
It was observed that the spacings tested affected biomass yields of the aerial parts of forest trees.Our results are corroborated by the findings of various Where: Means followed by the same lowercase letters do not differ between spacings, in the respective species; Means followed by upper case letters do not differ between species, in the respective spacings, at 5% of error probability according to Tukey's test.The average values of biomass produced in this work were lower than those reported in previous studies from the stands of younger ages.Lima et al. (2011) found an average biomass yield of the Eucalyptus Where: Means followed by the same lowercase letters do not differ between spacings, in the respective species; Means followed by upper case letters do not differ between species, in the respective spacings, at 5% of error probability according to Tukey's test.Effect of age and spacing on biomass... benthamii stem of 416 t ha -1 , at 6 years of age.Also, Brito et al. (1983) found biomass productions of 405.6 t ha -1 Eucalyptus Salina, and 518.2 t ha -1 for Eucalyptus grandis, at 10 years of age.
The lower initial biomass production of Ateleia glazioviana than that in the other species, both in terms of period and spacing (Figures 1 and 2), is because of its growth, which is considered to be slow or delayed, according to Carvalho (2003).However, due to its characteristic bifurcations and large canopy, this species produced a considerable amount of branch biomass (Figure 2).This variation is influenced mainly by the greater area available to the canopy in less dense plantings.
The highest values of biomass observed in treatments with higher planting densities confirm the results of Botelho (1998) and Oliveira Neto et al. (2003): relating planting density with the distribution of biomass in the components of the different species, we found higher biomass production per unit area in the smaller planting spacing, which was mainly owing to the greater number of individual plants.
Thus, decreasing trends of biomass production can be observed in the different parts of the plants as a function of increasing available area for plant growth.However, Müller et al. (2005) reported that over time the amount of wood in trees planted in different spacings tends to equalize, owing to growth stagnation in the denser plantations at younger ages and wider-spaced plantings at more advanced ages.
Wood biomass represented the largest proportion of biomass production.The increase of this proportion was observed throughout all age groups, except for in Ateleia glazioviana, which showed an irregular growth pattern than the other species.In relation to branch, leaf, and bark, an oscillation in the participation of these biomass components was observed throughout the age classes; however, there seemed to be no evidence of uniformity, although a trend of their respective proportions to decrease with stem growth was observed (Figure 4).
The more pronounced development of wood in Eucalyptus grandis compared with that in other components and species is in line with the results of Vieira et al. (2012).Their study described an increase of growth in these components, reaching more than 85% of the total aerial biomass at 6 years of age.
Correspondingly, Rondon (2002) observed that the majority of the aerial biomass of trees is within the trunk portion.Poggiani et al. (1983) and Pereira et al. (1984) found that on average 85% of the aerial biomass of Eucalyptus saligna at 8 and 9 years of age comprised of wood and bark, whereas the rest was accounted for by the canopy.Moreover, Santana et al. (1999) found an even more pronounced bias than those in the earlier studies, with 88% to 92% of the biomass concentrated in the stems of Eucalyptus saligna and Eucalyptus grandis at five sites with trees above 6.5 years of age.Similar results were found in Eucalyptus spp.by Leles (1995) and Barichello et al. (2005), who confirmed the relative contributions to total biomass by the different components in the following order: wood > branch > leaf > bark.Schumacher (1992) argues that during the initial development phase of a forest, much of the total carbohydrates is used for the production of biomass of the canopy, but over time, when individual canopies begin to compete with each other for space, biomass production in the trunk increases and that of the leaves and branches decreases gradually.However, Schikowiski et al. (2013) and Eloy et al. (2017b) emphasize the importance of the other components besides the stem in order to assess the total biomass of a tree and its variation, and to identify the potentially influencing factors.

CONCLUSIONS
Biomass production of different forest tree species is influenced by plant spacing and depends on the age after planting.
The increase in planting density is directly related to biomass production per unit area, i.e., denser spacings yield larger quantities of biomass than the less-dense plantings.
Forest species differ regarding biomass production.Eucalyptus grandis produced the largest quantity, followed by Acacia mearnsii, Mimosa scabrella, and Ateleia glazioviana.
Age affects the production and proportion of the biomass components.
Wood biomass represents the largest proportion of biomass production over time, with increasing percentage proportions throughout all age groups, followed by those of the branch, leaf, and bark.

ELOY E et al.
For a greater production of aerial biomass within a shorter duration, it is, thus, recommended to plant Eucalyptus grandis in spacings of 2.0 × 1.0 m.

Table 1 -
Analysis of variance of wood, bark, branch, leaf, and total biomass, of the four forest species, distributed in the four spacings, in the 1 st , 3 rd , and 5 th year after planting.