BIOMASS AND ENERGY YIELD OF LEGUMINOUS TREES CULTIVATED IN AMAZONAS

Abstract Energy forests emerge as an alternative to fossil fuels for energy production. The good performance of these forests should consider the selection of fast-growing species, high biomass productivity and energy yield. The aim was to investigate growth and energy yield of Acacia auriculiformis and Acacia mangium in a short-rotation plantation in the Amazonas. The energy yield was determined on 12 trees per species, from the results of biomass, calorific value and basic density. When 9 years-old, A. mangium had the highest growth rates in height (1.9 m yr -1 ) and DBH (2.5 cm yr -1 ). The greatest biomass productivity was observed in A. mangium (33.4 Mg ha -1 yr -1 ), which was 84% higher than A. auriculiformis (18.1 Mg ha -1 yr -1 ). Basic density (0.54 g cm -3 ) and calorific value (4,400 kcal kg -1 ) showed no significant differences between species. The energy yield of A. mangium (1,317 Gcal ha -1 ) was twice as of A. auriculiformis (684 Gcal ha -1 ). A. mangium has better energy performance, compared to the A. auriculiformis , and therefore could the most recommended for the formation of energy forests in disturbed areas in the state of Amazonas.


INTRODUCTION
World's uncertainty regarding future energy availability results from political instabilities, increase of oil price, climate changes and, mainly, from the constant threat of traditional fossil fuel sources running out (HOUTAR, 2010).As consequence, several countries are investing in new renewable energy sources, like energy forests (PÉREZ et al., 2014).However, these natural sources still represent a very small part of the global energy demand.
The global carbon dioxide emission (CO 2 ) will reach values of 40 Gt in 2030 (AGÊNCIA INTERNACIONAL DE ENERGIA (IEA), 2006).Between 1750 and 2011, emission of CO 2 caused by fossil fuels burning and production of concrete represented 67% of the total emissions (PAINEL INTERGOVERNAMENTAL SOBRE MUDANÇAS CLIMÁTICAS (IPCC), 2014).Incentives to energy forests as source of energy could positively modify this scenario, because it is estimated that the process of energy production from wood results in 60% less CO 2 emissions in the atmosphere than the emissions generated by the use of fossil fuels (HOUTAR, 2010).
Plantation of energy forests could also contribute to reduction of deforestation rates.Until 2012, deforested area in the group of states known as Legal Amazonas was about 753.000 km 2 , where the Amazonas state itself contributes with 5% of this amount (INSTITUTO NACIONAL DE PESQUISAS ESPACIAIS (INPE), 2014).The municipalities of Itacoatiara, Manicoré and Maués, which occupy 4 th , 5 th and 6 th place in the deforestation ranking of the Amazonas state, are between the municipalities that had the greatest production of wood in 2012 (INSTITUTO BRASILEIRO DE GEOGRAFIA E ESTATÍSTICA (IBGE), 2013;INPE, 2014).
Selection of exotic species for plantations in the Amazonas must be considered when the intention is to reintroduce deforested areas into productive processes, with the intent to reduce the exploitation pressure on native forests.These species, in general, are tolerant to different environmental conditions, like low quality locations in terms of nutrients availability and soil compaction, like for example pasture areas, which represent around 66% of the deforested area of the Legal Amazonas (INPE, 2014).
Potential of energy forests partly depends on productivity of the woody biomass of planted species, from characteristics of wood, mainly in terms of density and calorific power, from adaptation skills of the species to the different conditions in the location and from the silvicultural methods applied.Expectation is for the species to accumulate more biomass in a short period, with high density and high calorific power of the produced wood (PÉREZ et al., 2014).
The species A. auriculiformis A. Cunn.ex Benth.and A. mangium Willd may be adequate to formation of energy forests, because of their rapid growth rate and high wood biomass production (BARROS et al., 2009;KRISNAWATI et al., 2011).However, the best understanding of these species performance in terms of energy production in the Amazonas climate and soil conditions still need improvement.In this context, the present study aimed to answer three principal questions: i) Are there differences in terms of energy production between the species A. auriculiformis and A. mangium?ii) If such differences exist, which characteristics most contribute to them? and iii) Which of the species would be the more indicated to plant energy forests in the Amazon region?Thus, aim of the present work was to investigate growth and energy productivity of Acacia auriculiformis and Acacia mangium in shortrotation forest plantations in the Amazonas state.

Location and characterization of the experiment
The studied plantations are located between coordinates 03° 14' S and 60° 13' W. This area belongs to Embrapa Amazônia Ocidental, located in the Experimental Land of Caldeirão, Iranduba, AM.Annual average temperatures and rainfall, in the period from 1971and 2008were, respectively, 25.9 °C and 2,619 mm (ANTÔNIO, 2008) ) and climate, according to Köppen classification, is Afi type.Soil of the region belongs to the class of Yellow Latosols (ABREU et al., 2012).
Plantations of A. auriculiformis and A. mangium were made between November 2002 and February 2003.To do that, seven years old plantlets, with an average height of 30 cm produced in the nursery of Embrapa Amazônia Ocidental.Overall, 200 plantlets were planted per each species in 0.12 hectares plots (30 x 40 m).Spacing was 3.0 x 2.0 m.Plantlets were fertilized with triple superphosphate, only at the moment of plantation.After 30 days, plantlets went through replanting, and grass competition was controlled yearly.

Sampling
With the intention to define the number of trees to be sampled, DBH diameters of all trees where measured (DBH, measured at 1.30 meters from the ground), thus conducting a pilot inventory.Every single tree was considered as sample unit.Later, each plantation was divided into three DBH classes, lower (4.8 -10.6 cm), middle (10.6 -16.4 cm) and upper (16.4 -22.2 cm) for A. auriculiformis and lower (6.2 -17.2 cm), middle (17.2 -28.2 cm) and upper (28.2 -39.2 cm) for A. mangium.In each class, four trees were chopped, totalizing 12 trees from each species.

Growth in height and diameter
Growth variables considered were total height and DBH of the 24 selected trees.Height was measured after logging and values were obtained by a metric measuring tape with 1.0 cm resolution.DBH was measured by a diametric measuring tape, with 0.1 cm resolution.The average annual increment (AAI) in height and diameter was calculated for a period of 9 years.

Biomass
Alive biomass above the ground was quantified by a destructive method.Biomass components of trees were divided into leaves, thin branches (Ø < 10 cm), thick branches (Ø ≥ 10 cm) and stem.Material extracted from leaves and thin branches was, respectively, 3.0 and 5.0 kg of biomass, while thick branches and stem gave discs 5.0 cm thick, cut at 0% (base), 50% and 100% (top).Fresh biomass of each section was measured in field using a weighting scale type Micheletti MIC-2 with maximum load of 300 kg and a precision of 200g, calibrated before use with sampling weights.To determine dry biomass, materials were put in a forced ventilation oven with controlled temperature at 100 -105 °C until mass was stable.Next, samples were weighted with a digital scale with 30 kg maximum load and precision of 1.0g.Basing on fresh biomass data of the aliquots, for each component of the logged trees, dry biomass was calculated with the following equation (TÉO, 2009): where : B s (kg) = Dry biomass, B f (kg) = Fresh biomass.

Determination of calorific power and basic density
From the 24-logged trees, 5.0 cm thick discs were extracted to study the calorific power.These discs were taken from DBH height.Discs were wrapped in paper bags and taken to the Laboratory of Forest Products belonging to Embrapa Amazônia Ocidental.Next, from each disc, three cubes were extracted in the sapwood-to-core direction, weighting 0.5g, as recommended by the calorimeter's manual.These cubes allowed determination of the superior calorific power, using a calorimetric pump Alemmar type Cal 2k.
To study basic density, 5.0 cm thick discs along the stem were collected, sectioned at 0% (base), 50% and 100% (top).Disks were wrapped in paper bags and taken to the Laboratory of Forest Products belonging to Embrapa Amazônia Ocidental.In this laboratory, two wedges were cut from the opposite sides of each disk.Next, basic density of bark and wood components was determined basing on NBR 11.941 (ASSOCIAÇÃO BRASILEIRA DE NORMAS TÉCNICAS (ABNT), 2003).

Estimation of energy production and energy yield
Estimations of energy production and energy yield of A. auriculiformis and A. mangium were calculated basing on the following equations (CINTRA, 2009): where: PE (Gcal) = Energy production; PdE (Gcal ha -1 ) = Energy Yield; Bs (kg) = dry biomass; PCS (kcal kg -1 ) = Superior calorific power.

Experimental design and statistical analyses
Experimental design was completely randomized (CRD), with 2 treatments and 12 replications.Data were submitted to Shaphiro-Wilk and Bartlett's tests to verify respectively normality and homogeneity of variances.Next, data were submitted to analysis of variance (Anova One Way), and all the statistical analyses were performed using the program R (version 2.15.1).

RESULTS AND DISCUSSION
Height and diameter growth Acacia mangium, when 9 years old, presented the greatest DBH and height growths.This species reached an average DBH value 1.6 times bigger than A. auriculiformis (p<0.05;Table 1).The height and DBH average annual increments (AAI) of A. mangium were, on average, 33% higher than results obtained by A. auriculiformis (Table 1)   Table 1.Growth of Acacia auriculiformis and Acacia mangium in 9-years-old plantation in Iranduba, AM.Mean ± Mean standard error (n = 12).Tabela 1. Biometria de Acacia auriculiformis e Acacia mangium em plantio de 9 anos em Iranduba, AM.

Species
Means followed by the same letters in columns are not statistically different (Anova One Way, p>0,05).
This way, despite the difference between A. mangium and A. auriculiformis height and DBH results, both species proved to be adequate to plantations in the Amazonas climate and soil, since they presented significantly bigger growth rates compared to some native species indicated to formation of energy forests in this region.Forest tree species recommended for creation of energy production plantations must have, besides other characteristics (e.g.growth rate in monocultures, easy propagation, resistance to pests and diseases), a great growth potential.For species in field, growth potential is related to their efficiency in capture and use of primary resources like water, CO 2 , light and nutrients, and this efficiency is peculiar to every single species (SANTOS JR. et al., 2006).For this reason, one can observe different growth rates between different trees, even if they belong to the same genus.

Biomass
The biggest biomass stocks, considering all components, were observed in A. mangium (p<0.05;Table 2), being these stocks 2 times bigger compared to A. auriculiformis.The biomass annual average increment of A. mangium was 84% bigger than of A. auriculiformis.
Table 2. Biomass in the tree components of Acacia auriculiformis and Acacia mangium from 9-yearsold plantations in Iranduba, AM.Mean ± Mean standard error (n = 12).Tabela 2. Estoques de biomassa nos compartimentos arbóreos de Acacia auriculiformis e Acacia mangium em condição de plantio no município de Iranduba, AM.Média ± Erro padrão da média (n = 12).Difference in biomass accumulation between these two species may be related to adapting strategies used by each of them, because besides environmental conditions, other factors may determine biomass accumulation in forest plantations, like for example genetic potential of the species.There are differences between species in terms of biomass production efficiency per unit of nutrient absorbed, and this fact may be related to phenotypic plasticity aimed to help the performance of some species in locations with low availability of resources (SANTOS JR. et al., 2006).
Comparing biomass stocks of A. auriculiformis and A. mangium in the present study, with data from other works (KUMAR et al., 1998;SIREGAR et al., 2008), one can verify that results obtained are within the range of variation presented in literature.However, biomass stock of A. auriculiformis in the present work, is similar to biomass stocks accumulated by this species in poor soils, which allows to infer that its biomass stock, in the conditions of soil and climate of the Amazonas, was low, the opposite of A. mangium.
Both species presented significant differences in biomass distribution between the different tree components.Leaves represented the smallest stocks, and stems gave the greatest stocks (p<0.05; Figure 1).Biomass of leaves was on average 20 times smaller than stem biomass, twice bigger than thin branch biomass and none of the species presented thick branches (Figure 1).Figura 1. Distribuição percentual de biomassa nos diferentes compartimentos arbóreos de A. auriculiformis e A. mangium, aos 9 anos, em plantio em Iranduba, AM. Figure 1.Distribution of percentages of biomass in different components of the A. auriculiformis and A. mangium of 9-years-old plantations in Iranduba, AM.
Distribution percentage of biomass in the different tree components of the two species was similar.In A .mangium, 83 % of the total biomass was represented by stem, while 13% by thin branches and just 5% by leaves.Around 88% of the entire A. auriculiformis biomass was accumulated in stems, while 8% was in thin branches and 4% in leaves.Thus, the predominant biomass distribution order, in the different tree components of both species, was the following: stems > thin branches > leaves (Figure 1).
Plantation of energy forests must consider biomass distribution strategy of the species, besides growth rates (PÉREZ et al., 2014).Forest species with greater investment in the woody components as biomass distribution strategy, especially in stems, are better to form energy forests, because this component presents lower water contents, lower ash content and higher density, improving their energetic performance (PÉREZ et al., 2014).
Means followed by the same lower case letters in lines and upper case in columns are not significantly different respectively between components and species (Anova One Way, p>0,05).