A contribution to the identi fi cation of charcoal origin in Brazil II – Macroscopic characterization of Cerrado species

The Brazilian Cerrado is the richest savanna in the world. It is also one of the biomes more threatened in the country and a hotspot for conservation priorities. The main causes of deforestation in Cerrado are agricultural practices, livestock and charcoal production. Although charcoal has a minor impact, its consumption represents the deforestation of 16.000 Km2 of the Cerrado. To contribute for the biomes’s conservation it is very important to improve forestry supervision. Thus, in this work we present the macroscopic characterization of charcoal from 25 Cerrado’s species. We simulate the real conditions of forest controllers by using the magnifi cations of 10x, 25x and 65x. Likewise, the charcoals micrographs are all of transverse sections due to the larger amount of anatomical information. We also analyzed texture, brightness, vitrifi cation, ruptures and some special features. The species present several differences in their anatomical structure. Although some of them are very unique, this work does not intent to identify charcoals only by macroscopic analyses. But it might give directions to future identifi cation of genera or species. It also provides knowledge for government agents to verify the documents of forestry origin by fast analyzing a sample of charcoal itself.


INTRODUCTION
The Brazilian Cerrado is considered the richest savanna in the world, but it is one of the biomes more threatened in the country (MMA 2011).It is considered a hotspot for conservation priorities; it has 4,400 endemic plants which correspond 1.5% of global endemic fl ora (Myers 2000).Also, the Cerrado sensu lato contains 11,806 plant species occurring in (Lista de Espécies da Flora do Brasil 2014) representing 5% of the biodiversity in the world (MMA 2011).
The main causes of deforestation in Cerrado are: (i) agricultural practices; (ii) livestock; (iii) charcoal (MMA 2011).Historically, the massive occupancy of the biome began in 1920 due to THAÍS A.P. GONÇALVES et al. the coffee industry; around ten years later, the government promoted grants and technical assistance to livestock farming (Klink and Moreira 2002).It resulted in approximately 50% of Cerrado's area converted to agriculture and livestock (Klink and Machado 2005).To "contribute" with this reality there is the charcoal production.
Brazil is the world's largest producer of charcoal (FAO 2012).Almost 90% of charcoal goes to the iron and steel industry (Brasil 2012).Our country remains the only one producing iron from charcoal; comparing with coal, the iron from charcoal has better quality and does not contribute to environmental pollution (ABRAF 2013).
One of the worst problems of charcoal production is the illegal cutting of native species which is evaluated 30-35% of total output (IBGE 2010, ABRAF 2012).In 2005 almost 35% of native charcoal was produced from Cerrado's species (Duboc et al. 2007).Also, the charcoal consumption represents the deforestation of approximately 1,6 million hectares or 16,000 Km² of the Cerrado (MMA 2011).
In this context, the challenges for cerrado's conservation are especially: (i) illegal logging; (ii) counterfeiting documents, such as "document of forestry origin" (DOF); (iii) reforestation highly deficit; (iv) reinforcements in/applications of environmental laws; (v) diffi cult to make viable forest management Cerrado (MMA 2011).The Brazilian Government already did a voluntary national commitment to reduce 40% of the annual rates of deforestation in the Cerrado (Brasil 2013).To accomplish these goals, it is very important to improve forestry supervision.
To help the charcoal control, in this work we present the macroscopic characterization of 25 Cerrado species's carbonized.Also, we give some explanations about how the analysis must be done.It is essentially justifi ed by: (i) importance of Cerrado; (ii) the need to identify illegal charcoal; (iii) provide knowledge for government agents to verify the DOF by fast analyzing a sample of charcoal itself.
The carbonization process lasted 5 h, with a fi nal temperature of 450°C and heating rate of 1.66°C/ min; the samples remained at the fi nal temperature for 2 h (Muñiz et al. 2012).The resulting charcoal samples were manually broken and analyzed with a Zeiss Discovery V12 stereomicroscope according Gonçalves et al. (2014).The charcoal samples were deposited at the charcoal collection of 'Laboratório de Anatomia e Qualidade da Madeira' of Universidade Federal do Paraná (LANAQM/ UFPR) in Curitiba, Paraná State.
We use the recommendations of the IAWA Com mittee (1989) for descriptions and measurements.Also, others references for macroscopic wood analysis were used (e.g.Botosso 2009, FPL 2010).The vitrification degrees were analyzed according Marguerie and Hunot (2007).
The charcoals micrographs are all of transverse section, because this section has more information about the species.Also due to practical use, as forest controllers normally have portable magnifi er glasses that do not allow higher magnifi cations needed to observe the longitudinal sections.Figure 1 has micrographs with bars of 1 mm; the magnifications are 10x, simulating the most common portable magnifier glasses of forest controllers.Figure 2 has higher magnifi cations aiming to show some details; the bars are 100 and 200 μm; the magnifi cations are 25x, 65x and 100x, simulating better portable magnifi er glasses that can be easily fi nd in specialized markets.

RESULTS
The Table I shows the most important anatomical features to aid in charcoal identifi cation.The species Lithrea molleoides, Gochnatia polymorpha, Tabebuia aurea, Terminalia glabrescens, Caryocar brasiliense, Anadenanthera peregrina var.falcata, Copaifera langsdorffi i, Microlepis oleifolia, Ocotea corymbosa, Roupala montana, Tocoyena formosa, Styrax ferrugineus and Symplocos pubescens, has the fi rst and second degrees of vitrifi cation, seen just as a high level of brightness due to the cellular walls fusion.
We present the charcoals micrographs in Figure 1, the magnifi cations used here is similar to a 10x magnification of hand lens.The most important details are in Figure 2.

DISCUSSION
The charcoals anatomy descriptions of the analyzed species, agreed with the wood descriptions of Sonsin et al. (2014) thereby we could confi rm that by comparing our results with wood anatomy analyzes.
The fi rst features that are evaluated in macroscopic analyses in charcoal are texture, brightness and vitrifi cation.Comparing to wood, in charcoal we can't analyze color, odor and taste, and it is hard to obtain large sections to defi ne grain precisely.Also, density is other property that should be measure carefully, because it depends on many factors as temperature and/or time of carbonization.
The axial parenchyma is the most important feature in wood identifi cation (Metcalfe and Chalk 1950), though in charcoal is not always easy to observe -for example in the scanty (Ocotea corymbosa) and/or diffuse parenchyma (Lamanonia ternata).Banded parenchyma observed in Ficus guaranitica and different types of paratracheal  parenchyma, in a general way are better seen because of the contrast with fi bers.In addition, vessel arrangement when present, as in Gochnatia polymorpha, frequency of vessels when in a greater quantity, as in Symplocos pubescens or lower quantity, observed in Vochysia tucanorum and, in some cases great differences in vessel diameter, associated with the other anatomical features, such as included phloem and sometimes with sensorial characteristics, give valuable information to distinguish families or even genera.Besides that, some features are known to occur just in few families or are only seen in higher magnifi cations (under microscope).This kind of characteristic can initially help the government agents to separate families or simple check the DOF ("document of forestry origin") and recognize that the species listed in the document are not the ones analyzed.For example, the presence of very large rays which are observe in transversal section, observed here in the families Annonaceae, Ochnaceae, Primulaceae, Proteaceae (Fig. 1b, r, s, t).Another example is the presence of the features as scalariform perforation plates, traumatic canals and included phloem can be very helpful to separate families and even genera when joint with other anatomical characteristics and are better observed in higher magnifi cations.
However, some of these anatomical features might cause some misguiding in charcoal identification for workers with less experience, such as the presence of included phloem in Guapira noxia.As there are several ruptures in the included phloem which seems like a deformed vessel in transverse section.Nevertheless, one can separate it because some vessels remain with their walls intact.
The growth rings were observed in 84% of the species of this study.According to the studies of Alves and Angyalossy-Alfonso (2000), Barros et al. (2006), Worbes (1989Worbes ( , 1999) ) and Worbes et al. (2003) Tropical Forest in general have growth rings.The previous authors studied species from Atlantic Forest, Amazon Rain forest, Semi-deciduous Forest in Brazil and Venezuela.The presence of growth rings demarcated by marginal bands of axial parenchyma and fi ber zones were easier to observe in charcoals because of the large size of the cells.On the other side, marginal lines/bands of axial parenchyma and proximity of axial scalariform or reticulate parenchyma are not always easily observed.Sometimes the size of the sample or presence of ruptures might raise diffi culties to observe or hide the growth rings in charcoal.For example, the growth rings in the wood of Ocotea corymbosa (Sonsin et al. 2014) could not be observed in the charcoal because of the high occurrence of ruptures -in both radial and axial parenchyma.
Most studied species have texture fi ne to medium, such as Gochnatia polymorpha, Cordia sellowiana, Lamanonia ternata, Copaifera langsdorffi i, Leptolobium elegans, Myrcia bella and Symplocos pubescens.The characterization of texture in charcoal must be done carefully.It is due the possible ruptures caused by the process of manual breaking, e.g.: in species with large vessels or abundance of parenchyma.In such cases it might give an impression of coarser textures due to the irregularities in the surface.Therefore we recommend this analysis in charcoals recently broken with the surfaces as plain as possible.
The brightness in charcoal was observed in all studied species.Nevertheless, this feature is not of great value to identifi cation because, in general, this feature is seen in most species.Also, even if in transverse section it is not so evident; the radial section is very shinny.
Vitrifi ed elements were observed in most studied species, usually in degrees 1 and 2 (low brilliance-refractiveness; strong brilliance, respectively).According to Marguerie and Hunot (2007), this feature can be recognized by fused cellular cells; and also by the contrast between the white color and dark-shades of gray.It is important to emphasize that if the charcoal is too vitrifi ed, the cellular elements will be fused, causing a formation of a non-recognizable mass as observed by the previ-ous authors.Thus, environmental controllers might think that the sample is not of charcoal.Also, the white color can be observed in remaining content inside the cells and mineral inclusions.
As in wood, some qualitative features are best seen in higher magnifi cations, e.g.diffuse parenchyma; and others in lower, e.g.growth rings.In general, most of the data observed for the studied species, were well seen in lower magnifi cations.However, the practice is essential to develop the visual acuity for the identifi cation of features on charcoal.
Despite of the studied species did not present a pattern for the occurrence of ruptures as commented in a previous work (see Gonçalves et al. 2014), we observed ruptures of the axial and ray parenchyma cells in some areas of few woods.These ruptures are expected as the parenchyma tissue is weak and it has thin walls.

CONCLUSIONS
This is the fi rst paper with macroscopic characterization of charcoal's species from Cerrado.As in wood, we must analyze macroscopic before microscopic.The species analyzed had several differences in their anatomical structure and some of them are very unique and only observed in microscopic view.Otherwise, we strongly recommend higher magnifi cations for charcoal identifi cation.The main anatomical features that are used to identify the charcoal genus are axial parenchyma type, vessels distribution and diameter and rays width.Also, it is important to emphasize that in real cases, the possibility of take samples from the fi eld to the laboratory is not always an option.The present work might give directions to future identification of genera and/or species.Thus, it also provides knowledge for government agents to verify if the species listed in the documents (DOF) are the charcoal, by fast analyzing the sample itself.We hope this work can be use in practical activities to help the conservation of Cerrado's species.
the species studied.Numbers are related to the fi gures presented above.GR = growth rings.a = fi ber zones.b = marginal lines/bands of axial parenchyma.c = semi-ring porosity.d = marked by closeness of the narrow bands of scalariform parenchyma.e = variations within growth layers in the axial parenchyma, where the long aliform confl uent parenchyma tends to form irregular bands.-= absent or indistinct.Ø= vessel tangential diameter.Fb = Fibers.WT = fi ber wall thickness.MI = mineral inclusions. = present.( ) = rare or few.-= absent.Tx= Texture.fi = fi ne.md = medium.co = coarse.fb = fi brous.The parenchyma type with (*) is best seen in higher magnifi cation.The results are presented as mean values (minimum-maximum), standard deviation.