Introduction

Although in the long term the worldwide demand of tobacco generates many questions due to the health risks that its consumption entails, its production is the means of sustenance of a large number of families in the underdeveloped countries of Africa and Latin America (^{Siddiqui, 2001}; ^{Jimu et al., 2017}). In Colombia, this crop has over a century of history, and, at the national level, approximately 6 000 ha of land are planted with tobacco (^{MADR, 2016}). The department of Santander represents 40 % of this area, followed by Huila and Boyacá (36 %) (^{MADR, 2016}).

For drying the tobacco leaves, the small producers in the Third World countries use artisanal kilns fueled with firewood (^{Siddiqui, 2001}; ^{Jimu et al., 2017}). This flue cured or “barn-style” drying method is the most common in Santander, where Virginia tobacco is produced for the manufacture of cigarettes (^{Cerquera and Pastrana, 2014}; ^{MADR, 2016}).

Given their artisanal design, these kilns, fueled with firewood or charcoal, have a very low efficiency level (below 5 %); furthermore, they have unstable temperatures above the recommended values and an uneven dehydration of the leaves (^{Cerquera and Pastrana, 2014}; ^{Munanga et al., 2014}).

Considerations regarding the preservation of the native forests and the environmental pollution in tobacco-producing regions were unattended for several decades (^{Jimu et al., 2017}). I.e., according to the estimates, between 1.2 and 2.5 million hectares of forests are required every year worldwide for the purpose of curing tobacco; this represents 2 to 4 % of the global deforestation (^{Hu and Lee, 2015}). Due to environmental restrictions for coal combustion and to the higher cost of other systems for the small tobacco producers, it may be assumed that firewood will continue to be the main fuel.

Although in Colombia approximately 60 million hectares (52 % of the country’s territory) are covered with forests, deforestation rates are high (^{Armenteras et al., 2013}). Field identification and evaluation of the species utilized in traditional tobacco kilns may contribute to reforestation plans and to the development of technologies for improving their efficiency.

Based on the above, the present study had the following objectives: 1) to identify the species and quantify their calorific power when utilized as firewood at the use sites, and 2) to evaluate the initial in-field growth of eight native and exotic forest species with a high dendroenergetic potential.

Materials and Methods

The study comprised two phases, conducted during the years 2010 and 2011. In the first phase, the species utilized as firewood in traditional kilns were identified, and their gross combustion heat was quantified; the second phase corresponded to the assessment of eight forest taxa with a high dendroenergetic potential in three municipalities of Boyacá and Santander.

Initial growth of eight tree species with dendroenergetic potential

Once the surveys were conducted, two experiments were established in San José de Miranda and Enciso municipalities, department of Santander and Soata, department of Boyacá (Figure 1). The selected species stand out to a greater or lesser extent for their dendroenergetic potential, for the shade that they cast over the grasslands, for serving as living fences, and for being an additional source of food for animals and humans (^{Pérez et al., 2011}; ^{Olivares et al., 2011}; ^{Pinto et al., 2014}; ^{Díaz et al., 2014}; ^{Chitra and Balasubramanian, 2016}; ^{Márquez et al., 2017}). The treatments were prioritized with the participation of the local communities.

Municipio = Municipality; Parcela = Plot; Coordenada geográfica = Geographic coordinate; Altitud = Altitude; Unidad experimental = Experimental unit;

Figure 1 Geographic location of the established plots and field arrangement of the experiment (experimental unit). 

The study region has a tropical equatorial (Af) climate, according to Köppen’s classification; the rainfalls are concentrated in two periods: from April to May and from September to October; i.e., the region has a bimodal season regime. The utilized records of monthly temperatures and precipitations obtained through the Instituto de Hidrología, Meteorología y Estudios Ambientales, IDEAM (Institute of Hydrology, Meteorology and Environmental Studies) located at a distance of 5 to 10 km of the plots, were historical and specific for the study period (Figure 2).

Temperatura = Temperature; Precipitación = Precipitation

*No temperature record was available.

Figure 2 Climate rainfall and precipitation records and monthly temperatures of the study period (Dec_2010 to Apr/2011) and historical records (1979-2009 series) for San José de Miranda and Enciso (a), and Santander and Soata (b), Boyacá municipalities.  

As for the soils, they are characterized by medium to low contents of organic matter, exchangeable bases and available phosphorus; and neutral to lightly alkaline pH levels. According to the soil taxonomy of the United States Department of Agriculture (USDA) and FAO, they are classified as ustropepts, ustorthents and dystropepts (^{León and Coronado, 2003}; ^{Rodríguez et al., 2012}).

Results and Discussion

The availability of a particular species was proven to depend on three aspects: 1) offer by the small foresters of cold areas (> 2 000 masl); 2) utilization of the trees scattered within the tobacco-producing plots as living fences, and 3) firewood collected from Chicamocha, Servita and Tunebo rivers. Furthermore, we identified 20 multipurpose species used (Table 1) to fuel the kilns, among which E. globulus and E. grandis (57 %) are the most prominent, due to their easy availability and because of their provenance from the Concepción, Málaga and San Miguel municipalities, in Santander, and Soata, Boavita, El Espino and Susacón, in Boyacá.

The tropical region includes a wide variety of native taxa with a high timber potential for the recuperation of natural forests and the preservation of their biodiversity. However, silviculture in the country has focused on only a few species, usually on Eucaliptus spp. and Acacia spp. (^{Breugel et al., 2011}; ^{Zhang et al., 2016}). The present study proved Eucaliptus spp. to be the most frequently used. This accounts for the existence of small plantations in temperate and cold areas very close to the tobacco-producing zones, which makes them easy to acquire, compared to mineral coal obtained from more distant municipalities. Other utilized taxa were P. dulce (48 %), E. pendula (12 %), M. tinctoria (9 %), C. pittieri (8 %) and F. chinensis (8 %). In general, all the logs with a medium length of 50 to 80 cm and thickness of 0.5 to 20 cm are utilized in the process of curing tobacco 20 to 30 days after felling.

The residues of the materials left from the fires are incorporated to the prairies and to the crops after at least one year of decay.

The GCH ranged between 12.55 MJ kg^-1, for C. sinensis, and 18.77 MJ kg^-1, for E. pendulla. Its mean value for the rest of the species was 16.8 MJ kg^-1 (Table 1). ^{Pérez et al. (2011)} obtained values of 17 to 20 MJ kg^-1 for plantations of various eucalyptus species. The lower figures estimated in this study for E. globulus (15.91 MJ kg^-1) and E. grandis (17.55 MJ kg^-1) may be due to the acquisition of lower-quality or precocious logs by the producers.

Likewise, a study conducted in the Reserva de la Biosfera Selva El Ocote, (El Ocote Rainforest Biosphere Reserve) in Chiapas, Mexico, concluded that there was no relationship between the firewood value index (a combination of the calorific content, wood density, moisture content and amount of ash) and the preference for the 39 species utilized by the inhabitants, as only six taxa attained high values. The authors mention that the availability, closeness to home and viability may influence the choice of the species used for firewood (^{Márquez et al., 2017}).

In regard to the initial growth, the variance analysis evidenced a significant effect for the species-locality interaction on the variables stem diameter and height, both in tall (valor p=0.0183 y p=0.0137, respectively) and short species (value p=0.0421 and p=0.0001, respectively). 150 days after the establishment, E. grandis and E. globulus exhibited the best results for stem diameter, compared to the two other species, in three localities; however, their individual behavior was lower in the municipality of Soata, with 11.7 and 9.2 mm, respectively (Table 2).

P. guchapele and A. mangium exhibited the lowest values for this variable, although without significant differences between localities. Only San José de Miranda showed no statistical differences between the four short species. In Soata (5.4 cm) and Enciso (5.9 cm), P. dulce had lower values than the other taxa. No significant effect of the locality factor was observed (value of p=0.7830) (Table 2).

The tall species P. guachapele and A. mangium exhibited greater heights than the other two taxa in the three assessed localities. In Enciso, E. grandis and E. globulus exhibited their best performance, with 205.6 cm and 153.3 cm, respectively; these values were higher than those of the localities of San José de Miranda and Soata. P. guachapele and A. mangium registered no differences between the three municipalities (Table 3).i.e., the exotic taxa showed the greatest growth in the three localities.

Given the high vigor, the broad adaptability and the short period until the forest exploitation of eucalyptus, according to ^{Farias et al. (2016)}, this taxon may be assumed to have a greater acceptance in the studied tobacco-producing areas. However, the use of introduced species must be analyzed in depth. For example, the poor survival of Eucalyptus plantations located in southern China and, to a lesser extent, the emergence of twelve native tree species, are due to the allelopathic effects of their roots. Conversely, their topsoil promoted the initial growth of most of the taxa; therefore, differentiated management strategies are advisable (^{Zhang et al., 2016}).

The lower response in height for native species, may be due to their incipient domestication status and to lack of knowledge of their interaction with the environment, which hinders implementation of management techniques. However, they have other attractive ecological and environmental attributes.

In degraded soils of the southern Amazon, Tachigali vulgaris L.f. Gomes da Silva & H.C. Lima exhibited a greater survival, a growth rate thrice as fast and a topsoil production twice higher than those of E. urophylla and E. grandis. Furthermore, it suppressed the invasive species (^{Farias et al., 2016}). Another positive experience is cited in a joint plantation of E. grandis and P. guachapele on sandy soils, where the amount of N contributed to the soil and the mineralization rate of the residues increased significantly, even with a small contribution (11 % of the total deposited) of the legume (^{Carvalho et al., 2004}).

L. leucacephala is a forage species with a high protein content for the nutrition of ruminants and birds in grasslands, or as living fences, as is common in small tobacco farmsteads with a peasant economy. ^{Pinto et al. (2014)} assessed biomass production in Leucana collinsi Britton & Rose, with records of up to 2 490 kg ha^-1 of dry matter, 120 days after planting. P. guachapele, selected for this study, stands out for its high protein and fiber contents (18.9 % and 24.4% based on the dry weight, respectively) in its leaves; therefore, it is also recommended as a dietary supplement (^{Chitra and Balasubramanian, 2016}).

In Colombian long-haired sheep under grazing conditions, supplementation with G. ulmifolia, a tree species of the dry tropical forest, contributed to the maintenance and increase of weight gain, in contrast to the treatment without supplements, especially during the dry season (^{Díaz et al., 2014}). These additional qualities are of interest for the tobacco producers, because in Santander the sheep and goat chain has gradually become established, although its limiting factor is the low availability of water.

The results of the present study indicated a greater initial vigor of forest species for the edaphoclimatic conditions of the municipality of Enciso, without very marked differences between the localities of Soata and San José de Miranda. According to ^{Breugel et al. (2011)}, out of 47 native and 2 exotic tree species, 35 % exhibited differences in the final growth, in sites of high or low fertility and in humid or dry sites of Panama, two years after planting.

Likewise, the strong correlation (r= 0.62 to 0.91) of the climate conditions (temperature, rainfalls and accumulated solar radiation) on the growth in height was validated in E. urophylla × E. grandis (^{Elli et al., 2017}). On the other hand, selection and a successful protocol for the obtainment of high-quality seedlings during the nursery phase are crucial to restoration programs with native species, as documented by ^{Lu et la. (2016)}.

Conclusions

Tobacco growers of Boyacá and Santander use approximately 20 multipurpose species with different caloric values (12.5 to 18.8 MJ kg^-1). The species most frequently used are Eucaliptus spp. (57 %), Pithecellobium dulce (48 %), Escallonia pendula (12 %) and Manclura tinctoria (9 %). In general, the tall exotic taxa E. grandis and E. globulus exhibit a higher growth (mean height of 132.2 cm) than the native species P. guachapele (mean height of 40.9 cm). Finally, these studies must be continued until the forest exploitation and they must characterize the dendroenergetic and ecological potential of the evaluated species.