Original article

Eucalyptus sp. sawn wood yields in the Los Palacios agroindustrial base unit, Pinar del Rio, Cuba

0000-0001-7627-0152Ivania Pupo Carballo^{1  *}  , 0000-0002-9357-629XDaniel Álvarez Lazo²  , 0000-0001-8143-8466Ignacio Esteves Valdes³  , 0000-0002-4536-4938Damián Guerra Castellón²  , 0000-0003-3649-8245Rogelio Morejón Rivera⁴ 

¹Servicio Estatal Forestal de Los Palacios, Pinar del Río, Cuba.

²Universidad de Pinar del Río "Hermanos Saíz Montes de Oca". Pinar del Río, Cuba.

³Universidad Estatal del Sur de Manabí, Ecuador.

⁴Instituto Nacional de Ciencias Agrícolas, Carretera La Francia, km 1½, Los Palacios, Pinar del Río Cuba.

ABSTRACT

The present study is developed in the areas of the Los Palacios Agroindustrial Base Business Unit, with the objective of increasing the yield of Eucalyptus pellita F Muell, and Eucalyptus saligna Smith sawn wood from the use of a combination of treatments to reduce the effect of growth stresses and sawing schemes. 200 logs of 4 m were used and through the two-factor ANOVA analysis, it was found that there are no significant differences between the species on the Cracking Index; considering also that the best results are obtained in logs stored with irrigation for Eucalyptus saligna and standing ringed trees for both species. From the application of the elimination method in the linear regression analysis, the conicity variable is excluded for the construction of the prediction model using the diameter of the logs as the only independent variable in the two equations obtained, which are characterized by presenting high coefficients of correlation, determination and low standard errors of estimation. Taking into consideration the sawing methods, it is defined that the best yields of sawn wood are obtained from alternative tangential sawing combined with the treatment of standing ringed trees.

Key words: logs; splits; sawing; quality.

INTRODUCTION

The forestry industry is a necessary tool to make sustainable forestry production a viable objective. It is also an agent for sustainable development in its own right, increasing the socio-economic benefits of the sector.

^{Pupo et al. (2018)}, state that currently it is an objective of forestry policy to grow each year in wood production, mainly from plantations, alleviating the existing pressure on natural forests since the vast majority of species with high commercial value are distributed in these forests.

The demand for raw materials continues to increase and the supply of wood is even more limited, with the diversification of fast-growing species playing a fundamental role that, in addition to adapting to soil and climate conditions, are highly productive and have good quality wood, where the genus Eucalyptus stands out, coinciding in this aspect with ^{(França et al. 2019)}.

However, according to ^{Beltrame et al. (2016)}, the greatest technical limitation that this genre presents is the magnitude of the growth tensions, which is probably the most significant indicator, since high levels of tension (particularly longitudinal tensions) cause the defects that harm the most the performance and obtaining quality of sawn wood.

Growth tensions begin to be released from the moment of felling and appear as cracks at the ends of the logs. In sawn pieces, the remaining stresses also cause cracking and warping ^{(França et al. 2019} and ^{Silva et al., 2019)} . Therefore, the general objective of the present study is to increase the yields of Eucalyptus pellita and Eucalyptus saligna from the use of a combination of treatments to reduce the effect of growth stresses and sawing schemes.

MATERIALS AND METHODS

Methodology

The investigation was carried out at the Rigo Fuentes band sawmill, belonging to the La Palma Agroforestry Company, specifically in the "Los Palacios" Agroindustrial Base Business Unit located in the Popular Council of San Diego de los Baños, Los Palacios Municipality, Province of Pinar del Río, Cuba.

Los Palacios municipality is located in the south-eastern part of the province of Pinar del Río. It limits to the North with the municipalities of La Palma, to the South with the waters of the Gulf of Batabanó, to the West with the Consolación del Sur municipality and to the East with San Cristóbal, the territory extends from the elevations of the Guaniguanico Mountain Range to the south coast bathed by the Caribbean Sea and has a territorial area of 764.51 km².

The Agroindustrial Base Business Unit ¨Los Palacios¨ is located in the mountainous area of the municipality and has a forestry heritage of 12,249.60 hectares, with 199 workers, distributed in six brigades to attend to everything related to Forestry use and Forest industry.

Two species of genus Eucalyptus are used: Eucalyptus pellita and Eucalyptus saligna in plantations that cover 373.6 hectares of the 1451.70 hectares that constitute established plantations of the Forest Heritage of the Los Palacios Municipality.

Determination of the number of samples used in the research

Simple random sampling was used based on the use of 100 trees, with the diameter (D_1.30) as variable of interest; according to the methodology proposed by ^{Aldana et al. (2010)} and ^{Álvarez et al. (2020)}; which use the following equations Equation 1 and Equation 2:

$\text{n=}\frac{{\text{t}}^{\text{2}}{\text{s}}_{\text{x}}^{\text{2}}}{{\text{E}}^{\text{2}}}$ (1)

$\text{E=(LE*}\stackrel{\mathrm{-}}{\text{X}}\text{)}$ (2)

where: n = number of samples; 𝑡 2 = Student 's t-statistic squared; LE= limit of permitted sampling error.

The sample consists of 200 logs of 4 m length of each species, with a diameter between 21 and 30 cm. This number was divided into groups of 50 to apply the different treatments to reduce the rates of cracking of the logs and increase the sawn wood yields

Determination of the efficiency of the sawing process of Eucalyptus pellita and Eucalyptus saligna depending on the sawing schemes and treatments to reduce growth stresses

From the combination of three treatments to reduce the effect of growth stresses on the quality of the logs; combined with the application of different sawing methods, proposed by ^{Pedro et al. (2014)}, cited by ^{Juizo et al. (2018)}; ^{Carvalho et al. (2019)} to increase the efficiency of the E. pellita and E. saligna sawing process. The treatments are the following:

Treatment of stored logs with irrigation (T1)

This treatment consists of keeping the logs subjected to irrigation for 4 times a day, with the aim of maintaining a high humidity gradient in the wood to reduce the effect of internal growth stresses and at the same time make the sawing of the logs easier, according to (Pupo et al., 2018).

Treatment of stored logs with ring at the ends (T2)

It consists of making a 30 cm ring at each end of the logs, deepening 1/3 of the diameter, taking into account what was proposed by ^{(Pupo et al., 2018)}.

Pattern (T3)

Logs stored under normal conditions for comparison.

Girdling of standing trees (T4)

It consists of making a ring 30.0 cm wide above the felling height of the trees, deepening 1/3 of the diameter with the objective of cutting off the circulation of the sap, eliminating the cambium of the tree, until reaching the desired depth, for which it was necessary to wait three months ^{(Neto, 2017} and ^{Pupo et al., 2020)}. Next, a four-meter-long log is obtained located above the felling height and containing the ringing made.

The logs were transferred from the logging area to the Agroindustrial Base Business Unit "Los Palacios" where all the treatments were mounted. As an index of determining the effect of the methods applied to reduce growth stresses on Eucalyptus wood sp The Cracking Index is taken as a reference.

Determination of crack index

In the development of the treatments, the quantification of the cracks produced in the logs due to the release of growth stresses in terms of length (LG), depth (PG), width (AG) and number (NG). These data were processed with the facilities of the Microsoft Excel system where the means of each of the quantified parameters were determined (Figure 1), as well as the log cracking index (IRT), according to the treatment and species, using the equation used by Beltrame et al., (2015) (Equation 3)

log splitting index, %; a _i - maximum opening of the crack (i = 1...n), cm; C _i - size of the crack (pith-cortex), cm; D - average cutting diameter, cm.

For the construction of the models that allowed the prediction of the Cracking Index, the standard sample (T3) was taken, with the IR as dependent variables and the diameter in the center of the log and the conicity of the logs as independent variables. The selection of both variables is based mainly on conceptual or theoretical foundations ^{(Hair et al., 1999)}. To adjust the crack index estimation models, multiple linear regression was used, using the stepwise method.

After the time established in the applied treatments had concluded, the samples were transferred to the Martín González sawmill where the second phase of the research was carried out.

Due to the particularity of Eucalyptus sp regarding the high internal growth stresses and their subsequent defects in the sawn wood, two sawing schemes were applied to the samples, called (A)- tangential sawing with successive cuts and (B)- tangential sawing with alternative cuts, as represented in Figure 1; agreeing with ^{Carvalho et al. (2019)}.

Source: Pedro et al. (2014) cited by Juizo et al. (2018) y ^{Carvalho et al. (2019)}

Fig. 1.  - Sawing methods. A- Tangential sawing with successive cuts. B- Alternative tangential sawing. 

25 logs per treatment were processed to define the yields and quality of the sawn wood for the Eucalyptus species investigated. Sawn wood yields are obtained from the following mathematical expression, coinciding with ^{Leyva et al., (2017)} Equation 4.

Where: r_v - Total volumetric yield, (%); vma - Volume of sawn wood, (m ³⁾; Vt - Volume of wood in logs (m³)

The volume of total sawn wood in each log in the sawmills was determined based on the linear measurements obtained from sawn wood according to the expressions set out below Equation 5:

Where: - volume of sawn wood from a log, m³; - width, thickness and length of the piece i obtained from a log or group of logs, m; n- number of sawn pieces of a log

The calculation of the raw volume of each log was carried out using the Smalian formula, taking into consideration studies developed by ^{(Casagrande et al., 2019)}, as shown in the following equation (Equation 6):

$Vmb = \left(\right(g1 + g2) /2) * L$ (6)

Where: Vmb, volume of wood in bolus (m³); g₁: basal area at the fine end (m²);

g₂: basal area at the thick end (m²); L, length of the log (m).

Statistical analysis

The variables length (LG), depth (PG), width (AG) and number of cracks (NG) were analyzed to know if the theoretical assumptions were met (normal distribution and homogeneity of variances). A two-factor ANOVA was carried out, taking into account the behavior of each dependent variable in each methods or treatments applied in the investigated species, with the exception of the specific case of number of cracks that does not meet the theoretical assumptions since it does not follow a normal distribution and logarithmic transformations were applied.

Through the Duncan test at a level of 5 % probability of error, the existence of significant differences between the methods or treatments applied in each of the quantified variables was determined.

It has been made besides an analysis of variance to define differences between the species investigated and the methods or treatments applied to reduce the magnitudes of cracks (number, width, length and depth of cracks) and the Log Cracking Index (IRT) through an HSD Tukey test considering a value of P= 0.05

A paired samples t-analysis was used in the validation process of the regression models obtained for the prediction of the Cracking Index in the investigated logs.

RESULTS AND DISCUSSION

Analysis of treatments to reduce the effect of internal growth stresses in Eucalyptus pellita and Eucalyptus saligna wood

Figure 2 shows the behavior of the number of cracks per treatment depending on the species. Observing that the standard sample is the one with the greatest number of cracks, considering that no treatment has been applied to these logs to reduce the effect of growth stresses on the development of cracks at the ends; which directly affects the effectiveness of the mechanical transformation process of these logs in volume and quality of the processed wood; coinciding with ^{Braz et al. (2017)} (Figure 2).

Fig. 2.  - Number of cracks. EN- Eucalyptus saligna, Eucalyptus pellita, T1-Treatment 1, T2-Treatment 2, T3-Treatment 3, T4-Treatment 4 

In Figures 3 and 4, it can be seen that the best results were obtained in the treatments of logs stored with irrigation (T1) and in standing ringed trees (T4) for E. saligna Smith and in standing ringed trees (T8) for E. pellita (Figure 3 and Figure 4).

Fig. 3.  - Dimensions of cracks by treatment. EN- Eucalyptus saligna, T1-Treatment 11, T2-Treatment 2, T3-Treatment 3, T4-Treatment 4, LG-Length of cracks, AG-Width of cracks, PG-Depth of cracks 

Fig. 4.  - Dimensions of cracks due to treatment in Eucalyptus pellita. T5-Treatment 5, T6-Treatment 6, T7-Treatment 7, T8-Treatment 8, LG-Length of cracks, AG-Width of cracks, PG-Depth of cracks 

The results presented in Figures 3 and 4 made it possible to show a reduction in the dimensions of the magnitudes of the cracks, which is a consequence of a greater release of growth stresses; agreeing with ^{Müller et al. (2017)} and ^{Silva et al. (2017)}.

These results were demonstrated through a two-factor ANOVA taking into account the behavior of each dependent variable in each test or treatment carried out for each species.

When analyzing the behavior of each variable independently, it is summarized that in the variable number of cracks (NG) the interaction between the factors (treatments and species) is seen and the best treatments without significant difference between them are the treatments of stored logs with irrigation (T1) and standing ringed trees (T4) for E. saligna, where values of 3.25 cm and 2.0 cm respectively were obtained and in standing ringed trees (T4) for E. pellita with an average value of (1.66 cm) (Table 1).

In the case of the variable crack length (LG) there were no significant differences between the species or between the treatments. The treatment of logs stored with irrigation (T1) with 7.04 cm in E. saligna and 7.29 cm in E. pellita does not differ statistically from the treatment logs stored with a ring at the ends (T2) with 7.26 cm in E. saligna and 7.35 cm in E. pellita, indicating that the release of growth stresses were more evident in the appearance of cracks at the ends of the logs.

However, there is a significant difference between the treatment of standing ringed trees (T4) with the other treatments and the pattern (T3), considering this as the most effective to reduce cracks at the ends of the logs for both species with minimum values obtained of 4.65 cm in E. saligna and 4.67 cm in E. pellita (Table 1).

Table 1.  - Means of each variable in the different treatments evaluated 

Species	Treatments	NG	LG cm	AG cm	P.G. cm
Eucalyptus saligna	1	3.25d	7.04b	0.018a	0.045a
	2	6.25c	7.26b	0.017a	0.043a
	3	9.00ab	10.28a	0.019a	0.048a
	4	2.00d	4.65c	0.010b	0.010b
Eucalyptus pellita	1	8.00bc	7.29b	0.020a	0.049a
	2	7.50bc	7.35b	0.018a	0.043a
	3	10.25a	10.48a	0.020a	0.044a
	4	1.66d	4.67c	0.011b	0.010b

Note: Values with the same letter show that there are no significant differences between the treatments for P < 0.05.

This is a consequence of the fact that ringing causes the death of the trees, losing moisture, reducing the effect of drying stresses. Another reason is the partial release of growth stresses, causing a decrease in the magnitude of the cracks.

^{Silva et al. (2017)} obtained average cracking values by applying the treatment of standing ringed trees in E. europhylla (26.73 cm) much higher than those obtained in the present investigation. The width (AG) and depth (PG) of the cracks show similar behavior as in the previous variable. With minimum values obtained of (0.011 cm) in E. saligna and (0.010 cm) in E. pellita in the treatment of standing ringed trees, reaffirming its effectiveness.

Analysis of the determination of the Cracking Index

Through the two-factor ANOVA analysis, it was verified that there are no significant differences between the species on the Cracking Index; also considering the existence of a significant effect of the treatment factor on the Cracking Index (Table 2).

Table 2.  - Results of the analysis of variance for the log cracking index 

Variance analysis
Fountain	Sum of squares	Gl	Middle Square	F	P
Testing	0.415184	3	0.138395	27.40	0.0000
B: Species	0.00300312	1	0.0030012	0.59	0.4482
A x B	0.403344	3	0.0134448	2.66	0.0709
Residue	0.121225	24	0.00505104	.	-
Total	0.579747	31	.	-	-

^{Pupo et al. (2018)} and ^{Pupo et al. (2020)}, establish that the presence of cracks is a consequence of the presence of growth stresses and that is why the application of different treatments aims to reduce their effect, which develops with marked significance in the genus Eucalyptus, coinciding in this sense with ^{Silva et al. (2019)}.

Table 3 shows the existence of significant differences between the applied treatments, where T4 (girding of standing trees) is the one with the best performance with the lowest cracking index (0.15) in E. saligna and (0.26) E. pellita, although in the other treatments applied a tendency to decrease in relation to the pattern can be seen (Table 3) and coinciding with ^{Beltrame et al. (2015)}.

Table 3.  - Comparación de las medias de los Índices de Rajaduras de las trozas según los tratamientos aplicados 

TREATMENTS	IRT (%) E. saligna	IRT (%) E. pellita
1	0.32 b	0,37 b
2	0.29 c	0.29 c
3	0.56 a	0.48 a
4	0.15 d	0.26 da

Valores con la misma letra demuestran que no existen diferencias significativas entre los tratamientos para una P > 0,05.

In studies carried out on different clones of Eucalyptus sp by ^{Beltrame et al. (2015)}, the average value obtained is 0.46% although in some of them the results were between 0.14 and 0.34 %, very similar to those obtained in the present investigation.

Modelación matemática de la influencia de las variables diámetro y conicidad sobre la predicción del Índice de Rajadura de las trozas

It is very important to determine the effect of the characteristics of the tree on the wood values, in order to provide information for the selection of the models, which is why the log variables diameter and taper (CON) are used and to determine the degree of this relationship between the variables used in the estimation of the cracking indices in the logs was constructed in the correlation matrix shown in Tables 4 and 5.

As can be seen in Tables 4 and 5, the correlation coefficients of the variables used to estimate the cracking index in the two species studied are significant, which demonstrates the high degree of relationship between them (Table 4 and Table 5).

Table 4.  - Pearson parametric correlation matrix for Eucalyptus pellita 

-	Diameter	GO	WITH
DIAMETER Pearson correlation	1	-.857 ^**	-.0.45
Sig. (bilateral)	-	,000	.757
N	fifty	fifty	60
IR Pearson Correlation	-.857 ^**	-	.201
Sig. (bilateral)	.00	-	-
N	60	fifty	fifty
WITH Pearson correlation	-.045	.184	1
Sig. (bilateral)	.757	.201	-
N	fifty	fifty	fifty

**The correlation is significant at the 0.01 level (two-tailed)

Table 5.  - Pearson parametric correlation matrix Eucalyptus saligna 

-	Diameter	GO	WITH
DIAMETER Pearson correlation	1	-.857 ^**	-.0.45
Sig. (bilateral)	-	,000	.757
N	fifty	fifty	60
IR Pearson Correlation	-.857 ^**	-	.201
Sig. (bilateral)	.00	-	-
N	60	fifty	fifty
WITH Pearson correlation	-.045	.184	1
Sig. (bilateral)	.757	.201	-
N	fifty	fifty	fifty

Note: **The correlation is significant at the 0.01 level (two-tailed)

Using the independent variables (diameter and taper of the logs) with the greatest relationship with the dependent variable (IR), prediction equations for the log splitting index are obtained for the E. saligna and E. pellita and the compared models are seen in Table 6, with their respective correlation coefficients r, determination (r²) and standard error of estimation (s), obtaining a set of results that denote that there is a linear correlation between the cracking index of the logs, and the variables under study, reporting that there is an r square of 0.715 for E. saligna and 0.806 for E. pellita; which is considered strong. From the application of the elimination method in the linear regression analysis, the conicity variable is excluded for the construction of the prediction model (Table 6).

Table 6.  - Models to determine the log cracking index 

SPECIES	REGRESSION MODELS	R	R^2	YES
E. saligna	IRT= 1.474-0.046(Diameter )	0.845	0.715	0.09155
E. pellita	IRT= -0.606+0. 047(Diameter)	0.898	0.806	0.07087

The two equations constructed to determine the splitting index in logs have high correlation and determination coefficients and low standard errors of estimation. However, the best behavior is found in the model that corresponds to E. pellita, which has the best fit (the independent variables explain 89 % of the dependent variable) and the lowest value of the standard error of estimation.

Validation of the regression models obtained

Tables 7 and 8 show the comparison between the prediction of the IR variable in 50 % of new samples to validate with the real values of that variable, obtaining accuracy in the behavior of the proposed model as no significant difference was found between the real and estimated value of the models for E. saligna and E. pellita (Table 7) and (Table 8).

Table 7.  - T test for validation of the proposed model for IR in Eucalyptus saligna  

-	Paired differences					-
-	Half	Standard deviation	Mean standard error	95% confidence interval of the difference		t	gl	Yeah. (bilateral)
				lower	Superior
Pair 1 IR- Unstandardized Predicted value	,00173255	,079642299	,01592860	.03114246	,03460756	,109	24	,914

Table 8.  - T test for validation of the proposed model for IR in Eucalyptus pellita  

-	Paired differences					-
-	Half	Standard deviation	Mean standard error	95% confidence interval of the difference		t	gl	Yeah. (bilateral)
				lower	Superior
Par 1 IR- ajust	,06958	,22310	,04554	‘022562	,16279	,1,506	23	,146

The validation of the model guarantees its precise use in social practice for the prediction and control of the incidence of crack formation and its implication on the quality of the bolo wood and its influence on the yields and quality of the sawn wood.

Analysis of sawn wood yields based on the proposed sawing methods

Taking into consideration the experienced sawing methods, it was possible to establish that in the results presented in Table 9, the best yields were obtained for both species with the use of the sawing method with alternative tangential cuts and in the treatment of standing ringed trees where the lowest crack index values are obtained (Table 9).

Table 9.  -Yields of sawn wood for Eucalyptus sp. 

Treatments	GO- E. saligna %	GO E. pellita %	Rend - E. saligna CTCS %	Rend - E. saligna CTA %	Rend - E. pellita CTCS %	Rend - E. pellita CTA %
Storage with irrigation	0.32	0.37	46.34	49.98	45.75	46.95
End banding of logs	0.29	0.29	47.25	50.65	47.88	51.55
Pattern	0.56	0.48	42.19	44.14	43.21	42.76
Girdling of standing trees.	0.15	0.26	48.19	51.87	49.10	52.55

Note: IR- Crack index, CTCS- Tangential cut with successive cuts, CTA-Alternative tangential cut.

^{Anjos y Fonte (2017)}, in E. grandis and E. dunnii obtained yield values between 41.07 % and 43.85 %, 36.71 % and 41.02 % respectively, also lower than those obtained, although in E. saligna it obtained values between 43.87 % and 47.46 %, the latter very similar to the results obtained in both species with the use of the tangential sawing scheme with successive cuts, applying the ringing treatment at the ends of the logs.

Also, ^{Carvalho et al., (2019)}, in Eucalyptus sp obtained yield values of 47.08 % similar to the previous result and 53.43 % higher than those obtained. Melo et al. (2016), in studies carried out with native species, obtained yield results of 49.0 %, similar to those obtained in this research.

Demonstrating the need to take these approaches into consideration to obtain the best results in relation to the efficiency of the primary transformation process in sawmills for logs of E. saligna and E. pellita.

The simulation model is based on the solution of the two-dimensional cutting problem in an irregular space that could be extended to a third dimension. In its computational implementation, the algorithm uses the commercial information of the required wood orders to optimize the sawing pattern for a specific log.

CONCLUSIONS

Taking into consideration the experienced sawing methods, it can be established that the best yields are obtained for both species with the use of the sawing method with alternative tangential cuts combined with the treatment of standing ringed trees.

REFERENCIAS BIBLIOGRÁFICAS

ALDANA PEREIRA, E., 2010. Medición forestal [en línea]. La Habana: Félix Varela. ISBN 978-959-07-1321-7. Disponible en: https://books.google.com.cu/books/about/Medici%C3%B3n_forestal.html?id=Pv6OswEACAAJ&re%20dir_esc=y.

ÁLVAREZ LAZO, D., DOMINGOS SEBASTIAO, J.D., NAZARE BUANGA LELO, F.N., SILVA FUENTES, S., y ESTEVES VALDES, I., 2020. Análisis de la calidad del aserrado de maderas tropicales en Cabinda, Angola. Revista Cubana de Ciencias Forestales [en línea], vol. 8, no. 1, [consulta: 20/06/2023]. ISSN 2310-3469. Disponible en: Disponible en: https://cfores.upr.edu.cu/index.php/cfores/article/view/452 .

BELTRAME, R., GONZÁLEZ CADERMATORI, P., DELUCIS, R., GATTO, D.A., HASELEIN, C., y SCHNEIDER, P., 2016. Use of dummy variables to predict growth stresses in Eucalyptus ssp. Clones. Revista Árvore [en línea], vol. 40, no. 6, DOI 10.1590/0100-67622016000600019. Disponible en: https://www.researchgate.net/publication/313830949_Use_of_dummy_variables_to_predict_growth_stresses_in_Eucalyptus_ssp_Clones.

BELTRAME, R., LEMOS DE PERES, M., LAZAROTTO, M., GATTO, D.A., SCHNEID, E., y HASELEIN, C.R., 2015. Tensão de crescimento e sua relação com as rachaduras de topo em toras de Eucalyptus spp. Scietie Forestalis [en línea], vol. 43, no. 105, Disponible en: https://www.ipef.br/publicacoes/scientia/nr105/cap06.pdf.

BRAZ, R.L., OLIVEIRA, J.T. da S., ROSADO, A.M., VIDAURRE, G.B., y OLIVEIRA, J.G.L. de, 2017. Deformação Residual Longitudinal de Clones de Eucalyptus em Local Sujeito a Ventos. Floresta e Ambiente [en línea], vol. 24, [consulta: 23/06/2023]. ISSN 1415-0980, DOI 10.1590/2179-8087.028013. Disponible en: Disponible en: https://www.scielo.br/j/floram/a/M5MLQqsyWbzr5QksrRDscsd /?lang=pt .

CARVALHO, D., PEREIRA DA ROCHA, M., JUNIOR, R., y KLITZKE, R., 2019. Rendimento e variedade de produtos no desdobro de toras de Eucalyptus spp. Tecno-Lógica [en línea], vol. 23, no. 1, DOI 10.17058/tecnolog.v23i1.12215. Disponible en: https://www.researchgate.net/publication/331038746_Rendimento_e_variedade_de_produtos_no_desdobro_de_toras_de_Eucalyptus_spp.

CASAGRANDE STRAGLIOTTO, M., COSTA OLIVEIRA, A., CORRADI PEREIRA, B., y MELO FREITAS, J., 2019. Yield in sawnwood and residue utilization of Quarea paraenses Ducke and Erismauncitum Warn. Floresta [en línea], vol. 49, no. 2, Disponible en: https://pdfs.semanticscholar.org/867b/46d1bb2fceee9501a054d6cf4fe2778360cb.pdf

FRANÇA, M.C., FRANCISCO JUÍZO, C.G., PEREIRA DA ROCHA, M., JORGE KLITZKE, R., MOREIRA DA SILVA, J.R., CONTI JUNIOR, J.L.F., y MONNERAT CARVALHO BASSA, A.G., 2019. Qualidade da tora e da madeira de clones de Eucalyptus para utilização na indústria de madeira serrada. Revista Ciência da Madeira (Brazilian Journal of Wood Science) [en línea], vol. 10, no. 1, [consulta: 26/05/2023]. ISSN 2177-6830. Disponible en: Disponible en: https://periodicos.ufpel.edu.br/index.php/cienciadamadeira/article/view/12206 .

FRANCISCO JUÍZO, C.G., PEREIRA DA ROCHA, M., y FERNANDO BILA, N., 2014. Avaliação do rendimento em madeira serrada de eucalipto para dois modelos de desdobro numa serraria portátil. Floresta e Ambiente [en línea], vol. 21, no. 4, [consulta: 26/06/2023]. ISSN 1415-0980, DOI 10.1590/2179-8087.062213. Disponible en: Disponible en: https://www.scielo.br/j/floram/a/WXXvykXVHbHY6Hj9TYqNPDq/abstract/?lang=pt .

FRANCISCO JUÍZO, C.G., PEREIRA ROCHA, M., y LUCAS RAFAEL, A., 2018. Relação entre modelos de desdobro, rendimento e qualidade da madeira serrada de Pterocarpus angolensis. Scientia Agraria Paranaensis [en línea], vol. 17, no. 2, [consulta: 26/06/2023]. ISSN 1983-1471. Disponible en: Disponible en: https://e-revista.unioeste.br/index.php/scientiaagraria/article/view/17678 .

HAIR, J.F., 1999. Análisis multivariante [en línea]. España: Pearson Educación. ISBN 978-84-8322-035-1. Disponible en: https://books.google.com.cu/books/about/An%C3%A1lisis_multivariante.html?id=QV4INQAACAAJ&redir_esc=y.

LEYVA MIGUEL, I., ROJAS ROMERO, A., y SEGURADO GIL, Y., 2017. Determinación del rendimiento y calidad dimensional de la madera aserrada en aserríos en la Provincia de Guantánamo. Revista Cubana de Ciencias Forestales [en línea], vol. 5, no. 3, [consulta: 26/06/ 2023]. ISSN 2310-3469. Disponible en: Disponible en: https://dialnet.unirioja.es/servlet/articulo?codigo=6222088 .

MAGGI DOS ANJOS, R., y NAMIKATA FONTE, A., 2017. Rendimento de madeira serrada de espécies de Eucalyptus. Revista de Ciencias Agroveterinarias [en línea], vol. 16, no. 1, DOI 10.5965/223811711612017026. Disponible en: https://www.researchgate.net/publication/314970001_Rendimento_de_madeira_serrada_de_especies_de_Eucalyptus.

MISSIA DA SILVA, J.G., BAPTISTA VIDAURRE, G., MININI, D., FERREIRA OLIVEIRA, R., GONÇALVES ROCHA, S.M., y GOMES GONÇALVES, F., 2019. Qualidade da madeira de mogno brasileiro plantado para a produção de serrados. Scientia Forestalis [en línea], vol. 47, no. 121, Disponible en: https://www.ipef.br/publicacoes/scientia/nr121/cap01.pdf.

PIRES DE MORAES NETO, S., 2017. Madeira de Eucalipto para Serraria [en línea]. Brasil: Embrapa Cerrados. Disponible en: https://www.researchgate.net/publication/325497874_Madeira_de_Eucalipto_para_Serraria.

PUPO CARBALLO, I., ÁLVAREZ LAZO, D., y ESTEVES VALDES, I., 2020. Método para reducir los índices de rajaduras en trozas de Eucalyptus sp. Avances [en línea], vol. 22, no. 3, [consulta: 26/06/2023]. Disponible en: Disponible en: https://www.redalyc.org/journal/6378/637869117008/html/ .

PUPO CARBALLO, I., ÁLVAREZ LAZO, D.A., MOREJÓN RIVERA, R., ESTÉVEZ, I., y PRIETO RUIZ, L., 2018. Tratamientos para reducir los índices de rajaduras. Revista Cubana de Ciencias Forestales [en línea], vol. 6, no. 3, [consulta: 26/06/2023]. ISSN 2310-3469. Disponible en: Disponible en: http://scielo.sld.cu/scielo.php?script=sci_abstract&pid=S2310 -34692018000300260&lng=es&nrm=iso&tlng=es .

SILVA, J.C. da, LADEIRA CARVALHO, A.M.M., y FREITAS HOMEM DE FARIA, B. de, 2017. Methods for alleviation and reduction of effects of growth stresses in Eucalyptus urophylla. Revista Árvore [en línea], vol. 41, [consulta: 26/06/2023]. ISSN 0100-6762, 1806-9088. DOI 10.1590/1806-90882017000400003. Disponible en: Disponible en: https://www.scielo.br/j/rarv/a/q7CqKVjCWJBDfvGcr576Y7L/?lang=en .

VEREDIANA MÜLLER, B., PEREIRA DA ROCHA, M., JORGE KLITZKE, R., MOREIRA DA SILVA, J.R., y BAYESTORFF DA CUNHA, A., 2017. Produção de madeira serrada com cinco espécies de eucalipto resistentes à geada. Advances in Forestry Science [en línea], vol. 4, no. 4, [consulta: 26/06/2023]. ISSN 2357-8181. DOI 10.34062/afs.v4i4.5231. Disponible en: Disponible en: https://periodicoscientificos.ufmt.br/ojs/index.php/afor/article/view/5231 .

Received: May 12, 2023; Accepted: August 16, 2023

^*Autor para la correspondencia: [e-mail]

Los autores declaran no tener conflictos de intereses.

Los autores han participado en la redacción del trabajo y análisis de los documentos.