Influence of Tree Diameter and Height on the Physical and Mechanical Properties of Retrophyllum rospigliosii Wood

Abstract

Retrophyllum rospigliosii is a valuable conifer species from the Andean tropical forests, reaching diameters up to 2.5 m and heights of 45 m. Due to its high commercial demand and distinctive shape and size, its wood is highly sought after, leading to its classification as vulnerable on the IUCN Red List. This study evaluated the physical and mechanical properties of R. rospigliosii wood and how these properties vary according to tree diameter and height in plantation-grown specimens in the Cauca department, Colombia. Standard physical and mechanical tests followed international procedures to assess density, dimensional stability, and mechanical performance. The results showed stable wood density, with a basic wood density of 0.35 g/cm³, and a green density of 0.54 g/cm³. Volumetric shrinkage was 3.52% in the radial direction and 5.05% in the tangential direction, indicating good dimensional stability. Mechanical properties included a modulus of rupture (MOR) of 58.23 MPa, a modulus of elasticity (MOE) of 4702.23 MPa in static bending, and a compression strength of 54.08 MPa. Wood properties showed minimal variation across different diameter and height classes, indicating structural uniformity within the plantation. Given these characteristics, R. rospigliosii wood is suitable for non-structural applications such as furniture, moldings, and decorative items. Further studies should explore silvicultural strategies that enhance wood quality while ensuring sustainable management and conservation of this threatened species.

1. Introduction

Retrophyllum rospigliosii is a species of the Podocarpaceae family distributed in the tropical montane forests of Bolivia, Ecuador, Colombia, Peru, and Venezuela, at altitudes ranging from 1500 to 3750 m above sea level [1,2,3]. Under natural conditions, this species can reach heights of up to 45 m and diameters of 2.5 m [4,5]. However, due to deforestation and forest degradation [6,7], as well as its extensive use in fine carpentry and construction, more than 30% of R. rospigliosii populations have disappeared in recent decades [8]. As a result, R. rospigliosii is classified as a vulnerable species on the IUCN Red List [9].

However, most of the reforestation projects in Colombia, especially in the Andean region, are short rotation eucalyptus and pine monocultures (90,495 ha for eucalyptus and 25,438 ha for pines, respectively) [10]. As a response to the threat of R. rospigliosii, three reforestation pilot projects were established to ensure the species’ conservation, identify silvicultural practices for sustainable management, and explore potential wood utilization. These plantations were established between 1998 and 1999 by Smurfit Kappa Carton de Colombia and the National Federation of Coffee Growers of Colombia in the departments of Antioquia and Cauca [11,12]. Current cultural treatments in plantations of R. rospigliosii include fertilization at planting, weed control during early stages to minimize competition, and animal exclusion to protect young trees [11]. These practices, particularly early fertilization, have demonstrated significant increases in plantation growth [13]. Moreover, pre-germination treatments have achieved high germination rates, approaching 85%, which optimizes seedling production for reforestation efforts [14]. R. rospigliosii reached low asymptotic dimensions (21.6 cm in diameter, 18.1 m in height, and 159.4 m³ ha⁻¹ in volume), which is reflected in its low growth and carbon sequestration rates compared to other native or exotic species commonly planted in Colombia [11,12].

The commercial applications of R. rospigliosii wood are primarily determined by its density and mechanical properties. Wood density directly influences strength, stiffness, and durability as it determines mechanical performance and workability during industrial processing for structural and decorative applications [15,16]. Additionally, wood density impacts workability in processes like cutting, planning, molding, drying, and chemical treatment [17,18,19]. In coniferous species, tracheid wall thickness is the primary factor influencing wood density, with a direct correlation between increased wall thickness and higher specific gravity [20]. Mechanical properties, including static bending strength, parallel compression, and hardness, are key indicators of wood’s ability to carry loads, resist impact, and maintain its structural integrity under applied stress. Wood with a high modulus of elasticity is preferred for structural applications due to its enhanced rigidity and resistance to deformation and shear strength is particularly relevant for joints and assemblies subjected to sliding forces [21]. Furthermore, chemical composition, particularly lignin and cellulose content, influences wood quality. Higher lignin content enhances resistance to biological degradation and increases calorific value, while high cellulose content improves flexibility and workability [17,21]. Wood properties’ variation is influenced by both genetic and environmental factors, leading to significant differences in physical and mechanical characteristics [22]. These variations can occur between individual trees and within a single tree—radially from pith to bark and longitudinally from the base to the crown [22,23]. In addition, silvicultural practices have a profound impact on wood properties. For instance, appropriate tree spacing promotes optimal stem diameter development and rigidity, whereas high stand densities result in thinner, straighter stems suitable for specific structural applications [24].

Although R. rospigliosii is a high-value timber species, its industrial application remains limited due to insufficient research on the variability of its wood properties in plantations. Few studies have assessed the physical, mechanical, and chemical wood properties of R. rospigliosii [25,26,27,28,29], despite its ecological and commercial importance in the Andean forests of Colombia, Venezuela, Ecuador, Peru, and Bolivia. Previous research has suggested that young plantations of R. rospigliosii may exhibit relatively homogeneous wood properties within individual sites [29] but significant differences across sites [30]. The timber industry seeks greater uniformity in raw material properties to enhance product quality, optimize production efficiency, and improve processing techniques [22].

This study aimed to evaluate the physical mechanical properties of R. rospigliosii wood from a 22-year-old monoculture plantation in the Cauca department, Colombia. Specifically, the influence of tree diameter and height on these properties was analyzed, providing potential applications and informing silvicultural strategies for the sustainable management of the species.

2. Materials and Methods

2.1. Study Area

The R. rospigliosii plantation evaluated in this study is located in southwestern Colombia, between coordinates 2°28′0″–2°29′40″ N and 76°48′30″–76°50′0″ W, at an altitude of 1775 m above sea level. The site has an average annual precipitation of 2255 mm and a mean annual temperature of 19 °C, based on data from the Smurfit Westrock meteorological station. The plantation spans 8 ha and was originally established at a planting density of 1666 trees ha⁻¹. By the time of sampling, the stand was 22 years old, with an estimated remaining density of 300 trees ha⁻¹. Smurfit Westrock Colombia has managed it under conditions comparable to those of commercial conifer plantations in the Andean region of Colombia.

2.2. Sample Collection

A complete inventory of the plantation was first conducted to determine the diameter distribution of the trees, categorizing them into three diameter classes: Class I (8–19 cm), Class II (20–31 cm), and Class III (32–43 cm). Four trees from each diameter class were randomly selected and felled. Each tree was sectioned into ten equal-length logs, from which three segments were obtained: bottom, middle, and top, corresponding to the first, fourth, and eighth segments of the trunk. Finally, wood specimens were collected based on tree diameter and height for subsequent physical and mechanical property assessments (

Table 1. Tree diameter classes, diameter at breast height, and height of sampled trees.

Diametric Class	Tree	DBH (cm)	Height (m)
I (8–19 cm)	1	8.6	7.5
	2	7.7	6.1
	3	12.9	8.6
	4	15.1	9.4
II (20–31 cm)	5	18.5	9.0
	6	15.7	8.6
	7	26.7	13.4
	8	22.8	10.1
III (32–43 cm)	9	29.8	13.3
	10	29.3	14.3
	11	41.7	15.5
	12	36.7	14.0

).

2.3. Anatomical Description

The anatomical description followed the International Association of Wood Anatomists (IAWA softwood identification) [31]. Samples for anatomical description were randomly selected from three different trees at each diameter class and dimensioned to small wood blocks (1 cm³). Blocks were softened in a 1:1 solution of distilled water and glycerin at 60 °C to facilitate microtomy. Transverse, radial, and tangential sections (20 μm thick) were obtained using a Leica RM 123 rotary microtome (Leica Microsystems, Wetzlar, Germany). The sections were dehydrated through a graded ethanol series (70%, 50%, 35%, 15%) and subsequently rinsed in distilled water. Staining was performed using a 5% aqueous safranin solution for one minute, followed by progressive dehydration in ethanol (50%, 70%, 80%, 90%, and 100%). The sections were then cleared with xylene (50% and 100%), mounted on glass slides, and secured with Canada balsam. Anatomical features were examined under a Nikon ECLIPSE Ni light microscope (Nikon Corporation, Tokyo, Japan), and digital images were captured using a DS-Ri2 camera. Image analysis was conducted with ImageJ 1.47s software [32].

Tracheid length was determined using a maceration technique that combines the chemical separation of wood fibers, followed by fiber washing, staining, and subsequent measurement [33]. Increment cores (5 mm in diameter) were extracted from twelve selected trees using a Pressler increment borer. Previous studies have reported that maximum tracheid lengths in this species rarely exceed 3 mm, and only complete tracheids were considered for length measurement in this study. The cores were pooled into a composite sample and reduced to wood slivers (~4 g), which were subjected to maceration in a 1:1 solution of hydrogen peroxide and glacial acetic acid at 60 °C for at least one hour. The macerated tracheids were thoroughly rinsed with distilled water to remove any residual reagent, and then stained with a 1% aqueous safranin solution. Tracheid suspensions were mounted on glass slides for microscopic observation. Image acquisition was performed using a Nikon DS-Ri2 digital camera attached to a Nikon ECLIPSE Ni light microscope (Nikon, Tokio, Japan), and tracheid length measurements were obtained using ImageJ 1.47s software [32].

2.4. Determination of Physical and Mechanical Properties

Wood samples were classified by diameter class and height position before processing at the Wood Science Laboratory of the Forest Engineering Program, Faculty of Agricultural Sciences, University of Cauca. From these, a total of 36 samples from the outer part of the wood (near the bark), defect-free specimens (3 diameter classes × 3 height sections × 4 replicates) were prepared for each mechanical property following standardized ASTM testing protocols. Density was determined according to ASTM D2395-17 [34], calculating the ratio of oven-dry mass to volume at 103 ± 2 °C. Mechanical properties, including static bending, compression parallel to the grain, compression perpendicular to the grain, and shear parallel were evaluated following ASTM D143-22 [35] using a universal testing machine (WPM Werkstoffprüfsysteme Leipzig, Markkleeberg, Germany).

2.5. Determination of Cellulose, Hemicellulose, and Lignin Content

The chemical composition of R. rospigliosii wood was analyzed using an extractive-free, oven-dried composite sample obtained from the specimens used in mechanical testing. Extraction was conducted using a Soxhlet apparatus with a 2:1 toluene–ethanol solution, followed by hot water extraction, in compliance with TAPPI T 204 cm-17 [36] and TAPPI T 207 cm-22 [37] standards, respectively. Hemicellulose content was calculated as the difference between the holocellulose and cellulose fractions. Cellulose content was determined following ASTM D1103 [38] while lignin content was quantified in extractive-free wood according to TAPPI T 222 om-21 [39].

2.6. Statistical Analysis

Statistical analyses included a descriptive assessment of each wood physical and mechanical properties considered. Analysis of variance (ANOVA) was conducted to evaluate differences in properties across diameter classes (8–19 cm, 20–31 cm, and 32–43 cm) and tree height sections (base, middle, and apex) in R. rospigliosii trees. Tukey’s HSD test was performed to identify significant differences between diameter classes and longitudinal sections. Statistical significance was set at p ≤ 0.05. All analyses were performed using R software version 4.4.1 [40].

3. Results and Discussion

3.1. Anatomical Description

The wood of R. rospigliosii exhibits a yellowish-brown color with no distinctive odor or taste. It has a moderate luster, straight grain, and fine texture. Growth rings are well defined, showing a transition from gradual to abrupt between earlywood and latewood, with no resin canals observed. Axial parenchyma is diffuse and contains reddish deposits. Axial tracheids are quadrangular to hexagonal in cross-section, with smooth walls and uniseriate bordered pits, occasionally in an opposite arrangement. Rays are homogeneous, uniseriate, and exhibit cupressoid pits at the ray–tracheid interface (Figure 1). The anatomical features described in this study align with previous reports [41,42].

The average tracheid length was 1.83 ± 0.05 mm, classifying the wood as short fibered. These values are notably lower than those reported for the same species in natural forests (3.6 mm and 2.75 mm) [26,42] and below the 2.3 mm average observed in eight-year-old plantation-grown trees [29]. A 5 mm increment borer was used because there were already reports of tracheid lengths smaller than this size. However, as reported by Bergqvist [43], the tracheid length may be underestimated since the use of small-diameter increment cores increase the likelihood of tracheids being broken during sampling. Fiber length and key morphological parameters, including the slenderness ratio, flexibility coefficient, and Runkel ratio, determine wood’s mechanical properties for industrial applications, especially in pulp and paper production [23,44]. The fiber length obtained in this study classifies R. rospigliosii as a short-fibered species, making it comparable to hardwoods of the genus Eucalyptus, which are highly relevant in the pulp and paper industry.

3.2. Chemical Composition

The chemical composition of R. rospigliosii was 43.2% cellulose, 20.2% hemicellulose, and 28.7% lignin, consistent with the composition found in wood species commonly used for industrial purposes. Wood raw materials typically consist of 40%–60% cellulose, 10%–40% hemicellulose, 15%–30% lignin, and smaller proportions of extractives, proteins, and inorganic compounds [45,46]. The chemical composition of wood is a key determinant of its suitability for various industrial applications, particularly in paper production. For high-quality paper manufacturing, fiber materials must possess a high cellulose content, a low lignin proportion, and controlled levels of extractives and moisture to ensure optimal pulp properties.

The lignin content observed in this study falls within the lower range reported by various authors [47], comparable to that of other coniferous species, including P. michoacana, P. montezumae, P. oocarpa, P. pringlei, P. teocote, P. caribaea, P. patula, P. pseudostrobus, P. rudis, P. strobus, P. ayacahuite, P. leiophylla, and P. herrerae, which typically range between 25% and 30% [47,48,49,50,51]. While a more comprehensive evaluation of the physicochemical properties and industrial performance of R. rospigliosii is required to assess its potential for paper production, its relatively low lignin content suggests advantages for both pulping efficiency and bioconversion into liquid biofuels [52].

3.3. Physical Properties

The average density values obtained for R. rospigliosii were 0.54 g cm⁻³ for green density, 0.43 g cm⁻³ for air-dry density, 0.38 g cm⁻³ for oven-dry density, and 0.35 g cm⁻³ for basic density (

Table 2. Green, air-dry, oven-dry, and basic wood density (g cm⁻³) of 22-year-old R. rospigliosii by diameter class and height section *.

Tree Section	Wood Density (g cm−3)
	Green	Air-Dry	Oven-Dry	Basic
Diametric class
I	0.56 ± 0.01	0.44 ± 0.01	0.39 ± 0.01	0.36 ± 0.01
II	0.53 ± 0.02	0.43 ± 0.02	0.38 ± 0.01	0.35 ± 0.01
III	0.53 ± 0.02	0.41 ± 0.01	0.36 ± 0.01	0.33 ± 0.01
Height section
Bottom	0.55 ± 0.02	0.44 ± 0.02	0.38 ± 0.01	0.35 ± 0.01
Medium	0.53 ± 0.02	0.42 ± 0.01	0.37 ± 0.01	0.34 ± 0.01
Top	0.55 ± 0.01	0.43 ± 0.00	0.39 ± 0.01	0.35 ± 0.00
Average	0.54 ± 0.01	0.43 ± 0.01	0.38 ± 0.01	0.35 ± 0.00

* Average value ± standard error of the mean. Diameter classes I (8–19 cm), II (20–31 cm), III (32–43 cm). Height sections correspond to the lower part (bottom), middle (middle), and upper (top).

). These basic density values are similar to those reported for 22-year-old plantations in Colombia (0.34 g cm⁻³) [30] and 32-year-old plantations in Peru (0.35 g cm⁻³) [28]. Similarly, they are comparable to the values found in eight-year-old trees from a R. rospigliosii provenance trial in Colombia, where densities of 0.37 g cm⁻³ and 0.36 g cm⁻³ were recorded for the Jericó and Mesitas del Colegio provenances, respectively [29]. The basic density values recorded for this species are lower than those reported for natural forests in Ecuador (0.57 g cm⁻³ [27]) and Peru (0.41 g cm⁻³, [25]). Similarly, the densities are lower than those of other coniferous species cultivated and commercialized in Colombia, such as Hesperocyparis lusitanica (0.50 g cm⁻³) and Pinus patula (0.48 g cm⁻³) [53]. Analysis of variance showed no statistically significant differences in the average wood densities of R. rospigliosii across diameter classes or height sections. Because R. rospigliosii exhibits uniform density and falls within the typical range for conifers used in carpentry and non-structural construction [54], it holds potential for decorative applications and furniture manufacturing.

The total volumetric shrinkage of R. rospigliosii was 3.52% in the radial direction, 5.05% in the tangential direction, and 0.94% in the longitudinal direction (

Table 3. Shrinkage (%) of 22-year-old R. rospigliosii wood by diameter class and height section *.

Tree Section	Shrinkage (%)
	Radial Normal	Tangential Normal	Longitudinal Normal	Radial Total	Tangential Total	Longitudinal Total
Diameter class
I	1.76 ± 0.22 a	2.28 ± 0.22	0.65 ± 0.14 a	3.46 ± 0.43	4.98 ± 0.25	1.38 ± 0.25
II	1.23 ± 0.15 b	2.44 ± 0.20	0.41 ± 0.15 b	3.58 ± 0.36	5.01 ± 0.20	0.65 ± 0.22
III	1.02 ± 0.28 b	2.46 ± 0.27	0.49 ± 0.22 b	3.50 ± 0.41	5.25 ± 0.22	0.66 ± 0.33
Height section
Bottom	1.45 ± 0.19	2.57 ± 0.14	0.68 ± 0.15	3.41 ± 0.34	5.22 ± 0.36	1.27 ± 0.29
Medium	1.37 ± 0.27	2.61 ± 0.40	0.68 ± 0.15	3.59 ± 0.46	5.52 ± 0.20	0.97 ± 0.25
Top	1.37 ± 0.24	2.25 ± 0.21	0.20 ± 0.13	3.55 ± 0.43	5.34 ± 0.52	0.59 ± 0.26
Average	1.40 ± 0.13	2.38 ± 0.13	0.52 ± 0.09	3.52 ± 0.23	5.05 ± 0.13	0.94 ± 0.16

* Average value ± standard error of the mean. Diameter classes I (8–19 cm), II (20–31 cm), III (32–43 cm). Height sections correspond to the lower part (bottom), middle (middle), and upper (top). Different letters correspond to statistically significant differences.

). These values are lower than the average shrinkage values reported for the species, which range from 3.2% to 3.6% in the radial plane and from 5.8% to 7.3% in the tangential plane [25,27,28]. Analysis of variance revealed only statistically significant differences in shrinkage among diameter classes in radial and longitudinal normal. The dimensional stability coefficient was 1.77% for radial shrinkage and 1.52% for tangential shrinkage, classifying R. rospigliosii wood as stable and suitable for furniture, moldings, and marquetry [55].

3.4. Mechanical Properties

The mechanical properties of R. rospigliosii were assessed across diameter classes and tree height sections (

Table 4. Mechanical properties of 22-year-old R. rospigliosii wood across diametric classes and height sections *.

Tree Section	Static Bending			Compression Parallel to the Grain			Compression Perpendicular to the Grain	Shear Parallel
	Proportional Limit	Modulus of Rupture	Modulus of Elasticity	Proportional Limit	Modulus of Rupture	Modulus of Elasticity	Proportional Limit
	MPa
Diametric class
I	37.06 ± 2.07	62.37 ± 3.02	4336.11 ± 272.92	34.94 ± 2.93	53.99 ± 2.46	2139.79 ± 184.70	5.07.16 ± 0.57	4.64 ± 0.28
II	38.94 ± 1.93	58.09 ± 3.61	4748.24 ± 272.15	38.63 ± 3.17	55.42 ± 2.85	2450.70 ± 219.16	5.47 ± 0.24	4.33 ± 0.35
III	38.40 ± 2.26	56.30 ± 3.25	4818.47 ± 229.09	38.61 ± 3.62	53.09 ± 1.81	2335.02 ± 243.54	4.28 ± 0.68	4.19 ± 0.33
Height section
Bottom	40.11 ± 2.10	60.99 ± 3.73	5009.41 ± 325.36 a	38.17 ± 4.76	58.11 ± 2.59	2302.64 ± 309.65	4.72 ± 0.55	4.19 ± 0.36
Medium	38.04 ± 2.64	58.73 ± 3.21	4772.32 ± 144.81 ab	33.51 ± 3.18	53.00 ± 3.00	2054.02 ± 212.03	5.55 ± 0.41	4.34 ± 0.28
Top	36.73 ± 1.79	54.56 ± 3.62	4317.19 ± 292.07 b	39.38 ± 2.13	52.11 ± 1.38	2465.52 ± 145.04	4.68 ± 0.49	4.68 ± 0.29
Average	38.37 ± 1.24	58.23 ± 2.01	4702.23 ± 153.80	37.22 ± 1.85	54.08 ± 1.32	2292.88 ± 122.85	5.03 ± 0.28	4.38 ± 0.18

* Average value ± standard error of the mean. Diameter classes I (8–19 cm), II (20–31 cm), III (32–43 cm). Height sections correspond to the lower part (bottom), middle (middle), and upper (top). Different letters correspond to statistically significant differences.

). The average values obtained for static bending were 38.37 MPa for the proportional limit stress, 58.23 MPa for the modulus of rupture (MOR), and 4702.23 MPa for the modulus of elasticity (MOE). In parallel compression to the grain, the average values for the proportional limit stress, MOR, and MOE were 37.22 MPa, 54.08 MPa, and 2292.88 MPa, respectively. The proportional limit of the compression perpendicular to the grain was on average 5.03 MPa, and the shear strength parallel to the grain averaged 4.38 MPa. Analysis of variance revealed only statistically significant differences in the modulus of elasticity of the static bending among height sections. This suggests a relatively homogeneous distribution of mechanical strength within the plantation-grown trees, independent of these factors, which is beneficial for industrial processing.

Compared to previous studies, the proportional limit stress in parallel compression is higher than 15.98 MPa reported for 12-year-old R. rospigliosii plantations [42]. However, it is lower than the higher values recorded for this species in natural forests, where stress limits is 46.39 MPa [25,27]. Similarly, the MOE in compression parallel to the grain (2292.88 MPa) is lower than the 10,757 MPa reported in trees of this species from natural stands [25], suggesting differences in wood density and maturity between plantation and naturally regenerated trees. In comparison to commercial conifers commonly used for structural applications, R. rospigliosii demonstrates lower mechanical properties. The lower mechanical properties’ values in plantation-grown trees may be attributed to a higher proportion of juvenile wood, which is known to have lower density and reduced mechanical strength compared to mature wood. Species such as Hesperocyparis lusitanica (MOE = 8700 MPa, MOR = 65,100 MPa) [56] and Pinus patula (MOE = 7158 MPa, MOR = 85 MPa) [57] surpass R. rospigliosii in stiffness and strength. However, it remains competitive for non-structural applications, particularly against species such as Pinus tecunumanii and Pinus maximinoi, which are widely used in furniture, moldings, and interior applications. Overall, the absence of significant differences in mechanical properties between diameter classes and height sections suggests that R. rospigliosii maintains consistent mechanical behavior, independent of log position within the tree. This uniformity increases its suitability for applications like furniture, moldings, doors, windows, flooring, and wood composites, where material consistency is essential for industrial processing and performance.

4. Conclusions

This study demonstrates that R. rospigliosii from 22-year-old plantations exhibits uniform physical and mechanical properties across diameter classes and tree heights, with adequate dimensional stability but lower mechanical strength compared to commercial conifers. Despite its lower mechanical performance compared to other commercially dominant conifers, R. rospigliosii demonstrated favorable dimensional stability, which supports its suitability for non-structural applications that require consistency in raw materials. The research underscores the importance of evaluating lesser-known native species for sustainable forestry initiatives. However, since R. rospigliosii trees from 22-year-old plantations exhibit slightly lower mechanical and chemical properties than individuals of the same species from natural forests, future research should explore genetic selection, site-specific management strategies, and long-term performance assessments to enhance the industrial potential of R. rospigliosii and contribute to the diversification of forest resources in the Andean region.