Weed Control Increases the Growth and Above-Ground Biomass Production of Pinus taeda Plantations in Southern Brazil

Abstract

Pinus taeda plantations have been facing declining productivity in South America, especially due to competition for natural resources such as light, water, and nutrients. Competition with spontaneous vegetation in the early years is one of the main constraints on growth and biomass allocation in trees. However, the best method and timing for weed control and its impact on the productivity of Pinus taeda plantations are unknown. This study aims to evaluate whether weed control increases the growth and above-ground biomass production of Pinus taeda plantations in southern Brazil. This study was conducted at two sites with five-year-old Pinus taeda plantations in southern Brazil, with each being submitted to different weed control methods. This study was conducted in randomized blocks, with nine treatments: (i) NC—no weed control, i.e., weeds always present; (ii) PC—physical weed control; (iii) CC–T—chemical weed control in the total area; (iv) CC–R—chemical weed control in rows (1.2 m wide); (v) C6m, (vi) C12m, (vii) C18m, and (viii) C24m—weed control up to 6, 12, 18, and 24 months after planting; and (ix) COC—company operational weed control. The following parameters were evaluated: the floristic composition and weed biomass, height, diameter, stem volume, needle biomass, branches, bark, and stemwood of Pinus taeda. Control of the weed competition, especially by physical means (PC), and chemical control over the entire area (CC–T) promoted significant gains in the growth and above–ground biomass production of Pinus taeda at five years of age, particularly at the Caçador site. The results reinforce the importance of using appropriate strategies for managing weed control to maximize productivity, especially before canopy closure. In addition, the strong correlation between growth variables and the total biomass and stemwood indicates the possibility of obtaining indirect estimates through dendrometric measurements. The results contribute to the improvement of silvicultural management in subtropical regions of southern Brazil.

1. Introduction

Forest plantations covered approximately 294 million hectares worldwide in 2020, representing about 7% of the global forest area while supplying only 45% of the world’s industrial wood demand [1]. Forest plantations play a key role in global timber production, contributing significantly to the supply of wood, pulp, bioenergy, and other forest products, while reducing pressure on natural forests [2]. The Pinus genus is widely used in forest plantations, especially in temperate and subtropical regions, due to its rusticity, rapid growth, and multiple commercial uses. Pinus taeda L. is a species that is native to the southeastern United States and has remarkable socioeconomic importance for South American countries, exceeding 2.5 million hectares cultivated in Argentina, Brazil, and Uruguay [3,4,5]. The success of Pinus taeda plantations is due to their high adaptability to the subtropical characteristics of South America, such as favorable climatic conditions, advances in genetic improvement and forestry practices, and the ability of this plant to tolerate soils with low natural fertility in these locations [3,6]. Brazil stands out in the cultivation of Pinus taeda worldwide, with 1.9 million hectares cultivated and productivity (mean annual increment—MAI) exceeding 30 m³ ha⁻¹ year⁻¹ [2]. However, low productivity has been observed in recent years (<20 m³ ha⁻¹ year⁻¹) in southern Brazil [7,8], possibly due to limited availability of natural resources such as light, water, and nutrients [9,10].

Competition for natural resources is greatly intensified in the early stages of settlement (1st and 2nd years of age), when competition with spontaneous vegetation, especially grasses and other herbaceous or shrubby species, can reduce the efficiency of resource use by young trees [11]. Competition for light directly interferes with the photosynthetic rate [12] and apical growth [11], while edaphic competition compromises water and nutrient absorption, exerting negative effects on plant vigor and biomass production [13]. In addition, competition for light can alter the allocation of biomass between plant compartments, causing plants to prioritize stem elongation over thickening, which compromises the stability and commercial value of the wood [13]. In contrast, weeds also provide benefits to Pinus taeda plantations, such as increased biotic diversity, maintenance of moisture in the litter and soil, nutrient cycling, control of the erosion process, and water runoff [14]. Thus, details on weed control practices and their impacts on the growth and above-ground biomass production of Pinus taeda plantations are necessary to improve forest management and ensure high productivity in southern Brazil.

Weed control is a fundamental silvicultural practice for maximizing the growth and productivity of Pinus taeda plantations, especially in the first years after planting. Among the main methods are manual mowing and chemical control. However, control strategies, such as chemical control in the planting line or control up to a certain point, can contribute to reducing competition for light, water, and nutrients and thus promote better conditions for tree development [15,16] and reduce forestry operation costs. Pellens et al. [14] found that effective control of weed competition can increase the stem volume of Pinus taeda by >60% by 2 years of age in southern Brazil. However, Rubilar et al. [17] reported an increase of 56.7 m³ ha⁻¹ in the stem volume of Pinus radiata at 21 years of age in Chile, under weed control. However, studies that quantitatively assess the effects of different weed control regimes on Pinus taeda biomass production after canopy closure, which occurs at approximately 4 years old, are still limited. Therefore, more information is essential to improving technical recommendations and ensuring the success of commercial plantations, considering soil, climate, and operational specificities.

Thus, the present study aims to evaluate whether weed control increases the growth and above-ground biomass production of Pinus taeda plantations in southern Brazil. The hypothesis tested is that weed competition management, especially physical control, promotes a significant increase in tree growth and accumulated biomass when compared to the absence of control, due to enhancing the water and nutrient cycle of the soil.

2. Materials and Methods

2.1. Experimental Sites

This study was conducted on identical plots at two sites in southern Brazil, in the Caçador (Cac) (26°44′49″ S and 51°04′19″ W, 1030 m altitude) and Campo Belo do Sul (Cam) municipalities (28°00′22″ S and 50°51′23″ W, 957 m altitude), in the Santa Catarina state (Figure 1a–c). The sites are part of the experimental network of the Cooperative Research Program on Pinus in Brazil (PPPIB; http://www.ipef.br/pppib), coordinated by the Institute of Forest Research and Science (IPEF). The native vegetation of the region is predominantly mixed rainforest—Floresta Ombrófila Mista (Mata Atlântica biome). The prevailing climate in the region of the sites is classified as humid subtropical with mild summers (Cfb), according to the Köppen classification, and is characterized by evenly distributed rainfall, no dry season, average temperature of the hottest month below 22 °C, and occasional frosts [18].

The site areas were previously cultivated with Pinus taeda plantations that are 34–38 years old. The physical and chemical characteristics of the soil before this study started are shown in

Table 1. Main physical and chemical soil characteristics in the 0–0.2 m and 0.2–0.4 m layers at sites in southern Brazil.

Site	Depth	Clay	Silt	Sand	SOM	pH	P	K	Ca	Mg	H + Al	BS	CEC
	cm	g kg−1				–	mg dm−3		cmolc dm−3
Cac	0–20	653	241	106	45.0	4.9	3.3	73.7	1.4	0.3	13.3	2.0	15.3
	20–40	698	202	100	35.7	4.8	2.6	54.3	1.3	0.4	12.1	2.0	14.1
Cam	0–20	605	284	111	51.7	4.6	2.4	50.0	1.0	0.5	16.5	1.7	18.2
	20–40	606	251	143	38.0	4.5	1.9	34.7	0.5	0.3	16.0	1.0	17.0

Sites: Cac—Caçador and Cam—Campo Belo de Sul municipality, which are in southern Brazil. P: phosphorus; K: potassium; Ca: calcium; Mg: magnesium; BS: base saturation; H + Al: potential acidity; and CEC: cation exchange capacity. Clay, silt, and sand were determined by pipette method; pH in water is 1:1 ratio; SOM: soil organic matter determined by Walkley Black method; P and K were extracted by Mehlich-1; Ca, Mg, H + Al, BS, and CEC were extracted and determined by KCl 1 mol L⁻¹.

. The climate, soil, and management characteristics of the sites are shown below.

The Caçador site has an average annual rainfall of 1626 mm year⁻¹ and an average annual temperature of 16.7 °C (average of the last 30 years prior to this study) (Figure 1c). Data were obtained from the weather station in Caçador/SC, Brazil [19]. The relief of the experimental area is classified as moderately undulating, with soil classified as Oxisol [20]. On 8 April 2019, the site was manually planted with loblolly pine (Pinus taeda L.—WestRock) with a spacing of 2.5 × 2.5 m (1600 trees ha⁻¹) using 2nd-generation genetically selected material. Seedlings transplanted were 120 days-old, 40 cm in height, and 4.0 mm in stem diameter. The soil was prepared with subsoiling in rows at 60 cm depth.

The Campo Belo do Sul site has an average annual rainfall of 1760 mm year⁻¹ and an average annual temperature of 16.6 °C (average of the last 30 years prior to this study) (Figure 1c). The data were obtained from the weather station in Lages/SC, Brazil [19]. The relief of the experimental area is classified as undulating, with soil classified as Inceptisols [20]. On 3rd August 2019, the site was manually planted with loblolly pine (Pinus taeda L.—Mobasa) with a spacing of 2.5 × 3.0 m (1333 trees ha⁻¹) using 2nd-generation genetically selected material. Seedlings transplanted were 120 days old, 40 cm in height, and 4.0 mm in stem diameter. The soil was prepared with subsoiling in rows at 45 cm depth.

At both sites, planting techniques followed minimum cultivation practices, such as soil preparation in rows, minimal use of fertilizers, mechanical weed control, and maintenance of forest residues, without burning residues [21]. No fertilization was applied during this study, which is a common practice for Pinus taeda forestry in Brazil [22,23]. Chemical control of leaf-cutting ants (Atta spp. and Acromyrmex spp.) was carried out through the systematic application of granular baits throughout the experimental period, whenever necessary.

2.2. Experimental Design and Treatments

The experiment was designed in random blocks with 9 treatments and 3 blocks, totaling 27 plots that were evaluated (Figure 1d). The plots measured 20 × 40 m (width × length; 800 m²), with eight rows and 16 plants per row (totaling 128 trees per plot). All evaluations were performed in an internal plot measuring 10 × 30 m (width × length; 300 m²), with 4 rows and 12 plants in each row (totaling 48 trees evaluated).

This study was designed to identify the best method and the most appropriate time for controlling weed competition (

Table 2. Description of weed control operations and their respective abbreviations.

Treatment Abbreviation	Operation Description
NC	No control—with weeds present
PC	Physical weed control
CC-T	Chemical weed control in total area
CC-R	Chemical weed control in rows
C6m	Weed control up to 6 months after planting
C12m	Weed control up to 12 months after planting
C18m	Weed control up to 18 months after planting
C24m	Weed control up to 24 months after planting
COC	Company operational weed control

). The nine treatments were (i) NC—no weed control, i.e., weeds always present; (ii) PC—physical weed control; (iii) CC–T—chemical weed control in total area; (iv) CC–R—chemical weed control in rows (1.2 m wide); (v) C6m, (vi) C12m, (vii) C18m, and (viii) C24m—weed control up to 6, 12, 18, and 24 months after planting; and (ix) COC—company operational weed control. Physical and chemical weed control (PC, CC–T, and CC–R) were carried out continuously until the tree canopy closed, which occurred at approximately four years of age.

Weed control at different stages (C6m, C12m, C18m, and C24m) was carried out using chemical control across the entire area until the time established for each treatment; after this period, no further intervention was carried out. Chemical weed control was performed by applying herbicides in the following phases: (i) pre-planting with glyphosate (2.0 L ha⁻¹), Chopper (imazapyr—2.0 L ha⁻¹), or Fordor (isoxaflutole—0.2 L ha⁻¹), in a spray volume of 0. 5%; and (ii) post-planting with Missil (haloxifop-P-methyl—0.350 L ha⁻¹) or Esplanade (indaziflam—0.150 L ha⁻¹), in a spray volume of 0.5%. The herbicides were selected due to their broad spectrum of weed control in the study region. Weed control interventions and their timing were defined following the best internal practice guidelines and forest certification standards [24].

2.3. Floristic Composition and Biomass of Weed Species

At each site, the floristic composition of weed species was assessed at 6 months old. Sampling was conducted within the plots without any physical or chemical control, i.e., in the NC treatment (Figure 2a,b). The collection was carried out using quadrants installed at random locations within the plots (Figure 2c), at points previously defined in the rows and between rows of planting, using a template with dimensions of 1 × 1 m (width × length; 1 m²; Figure 2d), with six replicates per plot, in the three blocks, totaling 18 samples per site. All plant material inside the template was collected for subsequent species identification and determination of dry matter mass (Figure 2d). Weeds were classified into three categories according to their structure in the community: (i) abundant—species with a high number of individuals and high frequency (Figure 2e); (ii) dominant—larger species, with predominant canopy occupation (Figure 2f); and (iii) other weeds—those that did not fit into the previous categories, i.e., neither abundant nor dominant. The biomass stock of weeds, identified according to their structure in the ecological community at the respective locations, is shown in

Table 3. Biomass stock of weeds and their structure in the ecological community.

Site	Weed Structure	Weed Biomass	Main Species
		(Mg ha−1)
Cac	Abundant	1.98	Melinis minutiflora; Baccharis dracunculifolia; Baccharis uncinella; Cyperus esculentus
	Dominant	7.76	Brachiaria mutica; Cyperus esculentus; Austroeupatorium inulaefolium
	Other weeds	1.17
	Total	10.91
Cam	Abundant	6.21	Baccharis spp.; Melinis minutiflora; Ageratum conyzoides
	Dominant	9.67	Melinis minutiflora; Baccharis uncinella; Callisia repens; Miconia theaezans
	Other weeds	2.34
	Total	18.22

Sites: Cac—Caçador and Cam—Campo Belo de Sul municipality, in southern Brazil.

.

2.4. Tree Growth Assessment

Tree growth was assessed through annual measurements taken for all trees in the plots on each anniversary of planting at each site (between 2019 and 2024). The diameter at breast height (DBH—1.3 m above ground level) was measured on all trees in the plots using a tape measure, with an accuracy of 0.1 cm. Tree heights were measured on all trees in the plot using a Haglöf Vertex IV ultrasonic height meter (Haglöf Inc., Torsång, Sweden), with an accuracy of 0.1 m.

The stem volume with bark (Vol) was determined by rigorous cubing of three trees of average diameter in each treatment. The cubing points were determined based on relative heights, which are 0, 10, 20, 30, 40, 50, 60, 70, 80, 90, and 100% of the commercial height. The tip, the terminal portion of the trunk, was also evaluated by measuring the diameter at 50% of the length. The commercial height was determined considering the minimum usable diameter: Cac = 6 cm and Cam = 8 cm. Stem volume was calculated at five years of age, based on the DBH and height data for each tree, by determining the individual stem volume with bark and extrapolating to the hectare, considering the mortality of plants in each plot. The calculation of Vol was performed based on the equation proposed by [25] and is expressed in Equation (1) below:

$$Vol \left({\mathrm{m}}^3 {\mathrm{h}\mathrm{a}}^{-1}\right) = {\displaystyle \frac{{DBH}^2 \times \pi }{4}} \times H \times ff \times density$$

(1)

where Vol indicates the stem volume with bark per hectare; DBH indicates the stem diameter (in m); π is the mathematical constant pi (3.141593); H represents the tree height (in m); ff indicates the form factor of Pinus taeda (0.45); density refers to planting density (trees ha⁻¹), considering plant mortality within each plot.

2.5. Assessment of Tree Above-Ground Biomass

To evaluate the above-ground biomass production of trees under different conditions of weed competition, destructive sampling was performed at five years of age at each site. A total of 3 trees per treatment were evaluated in each of the three blocks, totaling 27 trees sampled per site. The trees were selected based on the arithmetic mean of the DBH, according to data from the forest inventory for the same year.

The selected trees were cut at ground level using a chainsaw, and the above-ground biomass was separated into compartments: needles, branches, bark, and stemwood. Then, the wet biomass of each compartment was weighed in the field using a portable scale with a maximum capacity of 50 kg and an accuracy of 10 g. For the needle and branch compartments, subsamples of approximately 250 g were collected from the lower, middle, and upper thirds of the crown and properly conditioned. A representative sample of needles was collected after homogenization of the material. The branch samples were obtained by collecting representative subsamples in relation to their diameter and position along the trunk. The stemwood was separated into five sections corresponding to 20% of the relative height, with 3 cm thick discs collected at the midpoint of each section, i.e., 10, 30, 50, 70, and 90% of the commercial height. From the trunk, discs were taken that were 3 cm thick, and these were separated into bark and stemwood and set aside. All subsamples were weighed in the field with a precision scale to obtain the wet weight, stored in paper bags, and sent to the laboratory, where they were dried in an oven with forced air circulation at 70 °C until they reached a constant weight. The moisture content was calculated based on the ratio between the wet weight and dry weight. The dry biomass of each compartment was estimated from the moisture content of the respective samples, and the values were extrapolated to the entire tree, to the plot, and, subsequently, to the hectare. The sum of all fractions resulted in the estimate of the total above-ground biomass of the trees.

2.6. Statistical Analysis

The data of growth (height, diameter, and stem volume) and above-ground biomass production (needles, branches, bark, stemwood, and total) were submitted to analysis of variance (ANOVA). The normality of the residuals and the homogeneity of the variance were evaluated using the Shapiro–Wilk and Bartlett tests, respectively. Results were considered significant at p < 0.05. The means of each variable were compared using Tukey’s test (p < 0.05). ANOVA and Tukey’s test were performed using the “ExpDes.pt” package in R [26]. To verify the correlation effects between the response variables, the growth and above-ground biomass production data were submitted to Pearson correlation and regression adjustment using “corrplot” [27] and “factoextra” [28], in the R software. The quality of the model fit was evaluated based on the coefficient of determination (R²), the root mean square error (RMSE), and inspection of the residuals. Scatter plots with the adjusted equations were used to visualize the relationship between the variables. All analyses were conducted in R software [29]. From this, growth variables were selected, and linear regressions were proposed to indirectly predict above-ground biomass production. After adjusting the equations, the predicted biomass values were compared with the actual observed data to verify the accuracy of the estimates and the applicability of the developed equations. Statistical analyses and the generation of scatter plots with the adjusted curves were performed in the R environment.

3. Results

3.1. Tree Growth

Plant height was not affected by weed competition control (Figure 3a,b). However, although not significant, physical weed control (PC) at the Cam site resulted in a 1 m increase in tree height compared to the control (9.77 vs. 10.73 m). Weed competition control influenced the stem diameter growth at both sites (Figure 3c,d). The lowest stem diameter values were found in trees grown without weed control (NC) at both sites. The stem diameter of the trees increased by an average of 32% under physical weed control (PC) and 28% under total area chemical control (CC–T). Mat competition control significantly affected the stem volume at the Cac site (Figure 3e), but the stem volume did not differ at the Cam site (Figure 3f). Physical weed control resulted in a 51% and 72% increase in stem volume at the Cam and Cac sites, respectively.

3.2. Tree Above-Ground Biomass

The above-ground biomass production at the Cac site was significantly affected by weed control (Figure 4a,c,e,g,i). Weed control up to 24 months after planting (C24m) increased the biomass production of needles, branches, and bark by 172, 225, and 116%, respectively, compared to no weed control (NC). Weed control by the physical (PC), chemical in total area (CC–T), chemical in rows (CC–R), and operational (COC) methods, as well as at 6 (C6m) and 24 months after planting (C24m), increased the stem biomass production by 72% on average (28.2 vs. 16.4 Mg ha⁻¹), compared to no weed control (NC). The total biomass production at the Cac site with weed control up to 24 months after planting (C24m) more than doubled (27.3 vs. 58.6 Mg ha⁻¹) compared to no weed control (NC). The above-ground biomass production in the compartments at the Cam site was not affected by weed control (Figure 4b,d,f,h,j). However, although not significant, physical weed control (PC) at the Cam site resulted in a 62 and 48% increase in stem and total biomass, respectively.

3.3. Proportion of Tree Above-Ground Biomass

The proportion of above-ground biomass production varied between the weed control methods and evaluated sites (Figure 5). The highest proportions of stemwood (62%) were found for chemical weed control in the total area (CC–T) at the Cam site, while the lowest proportions (48%) were observed for weed control up to 24 months after planting (C24m) at the Cac site. In contrast, the lowest proportions of needles (13%) were found for CC–T at the Cam site, while the highest proportions (19%) were found for C24m. The branches biomass varied between 14 and 23% of the total biomass production. The highest proportions of branches were observed for C24m. The bark biomass ranged from 8 to 11% of the total biomass production, with lower proportions being observed for company operational weed control (COC) at the Cac site and higher values for weed control up to 6 months after planting (C6m) at the Cac site.

3.4. Correlation Between Growth and Above–Ground Biomass of Trees

The growth data (DBH, H, and Vol) showed a strong correlation with the above-ground biomass production data in the different compartments (Figure 6). The DBH variable showed a high correlation with the stemwood, total, and bark biomass production, with values of 0.81, 0.80, and 0.70, respectively. For the H variable, a correlation of 0.62 was observed for stemwood. For the other compartments, a moderate to weak correlation was observed (<0.5).

3.5. Indirect Estimation of Above–Ground Biomass of Trees

The regression analysis between the measured and predicted biomass production was significant for the stemwood total biomass (Figure 7). In contrast, the regression analyses between the measured and predicted biomass production for the needles, branches, and bark were not significant. The correlation coefficients (R²) between the variables were 0.39, 0.38, 0.48, 0.65, and 0.65 for the needles, branches, bark, stemwood, and total biomass, and the regressions were 1.67, 2.18, 0.77, 3.41, and 5.91, respectively. The root mean square errors for the needles, branches, bark, stemwood, and total biomass regressions were 1.67, 2.18, 0.77, 3.41, and 5.91, respectively.

4. Discussion

The results demonstrate that both physical and chemical control over the entire area have a significant effect on the growth of Pinus taeda, especially in terms of DBH and stem volume increase. These results broadly corroborate the international literature, which highlights the importance of silvicultural practices, such as weed control, subsoiling, and fertilization, in increasing forest productivity, which has been extensively investigated in the United States [15,16,30,31,32,33,34]. In South America, consistent results have also been reported for Pinus spp., as observed by Rubilar et al. [17] in Chile and Pellens et al. [14] in Brazil. Physical weed control resulted in gains of 51% and 72% in stem volume at the Cam and Cac sites, respectively. These gains corroborate the results reported by Pellens et al. [14], who recorded a 61% increase in the individual productivity of Pinus taeda in southern Brazil when comparing areas with total weed control to areas without control.

Similar results were also reported in Argentina, with an 84% increase in stem volume in three-year-old Pinus taeda plantations subjected to intensive weed control [35]. This reinforces that interspecific competition, especially in the early years of development, is one of the main limiting factors for the growth of Pinus taeda in subtropical environments in South America. Positive responses may be related to reduced competition for essential resources, such as water, light, and nutrients, which is promoted by effective control of competing vegetation. During the early stages of development, Pinus taeda seedlings have an underdeveloped root system and reduced canopy, which makes them highly sensitive to the effects of interspecific competition [16,17]. Weeds, such as grasses and fast-growing herbaceous species, have a high capacity to absorb water and nutrients, especially in the surface layer of the soil, as well as a high capacity to form dense aerial parts, which limits light interception by Pinus taeda seedlings [35]. Campoe et al. [11] found that suppressed (non-dominant) Pinus taeda plants reduce light use efficiency, which reduces photosynthesis and thus above-ground growth. This reinforces that reducing competition allows seedlings to direct a greater proportion of assimilates toward growth in height and diameter, favoring both root system expansion and photosynthetic activity, which has already been verified by Kulmann et al. [13]. Studies show that removing competing vegetation results in greater soil water availability, reduced water deficit, and increased water use efficiency [14,30]. In addition, competition for light becomes critical in situations where there is high coverage of herbaceous and shrubby species. Under these conditions, the limitation of photosynthetically active radiation directly impacts the photosynthetic rate of seedlings, restricting biomass accumulation [12,15]. Thus, physical or chemical control of vegetation not only minimizes the effects of direct competition, but also creates a more favorable microclimatic environment, with reduced relative humidity at ground level, a lower incidence of pests associated with weed control, and improved soil temperature, favoring physiological processes that occur in Pinus spp. [17].

The results obtained herein demonstrate that weed control had a significant effect on biomass production in different parts of Pinus taeda at the Cac site, with significant increases in the biomass production of needles (172%), branches (225%), and bark (116%) compared to the treatment without control (NC). This pattern was maintained for stem biomass, which showed an average increase of 72% among the different control treatments, in addition to a 115% increase in total biomass (58.6 vs. 27.3 Mg ha⁻¹). This reinforces that interspecific competition for limiting resources, such as water, light, and nutrients, has a strong influence on biomass accumulation, especially in the compartments most sensitive to carbon limitation, such as the needles and branches, which are responsible for light capture and photosynthesis [16,30]. The proportionally greater increase in the biomass of the needles and branches suggests that, in the absence of competition, trees prioritize the development of tissues that are responsible for capturing resources, creating a functional basis that supports secondary growth (stem) in subsequent years. In addition, prolonged management up to 24 months (C24m) provided the greatest gains, which indicates that the critical period of competition for Pinus taeda in this environment extends beyond the first year of establishment. This finding is consistent with the literature, which suggests that Pinus taeda remains highly sensitive to competition in the first two years after planting, a period that is fundamental for defining its long-term growth potential [14,17]. In contrast, at the Cam site, no statistical differences in above-ground biomass production were observed, which may be related to local soil and climatic conditions, which may have been more favorable to the development of Pinus taeda even in the presence of competition. Studies report that high rainfall, such as the site average (1760 mm year⁻¹), enables greater water retention capacity, root deepening, and resistance to periods of water stress, resulting in less competitive pressure from weed control [9,15,36]. The difference in response between the sites highlights the importance of the local context in defining the effectiveness of weed competition management. Previous studies have reported that the response to weed control is highly dependent on soil characteristics, the competing vegetation composition, and climatic conditions, being more pronounced in environments where resources are more limited or where competitive pressure is more intense [30,35].

We found that the distribution of above-ground biomass among the different compartments of Pinus taeda varied significantly depending on the weed control treatments and the characteristics of the sites that were evaluated. The highest proportions of stemwood biomass were observed in the treatment with chemical control in the total area (CC–T) at the Cam site, reaching 62% of the total biomass. This result suggests that weed control in the total area favored a pattern of C allocation that was directed toward increasing the diameter and stem volume, which is desirable from a silvicultural point of view, given the economic importance of the trunk as the main commercial product [16,30]. This is possibly related to increased light use efficiency contributing to greater photosynthesis and increased C partitioning, as reported by Campoe et al. [11]. In contrast, the lowest proportions of stem biomass (48%) were found in the weed control treatment up to 24 months after planting (C24m) on the Cac site. This lower allocation to the stem can be explained by the plants’ greater relative investment in canopy compartments, such as needles and branches, which indicates a functional prioritization of the tree of structures related to capturing resources (light and CO₂), probably in response to the history of competition in the early stages or the more restrictive environmental characteristics of the Cac site [17,35,37]. The increase in the proportion of needles and branches in the treatments with prolonged control may reflect a compensatory strategy by the trees, which seek to maximize their leaf area in order to increase their photosynthetic capacity and thus sustain more vigorous subsequent growth of the stem. This allocation pattern has already been described in other studies with Pinus taeda, especially in situations of recovery after initial stress due to competition or in environments with lower nutrient and water availability [13,30].

The results showed a strong correlation between growth (DBH and height) and above-ground biomass production, particularly in the most economically relevant compartments, such as the stemwood and total biomass. The DBH variable stood out as the best predictor, with high correlations for stemwood (r = 0.81), total biomass (r = 0.80), and bark (r = 0.70). This confirms that the increase in diameter directly reflects the increase in stem volume and, consequently, the biomass accumulated in the trunk and the tree as a whole. This reinforces that DBH can be considered a robust and reliable variable for estimating the above-ground biomass in forest species, including Pinus taeda [17,30]. In addition, height showed a moderate correlation (r = 0.62) with stem biomass, but a weak or non-significant correlation with the other compartments. This may be due to the lower variability in height compared to the DBH, since, in Pinus taeda plantations, trees tend to grow more uniformly in height, while their diameter growth is more sensitive to local conditions of competition, management, and resource availability. Kulmann et al. [13] found an increase in DBH parallel to the increase in resources, especially nutrients, confirming the possibility of an increase in this variable. The regression analysis reinforces this trend, with highly significant models for estimating stem biomass and total biomass, as evidenced by the coefficients of determination (R²) of 0.65 for both and the low RMSE for the stem (3.41 Mg ha⁻¹) and total biomass (5.91 Mg ha⁻¹) in relation to their respective means. These values indicate that approximately 65% of the variation in stem and total biomass can be explained by the dendrometric growth variables. In contrast, the models for the needles, branches, and bark compartments were not significant, with R² values below 0.5, which reflects the high residual variability and low predictive capacity of these models. This reinforces that allometric models based on DBH and height are more accurate and robust for estimating stem and whole-tree biomass, while they have limitations when applied to crown components, which have greater structural variation and depend on specific characteristics of the tree’s architecture [17,30].

Therefore, for practical applications such as forest inventory, carbon stock estimates, and sustainable forest management, dendrometric variables, especially DBH, are efficient for predicting the commercial (stem) and total biomass. However, modeling the biomass of canopy compartments requires more specific approaches that possibly incorporate additional variables such as canopy projection area, leaf area index, or the use of technologies such as LiDAR and aerial photogrammetry to improve the accuracy of these estimates.

5. Conclusions

Controlling weed competition in the early stages of Pinus taeda cultivation is essential to promoting significant gains in growth and above-ground biomass production, especially in subtropical regions of southern Brazil. The physical control (PC) and total area chemical control (CC–T) methods were more responsive, resulting in average increases of 32 and 28% in the stem diameter, respectively, compared to the treatment without weed control (NC).

However, the results differ from site to site: (i) at the Cac site, the effects were more significant—vegetation control for 24 months (C24m) more than doubled the total biomass (58.6 vs. 27.3 Mg ha⁻¹). The stem biomass showed an average increase of 72% in the treatments with control; (ii) at the Cam site, although the effects were not significant, an increase of up to 62% in the stem biomass was observed under physical control. In addition, weed control affects the allocation of biomass between plant compartments. The proportion of stemwood reached 62% under the CC–T treatment at the Cam site, while the needles accounted for up to 19% under the C24m treatment at the Cac site. Furthermore, there was a strong correlation between the stem diameter and stemwood biomass (r = 0.81) and the total biomass (r = 0.80), which indicated that dendrometric variables can be used as good predictors of productivity.

Thus, silvicultural practices such as weed control until canopy closure represent an effective strategy for increasing the productivity of Pinus taeda plantations in southern Brazil. However, it is important to note that other relevant parameters—such as wood quality and phytosanitary aspects of the plantations—were not evaluated in this study, which may limit the scope of the conclusions presented.