Biomass Production and Nutritional Sustainability in Different Species of African Mahogany

Abstract

Wood from reforestation gains market value due to its sustainable and legal origin. Planted forests in Brazil play a crucial role in economic, social and environmental aspects, with Eucalyptus and Pinus dominating the timber sector. However, non-majority species, such as those of the Khaya genus, have attracted great commercial interest due to the quality of their wood, being seen as an alternative to Brazilian mahogany. This study aimed to evaluate the biomass production of Khaya spp. stands and the nutrient uptake impacts in different harvesting scenarios. The research area is in Reserva Natural Vale (RNV) in Sooretama, Espírito Santo state, Brazil. The study was conducted 9.5 years after the planting of the Khaya spp. monoculture at a spacing of five m × five m, and the base fertilization consisted of 150 g of yoorin thermophosphate and 15 g of FTE BR 12 per seedling. The seedlings were of seminal origin, coming from different regions of Brazil and corresponding to three species: Kkaya grandifoliola C.DC (Belém-PA), Khaya ivorensis A. Chev. (Linhares-ES) and Khaya senegalensis A. Juss. (Poranguatu-GO). K. senegalensis exhibited the highest percentage of bark, while K. ivorensis was found to have the highest percentage of leaves. The biomass of the stems and branches did not vary by species. The relative biomass proportions had the following order: branches > stems > bark > leaves. The stocks of Ca and Mg were higher for K. grandifoliola, exceeding those for K. senegalensis (22.1%) for Ca and for K. ivorensis (42.3%) for Mg. The lowest nutrient uptake occurred in the scenario in which only the stem was removed, with averages of 44.17, 10.43, 21.93, 52.59 and 9.97 kg ha⁻¹ for N, P, K, Ca and Mg, respectively. Compared to total biomass harvesting, this represents a reduction in export levels by 91.34% for N, 79.31% for P, 94.66% for K, 94.29% for Ca and 93.28% for Mg. The nutrient uptake assessment demonstrated that more conservative harvest scenarios resulted in lower nutrient losses, indicating the importance of forest management practices that prioritize soil and nutrient conservation. In summary, the findings of this study provide a solid basis for the sustainable management of Khaya spp., highlighting implications for productivity and nutrient dynamics on a small or medium scale.

1. Introduction

The intensive exploitation of native forest species has resulted in a significant reduction in the production of hardwoods, particularly in tropical and subtropical nations [1]. Furthermore, enormous tracts of continuous land will continue to be cleared due to the increase in international demand for and commerce involving agricultural products grown in tropical rainforest zones [2]. For example, the expansion of the Brazilian agricultural frontier, which includes processes such as the occupation of public or private forested lands, timber extraction, the introduction of livestock farming and the development of modern agricultural techniques, has resulted in the deforestation of approximately 8,558,237 ha in the last 5 years [3,4]. Currently, wood obtained from reforestation is valued in the global market due to its legalized and sustainable origin [5].

Planted forests play an important role in economic, social and environmental aspects. In Brazil, they correspond to an area of 10.2 million ha, are mostly plantations of Eucalyptus and Pinus and comprise 95% of the total timber supply to meet the industrial needs of the country [6]. However, plantations of non-majority species are generally found on small and medium scales, which together have great production potential, depending on the management of and practices used for the species [7,8].

In tropical conditions, species of the Khaya genus, belonging to the Meliaceae family, have aroused great commercial interest because they have excellent-quality timber for multiple uses, particularly for both domestic and international trade as a substitute for Brazilian mahogany (Swietenia macrophylla). In Australia, studies on the implementation and genetic development of K. senegelensis A. Juss. stands have already been conducted and they have shown potential for expansion as timber as their products are of high commercial value. On the other hand, plantations in Brazil using the Khaya genus focuses on three species (K. grandifoliola C.DC., K. ivorensis A. Chev. and K. senegalensis), covering approximately 50,000 ha, predominantly planted in the southeast of the country. These have shown good economic returns, adaptive characteristics, relative levels of pest resistance and good productivity [9,10,11,12].

In forest ecosystems, species play an important biogeochemical role, considering nutrients are stored in biomass compartments [13]. Moreover, the stock and redistribution of these nutrients can be influenced by edaphoclimatic factors, age, and soil fertility [14]. Despite African mahogany (Khaya spp.) being characterized by its adaptability to low-fertility soils, there is limited information on its biomass production and nutrient use efficiency (NUE) across different compartments, mainly those related to the sustainability of production after biomass harvesting [15,16].

Similar to eucalyptus cultivation, the majority of Khaya spp. plantations are built in areas with naturally low levels of soil fertility and water availability, requiring a high NUE [17,18]. This parameter links biomass production with physiological parameters, including nutrient absorption, assimilation, translocation and storage. This index also assists in selecting the species best adapted to local conditions, serving as a selection criterion alongside quality and productivity parameters [19].

In this context, understanding nutrient cycling is one of the parameters that must be evaluated to comprehend the effects of nutrition on a species of interest and to maintain forest productivity [20]. Adapted silvicultural practices and efficient harvesting systems are essential for nutritional maintenance throughout production cycles, as depending on the harvesting scenarios used, there is the removal of fractions with intense metabolic activity and, consequently, the leaching of essential nutrients [21,22,23].

Research that demonstrates the ability of forest stands to utilize the nutrients provided under tropical conditions must be carried out given the need to choose species capable of combining the conservation of natural resources, the growing demand for forest products with greater productivity in a short space of time and the export of nutrients through harvesting [24,25]. Therefore, we posited the following research questions: (i) Is there a difference in biomass production between Khaya spp.? (ii) Which biomass compartments have the highest concentrations and stocks of nutrients? (iii) Does maintaining branches and bark in a forest site promote a lower nutrient uptake? Therefore, the aim of this study was to evaluate the biomass production of Khaya spp. stands and the impacts of nutrient uptake in different harvesting scenarios.

2. Materials and Methods

The nutrient balance in forest ecosystems was explored by collecting soil and plant tissue samples to determine nutrient concentrations and stocks across different forest compartments (soil, litter and biomass). Soil samples were analyzed for chemical attributes, bulk density and macronutrient stocks, while accumulated litter and tree biomass were sampled to quantify nutrient concentrations and stocks in each forest component. The study simulated various forest harvesting scenarios, estimating the nutrient removal in each scenario and calculating the nutrient use efficiency (NUE) to assess the nutrient cycling in the Khaya spp. stands. The integration of these methods provided insights into nutrient dynamics and the impact of biomass harvesting on ecosystem nutrient balance.

2.1. Research Area

The research area is situated in the Reserva Natural Vale (RNV) in Sooretama, Espírito Santo state, Brazil. The region’s climate is classified as tropical with dry summer (Aw) according to the Köppen climate classification, characterized by a wet summer and dry winter. The average air temperature is 23.5 °C, with an average annual precipitation of 1294 mm [26]. The topography is mostly flat and the soil is classed as the Acrisol type, featuring a moderate A horizon and a textural B horizon [27].

Stand Characterization

The region of the Khaya spp. stands was originally inhabited by Eucalyptus spp. In the 80s, a monoculture of legume species was planted, followed by fallowing [28]. The planting of the Khaya spp. monoculture took place in 2013, at a spacing of 5 m × 5 m, through manually dug pits with dimensions of 30 cm × 30 cm × 30 cm and base fertilization consisting of 150 g of yoorin thermophosphate [CaO (18%)-MgO (7%)-P₂O₅ (17%)] and 15 g of FTE BR 12 [B (1.8%)-Cu (0.8%)-Mn (2%)-Zn (9%)-S (1%)] per seedling. The seedlings are of seminal origin, coming from different regions of Brazil and corresponding to three species: Kkaya grandifoliola (KG)—Belém/PA, Khaya ivorensis (KI)—Linhares/ES and Khaya senegalensis (KS)—Poranguatu/GO. The experiment consisted of a randomized complete block design (RCBD). Each block had three rectangular plots measuring 1250 m², with a usable area of 750 m² (15 m × 15 m) and a simple border, totaling 30 productive trees per repetition (Figure 1).

2.2. Soils

2.2.1. Collecting and Analyzing Samples

Disturbed and undisturbed soil samples were collected 9.5 years after planting for chemical characterization and measurement of organic carbon and total nitrogen concentrations (

Table 1. Chemical attributes of soil at different depths under Khaya spp. stands in Sooretama, Espírito Santo.

Species	Depth	pH	N	P	K	Ca	Mg	Al	H + Al	SB	CEC	BS	OM
	cm	H2O	g kg−1	-----mg dm−3-----		------------------------------Cmolc dm−3------------------						%	g kg−1
KG	0–20	5.91	1.59	2.33	26.00	1.66	0.51	0.00	2.19	2.25	4.44	50.57	24.4
	20–40	5.81	0.75	1.33	19.78	1.12	0.36	0.02	2.01	1.54	3.55	43.12	19.3
KI	0–20	5.76	1.87	2.33	23.67	1.33	0.43	0.00	2.22	1.84	4.06	45.17	22.4
	20–40	5.74	0.90	1.33	16.78	1.02	0.30	0.01	1.91	1.37	3.28	41.44	16.7
KS	0–20	6.07	1.79	2.67	26.11	2.29	0.60	0.00	1.73	2.97	4.71	62.15	26.1
	20–40	6.00	0.88	1.44	19.00	1.40	0.42	0.00	1.67	1.88	3.55	52.48	18.6

K. grandifoliola C.DC. (KG), K. ivorensis A. Chev. (KI), K. senegalensis A. Juss. (KS), pH in water, concentrations of phosphorus (P), potassium (K), calcium (Ca), magnesium (Mg), aluminum (Al), potential acidity (H + Al), sum of bases (SB), cation exchange capacity (CEC), base saturation (BS) and organic matter (OM) concentration.

). Disturbed samples at depths of 0–20 cm and 20–40 cm were collected for natural fertility analysis, with determination of phosphorus, potassium and sodium using a Mehlich-1 extractor; pH levels in water at a ratio of 1:2.5; H + Al using the SMP method; organic matter through oxidation with Na₂Cr₂O₇.2H₂O + H₂SO₄ 10 mol L⁻¹; and calcium, magnesium and aluminum using 1 mol L⁻¹ KCl extractor [29]. Total nitrogen analysis was performed using sulfuric acid titration in a method known as Kjeldahl digestion.

The Mehlich-1 method was chosen because of its advantages, including its ease of execution and low cost, which makes it viable for use in commercial soil analysis laboratories. This analytical procedure provides extracts by decantation and has been established in tropical locations, particularly in Brazil, for regional comparisons. In addition, the cation exchange capacity was calculated. It is defined as a soil’s total quantity of negative surface charges and measured commonly in commercial soil testing labs by summing cations (positively charged ions that are attracted to the negative surface charges in soil—Ca²⁺, Mg²⁺, K⁺ and Na⁺), as well as acid cations (H⁺, Al³⁺ and NH⁴⁺).

Soil samples were collected at five sampling points per plot (center and vertices) and per depth using a Dutch auger. These were then homogenized for the determination of their attributes in the laboratory, totaling 15 samples. For the undisturbed samples, three trenches were excavated in each plot to a depth of 40 cm, with one every 15 m segment, for a total of 27 trenches. The trenches were dug transversely inside the useable plot area, with a spacing of 25% relative to the planting rows. The sampling depths were as follows: 0–5, 5–10, 10–20, 20–30 and 30–40 cm. Soil density and macronutrient analyses were also conducted.

2.2.2. Bulk Density (BD) and Nutrient Stock

Soil bulk density was determined using the volumetric ring method with stainless steel rings, utilizing a TAI-type sampler (undisturbed sample auger) with a volume of 100 cm³, following the recommendations of Embrapa [30]. The stocks of macronutrients in the soil were calculated using Equation (1) [31,32].

$$\mathit{MS}=T \times \mathit{BD} \times {\displaystyle \frac{t}{10}}$$

(1)

where:

• MS = macronutrient stock in the soil, in Mg ha⁻¹;
• T = macronutrients’ concentration, in g kg⁻¹;
• BD = soil bulk density, determined as the average of three replications, in g cm⁻³;
• $T$= soil layer thickness, in cm.

2.3. Accumulated Litter

2.3.1. Litter Collection and Processing

The accumulated litter was sampled during June 2022. The samples were collected from the interior of the plots in each treatment. The accumulated litter consists of all organic materials present on the forest floor, including leaves, twigs, bark, and other miscellaneous materials in varying degrees of degradation. Thirty samples per plot were collected using a randomized technique, with a square wooden template 0.0625 m² in area.

After collection, the samples were stored in paper bags, labeled and then dried in a forced-air circulation oven at 65 °C until a constant weight was achieved. The dry weight of the material was determined using an analytical balance with 0.001 g precision. Subsequently, the samples were homogenized, ground in a Willey-type mill, sieved through a 20-mesh screen, stored in glass jars and sent to the laboratory for the determination of macronutrient concentrations.

2.3.2. Nutrient Concentration and Stock

Nitrogen was extracted using sulfuric acid digestion with titrimetric determination. The concentrations of P, K, Ca and Mg were determined by nitro-perchloric digestion [29,33]. The stock of elements was calculated using the equation provided by Cuevas and Medina [34]:

$$S_{litter}=\left[Nutrient\right]\times DML$$

(2)

where:

• $S_{litter}$: stock of nutrients in litter (g ha⁻¹);
• $\left[Nutrient\right]$: nutrient concentrations in litter (g kg⁻¹);
• $\mathrm{DML}$: dry mass of accumulated litter (kg ha⁻¹).

2.4. Biomass

2.4.1. Dendrometric Characteristics

A forest inventory was conducted at 9.5 years of age. The diameter at breast height ($\mathit{d}\mathit{b}\mathit{h}$) at 1.30 m from the ground, total height ($\mathit{H}\mathit{t}$) and commercial height ($\mathit{H}\mathit{c}$) of all trees within the usable plot area were measured using a mechanical caliper and a Vertex hypsometer. Commercial height was evaluated up until the first stem bifurcation.

Four trees per block were selected based on diameter range, totaling 12 plants per treatment. On each sample tree, circumferences were measured at heights of 0.1 m, 0.5 m, 1.3 m and 2.0 m; thereafter, they were measured every 1.0 m up to the commercial height.

The volume of the stem with bark (commercial) was calculated using the Smalian method. Based on the values of $\mathit{d}\mathit{b}\mathit{h}$, $\mathit{H}\mathit{c}$ and stem volume of the felled trees, regression models were adjusted to predict individual and total stand volumes [35] (

Table 2. Mathematical models for adjusting commercial volume equations in Khaya spp. stands in Sooretama, Espírito Santo.

Kopezky–Gehrhardt	$\mathit{v}={\mathit{\beta }}_0+{\mathit{\beta }}_1\mathit{d}\mathit{b}\mathit{h}+\mathit{\epsilon }$
Hohenadl and Krenn	$\mathit{v}={\mathit{\beta }}_0+{\mathit{\beta }}_1\mathit{d}\mathit{b}\mathit{h}+{\mathit{\beta }}_2{\mathit{d}\mathit{b}\mathit{h}}^2+\mathit{\epsilon }$
Husch	$\mathit{l}\mathit{n}\left(\mathit{v}\right)={\mathit{\beta }}_0+{\mathit{\beta }}_1\mathit{l}\mathit{n}\left(\mathit{d}\mathit{b}\mathit{h}\right)+\mathit{\epsilon }$
Brenac	$\mathit{l}\mathit{n}\left(\mathit{v}\right)={\mathit{\beta }}_0+{\mathit{\beta }}_1\mathit{l}\mathit{n}\left(\mathit{d}\mathit{b}\mathit{h}\right)+{\mathit{\beta }}_2(1/\mathit{d}\mathit{b}\mathit{h})+\mathit{\epsilon }$
Spurr without β0	$\mathit{v}={\mathit{\beta }}_1\left({\mathit{d}\mathit{b}\mathit{h}}^2{\mathit{h}}_{\mathit{c}}\right)+\mathit{\epsilon }$
Combined models of Spurr	$\mathit{v}={\mathit{\beta }}_0+{\mathit{\beta }}_1\left({\mathit{d}\mathit{b}\mathit{h}}^2{\mathit{h}}_{\mathit{c}}\right)+\mathit{\epsilon }$
Stoate	$\mathit{v}={\mathit{\beta }}_0+{\mathit{\beta }}_1\mathit{h}+{\mathit{\beta }}_2{\mathit{d}\mathit{b}\mathit{h}}^2+{\mathit{\beta }}_3\left({\mathit{d}\mathit{b}\mathit{h}}^2{\mathit{h}}_{\mathit{c}}\right)+\mathit{\epsilon }$
Naslund	$\mathit{v}={\mathit{\beta }}_0+{\mathit{\beta }}_1{\mathit{d}\mathit{b}\mathit{h}}^2+{\mathit{\beta }}_2\left({\mathit{d}\mathit{b}\mathit{h}}^2{\mathit{h}}_{\mathit{c}}\right)+{\mathit{\beta }}_3\left(\mathit{d}\mathit{b}\mathit{h}{{\mathit{h}}_{\mathit{c}}}^2\right)+{\mathit{\beta }}_4{{\mathit{h}}_{\mathit{c}}}^2+\mathit{\epsilon }$
Meyer	$\mathit{v}={\mathit{\beta }}_0+{\mathit{\beta }}_1\mathit{d}\mathit{b}\mathit{h}+{\mathit{\beta }}_2{\mathit{h}}_{\mathit{c}}+{\mathit{\beta }}_3{\mathit{d}\mathit{b}\mathit{h}}^2+{\mathit{\beta }}_4\left({\mathit{d}\mathit{b}\mathit{h}}^2{\mathit{h}}_{\mathit{c}}\right)+{\mathit{\beta }}_5\left(\mathit{d}\mathit{b}\mathit{h}{\mathit{h}}_{\mathit{c}}\right)+\mathit{\epsilon }$
Schumacher and Hall	$\mathit{l}\mathit{n}\left(\mathit{v}\right)={\mathit{\beta }}_0+{\mathit{\beta }}_1\mathit{l}\mathit{n}\left(\mathit{d}\mathit{b}\mathit{h}\right)+{\mathit{\beta }}_2\mathit{l}\mathit{n}\left({\mathit{h}}_{\mathit{c}}\right)+\mathit{\epsilon }$
Spurr	$\mathit{l}\mathit{n}\left(\mathit{v}\right)={\mathit{\beta }}_0+{\mathit{\beta }}_1\mathit{l}\mathit{n}\left({\mathit{d}\mathit{b}\mathit{h}}^2{\mathit{h}}_{\mathit{c}}\right)+\mathit{\epsilon }$

$\mathit{v}$ = volume (m³); $\mathit{d}\mathit{b}\mathit{h}$ = diameter at 1.30 m above the ground (cm); ${\mathit{h}}_{\mathit{c}}$ = commercial height (m); $\mathit{l}\mathit{n}$ = Neperian logarithm; ${\mathit{\beta }}_{\mathit{i}}$ = adjusted model parameters (i = 0, 1, 2…n); and $\mathit{\epsilon }$ = estimation error.

).

2.4.2. Biomass Quantification

Biomass was determined through the direct (destructive) method, whichinvolved segmenting each tree into stems, bark, branches and leaves [36,37]. The components were weighed separately using a digital hanging scale with a maximum capacity of 300 kg to obtain the wet weight.

For wood sampling, five discs approximately 5.0 cm thick were obtained at the base, 25%, 50%, 75% and 100% of the commercial height [22,38]. Bark samples were collected from the wood discs, resulting in a composite sample. Live branches and leaves were sampled from the crown´s lower, middle and upper thirds [39].

The moisture content and dry biomass weight of the compartments were determined from the fresh sample weights dried in a 75 °C oven. The dried stems with bark samples were weighed, and the compartments were individualized, as per Picard et al. [39] (Equation (3)):

$${\mathit{B}}_{\mathit{t}}={\displaystyle \frac{{\mathit{F}\mathit{W}}_{\mathit{c}}\times {\mathit{D}\mathit{W}}_{\mathit{s}}}{{\mathit{F}\mathit{W}}_{\mathit{s}}}}$$

(3)

where:

• $B_t$: total dry biomass of a given compartment (kg);
• ${FW}_c$: fresh weight of a given compartment (kg);
• ${DW}_s$: dry weight of the samples (kg);
• ${FW}_s$: fresh weight of the samples (kg).

The bark biomass of the stem was calculated using the average percentage of bark in this compartment. The dry biomass production per hectare for each tree compartment was calculated based on the individual dry biomass of each fraction and the number of trees per usable plot extrapolated per hectare. By summing the compartments, the individual dry biomass per species was determined.

The regression models were then updated to predict biomass based on $dbh$ and $H_c$ values, with the best fits for stems, bark, leaves, branches and total biomass selected using adjusted determination coefficient ($R_{adj}^2$) and the standard error of the estimate ($S_{yx}\%$) [40] (

Table 3. Adjusted equations and their respective statistics for estimating the biomass of Khaya spp. trees at 9.5 years old in Sooretama, Espírito Santo.

Species	Equation	${\mathit{R}}_{\mathit{a}\mathit{d}\mathit{j}}^2$	Syx%
Stems
KG	$Y=-0.1692+0.0739{dbh}^2+0.0171\left({dbh}^2{h}_c\right)$	0.968	9.5
KI	$Y=-56.5676+0.1234 {dbh}^2+13.2225h_c$	0.961	7.3
KS	$Y=-3.35436+0.44785h_c+0.08274{dhb}^2+0.01587\left({dbh}^2{h}_c\right)$	0.965	9.9
Bark
KG	$Y=1.5418+0.1804h_c+0.0294{dbh}^2+0.0011\left({dbh}^2{h}_c\right)$	0.841	16.8
KI	$Y=-20.607+0.991dbh+2.606h_c$	0.681	22.1
KS	$Y=7.9288-0.0239{dbh}^2+0.0111\left({dbh}^2{h}_c\right)$	0.774	14.2
Leaves
KG	$Y=28.5409-3.7886dbh+0.1391{dbh}^2$	0.685	35.3
KI	$Y=-118.2059+10.9818dbh-0.2127{dbh}^2$	0.396	23.2
KS	$Y=37.46-3.743dbh+0.0849{dbh}^2+0.0041\left({dbh}^2{h}_c\right)$	0.439	36.9
Branches
KG	$Y=-135.1929+14.7543dbh-0.0175{dbh}^2{h}_c$	0.566	48.3
KI	$Y=31.3925+0.3313{dbh}^2-0.0264\left({dbh}^2{h}_c\right)$	0.311	25.2
KS	$Y=128.8834+6.5241dbh-30.8537h_c$	0.807	19.5
Total Biomass
KG	$Y=-164.9034+19.1161dbh$	0.814	24.3
KI	$Y=-109.034+14.588dbh+3.603h_c$	0.681	16.4
KS	$Y=517.66-87.845h_c-0.515{dbh}^2+0.164\left({dbh}^2{h}_c\right)$	0.861	11.3

KG = Khaya grandifoliola C.DC.; KI = Khaya ivorensis A. Chev.; KS = Khaya senegalensis A. Juss.; $R_{adj}^2$ = adjusted coefficient of determination; $S_{yx}\%$ = standard error of the estimate in percentage; $Y$ = estimated biomass (kg tree⁻¹); $dbh$ = diameter at 1.30 m above the ground (cm); $h_c$ = merchantable height (m).

).

2.4.3. Nutrient Concentration and Stocks by Compartments and Harvest Scenarios

After drying, the samples were processed in a Willey-type mill with 20-mesh sieves and stored for subsequent chemical analysis and determination of macronutrient concentration in plant tissue. Nitrogen titration was performed by sulfuric digestion, and the concentrations of P, K, Ca and Mg were determined through nitro-perchloric digestion [29,33].

The nutrient stock per hectare of each tree fraction is a function of the product of the dry biomass of different compartments per hectare and the nutrient concentration in each corresponding fraction. Based on these data, the amounts of nutrients exported in different harvesting scenarios were estimated (

Table 4. Simulated forest harvesting scenarios based on nutrient uptakes.

Scenario	Description	Potential Use	Equation
C1	Total removal of aboveground biomass	Energy use	$NC \left(C1\right)={\displaystyle \frac{Nutrient in the soil+litter}{Nutrient in the crown+bark}}$
C2	Removal of stem biomass with bark and branches	Multiple use	$NC \left(C2\right)={\displaystyle \frac{Nutrient in the soil+litter+leaves}{Nutrient in stem with bark+branches}}$
C3	Removal of biomass from the stems with bark	Non-conservation logging	$NC \left(C3\right)={\displaystyle \frac{Nutrient in soil+litter+crown}{Nutrient in stem with bark}}$
C4	Removal of stem biomass	Conservation timber use	$NC \left(C4\right)={\displaystyle \frac{Nutrient in soil+litter+canopy+bark}{Nutrient in stem}}$

NC—potential number of harvests.

).

2.4.4. Nutrient Use Efficiency (NUE) and Potential Number of Harvests (NC)

To assess the nutritional sustainability of the Khaya spp. under different harvesting scenarios, estimates of the nutrient use efficiency (NUE) index for each nutrient were determined following the method outlined by Turner and Lambert [41]. NUE is defined as the sum of the dry biomass compartments per unit of nutrient stored in the fractions of each harvesting scenario (Equation (4)).

$$\mathit{N}\mathit{U}\mathit{E}={\displaystyle \frac{\mathit{B}\mathit{i}\mathit{o}\mathit{m}\mathit{a}\mathit{s}\mathit{s}}{\mathit{N}\mathit{u}\mathit{t}\mathit{r}\mathit{i}\mathit{e}\mathit{n}\mathit{t} \mathit{u}\mathit{p}\mathit{t}\mathit{a}\mathit{k}\mathit{e}}}$$

(4)

where:

• $NUE$: nutrient use efficiency (kg);
• $Biomass$: sum of the dry biomass fractions used in each harvesting system (kg ha⁻¹);
• $Nutrient uptake$: sum of the nutrient stock exported in each harvesting scenario (kg ha⁻¹).

Subsequently, the number of rotations for each scenario was estimated. To do this, the nutrient stock remaining in the area, including the stocks of nutrients available in the soil and accumulated litter, was divided by nutrient stocks removed by harvesting [19].

2.5. Statistical Analysis

The experimental design used was randomized blocks, with three treatments and three replications. The treatments consisted of three species of Khaya spp. (K. senegalensis, K. ivorensis and K. grandifoliola). Data were tested for homogeneity of variances and normality of residuals using the O’Neill Mathews and Shapiro–Wilk tests, respectively, at a 5% probability level. Upon meeting these prerequisites, analysis of variance (ANOVA) was conducted. To assess the effect of the species on biomass production, means were compared using Tukey’s test. To assess the effect of species and harvest scenarios, as well as their interactions, the Scott–Knott test was applied. These analyses were performed using the R environment version 4.4.1, with the ExpDes package, considering a p-value < 0.05 for statistically significant results. Furthermore, to verify the correlations between treatments and response variables, a principal component analysis (PCA) was also conducted in the R environment using the prcomp function as part of the stats package [22,42].

3. Results

3.1. Dendrometric and Volume Characteristics

The statistics of the volumetric equations for the Khaya spp. stands showed good fits (

Table 5. Adjusted equations and their respective statistics for estimating the commercial volume of Khaya spp. trees at 9.5 years of age in Sooretama, Espírito Santo.

Species	Model	Equation	${\mathit{R}}_{\mathit{a}\mathit{d}\mathit{j}}^2$	${\mathit{S}}_{\mathit{y}\mathit{x}}\%$
KG	Combined variable Spurr	$v=0.0258+0.000056\left(d^2{h}_c\right)$	0.948	11.4
KI	Meyer	$v=14.9705-1.4106d-2.4866h_c+0.033d^2-0.0054\left(d^2{h}_c\right)+0.2338\left(d{h}_c\right)$	0.945	10.6
KS	Stoate	$v=0.1842-0.0303h_c-0.000263d^2+0.00010\left(d^2{h}_c\right)$	0.968	5.1

KG = Khaya grandifoliola C.DC.; KI = Khaya ivorensis A. Chev.; KS = Khaya senegalensis A. Juss.; $v$ = commercial volume (m³ tree⁻¹); $R_{adj}^2$ = adjusted coefficient of determination; $S_{yx}\%$ = standard error of the estimate in percentage; $dbh$ = diameter at 1.30 m above the ground (cm); $h_c$ = merchantable height (m).

). The adjusted coefficients of determination were similar across species, with values between 0.94 and 0.96, while the standard error was 5.1%, 10.6% and 11.4% for K. grandifoliola, K. ivorensis and K. senegalensis, respectively.

The dendrometric and volume variables showed statistical differences between Khaya spp. (

Table 6. Dendrometric variables in a Khaya spp. stand at 9.5 years of age in Sooretama, Espírito Santo.

Species	$\mathit{S}$	$\mathit{d}\mathit{b}\mathit{h}$	${\mathit{h}}_{\mathit{c}}$	${\mathit{V}}_{\mathit{i}}$	$\mathit{V}$	$\mathit{M}\mathit{A}\mathit{I}$
	(%)	(cm)	(m)	(m3)	(m3 ha−1)	(m³ ha−1 year−1)
KG	93.33 A	21.28 B (±4.9)	5.68 AB (±2.6)	0.1622 AB (±0.01)	60.54 A (±6.2)	6.37 A (±0.6)
KI	88.89 B	22.59 A (±2.8)	5.88 A (±1.4)	0.2234 A (±0.01)	79.71 A (±21.7)	8.39 A (±2.3)
KS	93.33 A	21.66 AB (±3.5)	4.86 B (±0.9)	0.1423 B (±0.01)	53.12 B (±4.3)	5.59 A (±0.5)
Mean	91.85	21.84	5.47	0.1760	64.46	6.78

KG = Khaya grandifoliola C.DC.; KI = Khaya ivorensis A. Chev.; KS = Khaya senegalensis A. Juss.; $S$ = survival; $dbh$ = diameter at 1.30 m above the ground; $h_c$ = merchantable height; $V_i$ = individual volume; $V$ = total volume per hectare; $MAI$ = mean annual increment. Averages followed by different capital letters in the column differ statistically according to Tukey’s test at the 5% probability level.

). It was observed that the survival rate of K. grandifoliola and K. senegalensis was approximately 5% higher than that of K. ivorensis. On the other hand, the variables of diameter at 1.30 m height ($dbh$), commercial height ($h_c$), individual volume ($V_i$) and total volume per hectare ($V$) were better for K. ivorensis, with averages of 22.59, 5.88, 0.22 and 79.71, respectively.

3.2. Aboveground Biomass

Significant variations in the percentages of each biomass component are observed in Figure 2. The compartments of bark and leaves show differences between species (p < 0.05). K. senegalensis exhibited the highest percentages of bark, with an average of 10.16%, followed by the other species, with approximately 7%. On the other hand, K. ivorensis stood out for its higher percentage of leaves, being superior to K. grandifoliola and K. senegalensis by 48% and 70%, respectively. The branches did not vary by species, reaching an average of 53.55%, the highest among the components. No differences were observed for the stems, which averaged 32.85%.

The relative proportions of biomass reveal the following order of allocation: branches > stems > bark > leaves. Considering the crown as the sum of the elements above the stem, it is noted that approximately 59% of the biomass is allocated to these compartments, leaving only 41% for stems and bark.

The biomass produced per unit area is found in

Table 7. Estimated biomass per hectare (Mg ha⁻¹) for the different compartments in Khaya spp. stands at 9.5 years of age in Sooretama, Espírito Santo.

Species	Stem	Bark	Leaves	Branches	Total
KG	28.91 (±3.21)	6.99 B (±0.47)	4.08 B (±0.48)	49.88 (±1.52)	90.29 (±4.24)
KI	29.98 (±5.27)	6.09 B (±1.04)	7.58 A (±0.28)	43.03 (±2.87)	85.97 (±4.22)
KS	25.96 (±2.04)	8.24 A (±0.84)	2.09 C (±0.36)	44.86 (±3.02)	83.49 (±1.43)
Mean	28.28	7.11	4.58	45.92	86.58

KG = Khaya grandifoliola C.DC.; KI = Khaya ivorensis A. Chev.; KS = Khaya senegalensis A. Juss. Averages followed by different capital letters in the column differ statistically according to Tukey’s test at the 5% probability level. Values in brackets represent the standard deviation.

. It is observed that the studied species had different patterns of compartmentalization (p < 0.05). K. senegalensis had the highest biomass values for the bark compartment, followed by K. grandifoliola and K. ivorensis, with values of 8.24, 6.99 and 6.09 Mg ha⁻¹, respectively. For the leaves compartment, K. ivorensis had 7.58 Mg ha⁻¹, which was 46 and 72% higher than the values of K. grandifoliola and K. senegalensis, respectively.

In general, the pattern of the distribution of compartments by area is similar to the relative distribution in percentage, which was as follows: branches > stem > bark > leaves. Furthermore, no significant differences were observed in the total biomass production per hectare between the species, with an average of 86.58 Mg ha⁻¹.

3.3. Nutrient Concentration

The P and S concentrations showed no significant effect for the stem compartment among the species, displaying average concentrations of 0.37 and 0.85 g kg⁻¹, respectively. In contrast, K. senegalensis has higher concentrations of N, K, Ca and Mg in the stem compartment compared to other species, with averages of 1.87, 1.09, 2.51 and 0.48 g kg⁻¹, respectively (

Table 8. Macronutrient concentration of Khaya spp. stands at 9.5 years of age in Sooretama, Espírito Santo.

Species	Nutrient Concentrations (g kg−1)
	N	P	K	Ca	Mg	S
Stems
KG	1.48 B (±0.04)	0.45 (±0.12)	0.78 AB (±0.04)	1.44 B (±0.14)	0.32 B (±0.02)	0.81 (±0.52)
KI	1.37 B (±0.02)	0.33 (±0.02)	0.49 B (±0.03)	1.64 AB (±0.30)	0.28 B (±0.01)	0.91 (±0.65)
KS	1.87 A (±0.06)	0.33 (±0.05)	1.09 A (±0.25)	2.51 A (±0.54)	0.48 A (±0.02)	0.83 (±0.58)
Bark
KG	6.26 (±0.41)	0.47 (±0.04)	6.18 A (±0.84)	20.07 AB (±2.31)	1.80 A (±0.18)	0.77 (±0.20)
KI	6.84 (±0.62)	0.42 (±0.01)	4.74 B (±0.51)	21.61 A (±1.62)	1.21 B (±0.13)	0.71 (±0.16)
KS	7.76 (±0.55)	0.45 (±0.01)	7.22 A (±0.43)	15.78 B (±0.77)	1.73 A (±0.17)	0.93 (±0.11)
Leaves
KG	15.49 (±0.75)	0.85 (±0.05)	6.72 (±0.87)	16.66 (±3.34)	2.63 (±0.44)	2.00 (±0.29)
KI	14.60 (±0.20)	0.75 (±0.03)	5.73 (±0.77)	18.73 (±0.42)	2.29 (±0.43)	2.13 (±0.20)
KS	14.49 (±1.41)	0.81 (±0.07)	6.20 (±0.92)	13.81 (±1.53)	1.89 (±0.35)	1.55 (±0.35)
Branches
KG	8.06 (±1.75)	0.68 (±0.13)	7.24 (±0.80)	15.21 (±2.05)	3.21 A (±0.63)	1.15 (±0.32)
KI	5.92 (±0.35)	0.74 (±0.07)	6.51 (±0.09)	15.13 (±1.45)	1.80 B (±0.37)	0.86 (±0.21)
KS	8.66 (±0.45)	0.76 (±0.19)	6.93 (±0.86)	12.48 (±2.26)	2.47 AB (±0.57)	0.84 (±0.08)

KG = Khaya grandifoliola C.DC.; KI = Khaya ivorensis A. Chev.; KS = Khaya senegalensis A. Juss. Averages followed by different capital letters in the column differ statistically according to Tukey’s test at the 5% probability level. Values in brackets represent the standard deviation.

).

There was no significant difference in the concentrations of N, P and S in the bark. K. senegalensis continues to show the highest values of K (7.22 g kg⁻¹) and Mg (1.73 g kg⁻¹); however, K. ivorensis has a superior Ca concentration, with an average of 21.61 g kg⁻¹.

For the leaves compartment, the macronutrient concentrations did not differ among species. This same trend is found for N, P, K, Ca and S in the branches compartment. K. grandifoliola exhibited the highest Mg concentration in the branches, representing a superiority of approximately 44% relative to K. ivorensis (Table 8).

3.4. Nutrient Stocks

Nutrient stocks per hectare in different compartments are observed in Figure 3. In the stem compartment, it is noted that the stocks of P, Ca and S were not influenced by the species. In contrast, K. senegalensis and K. grandifoliola showed the highest stocks of N and K (45.72 and 25.63 kg ha⁻¹, respectively), followed by K. ivorensis (41.07 and 14.54 kg ha⁻¹). The species K. senegalensis also exhibited superior Mg stocks for the stem, presenting approximately 26% and 33% more than K. grandifoliola and K. ivorensis, respectively.

The stocks of Ca in the bark compartment were the highest, with averages of 134 kg ha⁻¹, yet they did not differ among the species. Furthermore, K. senegalensis had the highest stocks of N, P, K, Mg and S in this compartment. However, K. ivorensis stands out with the highest averages in the leaves compartment for all the analyzed macronutrients, particularly Ca, followed by K. grandifoliola and K. senegalensis.

For the branches compartment, N and Mg showed differences between the Khaya spp. K. grandifoliola and K. senegalensis exhibited statistically similar averages of 394.88 for N and 135.65 kg ha⁻¹ for Mg, whereas K. ivorensis reached N stocks with averages of 254.09 kg ha⁻¹ and Mg stocks around 77.85 kg ha⁻¹.

The total accumulations of N, P, K and S per hectare were not influenced by the species. The stocks of Ca and Mg were higher for K. grandifoliola, being superior in the order of 22.1% for Ca in relation to K. senegalensis and 42.3% for Mg in relation to K. ivorensis.

3.5. Nutritional Sustainability

Nutrient uptake decreased when more conservative harvesting scenarios were used (Figure 4a–e). The highest amounts of N, P, K, Ca and Mg exported occurred in S1 among the species, with averages of 510.01, 50.40, 411.03, 920.68 and 148.30 kg ha⁻¹, respectively. However, it was noted that K. grandifoliola and K. senegalensis showed similarities in exportation with S1 = S2 > S3 > S4 for all of the nutrients analyzed, while K. ivorensis showed the following exportation sequence: S1 > S2 > S3 > S4.

The lowest exports occurred in S4, with averages of 44.17, 10.43, 21.93, 52.59 and 9.97 kg ha⁻¹ for N, P, K, Ca and Mg, respectively. When compared to S1, there was a decrease in export levels of around 91.34% for N, 79.31% for P, 94.66% for K, 94.29% for Ca and 93.28% for Mg.

When analyzing scenario S1, it can be seen that the species did not differ statistically in terms of N, P and K export. However, K. ivorensis and K. grandifoliola stood out for their higher Ca exports, with averages of 1005.69 and 972.77 kg ha⁻¹, respectively. Analyzing scenario 2 (S2), K. grandifoliola and K. senegalensis showed the same levels of exports for N, K and Mg, differing from K. ivorensis. The lowest losses of K and Mg in scenario S3 occurred for K. ivorensis (43.75 and 15.66 kg ha⁻¹), followed by K. grandifoliola (65.93 and 21.71 kg ha⁻¹) and K. senegalensis (88.36 and 26.53 kg ha⁻¹). In scenario S4, the species did not differ in terms of P and Ca exports, reaching averages of 10.43 and 52.59 kg ha⁻¹, respectively. On the other hand, K. senegalensis exported the highest stocks of N, K and Mg when compared to K. ivorensis, which corresponds in the same order to a superiority of 15.58%, 49.37% and 32.93%.

The scenario with the removal of stem biomass (S4) showed the highest NUE values for N, K, Ca and Mg, with averages of 646.13, 1435.12, 576.79 and 2951.09 kg ha⁻¹, respectively. However, this same scenario was similar to S3 for P, with approximately 2751.06 kg ha⁻¹ (Figure 5a–e). The highest NUE for N was observed for K. ivorensis in S4, while K. senegalensis showed the lowest averages in scenarios S2 and S3. The efficiency of P use showed no differences between the Khaya spp. (Figure 5b). As for K, the efficiency varied between the scenarios, being the same between the species for S1 and S2, while for scenario S3 and S4, K. ivorensis stood out with the highest averages, 833.37 kg ha⁻¹ and 2064.14 kg ha⁻¹, respectively (Figure 5c). The species showed the same Ca use efficiencies in scenarios S1, S2 and S3, with averages of 95.62 kg ha⁻¹, 97.83 kg ha⁻¹ and 191.54 kg ha⁻¹. In S4, it was observed that K. grandifoliola and K. ivorensis were superior to K. senegalensis by approximately 38% (Figure 5d). Similarly, the NUE of Mg did not differ between the species in scenarios S1 and S2, but the highest efficiency for K. ivorensis in S3 (2306.29 kg ha⁻¹) and the lowest for K. senegalensis in S4 (2098.71 kg ha⁻¹) stand out (Figure 5e).

The removal of the stems (S4) promotes a greater number of rotations for all the nutrients, with averages of 22, 6, 27, 59 and 51 for N, P, K, Ca and Mg, respectively, whereas the removal of the aboveground biomass (S1) contributes to the fewest rotations (1, 0.3, 0.4, 2.3 and 2.5 in the same order of nutrients) (Figure 6a–e). Regarding Khaya spp., it was also observed that, in general, there were no significant differences in the number of rotations within each scenario (Figure 6b–e). However, when analyzing less conservational scenarios (S1 and S2), it was found that K. ivorensis was superior by 33% and 25% compared to K. grandifoliola and K. senegalensis for N rotations (Figure 6a).

The principal component analysis (PCA) explained the nutrient uptake data from each scenario based on aboveground biomass at approximately 66.09%, with 38.07% for the first component and 28.02% for the second component. It was observed that K. ivorensis is more associated with leaf biomass and is not associated with the export scenarios. On the other hand, K. grandifoliola and K. senegalensis showed opposite trends to K. ivorensis, with positive relationships of branches with Mg, P and Ca, in addition to bark being associated with N, K and Ca, respectively (Figure 7).

Furthermore, it was noted that the K. senegalensis species has an association with bark biomass and the nutrient Ca for scenarios 3 and 4. In general, based on the forest inventory, an excellent state of stem health was observed for the K. senegalensis population, whereas the stems in the populations of K. ivorensis and K. grandifoliola showed occurrences of phytosanitary problems such as cortex cankers. The development of cankers creates irregular surfaces on the bark and excessive development with protrusions. Possibly, the absence of cankers in the K. senegalensis species allows for better correlations between the bark biomass and the species. Another noteworthy point is that studies have shown an increase in Ca concentrations in bark when there are cankers. These factors are inversely correlated to bark biomass and Ca exports in scenarios 3 and 4 for K. senegalensis (Figure 7).

4. Discussion

4.1. Dendrometric Characteristics and Volume

The combined models of Spurr, Meyer and Stoate variables presented satisfactory results and were able to represent K. grandifoliola, K. ivorensis and K. senegalensis well. Although there are few data available for Khaya spp., the present study presents lowers statistics than those of Silva et al. [43], who obtained the best adjustments using the Schumacher and Hall model for K. ivorensis with 4 × 3 m spacing and aged between 30 and 59 months, achieving an $R_{adj}^2$ of nearly 96% and a standard error of 0.023 m³. These differences may be related to age and also to planting spacing.

Similar outcomes were also observed by Oliveira et al. [44], who assessed the performance of models to estimate the marketable and total volume of K. ivorensis. The authors identified the Schumacher and Hall model as the best volume adjustment equation for stands in Minas Gerais. Conversely, Santos et al. [45] mention that even though the Schumacher and Hall model presents good statistics (with an R² of 0.97 and an RMSE of 12.2%), the Husch equation was the most accurate in determining the volume for K. ivorensis stands, justified by its simplicity of use.

The results showed that the species directly influences the dendrometric and volume variables. Survival rates of 91% for K. ivorensis at 18 months were found by Siqueira et al. [46], while Opuni-Frimpong et al. [47] observed mortality rates of 0.65% and 7.26% for K. antotheca and K. ivorensis, respectively, at 48 months of age. On the other hand, lower survival rates than those reported in the present study were obtained by Silva et al. [48] in a silvopastoral system at 72 months, with an average of 67.5% for K. grandifoliola. These values can be explained by the edaphoclimatic conditions and management practices inherent to each stand.

The superiority of K. ivorensis in terms of other dendrometric variables is corroborated by Heryati et al. [49], who found dbh values ranging from 11.6 cm to 14.4 cm and average heights between 7.8 m and 10.6 m in stands at 60 months of age across different soil classes. This was also observed by Aminah et al. [50], who found dbh values of 18.8 cm and an average total height of 15.8 at 84 months in Malaysia, while Vidaurre et al. [51] reported averages of 25.86 cm in dbh and 14.94 m in height for the species at 19 years of age.

According to Pinheiro et al. [15], K. ivorensis is the species with the highest requirements in terms of solar radiation, water and nutrients. Therefore, as the study area presents good nutritional conditions and there are no impediments to root development, the species adapted well, favoring growth in diameter, height and, consequently, volume. Moreover, the survival rate of K. ivorensis was the lowest, at 88%. As a result, there is a lower plant density, which may have favored development due to reduced competition.

4.2. Aboveground Biomass

The partitioning of biomass is influenced by factors including age, soil and climate conditions, spacing and intrinsic species characteristics [20,40,52,53]. The branch component exhibited the largest contributions to the biomass partitioning of Khaya spp. The results demonstrate the potential of this fraction for both biomass production and its commercialization. According to Souza et al. [54], the residues from the K. ivorensis and K. senegalensis species constitute an opportunity for energy recovery, justified by the high calorific value and mechanical resistance levels, which are essential for the production of charcoal, pellets, and briquettes [55].

Another aspect to consider is the management practices for guiding the stem growth of these species, since this component is of the greatest commercial interest. It was observed that this compartment occupied the second position in biomass allocation, below that found for species of the Eucalyptus genus, which on average account for 80% of the aboveground biomass [37,56,57].

Leaves play an essential role in the photosynthetic processes of plants [58]. Due to its nutritional requirements, the leaf biomass of K. ivorensis is greater than that of other species. This likely serves as a survival strategy, since it requires a greater amount of photoassimilates for growth and development in the same environment [15,59,60].

Albuquerque et al. [61] report that K. ivorensis seedlings, when subjected to water stress, strategically use photoassimilates by allocating some the triose phosphates to maintain starch concentrations and others for cellular respiration, forming a survival technique. In contrast, a different condition is observed for K. senegalensis. According to Matos et al. [62], this species exhibits high stomatal control, a low leaf chlorophyll concentration and reduced transpiration. Such characteristics enhance its tolerance to water stress and lead to lower leaf biomass production and higher bark biomass.

The total biomass production reported by Guimarães et al. [63] was 104.5 Mg ha⁻¹, distributed among wood (63%), roots (14%), branches (11%), bark (8%) and leaves (4%) for E. dunnii stands with an initial density of 1428 plants per hectare at 4 years. On the other hand, studying a seasonal deciduous forest in Itaara, RS, Vogel et al. [64] observed an average aboveground biomass production estimated at 210 Mg ha⁻¹, with branches being the component having the largest contribution (48.8%), followed by wood (43.4%), bark (5.4%) and leaves (2.4%). In the case of Levan et al. [65], 11-year-old Acacia mangium stands show an average aboveground biomass of 175.17 Mg ha⁻¹, divided into wood (112.8 Mg ha⁻¹), branches (54.34 Mg ha⁻¹) and leaves (8.03 Mg ha⁻¹).

The differences between these studies and the present study can be attributed to the inherent characteristics of the Khaya spp., which have dense and rounded canopies, composed of thick and cylindrical branches, as well as their age and edaphoclimatic characteristics [66,67,68]. Furthermore, the results demonstrate that biomass production, especially that of bark and leaves, has a significant effect among the species. It is worth noting that these components represent, on average, 12 Mg ha⁻¹ of the biomass produced per area, indicating a fundamental role in biogeochemical cycling and potential nutrient uptake scenarios.

4.3. Nutrient Concentration

Potassium levels are variable in aboveground biomass, as it is a mobile nutrient readily redistributed to growing plant parts. According to Taiz et al. [69], this element plays an important role in regulating osmotic potential and is involved in enzyme activities related to respiration and photosynthesis. It is noted that the highest levels of K are found in the canopy (leaves and branches), although there are no statistical differences. However, for the trunk and bark compartments of the species K. senegalensis and K. grandifoliola, higher values and better nutrient use efficiencies were expected, since the leaf biomass is the smallest among the studied species.

In the bark compartment, the highest concentrations of Ca are observed. This is due to its low mobility in plant tissues, in addition to being a structural element of the cell membrane [70]. Similar results were found for Eucalyptus spp. stands by Guimarães et al. [63], González-Garcia et al. [71] and Resquin et al. [72]. Regarding Khaya spp., it is noted that K. ivorensis stands out for its Ca values in the bark due to its thickness and the occurrence of cortex cankers, which are a fungal disease caused by L. theobromae and manifest as protruding lesions on bark [61,73,74]. On the other hand, since it is not affected by cankers in the study area and despite having the largest bark biomass per area among the species, K. senegalensis shows the lowest Ca levels in this compartment, due to the redirection of this element to the trunk, which performs a structural function [69].

K. grandifoliola and K. senegalensis contribute the highest levels of Mg in the bark and branches. This pattern is different from that observed for the genus Eucalyptus spp., in which Mg is allocated in the bark and leaves [21,75]. This is probably due to a physiological response of the genus Khaya spp., due to leaf loss in certain seasons, especially cold months or in the absence of rain [76]. It is also noted that Mg levels in the leaves do not differ between species; however, as they have the smallest leaf biomass per hectare, K. grandifoliola and K. senegalensis use the Mg reserve in other compartments to reduce the loss of this element, surpassing the levels of the species K. ivorensis.

The lower concentrations of P and S can be attributed to the scarce availability of these nutrients in the soil. Studying clones of E. urophylla × E. globulus at 10 years of age, Viera et al. [77] found low P values in the soil, which was reflected in reduced nutrient levels in the biomass and litter compartments. Ludvichak et al. [78] also observed low values of P and S, not exceeding 1.32 g kg⁻¹ and 1.43 g kg⁻¹, respectively, for different biomass components in stands of E. urograndis and A. mearnsii at 9 years of age.

4.4. Nutrient Stocks

Generally, nutrient stocks in different compartments are influenced by both the biomass distribution and nutrient concentration [79]. In the stem, it is observed that K. senegalensis presents the highest stocks of K and Mg due to its concentrations. It is noted that these elements can be electrostatically bound or act as ligands to larger carbon-containing compounds, as is the case with the stem [69]. It is also worth mentioning that Khaya spp. have the capacity to store calcium since this nutrient is not redistributed to new plant parts, accumulating in older tissues.

The same trend was also observed for the bark compartment, for which K. senegalensis shows the highest biomasses and concentrations, thus reflecting superiority in the stocks of most of the evaluated macronutrients. Regarding Ca, as K. senegalensis presented the highest biomasses and, conversely, the lowest concentrations, this contributed to the non-differentiation between the species. However, it is emphasized that the bark compartment stores a large quantity of nutrients, especially K, Ca and Mg, and debarking is recommended during harvest to ensure nutritional sustainability [21,57].

Regarding the leaf compartment, for the species K. grandifoliola and K. senegalensis, lower stocks for all evaluated nutrients are noted. This is due to the physiological strategy of these species regarding leaf loss combined with greater leaf biomass production by K. ivorensis. This was corroborated by Alves et al. [80], who observed superiority in nutrient concentrations in leaves, but with low stocks in this compartment in Caatinga species. Schumacher et al. [20] report that nutrient stock does not follow biomass distribution; however, they highlight that the leaves are considered the metabolic center of the plant and in their study occupied the second position in nutrient storage, after bark.

Branches store a large quantity of nutrients, especially K and Ca, which do not differ between species and reach average values of 316.92 and 654.43 kg ha⁻¹, respectively. Obtained stocks around 200 kg ha⁻¹ for the stem and 30 kg ha⁻¹ for branches in both nutrients in Eucalyptus species at 6.7 years. This relation is inverted and below that found in the present study, justified by the production of branches in Khaya spp. Moreover, differences in Mg storage for the species are noted. Being a mobile element and highly necessary in the canopy for the structural function in the chlorophyll molecule [69], Mg is readily available in the branches, especially for K. grandifoliola and K. senegalensis, whereas for K. ivorensis, it is allocated in the leaves, justified by its greater biomass. These storage differences are explained by Viera et al. [81] as a response to the nutritional demands of the species and the nutrient availability from the soil.

Regardless of the compartments, it is observed that the total nutrient stocks by the Khaya spp. are high, with the stocks of P, K, Ca, Mg and S reaching averages of 50.40 kg ha⁻¹, 411.03 kg ha⁻¹, 920.68 kg ha⁻¹, 148.30 kg ha⁻¹ and 84.20 kg ha⁻¹, respectively. Such values are superior to those found by Dick and Schumacher [82] for Acacia mearnsii and Kulmann et al. [22] and Rocha et al. [83] for stands of Eucalyptus and Pinus.

4.5. Nutritional Sustainability

Harvesting all aboveground components becomes one of the most aggressive practices in terms of nutritional sustainability, as it extracts a considerable amount of nutrients from the forest site, without the possibility of reabsorption by the species [22]. The results show that in scenarios S1 and S2, K and Ca are the macronutrients most susceptible to exportation. This trend was also found for Eucalyptus stands, demonstrating that the use of mineral or organic fertilizers is necessary to maintain productivity in future rotations [21,63].

The scenario that presented the lowest nutrient uptake was S4, in which only the stem is removed. This fact is explained by the maintenance of residues in the field [57,84] and the capacity of other Khaya spp. compartments to store nutrients [40,85,86]. However, debarking the stem is a costly process, making it economically unfeasible for most rural producers. Thus, scenario S3 becomes an option as a harvest scenario for Khaya spp., despite nutrient uptakes.

Nutrient use efficiency is an important parameter for evaluating the intensity of nutrient uptake within a forest site and, consequently, in choosing silvicultural practices capable of replenishing such nutrients [41,87]. Conservational systems, such as S3 and S4, promote a high NUE due to the maintenance of different biomass compartments in the field associated with reduced nutritional stocks in the trunk and bark. On the other hand, scenario S1 proposes the removal of components with high nutrient concentrations (leaves and branches), reducing efficiency and increasing exports.

Superior results were found for harvesting systems that extract only the debarked stems or stems with below-ground biomass in eucalyptus clones [22,88]. Unlike species of the genus Eucalyptus, Khaya spp. presents higher nutrient stocks distributed along the branches and leaves compartments but have lower nutritional concentrations, which is reflected in the lower nutritional efficiencies compared to those of Eucalyptus (Figure 3) [21,89]. Moreover, this result demonstrates lower exports justified by the low nutrient stocks in the trunk and bark of Khaya spp., especially N, P, K, Ca and Mg.

The number of rotations was influenced by different harvesting scenarios. It is observed that in S1 and S2, systems that remove most of the aboveground biomass components have a low number of rotations due to the quantity of N, P, K, Ca and Mg exports. On the other hand, scenarios that maintain leaves and branches, S3 and S4, contribute to higher rotations justified by the permanence and availability of nutrients in the biomass compartments, in addition to improvements in microbiological functions in the soil [69,90]. This demonstrates that conservative harvesting scenarios contribute to the greater sustainability of forest sites and, consequently, productivity in advanced ages, as observed for eucalyptus [90,91], pine [92,93] and boreal forests [94].

The results of the present study confirm the need for differentiated management practices for Khaya spp., such as maintaining the canopy in the harvest area to favor cycling and the reestablishment of adequate nutrient levels throughout production cycles. Thus, cutting the canopy, chipping the plant material and maintaining it in the cultivation areas is recommended. This technique can directly influence the final cutting cycle, since the residues will be decomposed and reused [20,95].

The final harvest of Khaya species is around 15 to 20 years [66,96]. However, many producers carry out thinning to remove unwanted trees or thick branches to favor stem growth (S2). Nutritionally, this practice does not constitute a good option, given that the number of rotations for all evaluated nutrients is low, characterizing high nutrient uptakes and requiring the maintenance of residues in the forest site. Moreover, the superiority of K. ivorensis in nitrogen rotations in scenarios S1 and S2 is due to its capacity to store this nutrient in accumulated litter, avoiding exports through harvest.

Principal components facilitated relationships between nutrient uptake and Khaya spp. with a clear trend. The results demonstrated that the leaf biomass production associated with high nutrient stocks in the litter of K. ivorensis promoted reduced export levels by this species. Thus, silvicultural practices that retain leaves in the system can favor better nutritional sustainability, with elevated nutrient levels due to photosynthetic activity [81].

In contrast, two other groups were formed based on export scenarios. The first of these was represented by K. grandifoliola, with positive relationships of exports for P, K, Ca and Mg in the branch compartment for scenarios S1 and S2. This can be explained by the fact that Ca and Mg stocks were higher for K. grandifoliola; thus, export scenarios that promote the total removal of biomass or the stem with bark + branches favor high exports of these nutrients [21]. The second group is represented by the export of P, K, Ca and Mg in the bark compartment for K. senegalensis in scenarios S3 and S4. As previously mentioned, the highest stocks of N, P, K, Ca, Mg and S were found in the species K. senegalensis, and if removed from the system via harvesting, this is also reflected as a loss of these nutrients. Therefore, as long as it meets economic viability, debarking for this species becomes fundamental [97].

Therefore, it is crucial to understand the relationship of nutrient export and dynamics in the adopted production system and its implications for the nutritional sustainability of the forest ecosystem. Each harvesting system directly influences both the efficiency of nutrient use and the number of nutrient uptakes, especially K and Ca, which are especially susceptible to removal. Conservationist systems, such as S3 and S4, which maintain part of the biomass in the field, promote greater efficiency in nutrient use, resulting in a greater number of rotations and contributing to the long-term sustainability of the system. However, it is important to highlight some limitations of this study. The first is that the intrinsic characteristics of each species, combined with similar soil and climate conditions and management practices between the study areas, may have influenced the results, making it difficult to generalize them to other environments. Another potential limitation is the lack of information on the effects of longer production cycles and the replenishment of nutrients in the soil after multiple rotations. This suggests the need for long-term research to assess the sustainability of different harvesting scenarios over several production cycles.

5. Conclusions

The analysis of dendrometric characteristics, aboveground biomass, nutrient concentrations and stocks, as well as the nutritional sustainability of Khaya spp., reveals a complex interaction between environmental variables and the biological characteristics of these species. The superiority of K. ivorensis in metrics such as diameter, commercial height and individual volume suggests that this species may be more productive under cultivation conditions, which is relevant for forest management practices.

Regarding biomass, the significant differences observed between species indicate that biomass allocation is not homogeneous and that each species has specific growth strategies. The leaves, branches and bark compartments exhibit the highest nutrient concentration and stocks. However, K. senegalensis stands out in terms of bark percentage, while K. ivorensis presents a more significant allocation in the leaves. Utilizing these fractions as a nutritional supplement becomes a viable alternative for site sustainability, as it can contribute to biomass production over rotations. Therefore, these variations have direct implications on biomass production and nutrient dynamics, which may influence their competitiveness in forest environments.

The nutrient uptake assessment demonstrated that more conservative harvest scenarios resulted in lower nutrient losses, indicating the importance of forest management practices that prioritize soil and nutrient conservation. The data suggest that K. ivorensis and K. grandifoliola are more efficient in calcium export, while K. senegalensis excels when it comes to other nutrients. The nutrient use efficiency (NUE) varies among species and scenarios, showing that species choice can impact productivity and long-term sustainability.

In summary, the findings of this study provide a solid basis for the sustainable management of Khaya spp., highlighting the importance of diversity in species traits and their implications for productivity and nutrient dynamics on a small or medium scale. Understanding these interactions is essential for the development of management strategies that optimize both natural resource conservation and forest production, contributing to more sustainable and efficient forestry practices.