Study site

The study was conducted at the forest nursery of the Faculty of Forestry and Environmental Sciences, Universidad Juárez del Estado de Durango, located in Victoria de Durango, Durango, Mexico (24° 01’ 13” N, 104° 68’ 25” W), at an elevation of 1 890 m. The site has an average annual temperature of 17 °C (with the coldest month averaging 1.7 °C and the warmest month 31 °C) and an average annual precipitation of 500 mm (^{Reyes Rodríguez et al., 2024}).

Growth conditions

The plants grew for 11 months under three environmental conditions: a) a forest greenhouse covered with 720 µm milky white plastic and 50 % black shade cloth beneath the plastic film (September 2020-February 2021); b) 50 % black shade cloth (March-May 2021); and c) open field (June-July 2021). Ambient temperature and humidity were monitored using an Elitech® hygrothermometer (Table 1).

Substrate porosity

Total porosity, air-filled porosity, and water-holding capacity of the substrates were determined using the methodology proposed by ^{Landis et al. (1990)} (Table 2).

Treatments and experimental design

A total of nine treatments were evaluated, corresponding to combinations of three substrates prepared with different ratios of fresh sawdust (FS), composted pine bark (CB), and peat moss (PM), along with three application rates of Multicote® controlled-release fertilizer, with an approximate release period of eight months. The CRF formulation is Nitrogen (18 %) + Phosphorus (6 %) + Potassium (12 %) + Boron (0.03 %) + Copper (0.045 %) + Iron (0.4 %) + Manganese (0.055 %) + Molybdenum (0.01 %) + Zinc (0.06 %).

The treatments were arranged in a completely randomized experimental design with a 3 × 3 factorial structure (three substrates and three fertilizer rates), with four replicates per treatment. Each replicate consisted of one expanded polystyrene forestry tray (15 cm in length and 4 cm in upper diameter) containing 77 cavities, each with a volume of 170 mL.

Seedling production

Seeding was carried out on August 21, 2020. Prior to sowing, the cavities of the trays were treated with a solution composed of 10 L of water, 4 L of vinyl sealer, and 0.6 kg of Cu (OH)₂ (copper hydroxide), to promote chemical pruning of the lateral roots of the seedlings. During the production process, irrigation was applied three times per week. The irrigation intensity depended on the prevailing climatic conditions, but efforts were made to ensure that moisture was consistently present in at least the lower two-thirds of the substrate volume within the container to avoid limiting plant growth. Additionally, weeding was performed monthly. During the first three months of seedling growth, fungus gnats (Bradysia impatiens Johannsen) were present in the substrate mixtures; to control them, the insecticide Denim® was applied every two weeks at a rate of 2 mL∙L^-1 of water.

Evaluated variables

At 11 months of growth, 10 seedlings from the center of each experimental unit were evaluated. The substrate was removed from the root ball of each plant, and the following variables were measured: a) height, measured with a ruler graduated in millimeters, b) Root collar diameter, measured with a Truper^® Caldi-6MP digital caliper (measurement range: 0.01 mm to 150 mm), c) Dry biomass; individual whole-plant samples were placed in kraft paper bags and dried in a forced-air oven (ECOSHEL®, model 9052) at 75 °C for 72 hours, then weighed on an analytical balance (Velab^® VE-204) with a precision of 0.01 g. The Dickson Quality Index (DQI) was calculated using height, root collar diameter, and root, shoot, and total dry biomass.

The concentration and content of nutrients (N-P-K) were estimated from the total needle biomass of the sampled plants. For each treatment, 20 g of vegetative material were selected, obtained from 12 randomly chosen plants. The material was ground using a Retsch SM 300® mill and analyzed at the Laboratory of Centro Nacional de Investigación Disciplinaria en Relación Agua, Suelo, Plantas, Atmósfera (CENID-RASPA) of the Instituto Nacional de Investigaciones Forestales, Agrícolas y Pecuarias (INIFAP). Nitrogen (N) was quantified using the Kjeldahl method; phosphorus (P) by wet digestion with nitric-perchloric acid and colorimetry using ammonium metavanadate; and potassium (K) by atomic absorption spectrophotometry. Additionally, vector nomograms were developed to evaluate the effect of fertilization on aboveground dry biomass. This analysis considered the mean values per treatment for N content (X-axis), N concentration (Y-axis), and aboveground biomass (Z-axis), expressed in relative terms, with the 6 g∙L^-1 fertilization dose used as the reference and normalized to 100.

Regarding costs, in June 2025, three commercial price quotes were obtained for the inputs used as substrates (raw sawdust, composted bark, and peat moss) and for the Multicote® 8 fertilizer from suppliers located in Victoria de Durango, Durango, Mexico. Based on these prices, the unit production costs were estimated for each substrate mixture and fertilization rate evaluated.

Statistical analysis

A database was compiled in Excel using the recorded growth data. To analyze the results, one-way analysis of variance (ANOVA) was conducted with the PROC GLM procedure in SAS 9.4® (^{SAS System, Inc. 2009}). When significant differences were detected, Tukey’s multiple comparison test (p ≤ 0.05) was applied to determine the most effective treatments. The assumptions of data normality and homogeneity of variances were verified using the Shapiro-Wilk and Levene tests, respectively.

Morphological characteristics

The interaction between substrate and fertilization factors showed that treatment T7(50 % FS + 25 % CB + 25 % PM + 12 g∙L^-1 of Multicote^® controlled-release fertilizer) was the most favorable. This was followed by treatment T4, which used the same substrate mixture but a lower fertilizer dose (9 g∙L^-1). Other treatments with positive results included T2 and T8, both consisting of 35 % FS + 25 % CB + 40 % PM, with fertilizer doses of 6 and 12 g∙L^-1, respectively. The lowest values were recorded in treatment T3, which used a substrate of 50 % CB + 50 % PM combined with 6 g∙L^-1 of Multicote® (Tables 4 and 5).

Based on morphological quality standards for forest species used in reforestation programs, the Mexican Standard NMX-AA-170-SCFI-2014 (^{Secretaría de Economía, 2014}), in its normative Appendix C, states that P. durangensis seedlings aged 10 to 12 months should have a height between 10 and 25 cm and a minimum stem diameter of 4 mm. In this study, all treatments met the minimum criteria and exhibited the necessary morphological characteristics for use in reforestation; however, some treatments stood above others, allowing the identification of the most favorable substrate and fertilization combinations.

The results demonstrate that the use of FS and CB, along with PM, created favorable conditions in the root system for the growth of P. durangensis. Additionally, ambient temperature and humidity (15.9 to 23.4 °C and 36.7 to 57.5 %, respectively) were conducive to plant growth (Table 1). Porosity also played an important role in these results; only the substrate composed of 50 % FS + 25 % CB + 25 % PM met the recommended total porosity value, and the substrate with 50 % CB + 50 % PM was within the appropriate range for air porosity. However, the values for all three substrates did not exceed the recommended ranges for these variables by more than 5.5 % (Table 2). This indicates that the available irrigation water provided suitable conditions for assimilation by the plants, as observed by ^{Castro-Garibay et al. (2019)}, where Pinus greggii var. australis showed the highest amount of available water (30 %) with a substrate of 60 % sawdust + 20 % bark + 20 % PM, compared to a substrate of 60 % PM + 20 % vermiculite + 20 % perlite, which had a value of 19 %.

For P. durangensis, ^{Hernández-Velasco et al. (2022)} reported that plants reached 22.6 cm in height and 5.2 mm in basal diameter using a substrate composed of 50 % base mix (28 % PM + 10 % agrolite + 12 % vermiculite) and 50 % CB, with a particle size of 3 mm, along with nutrition based on controlled-release fertilizer and soluble fertilizer irrigation, after 15 months of greenhouse growth. In that study, the plants exceeded the attributes stipulated in the Mexican Standard NMX-AA-170-SCFI-2014, likely because the plants were older compared to those in the present study (11 months). In P. cooperi Blanco, ^{González et al. (2018)} analyzed four substrates based on FS, CB, and PM with controlled-release fertilizer and found that the substrate containing 46 % PM + 54 % CB + 8 g∙L^-1 of Multicote® contributed to better morphological performance. However, plants grown in the substrate with 30 % PM + 20 % CB + 50 % FS + 8 g∙L^-1 of Multicote® also exhibited favorable growth, and the substrate cost was 39.8 % lower.

Other positive experiences related to the use of sawdust as part of the substrate are reported by ^{Aguilera-Rodríguez et al. (2016)}, where a mixture of 70 % composted pine sawdust, 15 % pine composted bark, and 15 % vermiculite allowed the production of P. montezumae Lamb. plants with appropriate morphological characteristics. Likewise, ^{Hernández-Zarate et al. (2014)}, also working with P. montezumae, found that substrates composed of 70 % FS + 10 % PM + 10 % perlite + 10 % vermiculite and 10 % CB + 70 % FS + 10 % perlite + 10 % vermiculite resulted in similar and favorable morphological growth values compared to the control substrate (60 % PM + 20 perlite + 20 % vermiculite).

Regarding controlled-release fertilizers, ^{Heras-Marcial et al. (2023)} studied the effects of different doses on the growth of P. patula under nursery conditions and found that 8 g∙L^-1 of 8-month Multicote Agri® combined with 4 and 6 g∙L^-1 of 12-month Multicote® provided adequate morphological growth. N-P-K constitutes a group of essential nutrients that promote plant growth. N is fundamental in physiological processes such as photosynthesis and cell division; P is present in many compounds within plant cells, including sugars, phosphates, lipids, nucleic acids, and free nucleotides (^{Fathi, 2022}; ^{Taiz & Zeiger, 2006}); and K contributes to the osmotic potential of cells and root tissues, facilitating cell elongation and plant turgor (^{Larriva Coronel, 2003}). Undoubtedly, controlled-release fertilizers have been formulated based on plant needs, but they must be applied in the correct formulation and dosage. In the present study, Multicote® (18-6-12 N-P-K) at a dose of 9 g∙L^-1 was the most appropriate based on its effectiveness and cost. Conversely, ^{Martínez-Nevárez et al. (2023)} evaluated doses of 4, 6, and 8 g·L^-1 of the same fertilizer and found that the intermediate dose was the most favorable for the quality of P. cooperi plants grown under nursery conditions.

Nutrient assimilation

The nutrient concentration in the needles of P. durangensis plants showed statistically significant differences (p ≤ 0.05) among the three fertilization rates evaluated (Table 6). Regarding nitrogen (N), the primary element involved in height growth, the CRF at a rate of 12 g∙L^-1 of substrate produced the highest nutrient concentration values. However, when only 6 and 9 g∙L^-1 were applied, the concentrations still reached the recommended levels established by ^{Landis et al. (2010)} (1.3-3.5 %). In contrast, no significant differences (p > 0.05) were observed in P and K concentrations among fertilization rates. Furthermore, the values obtained were below the minimum recommended for conifers (0.2-0.6 % P and 0.7-2.5 % K; ^{Landis et al., 2010}), which may be attributed to the fertilizer used containing lower amounts of P and K than required.

Based on morphological growth and nutrient assimilation in the plants, the 9 and 12 g·L^-1 rates of the eight-month Multicote® CRF are considered adequate to meet the necessary requirements for use. However, since they do not produce the minimum recommended concentrations of P and K, it is advisable to use a CRF with a higher percentage of these elements-for example, a 17-17-17 or 20-20-20 formulation.

The nutrient indicators obtained in this study showed higher nitrogen and lower phosphorus and potassium concentrations compared to those reported by ^{González et al. (2018)} for P. cooperi. These authors reported N-P-K concentrations of 1.0 %, 0.2 %, and 1.0 %, respectively, when using substrates composed of 50 % forest soil (FS) mixed with PM and CB in different ratios, and applying 8 g∙L^-1 of Multicote^®. Although these values were below the recommended levels, the plant quality indicators were satisfactory. Aguilera et al. (2021) produced P. patula seedlings using 11 substrates containing 10 % vermiculite + 10 % perlite + 20 % CB + 60 % sawdust from various conifer and broadleaf species, combined with 4 g∙L^-1 of Multicote® 18-6-12 (N-P-K) and 4 g∙L^-1 of Multicote® 12-24-12 (N-P-K). The average N-P-K concentrations obtained were 1.39 %, 0.29 %, and 0.49 %, respectively. Aguilera et al. (2021) also reported similar values-1.44 %, 0.22 %, and 0.51 % of N-P-K for the same species, using plants 20 to 30 cm tall grown from natural regeneration. In both growth conditions, the evaluated variables showed satisfactory results.

The graphical vector analysis showed a predominant direction in relation to the aboveground biomass. CRF at doses of 6 and 9 g·L^-1 of substrate promoted greater luxury consumption for all three nutrients. Luxury consumption is indicated by the shift of each vector toward the upper right corner in each graph; in other words, there was a greater increase in both the concentration and content of each nutrient (Figure 1). According to ^{Ávila-Angulo et al. (2020)}, the goal of fertilization is to increase nutrient content in plant biomass, especially when plants are established in low-fertility soils, because this nutrient content acts as a reserve during the later field stage to support plant rooting while it adapts to the new site conditions.

Figure 1 Vector diagram for nutrient analysis in the aboveground dry biomass of Pinus durangensis (Multicote® CFR) treated with controlled-release fertilizer at doses of 6, 9, and 12 g∙L^-1 of substrate under nursery conditions. 

^{Zamunér et al. (2012)} found adequate mineral nutrient assimilation with 6 g∙L^-1 of CRF (15-9-12 N-P-K) and nutrient excess with 9 g∙L^-1 of Multicote®. In another study with P. montezumae, Aguilera et al. (2016) reported that high doses of Multicote® CRF (8 g∙L⁻¹) maximize nitrogen concentration.

Production costs due to substrate and fertilizer

According to commercial price quotes obtained in June 2025 in Victoria de Durango, Durango, Mexico (Table 7), the lowest costs for materials used as substrates and fertilizer corresponded to the plants in treatments T1 and T4 (0.49 and 0.57 MXN per plant, respectively). Treatments T7, T2, and T5 fall within an intermediate cost range of 0.62 to 0.72 MXN per plant, while the remaining treatments exceeded 0.77 MXN per plant, reaching a maximum value of 0.95 MXN per plant (T12); that is, nearly double the cost of the lowest-cost treatment (T1). This evaluation only includes the costs of substrates and fertilizer doses, which were the focus of the study; however, a comprehensive analysis should consider the entire production system, including infrastructure, production inputs, and labor.

The lower costs of treatments T1 and T4 were mainly influenced by the higher use of sawdust compared to peat moss, since the former is a low-cost material (0.15 MXN·L^-1) compared to peat moss (5.5 MXN·L^-1). Although the fertilizer dose used is another influencing factor, it was less determinant in the final cost per plant. Overall, the treatments that most promoted plant growth (T7, T4, and T2) fall within an intermediate cost range and are cheaper compared to treatment T6, which could be considered the control.

The results indicate that using sawdust in a higher ratio than peat moss produces a lower-cost substrate and plants with good quality (Table 5), which is consistent with ^{González et al. (2018)}, who studied the costs of substrate mixtures for P. cooperi including FS, CB, and PM. Those authors found a 39.8 % difference in substrate cost (50 % FS + 20 % CB + 30 % PM vs. 46 % PM + 54 % CB), due to the inclusion of sawdust-a low-cost material-and the reduced proportion of peat moss, which is more expensive. A similar outcome was reported by ^{Martínez-Casas et al. (2024)}, who found that costs for P. durangensis were reduced by 28.4 % when using Multicote® fertilizer combined with equal parts PM and bark, and by 34.6 % with the combination of PM (50 %) + bark (25 %) + sawdust (25 %) compared to the use of PM (50 %) + vermiculite (25 %) + perlite (25 %).

Currently, in Mexico, sawdust and bark are materials that are available and abundant in areas with commercial timber harvesting; these residues amount to more than 690 000 m³∙year^-1, representing between 12 % and 14 % of the annual sawn volume (^{Mateo Sánchez et al., 2023}). This data reinforces the point made by ^{Buamscha et al. (2012)}, who state that, in addition to producing quality seedlings, the best substrate is one that is readily available, located close to the nursery, and low in cost.