Study area

The essay was carried out in the nursery of the Facultad de Ciencias Forestales de la Universidad Juárez del Estado de Durango (School of Forest Sciences of the Juarez University of the State of Durango), located at coordinates 24°00'48.38" N and 104°41'03.64" W, at 1 860 m.

Production of the evaluated plant

Sowing was carried out the last week of November 2015, in expanded polystyrene trays of 77 cavities with 170 mL each; a mixture composed of 44 % base mix (54 % peat, 24.5 % vermiculite and 21.5 % perlite) and 56 % of composted pine bark was used as substrate.

In order to know the morphological condition of the plant at the beginning of the experimental stage, an assessment to a sample of 56 three-month old plants extracted at random from the experimental lot was made, previously to the application of the treatments; the average initial characteristics of the seedlings under evaluation were the following: 5.2 cm of height of the aboveground part, 0.82 mm of diameter to the neck and 0.175 g of dry weight.

The plant materials remained in a greenhouse for the first eight weeks, with a 720-gauge polyethylene plastic cover, protected against ultraviolet rays; the next four weeks grew in a 60 % shadow mesh environment, and finally, the last three weeks were out in the open. Table 1 shows the specific conditions of temperature, relative humidity and luminosity, which prevailed in the three conditions where the trial was developed. A Hobo^® data logger U12-012 was used to record the reference variables.

Evaluated treatments and experimental design

During the preparation of the substrate, seven treatments were evaluated, all based on the incorporation of 6 kg m^-3 of the controlled release fertilizer (from 8 to 9 months) Multicote^® 18N - 6P₂O₅ - 12K₂O + 2MgO + micro nutrients (Haifa Chemicals Ltd.). With the exception of treatment 1 (control), which only received irrigation water without fertilization, the other six treatments were supplemented with water-soluble fertilizers: Triple 19 (19 N -19 P₂0₅ -19 K₂O), Poly-feed^® (20 N -10 P₂0₅ -20 K₂O) and Peters professional^® (20 N -7 P₂0₅ -19 K₂O), applied in fixed doses (200 ppm) and exponential (100, 200 and 300 ppm) (Table 2).

From the variation in the reception of luminosity by the position of the greenhouse, the essay was established under an experimental randomized block design. Each treatment had four blocks and the experimental unit consisted of a tray of 77 cavities of 170 mL; the selected plants were the 45 of the center.

The water-soluble fertilizers were applied for 15 weeks and their amounts were calculated based on the nitrogen concentration of each: 1) Fixed dose (200 ppm), permanently for 15 weeks; 2) Exponential dose, 100 ppm for the first four weeks; the next four, it was increased to 200 ppm. For four more weeks it was increased to 300 ppm; during the last three weeks of the experiment, the doses were reduced in the exponential treatments, adding 300, 200 and 100 ppm per week (Table 2). This allowed the doses to be equal in the fixed and exponential options. Between each dose change, to drain possible excess salts from previous fertilizations, the plant was just irrigated and not fertilized.

The water-soluble fertilizer was incorporated twice a week with a manual watering can; in each experimental unit (tray) 3.5 liters of water were added with the corresponding fertilizer solution. The pH of water was maintained between 5 and 6, and when it was higher than that interval it was adjusted with diluted hydrochloric acid in the irrigation water; pH was measured weekly with an Ohaus ST10 potentiometer.

Response variables

At 134 days (4.5 months) of starting fertilization, when the plants were 7.5 months old, ten plants of each experimental unit were randomly selected. From each one, the height of the aboveground part (cm) was recorded with a graduated ruler (cm); the diameter at the neck of the root (mm) with a 14388 SURTEK ^® 122204 Trupper^® vernier; total dry weight, the weight of the aboveground part and root were quantified with an Ohaus PA214 scale from the samples dried in an oven (FELISA^® FE-291D) for 72 hours at 70 ^oC. With these variables, the Dickson quality index (ICD) was calculated and the concentration and content of N-P-K in the foliage of the plants was estimated with three samples composed of 5 g of needles per treatment.

The chemical analysis of the plant tissue was made in the Laboratorio de Fertilidad de Suelos y Química Ambiental del Colegio de Postgraduados Campus Montecillo, Texcoco, Estado de México (Soil Fertility and Environmental Chemistry Laboratory of the Montecillo Campus Postgraduate School, Texcoco, State of Mexico). The nitrogen content was obtained by the micro-Kjeldahl method (digestion with H₂SO₄), phosphorus by colorimetric determination with yellow vanadomolybdate complex and potassium with flame emission spectrophotometry, with a SprectrAA Atomic Absorption Spectrometer 220 Fast Sequential (^{Chapman y Pratt, 1979}).

To establish if there were significant differences (p ≤ 0.05) in the morphological variables and to make the analysis of variance (ANOVA), the statistical package SAS^® (Statistical Analysis System) version 9.0 (^{SAS, 2002}) was used and when such differences were confirmed, the Tukey means test (α = 0.05) was performed (^{SAS, 2002}).

The effect of the fertilization treatments on the nutritional status of the plants was evaluated by an ANOVA. Additionally, with the graphic vector method, these results were illustrated. The construction and interpretation of vector nomograms was made based on the methodology described by ^{Haase and Rose (1995)}.

To evaluate the differences between the costs implied by the use of each fertilizer combination during plant growth, the cost of the Controlled Release Fertilizer incorporated in each treatment was considered, as well as the amount and cost of the water -soluble fertilizer, applied during each watering.

Morphological variables

The height of the plants showed a dispersion of data in a range of 29.1 to 31.6 cm with treatment 2 located in the upper statistical group (p <0.05) (Table 3). ^{Rodríguez (2008)} indicates that a plant with a stem height greater than 25 cm and a neck diameter smaller than 4 mm is more susceptible to wind damage; however, the NMX-AA-170-SCFI-2016 Mexican Standard (^{Secretaría de Economía, 2016}) establishes that Pinus greggii, from six to eight months of age, must have a height of 25 to 30 cm, which coincides with the results obtained in the present investigation.

For neck diameter, data dispersion was 3.2 to 3.5 mm, -with significant differences (p <0.05) between treatments- with treatment 3, which is located in the upper statistical group and the only one to reach the established standards in the NMX-AA-170-SCFI-2016 for Pinus greggii (^{Secretaría de Economía, 2016}), which states that plants in the nursery must reach a minimum diameter of 3.5 mm to ensure satisfactory performance in the field (Table 3). In several species, it has been found that the diameter at the neck of the root is one of the main morphological characteristics that determines the adaptation of plants in the plantation site (^{Tsakaldimi et al., 2013}). In the case of the Dickson Quality Index, the values oscillated between 0.22 and 0.26, without significant differences between treatments (p <0.05) (Table 3).

In the production of root biomass, the control plant was the best, despite having only been fed with the Controlled Release Fertilizer, with 26.8 % more biomass with respect to the average value of the other treatments (Table 4); Apparently, plants having insufficient nutriment supply developed more root in their search, which also recorded ^{Soriano (2011)}, who described that the highest dose of fertilization in Pinus patula Schiede ex Schltdl. & Cham and P. devoniana Lindl. it disadvantaged the growth of the radical system. According to ^{Berendse et al. (2007)}, as well as ^{Camargo and Rodríguez (2006)}, under low fertility, the plants allocate greater biomass to the root to favor their growth.

Treatment 3 is also outstanding in the biomass of the stem and the needles; ^{Rendón et al. (2001)} indicate that the biomass generated in this part of the plant is related to its growth, since the stem works as a support and conduction organ, while the needles help metabolic activity. As the biomass of the aerial part is generally greater than that of the root, the same treatment produced the highest total biomass (Table 4).

^{Jacobs et al. (2009)} evaluated in Quercus ilex Lour the response to the application of nitrogen in doses of 0 to 200 mg per plant and concluded that a dose of 150 mg may be enough for the growth of plants; however, a higher dose may represent the optimal value to be loaded nutritionally, without toxicity. On the other hand, ^{Prehn et al. (2013)} observed that in Guindilla trinveris Gillies ex Hook. & Arn., when applying nitrogen in increasing doses, the aerial biomass increased up to 2 g plant^-1 and from then onwards, it decreased until causing its death. In the case of this essay, none of the doses had a toxic effect.

Physiological variables

The seedlings with the highest amount of nutrient reserves were those of treatment 3 while the control showed the lowest value (Table 5).

When taking as a reference the nutritional status of treatment 1, vector nomograms show that the six treatments with the addition of water-soluble fertilizer had a high consumption of nitrogen, phosphorus and potassium, which is a condition that should favor the fertilization routines, since this translates into plants with greater nutritional reserves (^{Uscola et al., 2015}). Treatments 3 and 7 show the best ratio content-concentration-total dry biomass (Figure 1).

Concentración relativa = Relative concentration; Biomasa seca relativa = Relative dry biomass; Contenido relativo de nitrógeno = Relative nitrogen content; Contenido relativo de fósforo = Relative phosphorus content; Contenido relativo de potasio = Relative content of potassium; Descripción de tratamientos =Description of treatments: 1 = FEC (Fast release fertilizer) + water; 2 = FEC+Triple 19, fixed dose; 3 = FEC + Triple 19, exponential dose; 4 = FEC+Poly-feed®, fixed dose; 5 = FEC+Poly-feed®, exponential dose; 6 = FEC+Peters professional®, fixed dose; 7 = FEC + Peters professional®, exponential dose.

Figure 1 Nomograms of nutritional status of nitrogen, phosphorus and potassium of Pinus greggii Engelm., 134 days after fertilization began. 

A high content of nutrient reserves, especially nitrogen, in Abies religiosa (Kunth) Schltdl. & Cham. has been related to a competitive advantage of plants in places with low fertility (^{Navarro et al., 2013}). The same manifested ^{Sanchez et al. (2017)} and ^{Buendía et al. (2017)} when working with Pinus greggii and Pinus leiophylla Schl. & Deppe, respectively.

In low -fertility conditions, inducing high contents of nutrient reserves to favor a better performance in the field is the premise of the exponential fertilization regime proposed by ^{Timmer (1996)}, which is supported based on nursery results; however, in later studies, it is convenient to evaluate if in fact the plants with better nutritional status have a favorable performance in the field, since in spite of the existence of works of this nature that show positive results with the exponential fertilization regime, as for example those of ^{Salifu and Timmer (2003)} and ^{Duan et al. (2013)}, there are others such as ^{Everett et al. (2007)} that indicate the opposite. Such controversy could be based on what was indicated by ^{Burgess (1991)}, who pointed out that in some species the addition of fertilizer in an exponential way does not always satisfy the needs of the plants, mainly during the initial stages of growth when the demands of nutrients are high and the supply is scarce.

Analysis of fertilization costs

2.587 kg of Controlled Release Fertilizer were used in the 2 156 individuals under evaluation (28 trays with 77 plants each); the average price per kilogram was $ 50.00 Mexican pesos. On the other hand, the total water-soluble fertilizer used during the trial was 1.638 kg of Triple 19 (19 N -19 P₂0₅ -19 K₂O), 1.560 kg of Poly-feed^® (20 N -10 P₂0₅ -20 K₂O) and 1.560 kg of Peters professional^® (20 N -7 P₂0₅ -19 K₂O), each applied in eight trays of 77 plants (equivalent to two treatments per type of fertilizer), with an average cost per kilogram of Mx$21.59, Mx $36.80 and Mx $86.36, respectively. Table 6 shows the average cost per plant due to the alternative fertilization used in each treatment during the 4.5 months that the trial lasted; also, an example of the cost of fertilization for 100 000 plants is presented.

The treatments with the highest cost due to fertilization were number 6 and 7, with $ 0.269 Mexican pesos per plant. On the other hand, treatment 1 was the least expensive, lacking water-soluble fertilizer, with $ 0.051 per plant, followed by treatments 2 and 3 with $ 0.108. If these costs are extrapolated to a production of 100 000 plants, there would clearly be significant statistical differences; treatment 1 would cost 5 100 Mexican pesos, while treatments 2 and 3 would cost 10 800 Mexican pesos and would increase in the other treatments, until reaching the most expensive treatments (6 and 7), with 26 900 Mexican pesos (Table 6), in a fertilization period of 4.5 months.

From the technical point of view, treatments 3 and 7 were within the same statistical group and are the two treatments that responded best to the application of controlled delivery fertilizer supplemented with water soluble fertilizers in terms of the morphological and physiological variables evaluated. However, from the financial point of view, the fertilization in treatment 3 costs less than half of the cost of treatment 7, so, taking into account both aspects, treatment 3 is considered the most recommended.