Pecan (Carya illinoensis [Wangenh] K. Koch) is a crop with a wide range of uses and high profitability. The area harvested of this crop by the world’s main producing countries (China, Iran, United States, Turkey and Mexico) is between 96,909 and 390,924 ha, with a clear upward trend (Food and Agriculture Organization of the United Nations [FAO], 2018). Mexico stands out as the world’s second largest pecan producer, with a production volume of 159,535 t. Among the states with the most production are Chihuahua, Sonora, Coahuila, Durango and Nuevo Leon, which together account for 94.3 % of this production (Servicio de Información Agroalimentaria y Pesquera [SIAP], 2018). Seventy percent of Mexico's total production is exported, mainly to the United States and China (Zhang, Peng, & Li, 2015).
Several studies report that water and nitrogen fertilization are the main limitations for obtaining higher yields and better nut quality (number of nuts per kilogram and kernel percentage) (Castillo-González et al., 2019; Flores-Córdova, Soto-Parra, Javier-Piña, Pérez-Leal, & Sánchez-Chávez, 2018; Retes-López, Nasaimea-Palafox, Moreno-Medina, Denogean-Ballesteros, & Martín-Rivera, 2014); poor management contributes to alternate bearing (Worley & Worley, 2000). Nitrogen is the nutrient with the highest demand and volume of application in pecan orchards (Zaragoza-Lira et al., 2011), since it has a direct relationship with leaf nutrient concentration (Acuña-Maldonado et al., 2003; Wells, 2011) and increased fertilization costs (Soto-Parra, Piña-Ramírez, Sánchez-Chávez, Pérez-Leal, & Basurto-Sotelo, 2016).
It has been observed that modifying the dose and timing of nitrogen application are favorable agronomic management practices for increasing the productivity of pecan trees (Sánchez et al., 2009). The use of organic sources of this element, such as compost and earthworm humus, has not shown a significant effect on the yield and quality of the nut, in addition to keeping the content of organic matter and nitrates unchanged (Soto-Parra et al., 2016), but they do represent an increase in the production cost due to nitrogen fertilization (Tarango-Rivero, Nevárez-Moorillón, & Orrantia-Borunda, 2009; Zaragoza-Lira et al., 2011).
Martínez-Rodríguez and Ávila-Ayala (2002) evaluated different N doses applied in northern Mexico to 'Western Schley' pecan trees in production, concluding that the 60 kg·ha-1 dose showed the highest yield and quality values of the harvested nut. Flores-Córdova et al. (2018) applied 226, 121 and 94 kg·ha-1 of N, P2O5, and K2O, respectively, and found a yield of 1.94 t·ha-1, 159 nuts·kg-1 and 59 % edible nut. Sánchez et al. (2009), in the municipality of Aldama, Chihuahua, Mexico, tested three nitrogen doses (160, 320 and 480 kg·ha-1), which they applied during pre-sprouting of the tree, fruit set and growth, aqueous and milky stages, ripening and postharvest, and reported that the 160 kg·ha-1 dose improves the nitrogen use efficiency (NUE), productivity and quality of the nut (139 nuts·kg-1 and 54 % kernel).
Yield response is associated with orchard management; however, in low production years there is an increase in kernel weight (Wells, 2010), where the supply of an essential element such as N should be used when leaf analysis reveals its deficiency (Medina-Morales, 2004). In addition to being an agronomic management practice that can contribute to increasing the yield of the main pecan production areas (Tarango-Rivero et al., 2009), knowing the optimum dose of nitrogen fertilization represents a decrease in the economic and environmental impact caused by the excessive application of chemical fertilizers (Zaragoza-Lira et al., 2011). Therefore, the objective of this research was to evaluate a single and split application of nitrogen in 'Western Schley' variety pecan, as well as its effect on leaf nutrient concentration, yield and nut quality.
Materials and methods
Location of the experiment, plant material and experimental design
The study was conducted from 2010 to 2014 in the so-called "La Florida" orchard, located in the municipality of Jimenez, Chihuahua, Mexico (27° 06’ 16’’ NL and 104o 56’ 09’’ WL, at 1,321 masl), with average annual temperature and precipitation of 18.6 °C and 369.8 mm, respectively. As plant material, we used 'Western Schley' pecan trees planted in 1989 in a real 12 x 12 m frame (69 trees·ha-1) in a xerosol type soil with the following characteristics: clay-silty texture (18.1 % sand, 40.6 % silt and 41.3 % clay), pH of 7.76, 0.75 % organic matter, 8.54 % carbonates and 1.0 dS·m-1 electrical conductivity. According to a previous chemical analysis of the soil, the following results were found (mg·kg-1): 28.5 total N, 21.5 P, 1,075 K+, 3,900 Ca2+, 300 Mg2+, 406 Fe2+, 692 Mn2+, 86 Zn2+ and 79 Cu2+. Water supply during the production cycle (March to October) was provided by means of a micro-sprinkler irrigation system, applying 1,300 L per tree in a 15-day interval between each irrigation. Weed control was done manually.
The factors studied were dose (100, 150 and 200 kg·ha-1) and mode of application (single and split, with the latter involving two applications of 50 % each) under a 3 x 2 factorial arrangement with six treatments. A randomized complete block experimental design with six replicates was used; the experimental unit was a tree with a height of 10 ± 1 m and a trunk diameter of 29.87 ± 2 cm (measured at the beginning of the study at 20 cm above ground level). The N source was ammonium sulfate (20.5 % N and 24 % S). Nitrogen fertilizer was applied in a band at a depth of 10 cm and 3 m from the trunk in the shaded area of the tree. The first application was made during the third week of March (flowering) in all treatments, and the second split application was made during the second week of June (fruiting). In each evaluation year, and in parallel to the total application of N, 100 kg·ha-1 of P and K were also applied, where the P source was phosphoric acid (49 % P2O5 and 1.61 kg·L-1 density) and the K source was potassium thiosulfate (12.6 % K2O and 1.46 kg·L-1 density).
In the evaluation years (2010 to 2014), during the third week of July, leaf samples were taken for nutritional analysis, this in accordance with the indications of Ojeda-Barrios et al. (2014). To do this, 40 pairs of healthy leaflets (absence of mechanical damage or presence of pests and diseases) from the current growth cycle were selected from each experimental unit, located in the middle of the tree canopy and the four cardinal points. The leaflets were washed with a 0.1 % phosphate-free detergent solution, followed by a rinsing with deionized water and drying at 80 °C in an oven (Heratherm™ VCA 230, Thermo Scientific™, USA). The samples were homogenized in a mill (Wiley®, USA) with 1 mm mesh. Total N was quantified by the Micro-Kjeldahl method (Novatech®, USA and Micro Kjeldahl Labconco®, USA) (Fernández, del Río, Abadía, & Abadía, 2006). The concentration of Ca++, Mg++, K+, Cu++, Fe++, Mn++ and Zn++ was determined from 1 g of the dry sample by triacid digestion with HNO3, HClO4 and H2SO4 (25 mL mixture at a 10:1:0:25 ratio). Analyte quantifications were performed with an atomic absorption spectrophotometer (Series 300, Thermo Scientific™, USA), and total P was determined with the vanadate-ammonium molybdate method (Ojeda-Barrios et al., 2014).
Yield, nut quality and nitrogen use efficiency
The harvest for each evaluation year (2010 to 2014) was carried out in the fourth week of November, where each experimental unit was mechanically vibrated and the nuts were collected, counted and weighed with a scale (Combo-Rhino-122, Rhino®, Mexico) with a sensitivity of 0.1 g, this to obtain the yield in kg·tree-1. Quality variables (number of nuts per kilogram and percentage of edible nut) were determined according to the method indicated by the Mexican Standard NMX-FF-084-SCFI-2009 for non-industrialized food products for human consumption, such as pecan (Secretaría de Economía, 2009). The number of nuts per kilogram was the product of the total number of nuts harvested per tree by their total weight. To obtain the percentage of edible nut, 300 g of nuts were selected per experimental unit, the shell was separated from the kernel (edible part) and the weight was obtained with a portable electronic scale (Scout Pro SP202, Ohaus®, USA) with a sensitivity of 0. 01 g; finally, the weight was divided by the initial value of the sample (300 g) and multiplied by 100. On the other hand, the NUE was determined according to the methodology proposed by Sánchez et al. (2009), which considers the ratio between the kilograms of nut harvested and the kilograms of N applied, both per hectare.
Data analysis was performed with information from the five years of the study. Normality was verified with the Kolmogorov-Smirnov test (Sokal & Rohlf, 1995), and when this assumption was not fulfilled, log base 10 was used for data transformation. Subsequently, an analysis of variance was performed with the randomized complete block experimental design, considering a factorial treatment design. In the characters where treatment effect was detected, Tukey's mean comparisons tests (P ≤ 0.05) were performed. In all cases, SAS statistical software version 9.3 (SAS Institute Inc., 2011) was used.
Results and discussion
The leaf concentration of total N, P and K+ showed no significant difference in relation to the N doses, the application mode or their interaction (Table 1), which was confirmed by the absence of interveinal chlorosis in mature leaflets. The intervals found in this study for N (27.4 to 28.9 g·kg-1), P (1.6 to 1.7 g·kg-1) and K+ (10.6 to 11.6 g·kg-1) are similar to those reported by Medina-Morales (2004) and Walworth, White, Comeau, and Heerema (2017) in pecan orchards in production, located in northern Mexico and the southeastern United States of America. Likewise, Martínez-Rodríguez and Ávila-Ayala (2002) found no statistical difference among treatments with the application of 60, 120 and 180 kg·ha-1 of ammonium sulfate in relation to the leaf concentration of total N. In this sense, the magnitude of the response of pecan to N fertilization does not show a trend based on the applied dose. However, the same does not occur with P and K, since their reaction is slow or undetectable (Smith, Rohla, & Goff, 2012), and its effect can be seen in kernel weight and percentage (Smith & Cheary, 2013), hence the importance of monitoring and making timely applications to maintain optimal levels of these elements (Pond et al., 2006).
|Dose and application mode (kg·ha-1)||Total N||P||K+||Ca2+||Mg2+||Fe2+||Mn2+||Zn2+||Cu2+|
|100single||27.4 az||1.6 a||11.2 a||20 a||4.0 a||140.3 a||539.0 a||36.9 ab||9.21 a|
|150single||28.8 a||1.6 a||11.4 a||17 b||4.0 ab||138.6 a||632.6 a||35.8 ab||9.5 a|
|200single||28.4 a||1.6 a||11.1 a||19 ab||4.0 ab||132.2 a||569.9 a||39.7 a||10.0 a|
|100split||28.9 a||1.6 a||11.6 a||18 ab||4.0 ab||122.1 a||411.8 a||35.9 ab||9.1 a|
|150split||28.6 a||1.7 a||10.8 a||19 ab||3.0 b||125.8 a||393.5 a||32.9 b||9.3 a|
|200split||28.6 a||1.6 a||10.6 a||19 ab||4.0 ab||133.2 a||472.8 a||36.8 ab||9.5 a|
Significant interaction between dose and application mode of N was detected for Ca2+, Mg2+ and Zn2+, where the highest leaf concentration of calcium and magnesium was obtained with the single application of 100 kg·ha-1, with values of 20 and 4.0 g·kg-1, respectively. Similar values of calcium (20.2 g·kg-1) and magnesium (4.94 g·kg-1) are reported by Pond et al. (2006) in pecan orchards established in Arizona, USA, as well as by Medina-Morales (2004) in northern Mexico, with 18.0 and 3.82 g·kg-1, respectively. Leaf calcium and magnesium concentrations found in this study can be considered adequate or normal, according to the concentration range of 15.7 to 24.2 and 3.9 to 5.0 g·kg-1, respectively, indicated for these elements by Pond et al. In this regard, Zaragoza-Lira et al. (2011) report that the presence of calcareous soils with high base saturation and a pH between 7 and 8 increases the absorption of calcium and magnesium. In the case of magnesium, a similar concentration was maintained throughout the evaluation years.
In this work, no significant interaction between dose and application mode of N was detected for the concentration of Fe2+, Mn2+ and Cu2+ in the leaflets. In this regard, Pond et al. (2006), in determining the standard nutrient concentration in 'Western Schley' trees grown in Arizona, report ranges (low, normal and high) of values for Fe2+, Mn2+ and Cu2+, and when comparing these ranges with the values obtained in this work, it can be seen that Mn2+ is in the normal range (104 to 673 mg kg-1). Fe2+ and Cu2+ values were similar to the reported ranges of 104 to 673 and 6 to 10 mg·kg-1, respectively. This may be associated with a slight decrease in the initial pH value (7.76) in the root zone of the trees, since many microelements, among them Fe2+, Mn2+ and Cu2+, are found in unavailable forms or in conditions that prevent their absorption when the pH is alkaline (Tarango-Rivero et al., 2009). The latter is associated with the prevailing aridity conditions in the main pecan production areas in Mexico (Flores-Córdova et al., 2018; Martínez-Rodríguez, & Ávila-Ayala, 2002; Medina-Morales, 2004).
Regarding Zn2+, the single application of 200 kg·ha-1 of N significantly increased the leaf concentration of this nutrient (39.7 g·kg-1), in relation to the 32.9 g·kg-1 obtained with 150 kg·ha-1 of N applied on a split basis (Table 1). In the particular case of pecan, after N, Zn2+ is the most important nutrient element, and its deficiency (20 mg·kg-1) is manifested by an excessive reduction in vegetative growth (rosetting) (Castillo-González et al., 2019), associated with the presence of alkaline soils as indicated by Ojeda-Barrios, Abadía, Lombardini, Abadía, and Vázquez, 2012. Heerema et al. (2017) point out that the sufficiency ranges for Zn2+ are between 20 and 50 mg·kg-1. On the other hand, Pond et al. (2006) indicate that the normal concentration is between 86 and 256 mg·kg-1, which exceeds the values found in this study, so they could be considered as insufficient.
Given the difference between what was observed and reported by other authors, it is important to note that during the evaluation no interveinal chlorosis, necrosis or presence of leaves with undulate rosette margins were observed (Castillo-González et al., 2019). Additionally, it is important to consider that the pecan tree shows an alternate behavior with respect to its production (Flores-Córdova et al., 2018), and in high production years, the demand for macro and micronutrients is greater and its leaf concentration shows a decreasing trend, contrary to what occurs in low production years (Smith, 2010; Smith et al., 2012; Wood, Conner, & Worley, 2003).
Yield, nut quality and nitrogen use efficiency
Nut yield and quality are widely associated with the contribution of N (Sánchez et al., 2009). In this study, after five production and evaluation cycles, significant interaction was observed for yield, number of nuts per kilogram and NUE; thus, the 100 kg·ha-1 N dose with both application modes showed the highest NUE value (P ≤ 0.05) (Table 2). The yield of 44.60 kg·tree-1, obtained with the 100 kg·ha-1 in a single application, was statistically higher than the 34.78 kg·tree-1 achieved with the 150 kg·ha-1 split application (Table 2). These results contrast with data reported by Castillo-González et al. (2019) of 15.67 ± 1.96 kg·tree-1 in 20-year-old pecan orchards.
|Dose and application mode (kg·ha-1)||Yield (kg·tree-1)||Number of nuts per kg||Edible kernel (%)||NUE (kgnut·kgN -1)|
|100single||44.60 az||194.8 a||57.1 a||31.10 a|
|150single||38.90 ab||186.7 b||58.7 a||18.15 b|
|200single||40.48 ab||187.1 ab||58.3 a||14.10 c|
|100split||43.06 ab||196.2 a||56.3 a||30.10 a|
|150split||34.78 b||186.9 b||58.3 a||16.23 b|
|200split||42.09 ab||188.6 ab||58.1 a||14.00 c|
Acuña-Maldonado et al. (2003), with the annual (October) application of 125 kg·ha-1 of N and split in 'Western Schley' pecan, found no significant differences related to the yield obtained per tree, which indicates that the use of this element could be associated with the vegetative sprouting of the following growth period rather than with the development of the fruit (Sánchez et al., 2009). Additionally, some researchers indicate that the fact that the trees remain with leaves until mid-October affects the accumulation of photoassimilates, which are favorable at the start of sprouting and during fruit set and growth (Acuña-Maldonado et al., 2003; Wells, 2011). The application of a lower N amount to the soil decreases its contamination by volatilization, leaching and denitrification processes (Soto-Parra et al., 2016); it also reduces production costs linked to the application of nitrogen fertilizers (Flores-Córdova et al., 2018).
Soil pH is another factor to consider, since it considerably affects the availability of nutrients and, consequently, root absorption. In this work, no changes in pH values were observed, which could be explained by the buffering capacity of the soil and the way in which the fertilizer was added in the experiment (Wells, 2011).
The average value of 194.8 nuts per kilogram obtained in trees fertilized with 100 kg·ha-1 of N in a single application at the beginning of the year was significantly higher compared to the 186.9 obtained with 150 kg·ha-1 applied in the split application mode in March and June. In contrast, Sánchez et al. (2009), with a split application of N (using ammonium sulfate as a source) based on the tree’s phenological stage (pre-sprouting, start of fruit set, fruit growth, aqueous stage, milky stage, fruit ripening and post-harvest recharge), report 139, 140 and 142 nuts per kg, for 160, 320 and 480 kg·ha-1 of N, respectively. What was found in this study may be associated with a decrease in size and lower quality (Wells, 2011). Also, the number of nuts per kilogram is a quantitative characteristic and directly influences weight (Smith, Wood, & Raun, 2007). According to NMX-FF-084-SCFI-2009, the values found correspond to the average size of the nut (171 to 210 nuts per kg).
Another important quality variable is the percentage of edible kernel, since it maintains a directly proportional relationship with the value obtained in the market, and is a valued characteristic among nut producers (Flores-Córdova et al., 2018). In this study, the values of the percentage of edible nut ranged from 56.3 to 58.7, showing no statistical difference due to the dose, the mode of N application or their interaction (Table 2). Flores-Córdova et al. (2018) and Sánchez et al. (2009) obtained similar results and, according to NMX-FF-084-SCFI-2009, the percentage of kernel obtained is classified as excellent.
As a result of the increased cost of nitrogen fertilizers and the pollution generated by their excessive application, it is essential to use indicators such as NUE, which is a measure of the productive efficiency of the pecan crop (Sánchez et al., 2009). In this study, data ranged from 14.00 to 31.10 kg of nut produced per kg of applied N, where single and split applications of 100 kg·ha-1 were statistically more outstanding (31.10 and 30.10 kg of nut produced per kg of applied N, respectively) (Table 2). The above, in practical terms, indicates the difference between applying 100 or 200 kg·ha-1, which in addition to not affecting the yield obtained represents a significant economic outlay for the producer (Ojeda-Barrios et al., 2012; Wells, 2010). Sánchez et al. (2009) report values of 16.91 kg of nut produced per kg of N applied, this with the split application of 160 kg·ha-1 of N in seven applications based on the 'Western Schley' pecan’s phenological stages.
The N dose and application mode did not affect the leaf concentration of total N, P, K+, Fe2+, Mn2+ or Cu2+. Significant interaction between factors was detected for the concentration of Zn2+ and the single applied N dose of 200 kg·ha-1, by significantly increasing the leaf concentration of this nutrient to 39.7 g·kg-1. With the single application of 100 kg·ha-1 of N, the highest values of yield (44.60 kg·tree-1) and nuts per kilogram (194.8) were achieved, while the kernel percentage remained unchanged. The maximum nitrogen use efficiency was obtained with the 100 kg·ha-1 dose in a single or split application, with 31.10 and 30.10 kg of nut produced per kg of N applied, respectively.