This research was carried out in two 9-year-old commercial orchards of avocado cv. Hass in Antioquia, Colombia, for three consecutive years (2019-2021). The avocado trees planted were grafted on creole rootstock with a scion bud of cv. Hass and a minimum of five growth points (standard procedure). The first orchard was located in the Rionegro municipality at 06°5'56.8''LN and 075° 26' 21.9" LW (2,200 meters above sea level (masl)). The second one was in El Peñol municipality at 06°11'28.4''LN and 075° 14' 34.4" LW (2,100 masl). The WatchdogTM 2000 portable weather stations (Spectrum Technologies, 3600 Thayer Court, 107 Aurora, IL 60504) recorded climatic variables in each location. Rionegro presented a mean temperature of 17.2 °C, a maximum of 23.8 °C, and a minimum of 13.0 °C, with an annual rainfall of 1,800 mm. In El Peñol, a mean temperature of 18.5 °C, a maximum of 23.0 °C and a minimum of 14.9 °C was recorded, with an accumulated annual rainfall of 1,921 mm. According to Belda et al. (2014), the region's climate is Cw subtropical dry-winter, according to Köppen's classification. The soil of the experimental area is representative of the region, being classified as an Andosol according to FAO World Reference Base classification (Delmelle et al. 2015).

A randomized complete block experimental design with three replications was used. The experiment unit consisted of 15 cv. Hass avocado trees and the treatments of six plant densities. The plant densities (trees ha^-1) evaluated were: 204 (7x7 m), 278 (6x6 m), 333 (6x5 m), 400 (5x5 m), 625 (4x4 m) and 816 (3.5x3.5 m).

Five harvest seasons were carried out between 2019 and 2021. Three main harvests (2019M, 2020M, and 2021M) and two secondary harvests (2020S and 2021S). The main harvest (February-March flowering period) was carried out between December and January (next year), and the secondary (mitaca) harvest (August-September flowering period) was carried out in June-July (next year).

The soil nutrient availability was carried out after each harvest (2019M, 2020S, 2020M, 2021S, 2021M). Subsamples below the tree canopy and at a depth of 0-30 cm were taken for chemical analysis. The results of the soil analysis are presented in Table 1.

Table 1

Four leaves per tree were selected for the analysis of leaf tissue, one at each cardinal point. The fifth leaf of the last growth flow was collected after each harvest (2019M, 2020S, 2020M, 2021S, 2021M). The leaf was mature but not senescent, without fruiting, healthy (without physical or chemical damage or affected by pests or diseases), and older than three months, according to Maldonado (2002). The leaves were washed with distilled water and dried at 60 °C for 48 h in an oven with forced air circulation, Memmert UL 80 (Memmert GmbH + Co. KG, Büchenbach, Germany), or until a constant weight was reached. Next, they were milled, and placed in paper bags, and the total contents of N (EPA method 351.3) where determinated (USEPA 1993), P, K, Ca, Mg, Cl, S, Fe, Cu, Mn, Zn (inductively coupled plasma atomic emission spectrometer iCAP 7000 Plus (Thermo Scientific, Waltham, MA)), and B according to modified NTC 5404 (ICONTEC 2011).

To determine the nutritional balance index (NBI), Salazar-García and Lazcano-Ferrat (2001) adjusted the Kenworthy methodology for the cv. Hass avocado was used. The nutrient leaf analysis results were used to determine the balance index (B) for each mineral element, according to Equations (1) and (2).

Equation 1. Suppose the value reported in the laboratory (X) was less than the standard value:

Equation 2. If the value reported in the laboratory (X) was greater than the standard value:

where, S = standard value, I = influence of the variation, P = percentage of the standard, CV = coefficient of variation and B = balance index. Standard values (S) and coefficients of variation (CV) were used according to Salazar-García and Lazcano-Ferrat (2001).

To evaluate the concentration of nutrients, 75 fruits were randomly taken per plant distance, and each harvest with approximately 24% of dry material. Subsequently, the fruit structures were individualized, removing the peel, pulp, seed coat, and seed. All the fruit components were dried until a constant weight was reached in forced air ovens at 60 °C. After drying, all fruit components were milled to determine N (EPA method 351.3) (USEPA 1993), P, K, Ca, Mg, Cl, S, Fe, Cu, Mn, Zn (inductively coupled plasma atomic emission spectrometer iCAP 7000 Plus (Thermo Scientific, Waltham, MA)), and B according to modified NTC 5404 (ICONTEC 2011), by treatment and harvest in each location.

Five avocado fruits were taken per tree (180 per treatment) for each harvest season (2019M and 2020S) to quantify the fatty acid content. For each fruit, the mesocarp (pulp) was removed, homogenized, and lyophilized to determine the content (grams of fatty acid per 100 grams of fresh pulp) of oleic, palmitoleic, palmitic, linoleic, eladic, stearic, and fatty saturated acid. A gas chromatograph coupled to a mass spectrometer (Agilent 7890/MSD 5975C) was used, equipped with a capillary column (ZB-5 Zebron of low polarity) to separate the compounds of interest.

Vitamin E content (g α-tocopherol/100 g fresh pulp) was determined from five avocado fruits per tree (180 fruit per treatment) for each harvest season (2019M and 2020S). The mesocarp (pulp) was removed, homogenized, and frozen for each fruit. A gas chromatograph coupled to a mass spectrometer (Agilent 7890/MSD 5975C) was used before derivatization using BSTFA.

A random selection of 75 fruits per tree was each harvested. It was weighted and individually characterized by caliber according to established by the FAO in the CODEX STAN 197-1995 Revision (Table 2) for export sizes (FAO 2011).

Table 2

The statistical analysis was performed using the agricolae package in the R project statistical program (R Core Team 2021). A two-way ANOVA was carried out for the plant density factor (204, 278, 333, 400, 625, and 816 trees ha^-1) and the harvest season factor (2019M, 2020S, 2020M, 2021S, and 2021M). Differences between means were evaluated through analysis of variance, followed by Tukey's HSD mean comparison test, with a probability greater than 95%. A Pearson correlation analysis was used to examine the relationship among the content of all nutrients in the fruit and leaves of avocado.

Table 3 shows the nutritional composition of avocado tree leaves, where there were significant differences due to plant density (P<0.001) and harvest time (P<0.001). The 333 and 400 trees ha^-1 plant densities presented the highest leaf concentrations of N, P, Ca, Mg, S, Zn, and B. For K and Fe, higher concentrations were observed in the density of 625 trees ha^-1, and Mn was present in higher concentrations in the density of 204 trees ha^-1. On the contrary, the highest and lowest densities (204 and 816 trees ha^-1) reached the lowest concentrations of the evaluated nutrients. The 2019M harvest presented the highest levels of nutrients for N, Ca, Mg, S, Fe, Mn, and Zn; the other nutrients do not show marked significant differences at a particular plant density (Table 3).

Table 3

In addition to the foliar nutrient diagnosis status and the amount of nutrients removed, the assimilation dynamics of the different nutrients during fruit development are important to precise fertilization management decisions (Silber et al. 2018; Salazar-García et al. 2019). Maldonado-Torres et al. (2007) set the nutritional standards for the cv. Hass avocado for a yield greater than 20 t ha^-1. Therefore, according to the results, the nutritional concentrations of N, K, Ca, Fe, Mn, and Zn were between optimum and high levels, while P was at low levels, and Mg and B were at low concentrations. Maldonado-Torres et al. (2007) stated that the low level of P determined in these soils (Andosols) might be associated with the fixing effect of allophane, which has sometimes retained up to 2,000 mg kg^-1. Similar results were found by Tamayo-Vélez et al. (2022), who evaluated the leaf nutritional content in avocado trees in the high Andean region in Colombia.

According to Salazar-García and Lazcano-Ferrat (2003), Maldonado-Torres et al. (2007) and Campos and Calderón (2015) the recommended reference levels of nutrients in leaves for Hass avocado vary for N (1.9 - 2.31%), P (0.11 - 0.18%), K (0.5 - 1.4%), Ca (1.0 - 3.0%), Mg (0.3 - 0.8%), Fe (85 - 114 mg kg^-1), B (30 - 240 mg kg^-1), Zn (20 - 150 mg kg^-1) and Mn (30 - 500 mg kg^-1). From these reference values and according to the mineral composition of the leaf, only values below the reference were observed for P (278 and 625 trees ha^-1), which indicates that despite having observed significant differences in the concentration of nutrients in the leaf in response to the planting densities, in general the observed values were within the reference ranges for an optimal condition for the Hass cultivar water (Maldonado-Torres et al. 2007).

There were significant differences in plant density in the nutritional balance index (NBi) for N, Ca, Mg, S, Fe, Mn, Zn, and B. While NBi for P and K did not present significant differences (P>0.05) for any of the plant densities evaluated. Similarly, the 2019M harvest season showed the highest NBi for all mineral nutrients except boron (Table 4).

Table 4

Balanced and timely nutrition is essential to ensure the yield and quality of avocados. The avocado cultivar, the management practices, and the environmental conditions where it is cultivated influence the concentration of nutrients in the leaves and fruits (Salazar-García et al. 2019). Hence, in assessing the nutritional status of plants, Manzoor et al. (2022) recommend using the nutritional balance index (NBI) to seek the plants' nutrient balance. Leaf nutritional content allows the avocado leaves, according to the nutrient balance index proposed by Kenworthy (1973), as scarce (15-49%), below normal (>49-83%), normal (>83-117%), above normal (>117-151%), and excess (>151-185%). According to our results, although there were differences between plant densities, the N, P, K, Ca, and Fe generally had normal balances, while Mg, S, and Mn were below normal, indicating possible nutritional deficiency without symptoms being observed. Similar results were reported by Tamayo-Vélez et al. (2022), who evaluated the effect of rootstock/scion compatibility in the same regions. These authors found that the N, P, Ca, Fe, and Zn showed normal NBi, while K, Mg, S, Mn, and B were below normal, indicating possible nutritional deficiency without symptoms being observed.

Plant density is known to affect plant traits and growth above ground significantly. Nevertheless, understanding the effects of sowing density on fruit and leaf nutrient accumulation is essential in determining an adequate fertilization program to guarantee acceptable yield and fruit quality (Hecht et al. 2019). More plants at higher plant densities require and take up more nutrients. This does not necessarily occur proportionally due to competition, which can be asymmetric when plants of different sizes compete. Instead, the competition might trigger similar root growth and canopy plasticity responses as low resource availability significantly changes the root and canopy system architecture and functioning (Hecht et al. 2019; Tun et al. 2018; Cano-Gallego et al. 2023) and soil elements availability (Reddy et al. 2014).

Fruit mineral composition: The plant density significantly does not affect the nutrients in the avocado fruit for the minerals K, Ca, S, and Fe. Conversely, the contents of N, P, Mg, Na Cu, Mn, Zn, and B were significantly affected by plant density. However, a homogeneous behavior was not observed in the variation of this variable; in general, the plant densities of 333 and 400 trees ha^-1 presented high contents of all nutrients in fruit, except for Cu, whose contents were medium (Table 5). A similar result was reported: nitrite content in the fruit is also the lowest in the low-density treatment in cucumber (Ding et al. 2022), and a positive relation was found between leaf N to pulp N concentrations (Arpaia et al. 1996).

Table 5

Regarding the nutritional composition in each fruit tissue, the highest contents of all the nutrients were observed in the seed coat and pulp. In contrast, the peel and seed only presented higher values than the rest of the tissue in Na levels. Similar results were reported by Tamayo-Vélez et al. (2022), who found, in general, the highest nutrient contents in the seed coat. The seed coat provides an interface between the embryo and the external environment, can explain altered processes in fruits during embryogenesis, dormancy, and germination, and has a role in determining seed size (Haughn and Chaudhury 2005).

The concentration of elements in avocado fruits was in decreasing order of K>N>Ca>Mg>P>S>B>Fe>Zn>Mn>Cu. In this regard, Reddy et al. (2014) found a similar element concentration in avocado cv. Hass and Fuerte (Mg>Ca>Zn>Fe>Mn>Cu). Since the mesocarp is the edible part of the avocado fruit, its quality at consumption maturity usually identifies whether adequate orchard management was carried out (Salazar-García et al. 2019). It suggested that avocado fruits are nutritional as a good dietary source of the micronutrients Cu and Mn, contributing 75 and 34% towards dietary reference intake for these elements by avocado fruit consumption (Reddy et al. 2014). Mesocarp concentrations of both N and Ca are relevant for postharvest fruit quality (Salazar-García et al. 2019). Fruit from trees high in N showed increased susceptibility to chilling injury and a shorter ripening time (Kruger et al. 2016). Otherwise, high Ca levels have decreased cold-induced disorders (Salazar-García et al. 2019).

Some antagonistic and synergistic relationship between the soil exchangeable minerals and mineral concentration in avocado cv. Hass fruit has been reported by Reddy et al. (2014). Soil Al excess reduces cv. Hass fruit Cu (-0.8) and Fe (-0.9), Fe excess reduces Cu (-0.9) and Zn (-0.7), Ca excess reduces Mn (-0.7), and Mg excess reduces Mg (-0.8). On the other hand, the synergistic effects include soil Al excess increase in the Mn (0.8) and Cu (0.9) fruit concentration, Soil Mg excess increase in Mn (1.0) and Cu (0.9) fruit concentration, and soil Ca excess increase in Mn (0.7), Ca (0.7) and Cu (0.7) fruit concentration. The local soil presented strongly acid soils with high organic matter content, high and very high Ca, Mg with medium to low ranges, and high K. Minor elements are reported to have very high contents of iron, medium to high for manganese and zinc, and medium to low for copper and boron (Tamayo-Vélez et al. 2022). Ding et al. (2022) demonstrated that changing plant density has significantly improved the quality of the cucumber fruit.

In the case of avocado cv. Hass, it has been observed that increasing the N concentration in the leaf leads to a higher N content in the pulp (Tamayo et al. 2018). This finding was further confirmed in this study, where the highest leaf and fruit N concentrations were observed at intermediate plant densities (333 and 400 trees ha^-1). This suggests that plant density can play a significant role in determining fruit quality, with intermediate plant densities leading to higher N concentrations in the fruit.

The correlation matrix shows that some elements are associated with others in avocado leaves and fruits (Figure 1). The conventional approach to interpreting a correlation coefficient correlation selected was a very strong correlation (greater than 0.900), strong correlation (between 0.700 and 0.899), moderate correlation (between 0.400 and 0.699), and weak correlation (between 0 and 0.399) (Schober et al. 2018). S is positive and strongly associated with Ca in leaves and moderately associated with Mg, Zn, and N. Mg is moderately associated with Ca. On the other hand, P and Fe show a negative moderate association. In addition, some elements, mainly Fe and K, show no or minimal correlation with others (Figure 1A). In fruit, Zn is positive and strongly associated with Mn and B and moderately associated with Fe, Ca, and N. B is moderately associated with N, Mn, and K. Conversely, no negative association was observed, and P shows no or minimal correlation with other (Figure 1B). Lazare et al. (2020) reported significant correlations in most nutrients with the others in cv. Hass leaves with the others. The strongest correlations (more than 0.7) were between Mg and Ca, P and Cu, Mg and Mn, N and P, and Mg and N.

Figure 1

Silber et al. (2018) demonstrated high Zn concentration in the productive organs of avocados probably because Zn is transported in the phloem, unlike the other elements that are transported in the xylem. Zn is involved in diverse plant functions such as photosynthesis, sucrose and starch formation, protein metabolism, membrane integrity, auxin metabolism, flowering, and seed production. Zn shows a significant correlation with N, P, and B in leaves, while in fruits, with Cu, Fe, and S (Figure 1).

Figure 2 summarizes the contents of fatty acids present in the pulp of the fruits for the plant densities (Figure 2A) and harvest seasons (2019M and 2020S) (Figure 2B). The fruit harvested from trees in high densities presented the highest concentrations (g 100 g^-1) of Palmitoleic acid (3.01), Palmitic acid (7.15), Elaidic acid (1.61), and saturated fatty acid (7.38). On the contrary, the plant density of 333 and 400 trees ha^-1 had the lowest contents for this same group of acids (Palmitoleic acid - 0.55 g; Palmitic acid - 1.63 g; Linoleic acid - 1.11 g; Oleic acid - 2.63 g; Elaidic acid - 0.44 g; and saturated fatty acid - 1.73 g). Regarding the harvest season, in the 2020S, higher content of Palmitoleic acid (2 g), Palmitic acid (5.38 g), Linolenic acid (1.86 g), Elaidic acid (1.17 g), Stearic acid (0.19 g), and saturated fatty acid (5.57 g) were recorded. The 2019M harvest only presented higher values for Oleic acid (2.63 g).

Figure 2

Vitamin E content is presented descriptively in Figure 3. The concentrations of vitamin E (mg α-tocopherol 100 g^-1) were similar in the plant densities of 204 (1.42), 278 (1.45), 400 (1.36), and 625 (1.415) trees ha^-1. The lowest values (0.905) were reached in the density of 333 trees ha^-1. Similar to what was observed for fatty acid concentration, a lower fatty acid level was observed at this plant density. Regarding harvest time, vitamin E contents were observed for 2019M (1.24) and 2020S (1.35). It has been reported that vitamin A contents rank for 1.78-3.5 mg α-tocopherol 100 g^-1 (Jimenez et al. 2020). Nonetheless, this study's overall vitamin E content was low (0.90-1.45).

Figure 3

For many years, the profitability of avocado production was measured in terms of total fruit yield. However, this parameter has lost importance due to the avocado market. Harvest time, size, and fruit quality (external and internal) are considered the main factors for successful avocado marketing (Salazar-García and Lazcano-Ferrat 2001). However, the percentage of avocado fruits by size showed no significant differences due to plant distances. The harvest factor significantly affected most of the percentage proportions of each caliber in the avocado fruit. Despite the effect of the harvest season, a homogeneous behavior was not observed in the variation of this variable (Table 6).

Table 6

Related to crop yield, it is reported that planting density affects avocado fruit yield. Cano-Gallego et al. (2023) found that yield per tree and number of avocado fruits per tree are negatively affected by the increase in planting densities. In addition, fruit quality parameters show better results at intermediate planting densities of 333 and 400 trees per hectare. The intermediate densities of 333 and 400 trees produced the highest fruits per tree (323) and the highest fruit yield (19 t ha^-1).