1. Introduction
Almonds are one of the most important Spanish dryland crops, especially in the Mediterranean area, although it is also cultivated in the interior part of the peninsula. This crop has good economic benefits, so its implementation as an irrigated crop is incentivized. The cost of production of almonds is low compared with other fruit trees as it does not require large amounts of water. In addition, this species is able to adapt to adverse soil and climate conditions without economic losses and can thus occupy areas that are not normally suitable for other crops. Additionally, under optimum conditions, almond fruits can be stored for long periods of time prior to exportation.
The production of almonds has progressively increased since the year 2000, reaching a current production level of 3 million tons (shelled) worldwide. The continent with the greatest production is the United States of America, producing almost a million tons, exclusively produced by the state of California. The second-highest producer is Spain, with an approximate production of 200,000 tons per year, followed by Iran, with 100,000 tons [
1]. Australia is an emerging producer, with a production level that is growing exponentially. In Spain, the autonomous community with the largest area of cultivation is Andalusia, with around 160,000 hectares, followed by the Comunidad Valenciana, with 93,000 hectares, and the Region of Murcia and Castilla-La Mancha, with 73,000 hectares each. Of the total surface used for almond cultivation in Spain, only a low percentage is irrigated, while most of the crops are dryland [
2].
As for the surface area used for almond cultivation, the data from various statistics databases indicate a period of decrease. The dichotomy observed between the yield and surface area is due to the considerable increase in the performance of this crop per hectare. The technical advances in crop yield are led by the United States of America, thanks to the acquisition of new varieties that are more productive, the implementation of irrigation and fertilization systems, improvements in the mechanical harvesting of almonds, and the use of high-density plantations. The rest of the world is beginning to copy the American manner of cultivation, so production will continue to increase in the coming years. In Spain, the main production problem is frost events, so important work is being conducted on genetic improvements to obtain late and extra-late self-fertilizing varieties. Additionally, great efforts are being made towards switching from the typical dryland crop to irrigated crops, whose difference in performance (Kg/ha) can exceed 1000 Kg/ha [
3]
In Spain, a wide range of almond varieties are cultivated, among which we find two French-origin cultivars, Ferraduel and Ferragnés, and a Spanish-origin cultivar, Garrigues. The general characteristics of Ferragnés include vigor, vertical growth with a slight spread, high productivity, high adaptability, and a soft shell [
4]; it flowers late, with a medium-fast entry into production [
5]. Ferraduel is a sister variety to Ferragnés, and it is also characterized by late flowering, at the same time as Ferragnés. The fruit is hard and amygdala-shaped, with no double fruits and a somewhat late maturation period. It has an average performance; however, in dry years, this decreases considerably [
5]. Garrigués is a Spanish variety, specifically from the Region of Murcia. It meets nutritional requirements as a vigorous tree that grows upright, which is highly branched, and it has a fast entry into production. It has a high productivity in irrigated coastal areas. It has an average period of flowering with abundant and staggered flowers and does not tolerate drought or a lack of humidity. In the interior areas, it is greatly harmed by frost, and production in dryland areas is very low [
5].
Nutritional requirements vary according to the age and phenological state of the plant.
Table 1 and
Table 2 are representative examples of almond fertilization. In almonds, fertilization is mainly conducted through the soil or foliar application. The most traditional, and currently most common, method is the direct application of nutrients to the soil. The fertilizers move into the soil solution and are absorbed by the roots. To improve the efficiency of soil fertilization, the application criteria or recommendations have to be followed. Fertilization must occur when the tree has vegetative growth activity and therefore a greater capacity to absorb nutrients. It is recommended that fertilizer application be split into two or three separate occasions throughout the period of vegetative growth [
6].
Currently, it is thought that the best way to fertilize woody crops is to take into account the nutritional state of the plant via mineral analysis of the leaves. This system is the best method for detecting low levels of nutrients before the onset of deficiency. The nutrient content can be interpreted from the foliar analysis through various methods. The range of normality or sufficiency (RN) methods are utilized frequently, but these are being replaced by more sophisticated methods, such as the Standard Deviation Index (SDI), DRIS (Diagnosis and Recommendation Integrated System), M-DRIS (Modified Diagnosis and Recommendation Integrated System), and CND (Compositional Nutrient Diagnosis). For almonds, however, at present, there is little information associated with the nutritional diagnosis. Moreover, the existing information is obsolete given the appearance of new varieties and the implementation of new fertigation techniques.
Fruit trees acquire nutritional stability in their adult state [
7]. Due to this stability, by diagnosing the foliar composition, it is possible to make adjustments in fertilization. It is important to specify the optimal foliar sampling period for each species, which must coincide with the period of nutritional stability. The compositional nutrient diagnosis method (CND) takes into account the association between a given nutrient and the geometric mean of the concentrations on the basis of dry matter recorded for the other nutrients (multivariate relationships), including those that are not determined analytically. This is considered the best way to express balance in plant tissue [
7,
8]. This methodology has been applied to fruit plants such as banana, orange, pear, mango, guava, etc., but it has not yet been applied to almond cultivation [
9,
10,
11,
12].
The main objective of the present work is to establish the ranges of normality (RN) and DRIS and CND norms for the interpretation of nutritional analysis of this crop using three of the more popular almond cultivars in Spain—Ferraduel, Ferragnés, and Garrigues. We want to test the hypothesis that RN, DRIS, and CND norms are variety-dependent; therefore, it is necessary to determine this norm for each specific variety as in almond trees the agronomic characteristic among varieties can be different for each one. In addition, this study also allows us to establish the best time period during the season for collecting foliar samples of these almond varieties. This knowledge will be useful to growers for the design of protocols for collecting samples and to the agronomic engineers and plant nutrition scientists to interpret and understand leaf analysis data. Usually, these types of norms are used mainly in the academic world since, although they are successful in agriculture, they require mathematical knowledge to obtain and interpret. However, new technologies including artificial intelligence (AI), neuronal networks, and cheap and easy to use decision support systems (DSS) could be powerful tools for implementation. Therefore, it is necessary to perform an experimental assay to obtain data to feed to AI and DSS, letting us create these norms and guidelines. In this work, RN, DRIS, and CND norms will be created in a high-yielding population referred to as the “reference population”.
2. Materials and Methods
The selection of experimental plots was conducted according to the criteria by De Oliveira et al. [
13] and Sumner [
14], who state that plots to determinate DRIS norms have to be commercial and homogenous, with high production performance. Our plots were located between the provinces of Albacete and Murcia (Spain,
Figure S1), with similar agroclimatic and edaphic conditions. The trees were cultivated under a drip irrigation system with the usual agronomic management in this area. The varieties studied were Ferraduel, Ferragnes, and Garrigues. The Ferraduel and Ferragenes orchards were located in Hellin (Albacete) and Garrigues in Jumilla (Region of Murcia). All orchards were separated by one kilometer. The orchard had a yield production of about 4000 kg/ha. The foliar samples were collected in the first half of May until the second half of September over two consecutive years, 2017 and 2018, with no significant differences observed between both years. The reported data are from 2017.
During each year, 10 sampling periods were established from May to September, with samples collected every fortnight. This produced 12 foliar samples with 20 mature leaves for every sample. In each sample, the contents of nitrogen, phosphorus, potassium, calcium, and magnesium were analyzed at Fertiberia laboratories (Spain). The leaves were briefly rinsed with deionized water, oven-dried at 60 °C for 48 h, weighed, and ground in a hammer mill. After that, the leaves were digested via nitric-perchloric acid (2:1) at 90 °C. Analysis of K, Ca, and Mg was carried out by atomic absorption spectrometry (Perkin/Elmer 5500). The total N was determined using the semi-micro Kjeldahl method. Total-P was measured colorimetrically using the molybdenum blue method described by Murphy and Riley [
15].
2.1. Determination of the Foliar Sampling Period
With the results of the foliar analysis obtained, it was possible to establish the best period of the season for taking leaf samples in the almond trees, as previously mentioned, as the nutrients reach a plateau during this collecting time period. The nutrient was considered stable when its concentration did not vary significantly during a sampling time period. A period of time was considered stable when every one of the five macronutrients (N, P, K, Ca, Mg) fulfilled this rule. An ANOVA and comparison of means was performed with each of the five nutrients and the ten sampling dates. Thus, the periods in which the foliar levels were stable was determined. The values of N, P, K, Ca, and Mg throughout the almond sampling period are shown together with the multiple range test, which allowed us to determine the ideal sampling date for each variety.
2.2. Diagnosis Method: Range of Normality (RN), DRIS Norms, and CND Guidelines
Range of normality: knowing the adequate periods for foliar sampling, the range of normality was created by only using the foliar analysis corresponding to July (the most ideal month for collecting samples). The ranges were determined using the mean and standard deviation. The “normal” interval was calculated as the arithmetic mean (m) ± standard deviation (s). For the interval “low”, “very low”, “high” and “very high”, the procedure was the same.
2.3. Acquisition of the DRIS Norms
To obtain the DRIS reference guidelines for the macronutrients for the almond cultivars studied, DRIS calculation was carried out using the Excel program as described in the literature [
16,
17]. For each nutrient, the mean, standard deviation, and the coefficient of variability were used in the sampling period in which these were stable (see above). All the calculations for all the possible relationships between nutrient pairs were performed (A/B; B/A and A × B) for all the macro-elements: nitrogen (%), phosphorus (%), potassium (%), calcium (%), and magnesium (%). The following norms were calculated: N/P, P/N, N × P, N/K, K/N, N × K, N/Ca, Ca/N, N × Ca, N/Mg, Mg/N, N × Mg, K/P, P/K, K × P, K/Mg, Mg/K, K × Mg, P/Ca, Ca/P, P × Ca, P/Mg, Mg/P, P × Mg, K/Ca, Ca/K, k × Ca, Ca/Mg, Mg/Ca, and Mg × Ca. The selection of the type of ratio was dependent on the variation coefficient, and those with the smallest value were selected. Calculation for the DRIS norms was based on references [
9,
14,
15].
In these equations, A/B is the tissue nutrient ratio of the plant to be diagnosed, a/b is the optimum value or norm for that given ratio, CV is the coefficient of variation associated with the norm, and SD is the standard deviation of the norm.
2.4. Acquisition of the CND Guidelines
The calculation of the CND guidelines was conducted according to the studies by Parent and Dafir [
8], Aitchison [
18], and Khiari [
19]. A dimensionless parameter named “Sd” was calculated as the sum of all the nutrients (N + P + K + …), plus an “Rd” value, which represents all the values that were not determined analytically. The sum of all the nutrients plus the “
Rd” factor must be 100, so “
Rd” is calculated with the following formula:
The parameter “
G” is calculated, which is the geometric mean of all the nutrients including the
Rd parameter. The formula for calculating
G is:
Once these two parameters are calculated (
Rd and
G), the “
Vx” is calculated for each nutrient by using Equation (3), where “
x” represents each nutrient analyzed:
By definition, the sum of all the components from a plant tissue must be 100%. Therefore, the sum of all the V × x must be zero. After defining the guidelines, the analytical results of a sample can be interpreted with the CND indices ([
20]:
where
Vi is a multinutrient variable in the planting field evaluated,
vi is the reference population average, and
SD is the standard deviation of the reference population.
2.5. Statistical Analysis
There were twelve replicates for each genotype per harvest date. All measured parameters (mean, standard deviation, coefficient of variation, and standard error) as well as the analysis of variance (ANOVA) and the Tukey test for the separation of means (p < 0.05) were analyzed statistically using SPSS Version 20.0 software (SPSS Statistical Package, Chicago, IL, USA).
4. Conclusions
The data of this experiment reported that the optimum period for foliar sampling for the Ferraduel, Ferragnes, and Garrigues cultivars is during July, as in this month, the leaf mineral concentration was becoming stable in all five nutrients analyzed. With respect to the nutritional state for the N, P, K, Ca, and Mg contents, there were differences among varieties. Ferraduel and Ferragnes are very similar varieties; however, they are different with respect to the Garrigues cultivar. The data suggest that Ferraduel and Ferragnes require a higher content of N, P, and Ca than Garrigues. In fact, the DRIS and CND norms for Ferraduel and Ferragnes are very similar between them, but different with regard to Garrigues. These norms could be introduced in the Decision Support Systems (DSS), through their corresponding algorithms, to management of the fertilization in almond orchards, which will increase the effective use of fertilizers, leading to a decrease in the amounts applied. In future assays, we will obtain new norms in function of different agronomic conditions such as water resources used for irrigation, irrigation vs. drought, etc.