Adaptation of Almond Cultivars in Majorca Island: Agronomical, Productive, and Fruit Quality Characteristics

Abstract

Almond cultivation has a long-standing tradition on the island of Majorca, traditionally practiced under rainfed conditions. Currently, new plantations are established with irrigation; however, due to present conditions and the impacts of climate change, water availability is limited. The Government of the Balearic Islands permits a maximum water supply of 3000 m³ ha⁻¹ per year for almond cultivation. In this study, a 6-year field trial was conducted to investigate the adaptation of fourteen almond cultivars obtained from different research centers under deficit irrigation practices in the soil and climatic conditions of Majorca Island. The cultivars had a significant effect on trunk cross-sectional area TCSA, cumulative almond in-shell yield, cumulative kernel yield, yield efficiency, and shelling percentage. The ‘Marta’ and ‘Tarraco’ cultivars exhibited the highest TCSA values. Regarding cumulative almond in-shell yield, ‘Glorieta’ and ‘Constantí’ produced the highest yields, whereas the lowest yields were observed in ‘Mardía’ and ‘Tarraco’. In terms of shelling percentage, ‘Ferragnès’ exhibited the highest value. Kernel quality parameters were cultivar-specific. Additionally, oleic acid content was strongly negatively correlated with linoleic acid content. All cultivars exhibited an unsaturated fatty acid content exceeding 90%, with ‘Masbovera’ showing the highest value.

1. Introduction

The almond tree (Prunus dulcis Miller [D.A. Webb], syn P. amygdalus Batsch) originated in the arid mountainous regions of southwestern and central Asia [1] and rapidly spread to the Eastern Mediterranean, subsequently reaching the Western Mediterranean regions [2,3,4], North America, and finally the Southern Hemisphere, including South America and Australia [5]. According to archaeological and historical evidence, the introduction of almonds to the Eastern Mediterranean occurred during the second millennium BC [6,7,8], with further evidence supporting the existence of an extensive almond trade in the region during the fourth century BC [9]. Almonds are cultivated for their edible kernels, which are used for direct consumption and in various almond-based products and confections [10,11]. Globally, it is one of the most important tree nuts, and currently, it represents the largest production among all commercial tree nut products. In 2022, world production of almonds in shell was 3.63 million metric tons, with the United States being the leading almond-producing country, accounting for 76% of global almond production [12,13]. However, in terms of crop area, the European Union controls 39% of the world almond-tree dedicated area. In Spain, the area harvested was 761,660 ha, with approximately two-thirds of this area primarily concentrated in the Mediterranean region (Catalonia, Valencian Community, Murcia, Andalusia, and Balearic Islands), producing 229,326 metric tons of fruit per year (approximately 62% of Spain’s almond production) [14]. In the Balearic Islands, almond production amounted to 2674 metric tons of fruit per year, with a dedicated cultivation area of 14,253 ha [15].

In recent years, the area dedicated to almond cultivation has decreased significantly. According to the Dry Fruits report [16], this decline in almond farming is associated with several factors, including the rapid urban expansion driven by tourism, the aging of plantations, phytosanitary issues, the vulnerability of early-flowering varieties to frost, and pollination deficiencies. Additionally, low profitability during the period from 2007 to 2011, exacerbated by competition with almonds from California and Australia, which triggered a severe price crisis in the sector, has further contributed to this decline.

Global climate change will result in significant alterations in global agriculture, particularly in Mediterranean regions [17,18,19,20]. The Mediterranean region will be especially susceptible to the impacts of climate change [21]. For the Balearic Islands, climate change scenarios indicate an increase in temperature, droughts, evapotranspiration, and heat waves, along with a decrease in average precipitation [22]. Climate change has become apparent in the Balearic Islands from the latter half of the 20th century to the early 21st century. An abrupt decline of 65 mm in average annual precipitation has been objectively observed since 1980. Additionally, model trends indicate that the decrease could reach 163 mm per century [23]. This fact will result in a decrease in the available water for both human consumption and agriculture. In the Balearic Islands, the government has regulated water usage with the aim of maintaining groundwater bodies in good condition [24].

Traditionally, almond orchards in Mediterranean regions were primarily found in non-irrigated areas, especially in dry regions, due to their capacity to withstand water stress [25,26,27]. However, most new almond plantations are now being developed in traditional irrigated areas and in dry regions that are being converted to irrigation. Nevertheless, the limitations of water availability and the competition from other sectors for water resources in these regions necessitate the implementation of deficit irrigation strategies for new plantations [28]. In Mediterranean regions, the authorities have regulated the maximum water allocations for crops. For example, in almond cultivation, the Hydrographical Confederation of the Guadalquivir River Basin and the Hydrological Plan of the Balearic Islands establish an allocation of 250 mm and 300 mm, respectively [24,29,30]. According to a study conducted by López-López et al. [31] in the southeast of Spain, the estimated annual statewide almond orchard evapotranspiration is 920–1220 mm. Therefore, under semi-arid Mediterranean conditions, full irrigation could exceed 1000 mm [32]. In conditions of water limitations, the water supplied is lower than the irrigation requirements to meet maximum crop evapotranspiration (ET_c). Under these conditions, with irrigation around 300 mm, kernel yields of 1500 kg ha⁻¹ can be obtained [32,33].

This study takes an innovative approach by focusing on Majorca, where almond cultivation has traditionally been rainfed, utilizing local cultivars. However, the region is increasingly adopting irrigation practices and choosing commercial cultivars. The research aims to provide critical insights into the adaptation of these cultivars to the specific climatic conditions of the Mediterranean, offering essential data to identify which cultivars are most suitable for planting in the region. Given that cultivar choice is likely to be a key factor in achieving sustainable production in a Mediterranean environment with diminishing water availability, the objective of this study is to evaluate the adaptation of fourteen commercial cultivars on the island of Majorca under deficit irrigation practices, focusing on yield and almond quality, to ascertain their suitability for future plantations.

2. Materials and Methods

2.1. Site Description

The field experiment was carried out from summer 2014 to autumn 2022 at the experimental plot of Finca Xorrigo (39°34′ N, 2°48′ E), Majorca (Spain). The main soil characteristics are summarized in

Table 1. Physical and chemical properties of the soil (soil depth: 0–40 and 40–80 cm) where the assays were carried out.

Soil Property	Soil Depth
	0–40 cm	40–80 cm
Sand (%)	24.00	28.00
Silt (%)	47.00	42.00
Clay (%)	29.00	30.00
Texture	Loam clayey	Loam clayey
pH H2O	8.21	8.24
Calcium Carbonate Equivalent (%)	18.82	35.53
Active lime (%)	5.68	9.89
Electrical Conductivity (1:5; 25 °C) (dS m−1)	0.23	0.19
Organic Carbon (Walkley & Black) (%)	1.59	0.73
N (total) (%)	0.13	0.07
P (Olsen) (ppm)	13.55	7.88
Exchangeable K (ppm)	238.00	158.00
Exchangeable Na (ppm)	131.00	86.00
Exchangeable Mg (meq 100 g−1)	1.59	1.62
Exchangeable Ca (meq 100 g−1)	2.72	10.61
Cation Exchange Capacity (meq 100 g−1)	15.16	11.24
Exchangeable Sodium Percentage (%)	3.76	3.33

. All determinations were conducted following the official laboratory methods of the Spain Ministry of Agriculture, Fishery and Food [34]. The climate of the region is typical Mediterranean (BSk, according to the Köppen-Geiger climate classification system [35], characterized by hot and dry summers and mild winters, with precipitation concentrated from autumn to spring.

2.2. Plant Material and Experimental Design

Fourteen cultivars were assessed (

Table 2. Genetics, origin, and their main agronomic characteristics of the fourteen cultivars assayed.

Cultivar	Genetics	Origin	Flowering	Self-Compatibility
‘Antoñeta’	‘Ferragnès’ × ‘Tuono’	CEBAS–CSIC 1, Spain	Late	Self-compatible
‘Belona’	‘Blanquerna’ × ‘Belle d’Aurons’	CITA 2, Spain	Late	Self-compatible
‘Constantí’	‘FGFD2’ × Open pollination	IRTA 3, Spain	Late	Self-compatible
‘Ferragnès’	‘Aï’ × ‘Cristomorto’	INRA 4, France	Late	Self-incompatible
‘Glorieta’	‘Primorskiy’ × ‘Cristomorto’	IRTA, Spain	Late	Self-incompatible
‘Lauranne’	‘Ferragnès’ × ‘Tuono’	INRA, France	Late	Self-compatible
‘Mardía’	‘Felisa’ × ‘Bertina’	CITA, Spain	Extra-late	Self-compatible
‘Marinada’	‘Lauranne’ × ‘Glorieta’	IRTA, Spain	Very late	Self-compatible
‘Marta’	‘Ferragnès’ × ‘Tuono’	CEBAS-CSIC, Spain	Very late	Self-compatible
‘Masbovera’	‘Primorskiy’ × ‘Cristomorto’	IRTA, Spain	Late	Self-incompatible
‘Penta’	‘SS 133’ × ‘Lauranne’	CEBAS-CSIC, Spain	Extra-late	Self-compatible
‘Soleta’	‘Blanquerna’ × ‘Belle d’Aurons’	CITA, Spain	Late	Self-compatible
‘Tarraco’	‘FLTU18’ × ‘Anxaneta’	IRTA, Spain	Very late	Self-incompatible
‘Vairo’	‘4-665’ × ‘Lauranne’	IRTA, Spain	Late	Self-compatible

¹ CEBAS-CSIC: Centre for Applied Soil Science and Biology of the Segura-Spanish National Research Council; ² CITA: Agri-Food Research and Technology Centre of Aragon; ³ IRTA: Institute of Agrifood Research and Technology; ⁴ INRA: National Institute for Agricultural Research.

). Twelve cultivars were sourced from three distinct Spanish breeding programs: ‘Antoñeta’, ‘Marta’, and ‘Penta’ from CEBAS–CSIC (Centre for Applied Soil Science and Biology of the Segura–Spanish National Research Council) [36,37,38]; ‘Belona’, ‘Mardía’, and ‘Soleta’ from CITA (Agri-Food Research and Technology Centre of Aragon) [39,40,41,42]; and ‘Constantí’, ‘Glorieta’, ‘Marinada’, ‘Masbovera’, ‘Tarraco’, and ‘Vairo’ from IRTA (Institute of Agrifood Research and Technology) [43,44,45]. Two cultivars were acquired from INRA (National Institute for Agricultural Research), France: ‘Ferragnès’ and ‘Lauranne’ [46,47]. All trees were grafted onto INRA GF-677 rootstock [48,49]. These cultivars were compared in a completely randomized design (CRD) with three replicate plots and four trees per plot. Trees were planted in 2014 with a spacing of 7 m × 6 m (238 trees ha⁻¹). Sample collection began during the fruiting period in 2017.

2.3. Cultural Practices

The crop was managed following the standard practices commonly disseminated in the area, adhering to typical commercial methods. All the trees were trained using an open vase system. Drip lines were utilized, and the regulated deficit irrigation (RDI) strategy was implemented using the monthly k_c values proposed by Girona [50] and the monthly reduction coefficients of ET_c from Salvatierra & Gómez [51]. This strategy involved covering 60% ET_c, with an average seasonal irrigation volume of 3000 m³ ha⁻¹ year, in addition to rainfall. Until the trees reached maturity, only the necessary amount of water was applied. Fertilization was calculated using a nutrient balance, considering the extractions proposed by Alonso et al. [52]. Fertilizers were applied through fertigation. Herbicide was applied under the trees, and vegetation cover was mowed between rows. Pest and disease control was carried out based on the onset of disease and pest symptoms, pest treatment thresholds, and a treatment calendar adjusted according to weather conditions.

2.4. Meteorological Data

Meteorological parameters (monthly maximum, minimum, and mean temperature, as well as rainfall) were recorded by the meteorological station of the Irrigation Agro-Climate Information System (SiAR). Data were collected monthly throughout the entire duration of the assay.

2.5. Vigor

Tree vigor was measured at the end of each growing season and evaluated based on the trunk cross-sectional area (TCSA; cm²). The four trees from each experimental plot of each cultivar were measured. Trunk diameter (TD; cm) was measured at 20 cm above the graft union. TCSA was calculated as:

$$TCSA={\displaystyle \frac{\pi \times {TD}^2}{4}}$$

(1)

2.6. Yield and Yield Efficiency

Almond fruits were harvested at the stage of maturity, determined when the mesocarp began the drying process, specifically at stage 87 of the BBCH (Biologische Bundesanstalt Bundessortenamt and Chemische Industrie) scale [53], as described by Bleiholder et al. [54], Agustí [55] and Sakar et al. [56]. Fruits from each tree were collected by mechanical shaking using commercial equipment, and the hulls were removed with a self-moving huller. The in-shell almond production (Y) from each plot was individually weighed. The results were expressed as the mean of each cultivar, both in kg tree⁻¹ and in kg ha⁻¹. The latter was estimated considering the planting densities indicated in Section 2.2. Cumulative yield (Y_C; kg tree⁻¹ and kg ha⁻¹) was calculated for the years 2017 to 2022 by summing the annual productions of each cultivar.

A homogeneous and representative sample of 1.00 kg of in-shell almonds (after dehulling) was selected for each cultivar and naturally dried for approximately three weeks (until reaching 6% kernel moisture). Subsequently, 50 individual almonds per cultivar were randomly selected to determine the almond weight (g). Dry weight was determined, followed by the measurement of shell and kernel weight, as well as the shelling percentage (SP; %). Additionally, the percentage of double kernels (%) per 1.00 kg sample was also assessed. The measurements were carried out using a digital balance with a sensitivity of 0.01 g. The percentage of kernel (%) was calculated as follows:

$$PK={\displaystyle \frac{Kernelweight}{In-shellalmondweight}}\times 100$$

(2)

Kernel yield was computed by multiplying the in-shell almond yield (kg tree⁻¹) by the SP (%). The theoretical kernel yield per hectare was calculated by multiplying the kernel yield per tree by the tree density indicated in Section 2.2. Cumulative kernel yield (kg tree⁻¹ and kg ha⁻¹) was measured for the years 2017 to 2022 by summing the productions of each year for each cultivar.

Cumulative kernel yield (g tree⁻¹) and cross-sectional area increase (cm²) were utilized to calculate the yield efficiency (YE; kg cm⁻²) for the years 2017 to 2022 [57]. YE was obtained using the following formula:

$$YE={\displaystyle \frac{Cumulativekernelyield}{TCSAincrease}}$$

(3)

2.7. Kernel Quality Parameters

2.7.1. Sampling

Physicochemical analyses were conducted during three consecutive harvests (2020, 2021, and 2022). For the physical parameters, 50 individual kernels per cultivar and year were randomly selected. For nutritional composition and fatty acids, a ground sample of 500 g of kernels was provided per cultivar and year (replication) and analyzed for each harvest.

2.7.2. Morphological Parameters and Weight

The main dimensions of the kernel (length [L], width [W], and thickness [T]) were measured using an electronic digital caliper with a sensitivity of 0.01 mm. The kernel weight was measured using a digital balance with a sensitivity of 0.01 g.

2.7.3. Moisture Contents

The moisture content (method no. 934.06) of all samples was determined according to the AOAC official methods [58].

2.7.4. Nutritional Composition

The nutritional analyses were carried out by the IBIB Biotechnological Institute, Palma, 07198, Spain. Carbohydrates and sugars were measured by UV absorption spectrophotometry, and the fatty acid profile (FAP) was determined by gas chromatography (GC) with a flame ionization detector (FID). The Kjeldahl method (N × 6.25) [59] was used to determine the protein content.

2.8. Statistical Analysis

When the data exhibited a normal distribution and homogeneity of variance, a factorial ANOVA was conducted to determine significant differences among cultivars. Models including cultivar and year, and the interactions between them as a fixed factor, were built to separate treatment effects for almonds in shell yield and kernel yield. Models including cultivar as a fixed factor were built to separate treatments for TCSA increase, cumulative almond in shell yield, shelling percentage, and cumulative kernel yield. Mixed models, with cultivar as a fixed factor and year as a random factor, were built to separate treatment effects for in-shell weight, kernel weight, kernel length, kernel width, and kernel thickness. Models including cultivar as a fixed factor were built to separate treatments for energy, fatty matter, total carbohydrates, sugars, protein, dietary fiber, ash, dray matter, saturated fatty acids, monounsaturated fatty acids, polyunsaturated fatty acids, trans fatty acids, linoleic acid, oleic acid, stearic acid, margaroleic acid, margaric acid, palmitoleic acid, palmitic acid, and myristic acid. The Duncan test was performed to evaluate the statistical differences between means (p ≤ 0.05) indicated by the ANOVA analysis. A principal component analysis (PCA) was carried out to identify the primary variations in chemical composition within the cultivars. In addition, hierarchical cluster analysis (HCA) was used to investigate the similarities and dissimilarities among the cultivars. Finally, correlations between the chemical parameters of kernels were evaluated using Pearson’s correlation coefficient. All data analysis were carried out using the IBM SPSS Statistics 23.0 [60].

3. Results

3.1. Meteorological Data

Average temperature and rainfall values throughout the entire experimental period reflect the typical climate of the Mediterranean region. The mean temperature for the assayed period was 17.6 °C, and the mean annual rainfall was 375 mm. The highest mean monthly temperature was recorded in August 2022, with a value of 29.1 °C, and the most intense rainfall episode occurred in November 2021, with an accumulated rainfall of 174 mm (Figure 1).

3.2. Vigor

The statistical analysis of the TCSA increase highlighted the significant effect of the cultivar. Among the almond cultivars, ‘Marta’ and ‘Tarraco’ appeared to be the most vigorous, exhibiting the highest TCSA values, each exceeding 440 cm². In contrast, ‘Belona’, ‘Penta’, ‘Mardia’, and ‘Soleta’ displayed the lowest TCSA values, around 250 cm² (

Table 3. Increase in trunk cross-sectional area, cumulative almond shell yield, and cumulative kernel yield of the different cultivars assayed throughout the whole experiment.

Cultivar	TCSA Increase 1 (cm2)	CAY 2 (kg tree−1)	CKY 3 (kg tree−1)	YE 4 (kg cm−2)
‘Antoñeta’	299.29 ± 13.87 b, 5	51.96 ± 2.95 d, 6	17.88 ± 1.03 de, 6	6.00 × 10−2 ± 4.90 × 10−3 cd, 6
‘Belona’	247.28 ± 13.87 a	40.20 ± 2.95 bc	11.53 ± 1.03 ab	4.70 × 10−2 ± 4.90 × 10−3 bc
‘Constantí’	314.95 ± 13.87 b	80.37 ± 2.95 f	22.44 ± 1.03 e	6.50 × 10−2 ± 4.90 × 10−3 d
‘Ferragnès’	364.11 ± 13.87 c	48.42 ± 2.95 cd	18.35 ± 1.03 de	5.00 × 10−2 ± 4.90 × 10−3 bcd
‘Glorieta’	303.69 ± 13.87 b	81.80 ± 2.95 f	24.97 ± 1.03 f	8.30 × 10−2 ± 4.90 × 10−3 e
‘Lauranne’	304.96 ± 13.87 b	55.55 ± 2.95 d	18.09 ± 1.03 de	5.90 × 10−2 ± 4.90 × 10−3 cd
‘Mardía’	241.14 ± 13.87 a	30.20 ± 2.95 a	8.83 ± 1.03 a	3.90 × 10−2 ± 4.90 × 10−3 b
‘Marinada’	322.02 ± 13.87 b	54.85 ± 2.95 d	18.74 ± 1.03 de	5.80 × 10−2 ± 4.90 × 10−3 cd
‘Marta’	456.57 ± 13.87 d	53.30 ± 2.95 d	16.14 ± 1.03 cd	3.60 × 10−2 ± 4.90 × 10−3 b
‘Masbovera’	385.72 ± 13.87 c	65.21 ± 2.95 e	18.83 ± 1.03 de	4.90 × 10−2 ± 4.90 × 10−3 bcd
‘Penta’	242.67 ± 13.87 a	49.28 ± 2.95 d	14.15 ± 1.03 bc	5.80 × 10−2 ± 4.90 × 10−3 cd
‘Soleta’	250.88 ± 13.87 a	36.65 ± 2.95 ab	11.49 ± 1.03 ab	4.60 × 10−2 ± 4.90 × 10−3 bc
‘Tarraco’	444.47 ± 13.87 d	30.67 ± 2.95 a	9.44 ± 1.03 a	2.10 × 10−2 ± 4.90 × 10−3 a
‘Vairo’	308.29 ± 13.87 b	66.21 ± 2.95 e	18.13 ± 1.03 de	5.90 × 10−2 ± 4.90 × 10−3 cd
ANOVA	***	***	***	***

¹ TCSA increase: trunk cross-sectional area increase; ² CAY: cumulative almond in shell yield; ³ CKY: cumulative kernel yield; ⁴ YE: yield efficiency; ⁵ Mean ± standard error of 12 replicates. ⁶ Mean ± standard error of 3 replicates. (***) denote p ≤ 0.001. Different letters indicate significant differences between cultivars (Duncan test, p ≤ 0.05).

). The ‘Ferragnès’ cultivar exhibited high vigor, with a TCSA increase of 364 cm², similar to that of ‘Masbovera’, which had a TCSA increase of 386 cm². No significant differences were observed between these two cultivars. ‘Lauranne’, the other reference cultivar, exhibited moderate vigor, similar to ‘Constantí, ‘Glorieta’, ‘Marinada’, and ‘Vairo’, with a TCSA increase of less than 325 cm², and no significant differences were observed among them.

3.3. Yield Production

Both the cultivar and the growing season had a significant effect on kernel yield. It is important to note that the statistical analysis indicated an interaction between cultivar and growing season, highlighting the relevance of both factors, particularly the growing season, when analyzing and interpreting the results. This interaction is clearly illustrated in Figure 2, where cultivars exhibit differential behaviors across the different growing seasons analyzed. Furthermore, the alternate bearing behavior exhibited by some cultivars is also evident.

The statistical analysis also indicated significant differences among cultivars in both cumulative almond in-shell yield and cumulative kernel yield (Table 3). When analyzing cumulative kernel yield, ‘Glorieta’ emerged as the best performer, with approximately 25 kg per tree. The production of the two reference cultivars, ‘Ferragnès’ and ‘Lauranne’, was lower, with a cumulative yield of approximately 18 kg per tree. However, it is important to note that, despite the lower yields of the reference cultivars, no significant difference was observed between them and the cultivars ‘Antoñeta’, ‘Marinada’, ‘Masbovera’, and ‘Vairo’. Thus, these cultivars can be considered to have rendered similar results. ‘Mardia’ and Tarraco’ were the worst performers, with a cumulative kernel yield of less than 10 kg per tree. In terms of yield efficiency, ‘Glorieta’ showed the highest value (8.30 × 10⁻² kg cm⁻²), while ‘Tarraco’ exhibited the lowest value (2.10 × 10⁻² kg cm⁻²).

As for shelling percentage, the cultivar also had a significant effect. ‘Ferragnès’ was the cultivar that showed the highest shelling percentage, at 38%, but it did not significantly differ from the ‘Antoñeta’ and ‘Marinada’ cultivars (

Table 4. Shelling percentage per cultivar.

Cultivar	Shelling Percentage (%)
‘Antoñeta’	36.09 ± 1.42 ef,1
‘Belona’	28.35 ± 1.42 bcd
‘Constantí’	25.42 ± 1.42 ab
‘Ferragnès’	38.25 ± 1.42 f
‘Glorieta’	31.28 ± 1.42 cd
‘Lauranne’	32.79 ± 1.42 de
‘Mardía’	21.91 ± 1.42 a
‘Marinada’	35.83 ± 1.42 ef
‘Marta’	29.87 ± 1.42 bcd
‘Masbovera’	29.27 ± 1.42 bcd
‘Penta’	27.92 ± 1.42 bc
‘Soleta’	29.27 ± 1.42 cde
‘Tarraco’	31.92 ± 1.42 de
‘Vairo’	27.86 ± 1.42 bc
ANOVA	***

¹ Mean ± standard error of 6 replicates. (***) denote p ≤ 0.001. Different letters indicate significant differences between cultivars (Duncan test, p ≤ 0.05).

). The lowest percentage was obtained by ‘Mardia’.

3.4. Kernel Quality Parameters

Table 5. Mean value of weight of almond in shell, kernel weight, length, width, and thickness of kernels evaluated.

Cultivar	Weight in Shell (g)	Kernel Weight (g)	Kernel Dimensions (mm)
			L 1	W 2	T 3	W/L Ratio
‘Antoñeta’	4.10 ± 0.17 bc, 4	1.67 ± 0.02 f	25.44 ± 0.14 de	17.08 ± 0.10 i	8.65 ± 0.07 d	0.67 ± 0.01 h
‘Belona’	6.43 ± 0.17 h	1.73 ± 0.02 g	25.48 ± 0.14 de	17.59 ± 0.10 j	8.40 ± 0.07 c	0.69 ± 0.01 i
‘Constantí’	4.38 ± 0.17 cd	1.23 ± 0.02 b	22.31 ± 0.14 b	14.98 ± 0.10 d	8.61 ± 0.07 cd	0.67 ± 0.01 h
‘Ferragnès’	5.07 ± 0.17 ef	1.84 ± 0.02 h	28.83 ± 0.14 h	15.30 ± 0.10 e	8.99 ± 0.07 e	0.53 ± 0.01 bc
‘Glorieta’	5.25 ± 0.17 fg	1.51 ± 0.02 e	26.83 ± 0.14 g	15.60 ±. 0.10 f	8.14 ± 0.07 b	0.58 ± 0.01 e
‘Lauranne’	3.71 ± 0.17 b	1.41 ± 0.02 c	26.15 ± 0.14 f	15.00 ± 0.10 d	8.07 ± 0.07 b	0.57 ± 0.01 e
‘Mardía’	5.65 ± 0.17 g	1.53 ± 0.026 e	25.11 ± 0.206 d	16.15 ± 0.10 g	8.61 ± 0.07 cd	0.65 ± 0.01 g
‘Marinada’	4.69 ± 0.17 de	1.42 ± 0,02 c	24.64 ± 0.14 c	14.95 ± 0.10 d	9.36 ± 0.07 f	0.61 ± 0.01 f
‘Marta’	3.77 ± 0.17 b	1.48 ± 0.02 de	26.77 ± 0.14 g	14.02 ± 0.10 b	8.47 ± 0.07 cd	0.52 ± 0.01 ab
‘Masbovera’	4.29 ± 0.17 cd	1.48 ± 0.02 de	26.99 ± 0.14 g	14.53 ± 0.10 c	8.18 ± 0.07 b	0.54 ± 0.01 c
‘Penta’	2.82 ± 0.17 a	0.97 ± 0.02 a	20.74 ± 0.14 a	11.62 ± 0.10 a	8.46 ± 0.07 cd	0.56 ± 0.01 d
‘Soleta’	4.74 ± 0.17 def	1.43 ± 0.02 cd	27.01 ± 0.14 g	13.84 ± 0.10 b	7.80 ± 0.07 a	0.51 ± 0.01 a
‘Tarraco’	6.65 ± 0.17 h	1.93 ± 0.026 i	30.01 ± 0.200 i	16.66 ± 0.01 h	8.49 ± 0.07 cd	0.56 ± 0.01 d
‘Vairo’	4.67 ± 0.17 de	1.39 ± 0.02 c	25.71 ± 0.14 e	15.81 ± 010 f	8.18 ± 0.07 b	0.62 ± 0.01 f
ANOVA	***	***	***	***	***	***

¹ L: length; ² W: width; ³ T: thickness; ⁴ Mean ± standard error of 150 replicates. (***) denote p ≤ 0.001. Different letters indicate significant differences between cultivars (Duncan test, p ≤ 0.05).

presents the mean values for in-shell almond weight, kernel weight, and kernel size collected per cultivar. Both cultivar and growing cycle had a significant effect on all parameters examined, except for kernel length, where the growing cycle did not exhibit a significant effect. Moreover, the statistical analysis indicated an interaction between cultivars and the growing cycle. Almond size indices were characterized by considerable differences between cultivars; in particular, kernel weight ranged from 2.82 g (‘Penta’) to 6.65 g (‘Tarraco’). The same cultivar, ‘Tarraco’, also had the longest kernels (30.01 mm). The cultivars under investigation were characterized by different kernel shapes, varying from round (‘Belona’, with a width/length ratio of 0.69) to elongated (‘Soleta’, with a width/length ratio of 0.51).

The results of the nutritional analysis of almond kernels are presented in

Table 6. Chemical composition and energy value of almonds examined at three different harvest times (2020–2022). The results are expressed per cultivar.

Cultivar	E 1	FM 2	TC 3	S 4	P 5	DF 6	Salt	A 7	DM 8	FM/P 9
	(kJ 100 g−1)	(g 100 g−1)	(g 100 g−1)	(g 100 g−1)	(g 100 g−1)	(g 100 g−1)	(g 100 g−1)	(g 100 g−1)	(g 100 g−1)
‘Antoñeta’	606.00 ± 7.34 abc, 10	52.43 ± 0.99 abcd	9.63 ± 0.58 a	6.33 ± 0.80 a	23.93 ± 0.68 abcde	11.73 ± 1.35 a	<0.5	3.37 ± 0.13 a	95.95 ± 0.21 a	2.21 ± 0.10 abcde
‘Belona’	627.00 ± 7.34 c	55.90 ± 0.99 d	9.05 ± 0.58 a	5.15 ± 0.80 a	22.05 ± 0.68 ab	11.20 ± 1.35 a	<0.5	3.13 ± 0.13 a	95.92 ± 0.21 a	2.54 ± 0.10 de
‘Constantí’	599.33 ± 7.34 ab	51.33 ± 0.99 a	8.77 ± 0.58 a	4.90 ± 0.80 a	25.63 ± 0.68 de	10.53 ± 1.35 a	<0.5	3.66 ± 0.13 a	95.84 ± 0.21 a	2.02 ± 0.10 ab
‘Ferragnès’	621.67 ± 7.34 bc	55.40 ± 0.99 bcd	7.80 ± 0.58 a	4.27 ± 0.80 a	22.83 ± 0.68 abcd	10.43 ± 1.35 a	<0.5	3.46 ± 0.13 a	96.00 ± 0.21 a	2.43 ± 0.10 de
‘Glorieta’	591.67 ± 7.34 a	51.90 ± 0.99 ab	8.67 ± 0.58 a	5.20 ± 0.80 a	22.50 ± 0.68 abc	11.20 ± 1.35 a	<0.5	3.61 ± 0.13 a	95.69 ± 0.21 a	2.32 ± 0.10 bcde
‘Lauranne’	612.67 ± 7.34 abc	53.50 ± 0.99 abcd	9.57 ± 0.58 a	5.03 ± 0.80 a	23.17 ± 0.68 abcd	10.97 ± 1.35 a	<0.5	3.51 ± 0.13 a	96.00 ± 0.21 a	2.31 ± 0.10 bcde
‘Mardía’	605.00 ± 7.34 abc	52.05 ± 0.99 abc	9.25 ± 0.58 a	4.95 ± 0.80 a	25.05 ± 0.68 de	9.85 ± 1.35 a	<0.5	3.07 ± 0.13 a	95.61 ± 0.21 a	2.08 ± 0.10 abc
‘Marinada’	599.00 ± 7.34 ab	53.17 ± 0.99 abcd	7.23 ± 0.58 a	4.87 ± 0.80 a	22.93 ± 0.68 abcd	11.43 ± 1.35 a	<0.5	3.41 ± 0.13 a	95.97 ± 0.21 a	2.32 ± 0.10 bcde
‘Marta’	628.33 ± 7.34 c	55.50 ± 0.99 cd	8.73 ± 0.58 a	5.40 ± 0.80 a	23.43 ± 0.68 abcde	11.00 ± 1.35 a	<0.5	3.38 ± 0.13 a	96.08 ± 0.21 a	2.37 ± 0.10 cde
‘Masbovera’	608.67 ± 7.34 abc	53.83 ± 0.99 abcd	8.20 ± 0.58 a	4.93 ± 0.80 a	22.83 ± 0.68 abcd	11.23 ± 1.35 a	<0.5	3.44 ± 0.13 a	96.25 ± 0.21 a	2.36 ± 0.10 cde
‘Penta’	620.67 ± 7.34 bc	55.07 ± 0.99 bcd	9.53 ± 0.58 a	5.83 ± 0.80 a	21.70 ± 0.68 a	12.37 ± 1.35 a	<0.5	3.13 ± 0.13 a	95.77 ± 0.21 a	2.54 ± 0.10 e
‘Soleta’	611.33 ± 7.34 abc	53.73 ± 0.99 abcd	7.43 ± 0.58 a	4.57 ± 0.80 a	24.50 ± 0.68 cde	11.33 ± 1.35 a	<0.5	3.05 ± 0.13 a	96.07 ± 0.21 a	2.19 ± 0.10 abcd
‘Tarraco’	602.50 ± 7.34 abc	51.45 ± 0.99 a	8.45 ± 0.58 a	5.65 ± 0.80 a	26.50 ± 0.68 e	11.90 ± 1.35 a	<0.5	3.42 ± 0.13 a	95.38 ± 0.21 a	1.94 ± 0.10 a
‘Vairo’	628.67 ± 7.34 c	55.37 ± 0.99 bcd	8.43 ± 0.58 a	4.43 ± 0.80 a	24.20 ± 0.68 bcde	10.07 ± 1.35 a	<0.5	3.11 ± 0.13 a	95.99 ± 0.21 a	2.29 ± 0.10 bcde
ANOVA	*	*	NS	NS	*	NS	-	NS	NS	*

¹ E: energy; ² FM: fatty matter; ³ TC: total carbohydrate; ⁴ S: sugars; ⁵ P: protein; ⁶ DF: dietary fiber; ⁷ A: ash; ⁸ DM: dry matter. ⁹ FM/P: fatty matter/protein ratio. ¹⁰ Mean ± standard error of 3 replicates. NS and (*), denote not significant and p ≤ 0.05, respectively. Different letters indicate significant differences between cultivars (Duncan test, p ≤ 0.05).

. The statistical analysis indicated significant differences among cultivars in energy, fat content, and protein content. ‘Vairo’, ‘Marta’, and ‘Belona’ exhibited the highest energy values, while ‘Marinada’, ‘Constantí’, and ‘Tarraco’ presented the lowest values. The Duncan post hoc test revealed significant differences between ‘Belona’ and ‘Tarraco’, as well as between ‘Marta’ and ‘Tarraco.’ In the case of fatty matter, the average value ranged from 51.33 g 100 g⁻¹ in ‘Constantí’ to 55.90 g 100 g⁻¹ in ‘Belona’. On the other hand, the total carbohydrate content was highest in ‘Antoñeta’ (9.63 g 100 g⁻¹), followed by ‘Lauranne’ and ‘Penta,’ with values of 9.57 g 100 g⁻¹ and 9.53 g 100 g⁻¹, respectively. Regarding sugar content, the results of ANOVA did not indicate significant differences between cultivars. ‘Antoñeta’ presented the highest value, while ‘Ferragnès’ had the lowest value, with values of 6.33 and 4.27 g 100 g⁻¹, respectively. In terms of protein content, the range varied from 26.50 (‘Tarraco’) to 21.70 g 100 g⁻¹ (‘Penta’). The Duncan post hoc test indicated significant differences between ‘Tarraco’ and ‘Penta,’ as well as between ‘Belona’, ‘Soleta’, and ‘Tarraco’. The dietary fiber contents of the samples ranged between 9.85 g 100 g⁻¹ (‘Vairo’) and 12.37 g 100 g⁻¹ (‘Penta’). Finally, the samples showed only a small variation in ash content, with values ranging from 3.05 g 100 g⁻¹ (‘Soleta’) to 3.66 g 100 g⁻¹ (‘Constantí’).

Table 7. Fatty acid composition of kernels collected at three different harvest times (2020–2022). The results are expressed per cultivar.

Cultivar	SFA 1	MUFA 2	PUFA 3	TFA 4	UFA:SFA 5
	(%)	(%)	(%)	(%)
‘Antoñeta’	8.91 ± 0.22 g,6	72.20 ± 1.15 a	18.89 ± 1.04 cd	<0.01	10.24 ± 0.32 a
‘Belona’	7.42 ± 0.22 abc	75.32 ± 1.15 abcd	17.26 ± 1.04 bcd	<0.01	12.49 ± 0.32 ef
‘Constantí’	7.99 ± 0.22 bcde	78.47 ± 1.15 de	13.54 ± 1.04 a	<0.01	11.53 ± 0.32 cde
‘Ferragnès’	8.24 ± 0.22 defg	77.50 ± 1.15 bcde	14.26 ± 1.04 ab	<0.01	11.13 ± 0.32 abcd
‘Glorieta’	7.32 ± 0.22 ab	77.36 ± 1.15 bcde	15.32 ± 1.04 abc	<0.01	12.66 ± 0.32 f
‘Lauranne’	8.16 ± 0.22 cdefg	74.54 ± 1.15 abcd	17.30 ± 1.04 bcd	<0.01	11.27 ± 0.32 abcd
‘Mardía’	8.16 ± 0.22 cdefg	77.73 ± 1.15 cde	14.11 ± 1.04 ab	<0.01	11.26 ± 0.32 abcd
‘Marinada’	8.08 ± 0.22 cdef	73.66 ± 1.15 ab	18.25 ± 1.04 cd	<0.01	11.40 ± 0.32 bcd
‘Marta’	7.58 ± 0.22 abcd	78.06 ± 1.15 de	14.37 ± 1.04 ab	<0.01	12.20 ± 0.32 def
‘Masbovera’	7.24 ± 0.22 a	79.77 ± 1.15 e	13.00 ± 1.04 a	<0.01	12.81± 0.32 f
‘Penta’	8.83 ± 0.22 fg	74.94 ± 1.15 abcd	16.23 ± 1.04 abc	<0.01	10.36 ± 0.32 ab
‘Soleta’	8.50 ± 0.22 efg	71.59 ± 1.15 a	19.91 ± 1.04 d	<0.01	10.81 ± 0.32 abc
‘Tarraco’	8.85 ± 0.22 fg	72.48 ± 1.15 a	18.67 ± 1.04 cd	<0.01	10.33 ± 0.32 ab
‘Vairo’	8.43 ± 0.22 efg	73.94 ± 1.15 abc	17.636 ± 1.04 bcd	<0.01	10.88 ± 0.32 abc
ANOVA	***	***	***	-	***

¹ SFA: saturated fatty acids; ² MUFA: monounsaturated fatty acids; ³ PUFA: polyunsaturated fatty acids; ⁴ TFA: trans fatty acids; ⁵ unsaturated fatty acids/saturated fatty acids; ⁶ Mean ± standard error of 3 replicates. (***) denote p < 0.001. Different letters indicate significant differences between cultivars (Duncan test, p < 0.05).

presents the results of the nutritional analysis of the kernels. The statistical analysis indicated significant differences among cultivars in saturated fatty acids, monounsaturated fatty acids, polyunsaturated fatty acids, and the unsaturated fatty acids/saturated fatty acids ratio. Saturated fatty acids ranged from 7.24% (‘Masbovera’) to 8.91% (‘Antoñeta’). ‘Masbovera’ was the cultivar with the highest content of monounsaturated fatty acids, while ‘Soleta’ had the highest content of polyunsaturated fatty acids. Finally, ‘Glorieta’ (12.66) and ‘Masbovera’ (12.81) were the cultivars with the highest unsaturated fatty acid/saturated fatty acid ratio, whereas the ‘Antoñeta’ cultivar recorded the lowest.

The fatty acid compositions of the experimental cultivars are illustrated in Figure 3, expressed as a percentage of total fatty acids. The study revealed substantial variations among the cultivars, with significant effects on all the fatty acids determined. In all cultivars, the major fatty acids identified were oleic and linoleic acids, followed by palmitic acid. These three fatty acids dominated the lipid composition. The ranges for these three fatty acids were 37.10% (‘Tarraco’) to 42.82% (‘Marta’) for oleic acid, 6.40% (‘Constantí) to 9.46% (‘Soleta’) for linoleic acid, and 2.70% (‘Belona’) to 3.68% (‘Marinada’) for palmitic acid.

A principal component analysis (PCA) was performed to determine the relative importance of the different chemical parameters and the variability within the studied cultivars. The PCA was conducted using the collected data (nutritional analysis and the principal fatty acids). The analysis revealed that the two main components explained 46.56% of the variability observed among the studied cultivars (Figure 4). PC1 (28.99%), shown in red, was positively and significantly related to fatty matter, energy, stearic acid, and linoleic acid, while being negatively and significantly related to palmitoleic acid and salt. PC2 (17.57%), shown in green, was mostly explained by margaroleic acid, oleic acid, and total carbohydrate.

Three groups were differentiated according to these parameters (Figure 5). ‘Glorieta’, ‘Lauranne’, and ‘Mardía’ formed the first group. The second group included ‘Marinada’, ‘Masbovera’, ‘Tarraco’, ‘Constantí’, and ‘Penta’. Finally, the third group comprised two subgroups: one consisting of ‘Ferragnès’ and ‘Soleta’, and the other including ‘Belona’, ‘Marta’, ‘Antoñeta’, and ‘Vairo’.

4. Discussion

In the Mediterranean region, water resources are limited. This, combined with climate change scenarios, reduces water availability for plants [23,61]. Recently, there has been growing interest in almond cultivation under irrigation, primarily due to the limited profitability associated with rainfed almond farming [33]. According to the Hydrological Plan of the Balearic Islands, almond orchards are permitted a maximum irrigation volume of 3000 m³ ha⁻¹ year [24]. Under these conditions, the study of almond cultivars under deficit irrigation becomes increasingly important. There is a growing interest in understanding which cultivars are better adapted to Majorca under these conditions. The study presents a comparison of 14 almond cultivars to determine which cultivars are best suited to these conditions in terms of agronomic parameters and almond quality.

The vigor of the trees was significantly influenced by the cultivar. In our study, the ‘Marta’ and ‘Tarraco’ cultivars exhibited the highest vigor. For the ‘Marta’ cultivar, Egea et al. [36] obtained similar results, while ‘Tarraco’ had previously been described as a cultivar of moderate vigor [44]. ‘Ferragnès’ and ‘Masbovera’ also exhibited strong vigor. Vargas et al. [44] obtained similar results for ‘Ferragnès’; however, they classified the ‘Masbovera’ cultivar as very vigorous. The ‘Antoñeta’, ‘Constantí’, ‘Glorieta’, ‘Lauranne’, ‘Marinada’, and ‘Vairo’ cultivars showed moderate vigor. According to Socias i company et al. [62], adult trees of ‘Mardía’ show intermediate vigor, but our results classify this cultivar, along with ‘Belona’, ‘Penta’, and ‘Soleta’, as having low vigor. However, these cultivars have previously been described as having intermediate vigor [37,41,42,63]. This decrease in vigor could be attributed to deficit irrigation, a factor that has been demonstrated for other cultivars by different authors [64,65].

The cultivars ‘Ferragnès’ and ‘Glorieta’ exhibited early bearing precocity in production, with similar results reported by Malagón et al. [66]. ‘Lauranne’ and ‘Vairo’ also demonstrated very early production, with an accumulated kernel yield in the first three productive years exceeding 5 kg tree⁻¹. In our results, it can be observed that the ‘Antoñeta’ and ‘Marta’ cultivars exhibited alternate bearing behavior. A special mention should be made of ‘Glorieta’, the cultivar that showed the highest cumulative kernel yield, approximately 25 kg tree⁻¹. Additionally, the ‘Constantí’ cultivar also had an accumulated kernel production of over 20 kg tree⁻¹. Similar results were obtained for the ‘Glorieta’ cultivar under similar conditions by Miarnau et al. [67]. The yield efficiency results obtained in this study, an important factor in modern agriculture, were similar to those reported by Maldera et al. [68] and Alonso et al. [63] for the ‘Belona’, ‘Lauranne’, ‘Mardía’, and ‘Soleta’ cultivars. However, the lowest yield efficiency value exhibited by ‘Tarraco’ is attributed to its strong vigor and moderate production.

In California, the annual potential almond evapotranspiration (ET_c) was reported as 900–1350 mm [69,70], with an average yield of shelled almonds of 2480 kg ha⁻¹ [13]. In the southeast of Spain, the ET_c is similar, ranging from 920 to 1220 [31]. Under these conditions, with irrigation around 300 mm, a kernel yield of 1500 kg ha⁻¹ can be obtained [31,33]. In the last growing season, considering a tree density of 238 trees ha⁻¹ (7 m × 6 m), ‘Masbovera’ and ‘Glorieta’ yielded almost 2000 kg ha⁻¹ of kernels, specifically 1824 and 1934 kg ha⁻¹, respectively. ‘Marinada’, ‘Marta’, and ‘Lauranne’ also exceeded a kernel yield of 1500 kg ha⁻¹. The results of the ‘Marta’ cultivar were consistent with those of Miarnau et al. [67], who suggested that under deficit irrigation it could achieve a production of about 1850 kg ha⁻¹, and with García-Tejero et al. [71] who obtained similar yields between the ‘Marta’ and ‘Lauranne’ cultivars. The ‘Marinada’ cultivar, characterized by moderate vigor, can be planted with a spacing of 6 m × 6 m (278 trees ha⁻¹) and could achieve a kernel yield of 1794 kg ha⁻¹, a yield similar to that of the ‘Masbovera’ cultivar.

The shelling percentage is strictly related to the cultivar [68]. A high shelling percentage contributes to improving crop input transformation [72]. Most of the cultivars showed mean values consistent with other studies [36,37,44,63,73,74,75,76,77,78]. The ‘Antoñeta’, ‘Marinada’, and ‘Masbovera’ cultivars had a slightly higher percentage than those reported by Egea et al. [75], Lordan et al. [74], and Vargas et al. [44]. However, ‘Marta’ (29.87%) and ‘Constantí’ (25.42%) exhibited slightly lower shelling percentages than those reported by Egea et al. [75] and Vargas et al. [44], which were 32% and 27%, respectively. Additionally, Egea et al. [75] reported that irrigation did not have a significant effect on shelling percentage. Similarly, Gutiérrez-Gordillo [79] did not observe significant differences for the ‘Marta’ and ‘Lauranne’ cultivars. Therefore, the differences observed between our study and others are probably due to the adaptation of the cultivars to the terroir of Mallorca.

With regard to the almond in-shell and kernel size and weight, these are cultivar-specific traits [80,81]. In general, the kernel weight values, ranging from 0.97 (‘Penta’) to 1.93 (‘Tarraco’) g kernel⁻¹, were relatively high. Socias i Company et al. [80] indicated that the range of kernel weight typically varies between 0.5 and 1.5 g, with a preference for kernels exceeding 1.2 g for most applications. In our case, all cultivars exceeded this weight, except for the ‘Penta’ cultivar. Most cultivars showed similar values or even higher values compared to those reported by other authors under full irrigation [82,83]. The improvement in fruit unit weight when deficit irrigation is applied was also observed by Gutiérrez-Gordillo et al. [82]. These results support the possibility of enhancing fruit size under deficit irrigation, thereby offering an additional benefit in terms of fruit marketability and consumer satisfaction [84].

In general, almond kernels are a rich source of lipids, primarily composed of mono- and polyunsaturated fatty acids. Our results for lipid fractions aligned with those reported by Kodad [84] and Pérez-Sánchez and Morales-Corts [81]. Protein constituted the second most significant chemical component in almond kernels, following the lipid fraction. In our study, the protein content for each cultivar was higher than that reported by Kodad [84] for Spanish almond cultivars. Similarly to Alessandroni [85], protein content was found to have a negative correlation with lipid content (r = −0.63 **). The ratio lipids/protein ratio is highly important in processing because water absorption by almond paste relies on the equilibrium between these two constituents [84]. Similar to Lipan et al. [86], all cultivars except ‘Tarraco’ exhibited a ratio higher than 2. Dietary fiber constituted the third major chemical component in almond kernels [82]. The average content was consistent with those obtained by Lipan et al. [86]. The positive effects of these components on digestion, the intestinal microbiota, and cholesterol levels have been reported [87,88,89]. Kodad [84] reported that for Spanish almond cultivars, the carbohydrate content ranged from 1.8% to 7.6% of the fresh weight. However, in our study, the content was slightly higher, ranging between 7.23% (‘Marinada’) and 9.63% (‘Antoñeta’). Among the carbohydrates found in almonds, only sugars, starch, and certain sugar alcohols can be broken down, absorbed, and metabolized by humans to serve as an energy source [90]. Regarding sugars, our results were consistent with those of Yada et al. [90] for cultivars from Spain, Portugal, and California. It should be noted that small amounts of sugar are sufficient to impart a sweet taste to the kernels [91]. The almond kernel is also considered a valuable source of mineral elements that significantly contribute to human health [84]. Mineral content is sometimes expressed in general terms as ash content, with kernels containing approximately 3% (fresh weight) [90]. This parameter exhibits limited variability. In our study, all cultivars exhibited an ash content within the range of 3.05% to 3.66%. Various authors [87,92] have also documented similar variability for this parameter in Spanish almond cultivars, ranging from 3.05% to 3.45%. Regarding energy, the elevated nutritional value of almond kernels primarily stems from their substantial lipid content, which serves as a significant calorie source [93]. Lipid content was highly correlated with energy levels (r = 0.93 **). Similar results were obtained by Pérez-Sánchez and Morales-Corts [81]. However, all the cultivars in our study exhibited a higher energy content than those reported by Gradziel et al. [94] and Pérez-Sánchez and Morales-Corts [81].

Although the total lipid content was high, the dominance of oleic and linoleic fatty acids is nutritionally favorable [95]. Concerning the variables used as indices of oil stability and resistance against rancidity, the oleic/linoleic acid (O/L) ratio for each cultivar was higher than that reported by Socias i Company et al. [63]. These higher values could be attributed to the edaphoclimatic conditions of Majorca. Elevated O/L ratios indicate increased almond quality, and according to our results, this suggests that Majorcan almonds exhibit superior quality. Similarly to other studies [95,96,97,98,99], oleic acid content was strongly negatively correlated with linoleic acid content (r = −0.97 **). Furthermore, palmitoleic acid correlated positively with margaric acid (r = 0.63 *) and margaroleic acid (r = 0.67 **). We also found a negative correlation between palmitoleic and stearic acid (r = −0.63 *).

PCA results showed that the first three components explained approximately 64% of the observed variability among the studied cultivars. These results are consistent with those obtained by Lipan et al. [86] for almond cultivars in the Mediterranean area. These findings highlight the relevance of stearic acid, linoleic acid, margaroleic acid, oleic acid, energy, and total fat content as discriminant parameters that differentiate almond cultivars. Other authors have obtained similar results in previous studies. In the PCA biplot, the short distance between cultivars indicated their similarity based on the chemical parameters of the kernel, while the long distance indicated dissimilarity. Finally, the dendrogram separated the cultivars into three groups, confirming the observations of the PCA. Within these groups, a certain variability could be observed.

The correct selection of cultivars has significant economic implications for almond producers in the region of Majorca. Choosing cultivars that not only adapt well to local edaphoclimatic conditions but also optimize water use efficiency and maximize fruit quality can lead to greater profitability. Cultivars that offer high almond yields, efficient production, and superior kernel quality can reduce operational costs and enhance competitiveness in an increasingly demanding global market. Additionally, the adaptation of cultivars to deficit irrigation conditions can mitigate the risks associated with climate variability, providing greater economic stability in the long term. Therefore, informed decisions regarding cultivar selection are crucial to ensuring the sustainability and profitability of almond orchards in a changing climate.

This study was conducted under deficit irrigation conditions, which, while reflecting the current water limitations in the region, may not be fully generalizable to other areas with different water availability. Future research should explore a broader range of water deficit scenarios to better understand the performance and quality of cultivars under varying levels of water availability. Additionally, it would be valuable to continue testing new cultivars in the context of climate change to provide producers with information that can assist them in selecting the most suitable cultivars. Despite these limitations, the findings offer valuable insights into the agronomic performance and quality traits of almond cultivars, providing a solid foundation for further research aimed at optimizing almond cultivation in Mediterranean climates.

5. Conclusions

The results confirm that the choice of cultivar is a key factor in almond cultivation. Based on all the obtained results, the cultivars have been ranked from highest to lowest agronomic interest. The recommendations for each cultivar are as follows: (i) ‘Glorieta’ is characterized by a high cumulative almond in-shell yield and high yield efficiency; (ii) ‘Constantí’ is characterized by a high cumulative kernel yield and kernels with a high oleic/linoleic acid (O/L) ratio, low fatty matter, and high protein; (iii) ‘Masbovera’ is characterized by high vigor, almonds with low carbohydrate content, and high O/L ratio; (iv) ‘Marinada’ is highly recommended due to its medium vigor, high shelling percentage, and low cultural practice requirements, making it suitable for plantations with narrow spacing; (v) ‘Vairo’ is characterized by good yield efficiency, a low shelling percentage, and kernels with low sugar content and low dietary fiber; (vi) ‘Lauranne’ is characterized by good yield efficiency similar to ‘Vairo’, medium vigor, good productivity efficiency, and almonds with high carbohydrate content; (vii) ‘Ferragnès’ is characterized by a high shelling percentage and kernels with high weight and a high O/L ratio; (viii) ‘Antoñeta’ is characterized by a high shelling percentage and kernels with high width, high sugar content, and high dietary fiber; (ix) ‘Marta’ is highly vigorous with high O/L ratio kernels; (x) ‘Penta’ is characterized by low-weight kernels and a high fatty matter/protein (FM/P) ratio; (xi) ‘Belona’ is characterized by a high width/length (W/L) ratio kernel and an FM/P ratio similar to ‘Penta’; (xii) ‘Soleta’ has a low cumulative kernel yield similar to ‘Belona’ and kernels with high length, low width, low thickness, and low total carbohydrate content; (xiii) ‘Tarraco’ is characterized by high vigor and low productivity efficiency similar to ‘Marta’, with high weight and length kernels, low fatty matter, and high protein content; (xiv) ‘Mardía’ is characterized by almonds with high protein content and a high O/L ratio; however, its low vigor, low yields, and poor shelling percentage indicate that it has not adapted well to the edaphoclimatic conditions of Majorca. In general, the results provide valuable information about the adaptation of cultivars to the terroir of Majorca, offering insights that can support producers’ decisions. Continued research with new cultivars is necessary to further provide this valuable information to farmers.