3.1. Climatic Conditions, Physiological Response, and Final Yield
The climatic conditions during the irrigation period were characterized by mild average temperatures (19.8–26.9 °C) and low rainfall. The cumulative ET0 and rainfall were 840 and 85 mm, respectively. Maximum evapotranspiration rates occurred in June (203 mm), July (239 mm), and August (170 mm), whereas rainfall was concentrated during April (71.2 mm) and October (10 mm). Thus, according to the climatic conditions, the total amount of irrigation received for FI, SDI75, and SDI65 at the end of the season was 7700, 5744, and 5159 m3·ha−1, respectively.
In relation to the crop water status, significant differences in Ψ
leaf were found among cultivars and irrigation treatments (
Table 1), taking into consideration not only all the data but also the daily measurements. In this regard, according to three-way ANOVA for repeated measures, significant differences (
p < 0.01) were found for timing evolution, cultivars, and irrigation treatments. Focusing on the cultivars,
cv. Marta registered the highest average Ψ
leaf value (−1.55 MPa), which was significantly different from
cv. Guara (−1.80 MPa) and
cv. Lauranne (−1.71 MPa), without differences between them. Regarding the irrigation dose, FI registered an average value of −1.55 MPa, which was significantly higher than those of SDI
75 (−1.76 MPa) and SDI
65 (−1.71 MPa), and the SDIs were similar. The interactions between irrigation and cultivars showed the highest Ψ
leaf value for
cv. Marta under FI (−1.45 MPa) and the lowest for
cv. Guara under SDI
75 (−1.91 MPa) and SDI
65 (−1.90 MPa). A very similar trend was observed for each measuring day, which reflects that
cv. Marta had a different water status than the other cultivars, whereas SDI treatments were similar.
While the almond quality parameters are determined by water stress imposed during the kernel-filling period, final yield responded to the level of water stress produced over the entire period of crop development (including the first stages of vegetative and fruit growth, or even the postharvest conditions from the previous season) [
29]. In our case, SDIs were applied, hence, water stress was imposed during the whole season (even during the postharvest period of the previous year, 2018). In response to this strategy,
cv. Guara registered the highest yield reduction in SDI
65 (17% lower than FI), while the yield in SDI
75 was significantly similar to the yield in FI (
Table 2). The yield reductions obtained with
cv. Guara were in agreement with the Ψ
leaf values previously registered by this cultivar (
Table 1). In contrast, the yield of
cvs. Marta and Lauranne was not significantly affected by SDI treatments (
Table 2); these cultivars showed the best results in terms of Ψ
leaf (
Table 1), as has been previously discussed.
This differential response between cultivars is not new, as is it has been proved by other authors. Gomes-Laranjo et al. [
11] reported different physiological responses among cultivars when they were subjected to different DI strategies. The authors concluded that
cv. Lauranne was less sensitive to water restrictions than Ferragnès. More recently, authors such as Miarnau et al. [
30] highlighted that under the SDI strategy, with a total amount of water around 200 mm, the nut yield for
cvs. Guara and Marta amounted to 1200 and 1900 kg·ha
−1, whereas under FI conditions (750 mm) the values were 2800 and 3600 kg·ha
−1, respectively. All these results suggest that
cv. Marta may mitigate the water stress by means of an internal mechanism, yielding more than
cv. Guara, as was observed in the present work.
3.2. Antioxidant Activity and Total Phenolic Content
The results of antioxidant activity (AA) and total phenolic content (TPC) are shown in
Table 3. Highly significant effects (
p < 0.001) were found in response to the applied irrigation treatments and cultivars.
AA was studied through two methodologies: ABTS•+ and DPPH•. The ABTS•+ results showed significant differences among cultivars but not irrigation treatments, with cv. Marta showing the highest values. Moreover, it was notable that the highest values of ABTS•+ were for cv. Guara under SDI65, followed by cv. Marta under FI and SDI65 strategies. These results evidence the importance of the cultivar in the antioxidant activity of almonds, which can be increased by using DI strategies, as has been observed for cv. Guara under SDI65. The DPPH• assay corroborated the results obtained using ABTS•+, showing that cv. Marta recorded the highest AA. In terms of irrigation strategies, it was observed that AA increased with the highest level of stress (SDI65). Regarding the interaction irrigation × cultivar for DPPH•, cv. Marta under SDI65 registered the highest values, followed by cv. Marta SDI75 and FI. Overall, the obtained results highlight how cultivar and irrigation can positively affect almond’s antioxidant activity.
For TPC, the highest values were also reached with SDI treatments and, in terms of cultivar, Lauranne had the highest value. Regarding the interaction irrigation × cultivar, the highest values were shown for
cv. Guara SDI
65, followed by
cvs. Lauranne SDI
75 and Marta SDI
65; this trend was consistent with the water-stress values found in this experiment. These findings agreed with Lipan et al. [
16], who found a positive correlation between TPC and imposed water stress. Antioxidant activity and TPC are parameters of great importance for health properties. Ros et al. [
31] and Lopez-Uriarte et al. [
32] reviewed a total of 21 clinical studies that evaluated almond’s antioxidant activity, reporting numerous cardiovascular health benefits such as decreased blood pressure and visceral adiposity. Polyphenols contribute to almond color and astringency and increase its shelf life [
33]. The highest polyphenol concentration is found in almond skin, and many times this is eliminated through the operations of processing [
34]. Moreover, almond skin also has antimicrobial properties, which are promoted by synergistic interactions between phenolic acids and flavonoids and involved in fighting against diseases caused by
Salmonella enterica,
Listeria monocytogenes, and
Escherichia coli [
35]. The main contributions of almond polyphenols to health are in reducing inflammation and type 2 diabetes [
36]; they also have antiproliferative and antitumoral effects, including proanthocyanins, the major polyphenol found in almonds.
The relationship between cultivar and antioxidant activity in almond has been investigated [
33], and it was concluded that each cultivar has its own characteristic antioxidants. Considering our results,
cvs. Marta and Guara had greater amounts of antioxidants than
cv. Lauranne, as shown by ABTS
•+ and DPPH
•. The values obtained by ABTS
•+ and DPPH
• may be a result of the activity of tocopherols rather than polyphenols, since
cv. Lauranne had the highest TPC index. It is well known that food antioxidant activity can be given by polyphenols but also by other compounds such as tocopherols, which can increase under water-stress conditions, as previously reported by Zhu et al. [
37] and Lipan et al. [
34], who demonstrated the importance of DI strategies to increase the antioxidant activity and TPC in raw almonds.
3.3. Organic Acid and Sugar Content
In relation to organic acid results, strongly significant effects were observed between irrigation treatments and cultivars. That is,
cv. Guara showed the highest amounts of all studied organic acids, followed by
cvs. Lauranne and Marta (
Table 4). Thus, as described for antioxidant activity and TPC, the cultivar factor was of great importance. In relation to the imposed irrigation strategy, it is noticeable that SDI
75 treatment obtained the highest values in all organic acids and, consequently, in total organic acid amount.
Regarding the interaction irrigation × cultivar, it can be highlighted that all cultivars were positively influenced by SDI
75. The interaction of
cv. Guara and SDI
75 treatment registered the highest amount of organic acids except fumaric acid;
cv. Lauranne × SDI
75 had the highest value of fumaric acid. Similar results were reported by Lipan et al. [
16] for
cv. Vairo, which showed a clear relationship between water stress and total organic acid content, although other authors concluded that water stress did not have such effects on these compounds [
12,
38]. Comparing these results with those obtained for Ψ
leaf, it is noticeable that the best results in relation to organic acid content were found in the cultivar that registered the highest water-stress values, which is in line with the results of Lipan et al. [
22], who concluded there was a direct relationship between imposed water stress and organic acid content for these cultivars (Guara, Marta, and Lauranne).
Sugar content was also highly influenced by the cultivar and the irrigation strategy (
Table 5). All studied sugars showed a clear response to the irrigation treatments and cultivars, except for maltoheptaose. In terms of irrigation treatment, SDI
75 and SDI
65 showed the highest total sugar content of 62.9 and 62.2 g·kg
−1, respectively. In terms of cultivar, Guara (65.5 g·kg
−1) and Lauranne (63.7 g·kg
−1) had the highest amount of sugars. Total sugars, and specifically sucrose, tend to be higher as water stress increases, which is in line with other studies [
39] that observed that total sugars and sucrose increased with water stress. Relating to the interaction of irrigation × cultivar, it can be highlighted that all cultivars were positively influenced by DI, especially by SDI
75, as occurs with organic acids. For maltrotriose and sucrose, the best combination was observed in
cv. Guara under SDI
75; for glucose, the best combination irrigation × cultivar was in
cv. Marta SDI
65; and for fructose,
cv. Lauranne under the SDI
65 strategy reached the highest value. According to Sánchez-Bel et al. [
38], sucrose is the principal sugar in almond cultivars, due to its preferential production and accumulation in the almond during ripening, and probably its synthesis and accumulation would be influenced by water stress. In this line, there is a strong effect between water stress and the sugar composition of nuts. During the water-stress period, almond leaves begin to lose turgor due to dehydration and, to avoid this, the tree closes the stoma by decreasing the transpiration process. Therefore, to restore the osmotic balance, the tree concentrates sugars to recover the turgor that was lost by dehydration. This is in line with Prgomet et al. [
40], who found in a two-year experiment that non-irrigated trees accumulated more leaf water soluble sugars than control trees to maintain cell turgor, similar to what occurred in almonds, as previously discussed [
16,
21,
22].
3.4. Fatty Acids
Almond’s fatty acid profile has a broad spectrum, which makes this product a food that is a good energy source and does not increase cholesterol levels [
41]. A total of 25 fatty acids were identified in this experimental work, classified as saturated fatty acids (SFA) (
Table 6) and unsaturated fatty acids (
Table 7), which are subdivided into monounsaturated fatty acids (MUFA) and polyunsaturated fatty acids (PUFA). The almond lipid fraction identified was mainly composed of oleic acid (C18:1n9), linoleic acid (C18:2n6), palmitic acid (C16:0), and stearic acid (C18:0). These were significantly affected by water stress, which increased their concentration in both SDI treatments compared to the FI treatment. In particular, oleic and linoleic acids increased 6.0% and 10%, respectively, in SDI treatments.
Regarding the cultivar, Guara had the highest amount of oleic acid, while cv. Lauranne had the highest amount of linoleic acid (6.2% and 3.8% more than Marta and Guara, respectively).
The interaction irrigation × cultivar showed that cv. Guara in SDI75 and cv. Lauranne in SDI65 led to the highest content of palmitic acid; whereas cv. Marta in the three treatments and cv. Guara in SDI65 and cv. Lauranne in both SDI treatments showed the highest amounts of oleic acid.
The best results in terms of the oleic/linoleic (O:L) ratio were found for SDI
75 (this index improves as its value decreases), while in terms of the cultivar, the lowest values were obtained for
cvs. Guara and Lauranne. Linoleic acid is related to the stability of the oil. Thus, the lower the amount, the greater the stability [
42]. This fatty acid is essential for humans and cannot be synthesized by itself. Thus, although stability is reduced, the increase in linoleic acid by deficit-irrigation treatment gives the almond added value since this acid plays a fundamental role in the death of cardiac cells [
39]. Additionally, as water stress increases, total phenolic content and antioxidant activity increase, which could help in protecting against lipid oxidation.
Regarding the total monounsaturated fatty acid (MUFA) and polyunsaturated fatty acid (PUFA) content (
Table 8), these values were higher in both SDI treatments;
cv. Marta had a higher amount of MUFA and
cv. Lauranne of PUFA. These results agree with those in the study of Lipan et al. [
22], which found the same relation of these cultivars under different irrigation strategies. In contrast, in other cultivars, such as
cv. Vairo, decreased MUFAs and increased PUFAs with increasing water stress after three years of experimentation was highlighted by Lipan et al. [
39]. In other nuts, such as pistachio, these compounds were not affected by water stress [
43]. However, significant differences were found with values higher than 50% of MUFA and 30% of PUFA depending upon the pistachio cultivar used. In olive
cv. Arbequina, Garcia et al. [
44] reported an increase in the MUFA/PUFA ratio in the RDI treatment compared to the control treatment due to the desaturation of oleic acid with high stress and consequent linoleic formation. Bitok and Sabaté [
45] reported that products rich in MUFA and PUFA could contribute to the prevention of coronary heart and cardiovascular diseases as well as diabetes and obesity. To prevent these diseases, the US Food and Drug Administration recommends 42.5 g of almond [
46]. This provides greater functionality for almonds cultivated under water-stress conditions. The PUFA:SFA ratio provides information about whether a diet is atherogenic or could promote coronary heart disease. In this case, this ratio had a high relation with irrigation, cultivar, and irrigation × cultivar, and was higher in SDI
75 with
cv. Lauranne.
3.5. Relationship between Crop Water Status at Different Stages and Nutritional Composition of Almond
With the aim of finding the potential relationships between Ψ
leaf and quality parameters and estimate which of them were more affected by imposed water stress, PCA was done (
Figure 1).
According to the obtained results, the two main components (F1 and F2) explained 41.9% and 21.4% of total variance. As observed, samples of each irrigation treatment were grouped mainly separately, except for SDI75 for cvs. Lauranne and Marta grouped together in FI treatments, and totally opposite to SDI65 treatments. In this regard, SDI65 samples were mainly surrounded by sucrose, glucose, fructose, SFA, and total fatty acids. In addition, cv. Guara under SDI75 was characterized by total organic acids (especially malic, citric, and tartaric), maltoheptaose, maltotriose, and the PUFA/MUFA ratio. Finally, FI samples were characterized by Ψleaf and the (MUFA + PUFA/SFA) ratio.
Significant relationships were observed between Ψ
leaf values and the quality parameters that had previously shown significant differences between irrigation treatments. No significant correlations were obtained between Ψ
leaf and antioxidant activity (in terms of ABTS
•+ and DPPH
• index with
r = −0.10 and 0.08, respectively). However, a significant correlation was observed for TPC (
r = −0.59). On the contrary, Lipan et al. [
39] found a significant relationship between accumulated water stress and the ABTS
•+ index (
r = 0.79), which was not detected in the present study, probably because this experiment was developed with different cultivars.
More interesting were the Pearson’s coefficients among Ψ
leaf, organic acids, and sugar content. Within the organic acids, fumaric showed the best correlations (
r = −0.96 **), whereas for the sugar profile, the best relationships were found for sucrose (
r = −0.85 **), fructose (
r = −0.92 **), and total sugars (
r = −0.86 **). Similar results were reported by Lipan et al. [
39], who noted improvements in terms of sucrose (
r = −0.42 **), fructose (
r = −0.36 *), and total sugar content (
r = −0.39 **). Thus, according to these findings, high water-stress levels (lower values of Ψ
leaf) would be accompanied by higher sugar and fumaric acid content.
Finally, the fatty acid content increased directly with the imposed water stress, with SFA (
r = −0.88 **), PUFA (
r = −0.80 *), and total fatty acid content (
r = −0.84 **) showing the highest correlation with Ψ
leaf, whereas for MUFA these relationships were not significant (
r = −0.51). In line with these results, the most important variations in the fatty acid composition was observed for the oleic/linoleic (
r = 0.46) and (MUFA+PUFA)/SFA (
r = 0.71 *) ratios. These results are in agreement with those reported by Lipan et al. [
39], who found strong relationships between Ψ
leaf values and the oleic/linoleic ratio (
r = −0.82 ***), SFA (
r = 0.81 ***), MUFA (
r = −0.84 ***), PUFA (
r = 0.83 ***), and PUFA/MUFA (
r = 0.85 ***).
Even though the present work corresponds to a single year, the relationships found between the studied quality parameters and imposed water stress are in accordance with similar works developed over a single year or for long-term experiments. As already mentioned, authors such as Egea et al. [
12], in a two-year experiment with
cv. Marta, concluded that DI strategies did not affect the final content of sugars and organic acids. Moreover, Lipan et al. [
39], in a long-term experiment on
cv. Vairo, found significant relationships between quality parameters and imposed water stress. Those authors reported significant increases in fructose, sucrose, and total sugars when water stress was imposed, and similarly for antioxidant activity, SFA, and PUFA and decreased O/L rate, all in accordance with the results obtained in the present work. In addition, Lipan et al. [
22] reported significant improvements in organic acid and sugar content in the same cultivars considered in this work but subjected to RDI strategies, along with reductions in the oleic/linoleic ratio and increases in the PUFA, PUFA/SFA, and PUFA/MUFA ratios. Thus, despite presenting results for a single year, there is a clear pattern between the obtained results and other recently published results, reinforcing the hypothesis that DI strategies improve some the most relevant quality parameters related to the nutritional composition of raw almonds.