As the global population grows, so does the demand for food [1]. A balanced diet containing a broad spectrum of nutrients is important for human health [2], [3]. Animal-pollinated crop species provide key nutrients valuable for human health, including an estimated 74% of all globally produced lipids and 35–65% of vitamin E [4]. The benefits of animal pollination for crop production have been documented for many crop species [5]–[7], but the effects of animal pollination on the nutritional composition of crops and other measures of quality such as fruit shelf life and therefore its commercial value have just begun to be investigated [8], [9].

Animal pollination may become a limiting resource if the growth of crops reliant on pollination outpaces the growth in the number of honey bee hives [10] and agricultural intensification and habitat loss continue to negatively affect wild bees [11]–[13]. Human population growth and climate change are predicted to increase the strain on the resources required for food production (particularly water) [14]. While the effects of plant resource shortages have been documented on the yield and development of crops [15]–[17], the effects of the availability of water and other plant resources on nutritional composition are largely unknown. How pollination and resource availability both singly and jointly impact a crop’s nutritional composition is thus a pressing question, particularly in light of the growing demand for certain crops based on their purported nutritional value. Almond (Prunus dulcis [Mill.] DA Webb) is an example of a commodity where advertising has focused on the potential health benefits of its consumption [18]. Clinical trials have shown almonds to be cardio-protective [19], an effect that is attributed to their high monounsaturated fat content. The primary monounsaturated fatty acid present in California-grown almonds is oleic acid, an omega-9 fatty acid accounting for 58–74% of the total fat content [20]. Almonds also contain linoleic acid, an omega-6 fatty acid, and very small amounts of linolenic, an omega-3 fatty acid [20]. The health effects of omega-6 fatty acids are controversial, with evidence to suggest that the high omega-6 to omega-3 ratio of the Western diet is a contributor to cardiovascular disease, cancer and diabetes [21]. Almonds are also valued for their vitamin E (α-Tocopherol) content as it is an antioxidant which helps protect cell membranes from peroxidative damage [22].

Since almond is a highly pollinator-dependent crop and a large consumer of water and fertilizer [23], its production may be sensitive to fluctuations in these resources. Furthermore, approximately eighty percent of the world’s almonds are produced in California [24], a state where climate change is expected to reduce water availability [25], [26]. The aim of this study was to investigate if pollination and plant resource availability (water and fertilizer) influence the nutritional composition of almonds. We exposed whole almond trees to different pollination treatments (the exclusion of pollinators to allow for self-pollination only and hand cross-pollination with compatible pollen to achieve high levels of cross-pollination) and different resource treatments (reduced water and no fertilizer). We investigated whether self-pollination and reduced plant resources (fertilizer and water), when experienced by whole trees, influenced the nutritional composition of almonds, and if there were interacting effects. Since almond is highly pollinator dependent [23], we hypothesized that the impact of self-pollination on the nuts’ nutritional content would be of similar strength to that of the availability of water and fertilizer, and that pollen and resource availability would have interacting effects.

The study was conducted in 2008, in a 3.2 ha almond orchard in the Sacramento Valley, Northern California (122°2′1.925″W, 38°55′19.372″N, WGS 1984: the owner of the land gave permission to conduct the study at this site). Our study orchard contained trees of the most popular variety for production, Nonpareil, grafted onto peach rootstock (Prunus persica (L.) Batsch) in 2005 and planted in 2006 (third leaf planting). Other tree varieties compatible with Nonpareil were located 100–300 meters away, including Mission (100% compatible), Carrion and Wood Colony (50% compatible). Honey bee hives were placed in the orchard during bloom with the eight nearest hives being 300–350 meters away. As part of the grower’s pollination strategy, all hives in the orchard had Padre pollen (100% compatible) placed at the hive entrance to maximize the transport of compatible pollen to the Nonpareil trees.

Linoleic acid (an omega-6) is the predominant polyunsaturated fatty acid in almonds, so the values for linoleic acid and total polyunsaturated fatty acids were highly correlated (supporting information S3). Similarly, as the predominant monounsaturated fat, levels of oleic acid were correlated with the total amount of monounsaturated fat and total fat. We will therefore present the results for linoleic and oleic acid but not for monounsaturated fatty acids, polyunsaturated fatty acids or total fat. Levels of linolenic acid were below the threshold for accurate quantification (<0.04 g/100 g).

The overall nutritional composition of the almonds differed between treatment groups (Fig. 1, stress 10%). This was confirmed by the permutational multivariate analysis of variance, which found that 36% of the difference in nutritional composition was related to the treatment group (R² = 0.36, F₃ = 2.7, P = 0.026). The variation in nutritional composition did not differ between treatment groups (F₃ = 1.3, P = 0.339).

We found differences between the eight treatment combinations in the levels of vitamin E (F₇ = 20.4, P<0.001), oleic acid (F₇ = 5.0, P = 0.001) and linoleic acid (F₇ = 5.0, P = 0.001). Vitamin E was highest in the self-pollinated nuts and was lowest in the nuts from treatments combined with hand cross-pollination, especially cross-pollinated with no fertilizer (Fig. 2a). However, oleic acid was lower in the self-pollinated nuts of the single treatments (Fig. 2b), and lower still in self-pollinated nuts combined with reduced water and no fertilizer. Linoleic acid was highest in the self-pollinated nuts combined with no fertilizer and lowest in nuts from cross-pollinated trees (Fig. 2c). The proportion oleic to linoleic acid (F₇ = 7.1, P<0.001) varied between treatments (Fig. 2d). It was lower in self-pollinated nuts and decreased further when self-pollination was combined with no fertilizer.

While the adverse effects of pollination limitation and resource scarcity on fruit set and crop development have been documented [15], [17], [28], this is the first study to our knowledge, to test if pollination and plant resources act alone or in combination on the nutritional composition of a crop. The effect of self-pollination on the nutritional composition of Nonpareil almonds was greater than the effect of reduced water and fertilizer. The health benefits of almonds are attributed to their fat composition and vitamin E, and these were both affected by the pollination treatments.

We found the highest oleic to linoleic ratio in almonds in cross-pollinated trees, and the lowest ratio in pollinator-excluded trees. Almonds with a high oleic to linoleic ratio would be most favorable to consumers seeking health benefits. The high oleic acid content is credited for the cardio-protective effect of almonds [19], [29]. Although linoleic acid is an essential fatty acid that the body cannot synthesize, rising dietary levels of omega-6 fatty acids are believed to contribute to chronic diseases in Western cultures [21]. In terms of crop characteristics, a high oleic (mono-unsaturated fatty acid) to linoleic (an omega-6 polyunsaturated fat) acid ratio is desirable as it improves the stability of the fats against rancidity and therefore increases the almonds’ shelf life [27]. A positive effect of bee-pollinated flowers on shelf life was recently demonstrated for strawberry production, since bee pollination leads to firmer fruits that last longer [8]. Further, Klatt et al. [8] show that bee pollination increased fruit weight and redness and also reduced sugar-acid ratios and therefore produced higher commercial grades. Together these two effects increased the value attributed to bee-pollinated fruits. Other studies showed that oil content in oilseed rape (Brassica napus var. SW Stratos™) [30] and sugar content in mandarin orange (Citrus reticulata Blanco) [31] were improved by pollination. Negative effects of limited pollination have included lowered calcium concentrations (associated with storage disorders) in Braeburn apples (Malus domestica Borkh) [32], although this effect seems to be variety specific [9] and lowered fruit weight and length in self-pollinated Cucurbita moschata, cv. Piramoita [33]. Consistent with previous studies, our overall findings were that cross-pollinated trees bore fruit with the most favorable nutritional content. A notable exception in our study was vitamin E (α Tocopherol), which was higher in self-pollinated almonds.

The mechanism(s) for the impact of cross- versus self-pollination on nutrients is as yet unknown. Self-pollination in almond may alter the nutritional composition of the nuts through a reduced development rate. Higher levels of linoleic acid are found in the earlier stages of oil accumulation in almond, with levels declining in the later developmental stages as oleic acid increases [34]. The self-pollinated almonds may develop more slowly than the cross-pollinated almonds. Compatible pollen tubes were found to grow more quickly in almond than self-pollen tubes, and reached the ovary earlier [35]. In addition, self-pollination was found to delay megasporocyte differentiation [36]. If self-pollinated almonds develop more slowly, at harvest time they may not have accumulated as much oleic acid or lost as much linoleic acid as cross-pollinated almonds. Given that self-pollen tubes grow more slowly [35], in self-pollinated flowers there may also be an effect of embryo sac degeneration on the fruit’s development and nutritional composition. While the mesh cages used to exclude pollinators from the trees could also have played a role in slowing fruit development by reducing the amount of sunlight the trees received, we would expect this to be minor due to the relatively short length of time the cages were covering the trees. Klatt et al. [8] explained their findings of increased strawberry shelf life with bee pollination as deriving from the increased amount of fertilized achenes per fruit (strawberry is an aggregate accessory fruit, consisting of many achenes each containing a single seed) when pollinated by bees. These achenes both create firmness structurally and produce the plant hormones that prevent fruit softening. It remains to be tested for any kind of crop if plant hormone production differs in self- versus cross-pollinated fruits or seeds.

In contrast to the pollination treatment, the fertilizer and water treatments had the potential to alter the resources available to the trees after the flowers had finished blooming. However, the limited impact of the fertilizer regime on almond’s nutritional composition suggests that sufficient resources may have been stored in the soil from prior fertilizations. An effect of reduced resources such as fertilizer on almond composition may only be detected after multiple years with no fertilizer; especially as perennial almond trees may also store resources [17]. Reductions in irrigation have been found to reduce almond production and yield during the drought year [23] and in subsequent years [37]. The lack of impact of reduced water on the nutritional content of the almonds is supported by a review of almond composition studies, which found that irrigation had little effect on the oil content [38].

We did not analyze nutrients of almonds from open-(bee) pollinated trees not treated with reduced water or no fertilizer in the same study year. In a previous study [23], Klein et al. investigated the weight and number of nuts produced by these same almond trees as well as open-pollinated trees and found that self-pollinated trees produced the fewest, heaviest nuts. Open pollinated nuts were intermediate with respect to weight and number, meaning that hand cross-pollinated trees produced larger numbers of smaller, lighter nuts. Nut weight and number was strongly related to the pollination treatments and the plant resource treatments had little influence [23]. This led us to hypothesize that the effect of self- versus cross-pollinated nuts could be indicated by nut size. We therefore analyzed the same nutrients for small and large open-pollinated nuts collected randomly in the experimental orchard under normal water and fertilizer regimes a year after the experiment took place. However, contrary to our expectations we found that the nutrient content of large and small nuts were not significantly different (data not shown). We conclude that although pollination can influence fruit weight, the weight of a fruit seems not to influence the amount of nutrients per unit weight in the nuts. Since these two processes are decoupled, this suggests that pollination directly influences the ratio of these nutrients that are important for human health.

The nutritional composition of crops may vary under a variety of interacting factors including agricultural practices, pollination and climate. The combined resource manipulations in this study showed little interactive effect on the fat composition. For vitamin E however, the combined resource and pollination treatments showed different levels of vitamin E than would have been expected from the effects of the single treatments. This suggests that changes in pollination could have unexpected effects on crop nutrition due to interactions with other inputs. In tomato (Lycopersicon esculentum L.), the benefits of arbuscular mycorrhizal fungal colonization for fruit quality (lower acidity, greater ascobrbic acid content) were found to be more pronounced when the plants were water stressed [39].

The present study demonstrates that pollination can positively change the nutritional value and commercial quality of almonds and that the effects are even stronger in combination with reduced availability of water and especially of nutrients applied to the soil. The potential nutritional benefits from cross-pollination deserve further study and may increase the current estimates of the economic and health value of pollination service to crops [4], [8], [40], [41].