The security of food production depends on the function of many synergistic ecosystem services such as soil fertility, pest control, pollination, and flood and erosion control (Kremen and Miles, 2012; Tscharntke et al., 2012; Bommarco et al., 2013). Widespread adoption of farming practices that support these services is only feasible when those practices do not inhibit economic sustainability or overextend logistical resources (e.g., time, knowledge, labor) for the farmer. It is, therefore, critical that we determine farming practices that simultaneously benefit ecosystem services, food production, and farmer livelihoods. Practices that increase on-farm vegetative diversity, such as polyculture (growing multiple crops within a field), cover cropping, hedgerow installation, or natural habitat restoration, often improve ecosystem function with neutral or even positive effects on crop yields (Rosa-Schleich et al., 2019; Tamburini et al., 2020). For example, crop diversification has been shown to improve soil health (Beillouin et al., 2021) and natural pest control (Iverson et al., 2014), while hedgerows around fields and natural habitat patches in the surrounding landscape can enhance natural predator and pollinator communities (Kremen and Miles, 2012; Karp et al., 2016; Albrecht et al., 2020; Tamburini et al., 2020).

The importance of maintaining robust pollinator communities for food production is now well-established and recognized by many international government sectors (see http://www.fao.org/pollination/major-initiatives/en/ for a list of global and national pollinator initiatives), including the Pollinator Health Task Force in the US (Pollinator Health Task Force 2015). Reliable pollination services maintain food diversity, marketability, production stability, and even shelf life (Steffan-Dewenter et al., 2005; Klein et al., 2007; Gallai et al., 2009; Potts et al., 2010; Eilers et al., 2011; Tscharntke et al., 2012; Wietzke et al., 2018). Roughly 30% of all foods are dependent on pollination to some degree, and those foods provide the majority of our lipids and important vitamins and minerals (Eilers et al., 2011). Yet, as consolidation and industrialization continues across various agricultural production systems, important pollinator habitat is lost which negatively impacts beneficial native bee communities and increases reliance on the managed honey bee (Apis mellifera) (Kremen et al., 2004; Kremen and Ostfeld, 2005; Potts et al., 2005; Goulson et al., 2008; Kohler et al., 2008; Ricketts et al., 2008; Rundlöf et al., 2008; Dicks et al., 2010; Roulston and Goodell, 2011).

The value of crop pollination by native bees is often underestimated, despite their important contribution to food production (e.g., Kremen et al., 2002; Klein et al., 2007; Winfree et al., 2007; Albano et al., 2009; Brittain et al., 2013b; Garibaldi et al., 2013). While crop pollination is often managed using honey bees in North America, Europe, and other regions, robust and diverse communities of native bees can provide more effective pollination to many crops than honey bees alone (Albano et al., 2009; Klein et al., 2012; Garibaldi et al., 2013). The pollination services provided by a highly abundant single pollinator species, such as the honey bee, could be insufficient when physical, phenological, or behavioral mismatch occur at some or all points in time. In these situations, the crop would benefit from the functional complementarity that a diverse pollinator community may offer; different pollinator species can provide unique, complementary roles that together yield higher function (Chagnon et al., 1993; Blüthgen et al., 2006; Hoehn et al., 2008; Rader et al., 2013a,b). Functional complementarity can occur temporally, when different pollinator species are present at different times of the day, season, or year (Blüthgen and Klein, 2011; Pisanty et al., 2016), or spatially, when different bees forage at different locations on a plant or flower [e.g., outer verses inner flowers, Blüthgen and Klein (2011); Brittain et al. (2013a), different parts of the flower receptacle, Chagnon et al. (1993)]. Thus, promoting diverse pollinator communities that span a range of functionality should result in better quality and more resilient pollination services (Grab et al., 2017). Specifically in strawberry, it has been observed that honey bees tend to visit only the top of the flower receptacle, while other native bees often crawl around the flower base, leading to more complete pollination of the achenes and, consequently, less malformed and heavier berries (Chagnon et al., 1993). Further, honey bee visits have led to smaller berries in strawberry, despite adequate amount of pollen deposition, probably due to their methodical foraging within rows, which reduces cross-pollination among varieties when they are present in a farm field (MacInnis and Forrest, 2019).

One of the most effective ways to build or restore robust pollinator communities is to increase vegetative diversity at both landscape and local scales (Ricketts et al., 2008; Winfree et al., 2009; Klein et al., 2012; Kremen and Miles, 2012). At a large scale, farms with greater proportions of natural vegetation in the surrounding landscape have a higher abundance and diversity of native bees (Steffan-Dewenter et al., 2002; Kremen et al., 2004; Klein et al., 2012; Mandelik et al., 2012; Kennedy et al., 2013; Martins et al., 2015). Similarly, studies in Europe and North America have found positive, landscape-scale effects of crop compositional and configurational diversity on pollinator regional species richness (Sirami et al., 2019). At a local scale, leaving areas of undisturbed land on the farm such as field margins or boundaries, drainage ditches, native vegetation remnants, and hedgerows, can provide important habitat to unmanaged pollinators and, in turn, increase pollinator abundance and richness (Hannon and Sisk, 2009; Klein et al., 2012; Kennedy et al., 2013). Recent syntheses comparing farms with more diversified vs. more simplified types of farming found agricultural diversification to have important positive effects on pollinator abundance, richness, and crop pollination (Tamburini et al., 2020; Sánchez et al., 2022). However, while some of these studies looked at aspects of crop diversification, such as crop rotational diversity (planting crops in a sequence over time), agroforestry (including a tree and a shrub or ground layer) or intercropping (growing two crops together in the same row to facilitate inter-species interactions), they did not examine how polycultures (i.e., growing multiple crops in separate blocks of rows within a farm field) can contribute to pollinator abundance, diversity, and function, due to the paucity of such studies.

In California, polyculture is practiced in the Northern and Southern Central Valley and in the Central Coast growing regions of the state, often but not always on organic farms (Chaplin-Kramer and Kremen, 2012; Forrest et al., 2015; Guzman et al., 2019). While organic farming generally supports abundant and diverse insect communities (Holzschuh et al., 2008), variation in farm-scale crop diversity and input use exists across organically managed farms (Tscharntke et al., 2021). Working within a United States Department of Agriculture (USDA) certified organic management system that reduces chemical variables that influence pollinators, we compare field specialization (monoculture) to field diversification (polyculture) to specifically explore whether crop diversity within a field may better support native bee populations and, if so, whether this may lead to higher provisioning of pollination services. We expected that, if crop diversity better supports native pollinators, polyculture farms would increase native bee abundance and diversity, thereby, increasing crop yields. If the benefits of local on-farm crop diversity can substitute for the known benefits of landscape-scale natural habitat, then polyculture farming may provide a pathway to restoring critical services provided by native pollinators, while optimizing the amount of land left in production.

We collected/observed 2, 422 honey bees, 1, 699 native bees, and 2, 399 non-bee visitors. Our native bee specimens comprise 41 species. Many of these native species were rare, with 27 of the 41 collected fewer than 5 times. Supplementary Table 3 presents a complete list of collected species and information about how and where they were collected.

We found higher native bee abundance and richness at polyculture farms than monocultures (effect of polyculture on abundance = 0.69, P < 0.01 and richness = 1.82, P < 0.01, Table 1, Figure 1A, and Supplementary Figure 2A). Further, both native bee abundance and richness were higher at farms with more surrounding natural cover (effect of natural cover on abundance = 3.08, P < 0.05 and richness = 1.92, P < 0.001, Table 1) and this effect was stronger at monoculture farms (interaction terms were both negative, Table 1, Figure 1B, and Supplementary Figure 2B; note that the interaction between farm type and natural cover is only statistically significant for native bee richness). In contrast, we found no evidence that farm type or natural cover impacts honey bee or non-bee abundance (Table 1). We found no strong effect of collection method, indicating that, for native bee abundance or richness and for non-bee abundance, both nets and pans collect roughly similar communities in this system. Many fewer honeybees were collected in pans than were counted foraging which is not surprising, given the large number of honeybees foraging at any given time.

Excluding pollinators led to smaller berries (Table 2) and berries that were more likely to be malformed and non-marketable, while excluding both pollinators and wind intensified these effects (Figure 2 and Table 2). We did not observe a significant effect of farm type or surrounding natural cover on berry malformation, or any evidence of interaction effects between these variables.

Translating the above results into real-world numbers: at both monocultures and polycultures, berry weight decreases from 20 to 18 g per berry when pollinators are excluded and to 16 grams per berry when both pollinators and wind are excluded. The probability that a berry is malformed at a farm at the average level of natural cover more than doubles (from 0.18 to 0.39) when pollinators are excluded and more than triples (from 0.18 to 0.62) when both pollinators and wind are excluded (Figure 2).

Finally, we found that honey bee abundance increases the probability of berry malformation while native bee species richness decreases it (Table 3) but we found no such effects for berry weight.

For the future of farming to be sustainable, farmers need economically feasible options for practices that maintain or improve natural resources beneficial for farming operations. Our study corroborates the body of literature that shows that vegetative diversity at both the landscape and the field level can indeed enhance the natural resources that provide ecosystem services, in this case pollinators. However, the amount of natural habitat surrounding a farm is largely out of farmer control and, therefore, not usually a feasible diversification option (Morandin and Kremen, 2013). Lack of surrounding natural habitat may be particularly true in agriculturally intensive regions where natural habitat is limited or non-existent. It may also prove impossible for farmers to restore or conserve habitat parcels if they lease rather than own land and have little incentive to invest in long-term diversification, despite the benefits it can provide (Chapman et al., 2022). In either case, farmers may not have access to parcels near natural habitat.

While we have focussed on landscape level natural habitat, non-crop habitats such as floral strips, flowering hedgerows and windbreaks (which provide nesting habitat) were also present across both monoculture and polyculture sites. Floral strips and flowering hedgerows were typically more abundant at polyculture sites, while non-flowering windbreaks were typically more abundant at monoculture sites (Supplementary Table 1). These semi-natural habitats may impact the observed pollinator communities and it is possible that this may confound the effects of field diversification and abundance of semi-natural habitats in the fields surroundings.

A hopeful outcome of this study is that diversifying crops on a farm, which is largely in control of farmers, can potentially substitute for large scale adjacent natural habitat in supporting abundant native pollinator communities. Farms with greater crop diversity had significantly more native bees visiting strawberry and these visits came from a significantly greater number of species. It is possible that, in polyculture, different crops bloom at different times and this temporal complementarity provides greater phenological coverage for a greater number of pollinator species' flight seasons. In contrast to native bees, honey bees and non-bees were equally abundant across all sites. The former is not surprising given that honey bee numbers in this landscape are largely a function of where hives have been placed to provide pollination for other (non-strawberry) crops. Overall, local crop diversity on farms had an important and positive impact on pollinator communities.

It is notable that the results from this study also highlight the benefits of diversifying crops (within) the organic system. Many studies focus on the comparison of conventional to organic agriculture without recognizing or describing the variation in farm-scale diversification that exists within organic (but see Ponisio et al., 2015; Tscharntke et al., 2021; Lu et al., 2022). Organic food is produced on a range of farm sizes and levels and forms of on-farm diversification, from specialized monocultures to highly diversified systems, both over space and time. While organic management has been shown to support biodiversity better than conventional systems overall (Holzschuh et al., 2008; Kremen and Miles, 2012), our results show that, within organic agriculture, crop-diverse farms will support biodiversity better than specialized monocultures. Since 2017, in the United States, USDA organic certification requires farmers to maintain or improve biodiversity (National Archives and Records Administration, 2018), and this study suggests that adding crop diversity can assist in fulfilling that mandate.

Pollinator exclusion experiments revealed that insect pollination is important for strawberry production, even though most fields hosted only a single variety (Albion) consisting of genetically-identical (clonal) individuals. Pollinator visitation nonetheless can increase the probability of producing marketable fruits by nearly 20%. Fruits resulting from flowers that were open to pollinators were significantly bigger and better shaped. Fruits from flowers where both wind and pollinators were excluded were the smallest and the most likely to be malformed, suggesting that wind may also play an important role in pollination. While strawberries are self compatible, berries from open flowers are often larger and more marketable (Wietzke et al., 2018). That is, the flowers require a vector to move the pollen within the flower to facilitate self-pollination. Our results suggest that wind may move the flowers in a way that allows self pollen to drop onto the receptacle and facilitate selfing. It is also likely that any floral visitor can help facilitate selfing, which is why future work should examine the role of non-bee visitors.

Our post-hoc analyses linking pollinator community metrics to berry weight and marketability revealed that all insect visits are likely not equivalent, with higher native bee richness lowering the likelihood of berry malformation and higher honey bee abundance actually increasing it. Previous studies (Chagnon et al., 1993) and our field observations have found that honey bees tend to visit only the top of the strawberry receptacle, potentially providing incomplete pollination. Because native bees often visit other parts of the receptacle and, in particular, the base of the flower, the native bees may complement honey bees to yield more complete pollination and, ultimately better formed and more marketable berries. In contrast, when honey bees are the dominant strawberry pollinator, complete pollination may not always occur. Further, when multiple varieties are present, honey bee visits may saturate stigmas with self or clonal pollen from the same variety, leading to lower weights or marketability than visits from wild pollinators, which forage more erratically and tend to cross-pollinate among varieties more frequently (MacInnis and Forrest, 2019, 2020). Instead in our study system, pollen transported by bees was most likely from clonally-related plants.

While our results indicate that pollinators (and wind) significantly improve strawberry marketability, we did not find that sites with greater native pollinator abundance or diversity produced more marketable fruits. We also did not find a direct effect of local or landscape diversification on fruit marketability. Our dataset may have been too small or contained too little variation within the open pollinated flowers to detect an effect. The vast majority of flowers that were open to pollination resulted in a marketable fruit, regardless of local or landscape level diversification for each site. It is possible that, while the benefits of pollinators can be demonstrated by exclusion, pollinator limitation does not exist for strawberries at our selected study sites, where both honey bees and native floral visitors were abundant. Larger differences might also have been found if multiple varieties were planted in close proximity, which can lead to increased fruit weight and quality (MacInnis and Forrest, 2019, 2020).

Our work suggests that local and landscape diversification support larger and more diverse pollinator communities and those pollinators provide an important service to strawberry production, even in a clonal production system. In this system, field diversification is a promising management practice that provides environmental benefits and is more feasible for farmers than conserving or restoring large swaths of natural habitat that are often not part of the land they manage. Additionally, field diversification is a practice that may be executed by conventional and organic farmers alike (Tscharntke et al., 2021). Field diversification likely improves conditions for pollinators by providing multiple sources of pollen and nectar that support different species, or support the same species at different times (Guzman et al., 2019) and will benefit both. While our findings support the role of diversification in moving toward sustainable food production, moving from a highly specialized farming operation to one with multiple crops will be met with many challenges. Further, many recent syntheses have shown additive or interactive benefits between local diversification and surrounding semi-natural habitat (Kennedy et al., 2013; Sirami et al., 2019). Future studies should consider multiple and simultaneous ecosystem services to quantify comprehensive benefits of crop diversity against the costs.

The original contributions presented in the study are included in the article/Supplementary material, further inquiries can be directed to the corresponding author.

AS and CK conceived and designed the study. AS carried out fieldwork. LM'G and AS conducted statistical analyses. All authors wrote and edited the manuscript. All authors contributed to the article and approved the submitted version.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. The reviewer GT declared a past co-authorship with the author CK to the handling editor.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.