Wildflower plantings promote blue orchard bee, Osmia lignaria (Hymenoptera: Megachilidae), reproduction in California almond orchards

Abstract Concerns over the availability of honeybees (Apis mellifera L.) to meet pollination demands have elicited interest in alternative pollinators to mitigate pressures on the commercial beekeeping industry. The blue orchard bee, Osmia lignaria (Say), is a commercially available native bee that can be employed as a copollinator with, or alternative pollinator to, honeybees in orchards. To date, their successful implementation in agriculture has been limited by poor recovery of bee progeny for use during the next spring. This lack of reproductive success may be tied to an inadequate diversity and abundance of alternative floral resources during the foraging period. Managed, supplementary wildflower plantings may promote O. lignaria reproduction in California almond orchards. Three wildflower plantings were installed and maintained along orchard edges to supplement bee forage. Plantings were seeded with native wildflower species that overlapped with and extended beyond almond bloom. We measured bee visitation to planted wildflowers, bee reproduction, and progeny outcomes across orchard blocks at variable distances from wildflower plantings during 2015 and 2016. Pollen provision composition was also determined to confirm O. lignaria wildflower pollen use. Osmia lignaria were frequently observed visiting wildflower plantings during, and after, almond bloom. Most O. lignaria nesting occurred at orchard edges. The greatest recovery of progeny occurred along the orchard edges having the closest proximity (80 m) to managed wildflower plantings versus edges farther away. After almond bloom, O. lignaria nesting closest to the wildflower plantings collected 72% of their pollen from Phacelia spp., which supplied 96% of the managed floral area. Phacelia spp. pollen collection declined with distance from the plantings, but still reached 17% 800 m into the orchard. This study highlights the importance of landscape context and proximity to supplementary floral resources in promoting the propagation of solitary bees as alternative managed pollinators in commercial agriculture.


| INTRODUC TI ON
The California almond (Prunus dulcis Mill.) industry relies heavily on the availability of honeybees (Apis mellifera L.; Hymenoptera: Apidae) to meet the pollination demands of their orchards (Traynor, 2017). These demands have grown in recent years as the amount of almond-bearing acreage now exceeds 470,000 ha (CDFA, 2019), requiring over two million honeybee hives annually during bloom (Goodrich & Goodhue, 2016), and accounting for 73% of the U.S. honeybee population (as of January 2017; Goodrich, 2018).
Consequently, it has become increasingly more difficult for commercial beekeepers to meet the pollination demands of the industry (Aizen & Harder, 2009;Seitz et al., 2016;Ward, Whyte, & James, 2010), which is compounded by persistent stressors impacting honeybee health and survival (vanEngelsdorp et al., 2009). Incorporating Integrated Crop Pollination strategies that supplement orchard pollination with alternative bee species may become necessary to bridge the widening gap between honeybee colony supply and demand (Bosch & Kemp, 2001;Isaacs et al., 2017;Wesselingh, 2007).
The blue orchard bee, Osmia lignaria Say (Hymenoptera: Megachilidae), native to North America (Rust, 1974), has been implemented effectively as a pollinator of commercially managed orchard crops, including apples, cherries, and almonds (Bosch & Kemp, 1999;Sheffield, 2014;Torchio, 1979Torchio, , 1985. When employed in commercial almond orchards, research shows that O. lignaria copollination with honeybees in almond orchards significantly increases fruit set versus when either pollinator is implemented alone (Brittain, Williams, Kremen, & Klein, 2013;Pitts-Singer, Artz, Peterson, Boyle, & Wardell, 2018). These results hold true in both semifield and openfield studies ranging from 20 × 13 × 3 m enclosed cages (Brittain et al., 2013) and up to 4-ha tracts of open almond orchard . Presently, the greatest challenge for the successful implementation of this alternative pollinator is their limited supply, coupled with their high cost. Sustainable in-orchard reproduction of O. lignaria in commercial orchards for use during the following year is not always achieved (Artz, Allan, Wardell, & Pitts-Singer, 2013, 2014. Except for the occasion when in-orchard progeny recovery exceeds the number of O. lignaria initially released Pitts-Singer et al., 2018), most O. lignaria currently available for distribution are captured from natural environments, which may have repercussions on native populations and their contributed ecosystem services to wildlands (Tepedino & Nielson, 2017). Additionally, trapping bees is labor-intensive, and management practices for the processing and cleaning of cocoons to eliminate pests and diseases can be costly. Therefore, initial retail costs of O. lignaria for commercial pollination exceed costs associated with hiring contracted honeybee pollination services at recommended stocking rates (Koh, Lonsdorf, Artz, Pitts-Singer, & Ricketts, 2018). Undoubtedly, in-orchard management practices must improve to support higher rates of O. lignaria reproduction for this species to become a viable alternative or supplement to honeybee almond pollination.
Almond blossoms are only available to foraging bees for 2-3 weeks of the year, which does not fully accommodate the 4-6-week life span of foraging O. lignaria. This limits the foraging period of O. lignaria in commercial orchards, where intense chemical control of weeds and other vegetation prevents pollinator access to supplementary floral resources that may extend their reproductive season and improve their overall nutrition. Nutritional limitation is one proposed explanation for the widespread decline of pollinator populations, particularly where monocultures dominate the landscape and offer only one or two mass floral resources for foraging bees (Brodschneider & Carlsheim, 2010).
Our objective was to determine the impact of managed wildflower plantings installed adjacent to commercial almond orchards on O. lignaria reproductive success. In total, 28.8 ha of commercial almond orchards were supplemented with managed O. lignaria, at variable distances from managed floral plantings. We hypothesized that access and proximity to alternative floral resources would improve O. lignaria nesting and reproduction during (and following) almond bloom. Three plots of previously fallow land, alongside a commercial almond orchard, were seeded with wildflower species known to overlap with (and extend beyond) almond bloom, providing diverse and extended floral resources for nesting O. lignaria through 2015 and 2016. Visitation to wildflower plantings, rates of bee reproduction, progeny outcomes, and pollen composition of representative provision masses were compared across six discrete distances (or "zones") from the maintained wildflower plantings to verify the use of alternative floral resources.

| Experimental layout and O. lignaria management
This study was conducted across a 2.12 km 2 swath of commercial  Lundin et al. (2017). In summary, the plantings started flowering slightly before almonds, peaked during almond bloom, and extended ca. 4 weeks beyond almond bloom.
Phacelia ciliata dominated the wildflower plantings in both years and provided 96% of the floral area (Lundin et al., 2017). Orchard understories were kept bare via chemical control of weeds and vegetation, so that almost no competing floral resources other than those provided by almond trees and the wildflower plantings were available to foraging bees. Some wild mustard, grasses, and other drought-tolerant pollen sources appeared during late almond bloom in fallow land surrounding the wildflower plantings, although no species overlap between weeds and managed plantings occurred (Lundin et al., 2017).
Osmia lignaria were released in six equally spaced 1.6-ha orchard regions within paired 16.2-ha blocks. These orchard regions, or "zones," varied in their proximity to the wildflower plantings and orchard edges across the three replicates, ranging from 80 m (zone A1) to 580 m (zone C3) ( Figure 1; Table 2). Within each zone, 40 nest boxes were distributed uniformly, in accordance with best management practices (Bosch & Kemp, 2001;Koh et al., 2018). Onehundred nesting tunnels were installed in each nest box to meet the recommended density of at least two nesting tunnels per female released (Bosch & Kemp, 2001). Nest boxes were folded corrugated plastic boxes (21.5 × 20 × 25.5 cm; Figure 2a nest-building. In 2016, water trucks delivered water to soil along F I G U R E 1 Aerial view of experimental orchards in Lost Hills, Kern Co., California. Three wildflower strips (indicated by white floral squares) were planted and maintained west of orchards through 2015 and 2016 almond bloom. Osmia lignaria were released into one of six zones (orange rectangles; lettered A1-C2), 1.6 ha in size, with associated nesting materials. "X" denotes areas of limited bee releases to evaluate pollen provision composition at extreme distances (800 m) from the wildflower plot (2016 only) Nest boxes were left in the orchard until late April in 2015 to allow larval development to late instars prior to their removal (Bosch & Kemp, 2001). In 2016, nest boxes were removed earlier (mid-April), due to the scheduled destruction of experimental orchards prior to harvest. Upon their removal from the orchard, nest boxes were held in a warehouse at ambient temperature until mid-August for both years.
During this time, offspring continued their development to adulthood in enclosed cocoons, the developmental stage at which they enter winter diapause. Upon reaching the adult stage, bees were introduced to a 4°C incubator for overwintering and held until the following spring.

| Osmia lignaria visitation
Osmia lignaria visitation to wildflower plantings and in fallow plots was monitored at five time points in 2015 and 2016 to confirm their use of planted floral resources. Time points selected were representative of peak almond bloom, late bloom, 1 week following bloom, 2-3 weeks following bloom and 4 weeks following bloom. These coincided with assessments of floral area at wildflower plots and within adjacent, fallow fields (as reported in Lundin et al., 2017). Osmia lignaria visitation to all plots wereas recorded along two 50-m transects at each time point, using methods described in Lundin et al. (2017) and detailed in Appendix S2. Visitation rates, summed across season, were compared using a generalized linear mixed model with a negative binomial distribution and log link function via PROC GLIMMIX in SAS 9.2 (SAS Institute, 2008) with treatment (wildflower or control) as a fixed factor and the replicate pair of wildflower and control plots as a random factor. Each year was analyzed separately.

| Reproduction and progeny outcomes
In August 2015 and 2016, total bee reproduction from the nesting tunnels was assessed using digital X-radiography (6-s exposure at 22 kVp, Faxitron 43804N; Faxitron Bioptics). The resulting images provided a full census of bee reproduction and progeny outcomes.
Metrics from the census included the number of cells with viable progeny, proportional mortality of cells, female-to-male sex ratio, average cells produced per tunnel, proportion of bees that died during development, proportion of cells occupied by parasites and/ or scavengers, and the proportion pollen ball, which are cells with

| Pollen provision composition
To

| Visitation
Osmia lignaria were observed frequently in the wildflower plantings throughout 2015 and 2016 during and after almond bloom ( Figure 3).

| Nesting
More nesting occurred, and more cells were produced, in 2016 ver-  (Table   S1). The highest overall nesting occurred at orchard edges (zones A1 and A2), with lower nesting in the orchard interior (Figures 4 and   5; Tables S4 and S5). Despite notable differences in the geographic distances separating "edge" zones A1 (80 m) and A2 (410 m) from the wildflower plantings, no significant difference in nesting was determined for 2015 or 2016 (Table S2). Likewise, no differences were detected for nesting over time for any combination of interior zones (B1-C2) for either year (Table S2).  Figure 4). Nesting at long-distance sites ( Figure S2, Appendix S3) did not continue after bloom, and considerably less nesting occurred overall.

| Reproduction and progeny outcomes
Rates of predation and parasitism were low across all zones, accounting for about 3% of the mortality observed in both years (Tables S4 and S5). During both years, more cells were recovered from A1 than in any other zone, while A2 generated the second highest number of live cells recovered (Figure 5a). In 2016, mortality was highest in zones further from the wildflower planting  (Table S3).

| Pollen provision composition
Osmia lignaria frequently used pollen from almond trees and the wildflower plantings across all zones (Table 3 and Table S6). By far, the most abundant pollen grains counted were Phacelia spp. and almond grains; any other pollen grains identified were pooled and presented as "other" in  len provisions, we confirmed that nesting females collected pollen from the wildflower plantings from up to 800 m away. Most nest completion occurred at orchard edges (zones A1 and A2), and significantly, more cells were recovered from A1 over A2, which is likely a consequence of closer proximity to wildflower plantings.

| D ISCUSS I ON
We conclude that planting and maintaining wildflowers is a promising strategy for integrating O. lignaria pollination into existing orchards, particularly in landscapes with limited alternative floral resources.
More progeny were recovered from nesting locations along orchard edges without adjacent forage plantings (zone A2; 410 m) than in the orchard interior, even when interior nests were proximally closer to the wildflower plantings (zones B1 and C1; 240 and 400 m, respectively). Even more remarkable was that no significant differences in nest completion over time were detected between A1 and A2 (both years), suggesting that, in this case, proximity to orchard edges is better predictors of managed O. lignaria nesting than relative proximity to alternative floral resources. The availability of open land with blooming wildflowers along orchard edges may have promoted bee nesting by mitigating some of consequences, for example, reduced floral diversity, of agricultural intensification in the area (Brodschneider & Carlsheim, 2010). Further, it is likely that almond blossoms and wildflowers along edges receive more sunlight and visibility, which could make them more attractive to foraging bees (Burgess, Kelly, Robertson, & Ladley, 2005). Additionally, it is well known that bees and other insects rely on visual landmarks for navigation across a given landscape (Collett & Collett, 2002). Possibly, abrupt orchard edges may have provided discrete visual cues for O. lignaria navigation, which promoted nesting near orchard edges.
It is also critical to remember that although nesting over time did not vary between A1 and A2, significantly more viable cells were recovered from A1, which had the closest proximity to the plantings. This Many of the significant trends observed from progeny outcomes were not reproducible between years. For example, female-to-male ratios were significantly higher in zones further away from the wildflower plantings in 2016, while no such effect was observed in 2015.
Female bees intrinsically have a higher pollination value, as they tend to live longer and need pollen and nectar resources for nest-building, so understanding factors driving sex ratios in this system is important. Overall, sex ratios favored more females in 2015 versus 2016.
Poorer overall nesting could bias sex ratios toward females, since female eggs are typically laid first in the nesting tunnel (Bosch & Kemp, 2001