Crop pollination is critically important for farmers and it’s a major revenue source for many beekeepers. Depending on the crop and location, farmers currently pay from ~$30-250 per hive and stocking densities are recommended at ~1-4 hives per acre for the field where they’re placed.
Given the importance of pollination, it’s equally important to assess whether stocking-density recommendations are sound. Most fields where pollination-dependent crops are grown exist within mixed agricultural landscapes where one farmer’s field may be next to another farmer’s field, which may be next to a non-crop field or maybe even a section of woods. Since honey bee foragers commonly travel a kilometer or more to collect pollen and nectar, should farmers who rent honey bees consider their surrounding landscape and hives beyond their fields?
In other words, should farmers consider landscape-scale hive density instead of field-scale hive density if their goal is to maximize crop pollination? This is the topic for the sixty-sixth Notes from the Lab, where I summarize “Landscape-level honey bee hive density, instead of field-level hive density, enhances honey bee visitation in blueberry,” written by Maxime Eeraerts and colleagues and published in the Landscape Ecology [2023].
For their study, Eeraerts and colleagues monitored 16 blueberry fields in western Washington during pollination in 2021 (see Figure 1). Fields ranged in size from 8.9 to 25.2 acres, with an average size of 16.8 acres. Three types of data were taken at each location. First, all honey bee hives at a focal field and within a 1 km radius of the field were located and mapped via information provided by farmers, beekeepers, extension agents, and personal experience. In total, 6,340 hives were located across the 16 sites, which ranged from 1.4 to 10.2 hives per acre at focal fields and 16 to 920 hives per 1 km radius.
Second, the composition of land within a 1 km radius of each location was determined using the National Land Cover Database and Cropland Data Layer. The amount of focal crop (blueberry) and semi-natural habitat (SNH; shrubland, extensive grassland, woody and herbaceous wetlands, and forest) were quantified. Semi-natural habitat was quantified because this habitat type is often associated with wild bee abundance during crop pollination, and it can also potentially influence honey bee visitation to crop flowers (more on that later). The amount of blueberry fields and SNH in the surrounding landscape ranged from 6.8-38.0% and 0.0-32.9%, respectively. Other land-use types in the surrounding landscapes were non-blueberry arable crops, pastures, and other areas (e.g., water bodies, urban areas).
Finally, surveys of honey bees and wild bees foraging at blueberry flowers were conducted in each field during peak bloom using timed transects. In addition, pollination was measured at each site by comparing the number of berries and berry weights between bushes with open flowers (exposed to pollinators), bagged flowers (pollinators excluded), and hand-pollinated flowers (exposed to pollinators and with supplemental pollen provided via a fine paintbrush).
So, what did they find? What shaped honey bee and wild bee visitation to blueberry flowers? For honey bees, landscape-scale hive density was the best predictor of visitation to blueberry flowers (see Figure 2B). Landscape-scale hive density was a better predictor of visitation to blueberry flowers compared to field-scale hive density, and it was tightly linked to the percentage of land used for blueberry production within a 1 km radius (see Figure 2A). Together, these results indicate that when additional hives are at nearby neighboring blueberry farms, honey bee visitation to blueberry flowers is increased at a focal farm, which may improve pollination.
Conversely, wild bee visitation to blueberry flowers decreased when a greater percentage of land was used for blueberry production within a 1 km radius (Figure 2D). For wild bees, the best predictor of visitation to blueberry flowers was the amount of semi-natural land within a 1 km radius (Figure 2C). This means having greater amounts of semi-natural habitat surrounding blueberry fields increases the number of wild pollinators, which may improve pollination.
Did honey bee or wild bee visitation predict blueberry pollination? Honey bee visitation did. Greater honey bee visitation was associated with greater seed number per berry, though neither fruit set nor berry weight were strongly predicted by honey bee or wild bee visitation. Honey bees were the dominant pollinator in the blueberry fields monitored in this study with 1,155 individuals caught in total, while only 54 wild bees were caught on blueberry blossoms. In other words, even though wild pollinators can be more effective pollinators that honey bees on a per-visit basis, there were nearly 20 times fewer wild bees compared to honey bees, which meant that honey bees were doing the vast majority of pollination.
Did semi-natural habitat within a 1 km radius reduce honey bee visitation to blueberry flowers? No. Some farmers worry about non-crop flowers potentially impacting crop pollination by drawing pollinators away from the crop, but this concern was not supported in this study. Instead, there were benefits of having semi-natural habitat in the landscape. First, semi-natural habitat was clearly beneficial for wild bees. Second, other studies have shown that honey bees benefit from semi-natural habitat for two reasons: Additional pollen sources can provide improved nutrition, and can dilute pesticide exposure during crop pollination (as long as the pollen and nectar from semi-natural habitat isn’t contaminated with pesticides).
Well this is interesting. What does it mean for farmers who want to maximize pollination in their blueberry fields? The most important insight from the authors’ paper is that hive densities surrounding an individual focal field matter. This means that an individual field may benefit from hives at neighboring farms that are within 1 km of that field. Alternatively, if there are few hives within 1 km of a focal field, the field is ….
