I promised I was going to talk about things other than nuclear power, and lo: so it shall be written, so it shall be done. Here, I present the first half of a paper I produced earlier this year, dealing with the fate of bumblebees. I'm quite fond of the fuzzy little beasts.
While the mainstream press has been primarily focused on the precipitous decline of European honeybees (Apis melliferra) and the emergence of Colony Collapse Disorder, wild native bees have also been suffering significant losses in numbers and range ( reviewed by Spivak et al. 2011, Williams and Osborne 2009). There are some 4,000 species of bees native to North America, including bumblebees (Bombus spp), leafcutter bees (Megachile spp), mining bees (Andrena spp), mason bees (Osmia spp), and others, many of which are important pollinators of New World crops (Spivak et al. 2011). I will focus primarily on Bombus species, as these are among the most studied of wild bees, and include species that are reared commercially worldwide for use as pollinators of plants such as tomatoes (Solanum lycopersicum) and peppers (Capsicum spp) in greenhouses.
Across North America and Europe, wild bee losses have not been consistent, and the factors causing them are uncertain. Some species appear to be more or less unaffected, while others have suffered large losses, possibly due to species-specific ecological processes (reviewed by Murray et al. 2009). Three historically common, closely related North American Bombus species (B. occidentalis, B. affinis, B. terricola) have been shown to have large reductions in their home range (23-87%) and relative abundance, while other species have remained stable in their distribution and abundance (Cameron et al. 2011).
Potential Sources of Decline
The cause of bumblebee declines remains unclear, though certain factors are implicated by current research: (1) importation of parasites carried by commercially reared bumblebees, (2) sublethal effects of exposure to pesticides, (3) loss of appropriate habitat, and (4) the phenology and foraging patterns of the bees themselves and how they interact with the previous factors (Williams et al. 2009; Williams and Osborne 2009; Winfree 2010).
Global movement of bumblebee parasites, and subsequent infection—a phenomenon known as pathogen spillover, where native bees become infected by pathogens carried by imported bees—has been reported in Japan (Goka et al. 2006), where a European strain of the tracheal mite Locustacarus buchneri showed evidence of having infected native Japanese bees. The Japanese variety of the mite may also have been exported to Europe with Japanese bees (B. ingitus) that were commercialized, and a global exchange of genes for the parasite maybe underway. The details and extent of this genetic exchange are currently unclear, due to a poor understanding of the life cycle of the mite, and uncertainties in the degree of relatedness between the Asian and European populations.
Evidence from studies in Canada show that the bumblebee pathogens Crithidia bombi (a protozoan) and Nosema bombi (a microsporidium) are transmitted from imported bees used in greenhouses (primarily Bombus impatiens and B. occidentalis) to native bumblebees. Native bees show high infection rates in close proximity to greenhouses that import bumblebees (6.3 – 18.4% for N. bombi, and 5.3-75% for C. bombi, depending on species of bee), whereas areas geographically removed from greenhouses showed no signs of infection by C. bombi and lower rates of infection by N. bombi (< 5%) (Colla et al. 2006). These results stand in strong contrast to those from Massachusetts, where C. bombi was very prevalent among native bees at two sites that had no commercially reared bees nearby ( 45% and 65%), and N. bombi was also present at slightly higher rates (14% and 5%) (Gillespie 2010). Rates varied by bee emergence, with earlier emerging bees having lower infection rates than those that emerged later.
Attempts have been made to model the effects of pathogen spillover and results suggest that parasites that can be spread by contact with flowers visited by infected bees (including C. bombi) would spread rapidly near greenhouses hosting infected bees, and would expand in range at an increasing rate over time (Otterstatter and Thomson 2008). The pathogenic ‘wave’ of infections was not seen in Canada, potentially due to the short lifecycles of local bee species (constrained by the Canadian summer), or because the study design did not sample bees at a small enough temporal scale, and only surveyed active bees (infected bees tend to become sluggish and may not leave their nest). This model does present a potential explanation for the high infection rates of bees in Massachusetts, however, and the most precipitously declining North American species show higher infection rates for N. bombi than stable species (Cameron et al. 2011).
There has been some concern and controversy regarding the impact of pesticides on honeybees, in particular neonicotinoid pesticides that have entered the market in the last several decades. Much of the controversy has involved a highly critical critique of the neonicotinoid pesticide clothianidin by EPA scientists that appears to have been suppressed by the agency (Philpott 2010). The impact of these pesticides on bees generally is not yet clear, and there is significant disagreement within the literature regarding the nature and degree of the effects of neonicotinoid pesticides on bumblebees. There is also not a great deal of consistency in methodology, making the direct comparison of studies difficult. Although the research results from pesticide studies are somewhat conflicting and obviously incomplete in scope, several independent studies suggest that bees may experience negative impacts (primarily a decrease in foraging efficiency) due to sublethal exposure.
Bumblebees are exposed to neonicotinoids when they are expressed in pollen and nectar in plants grown from treated seed. Used at the registered dose for sunflowers (Helianthus annuus), imidacloprid showed no significant negative effects on B. terrestris when bees gathered 98% of their nectar and ≈25% of their pollen from the treated sunflowers, though residues of the pesticide were detectable in their honey ( 0.01 mg/kg; Tasei et al. 2001). Unfortunately, the actual concentrations present in the nectar and pollen of the sunflowers were not determined, and thus there was no quantified level of exposure.
Similarly, clothianidin was found not to have any observable negative impacts on B. impatiens at doses up to 36 ppb (in pollen and sugar water), a dosage believed to be higher than levels found in seed-treated crops (Franklin et al. 2004). However, at a 30 ppb dosage of imidacloprid (again believed to be higher than levels found in treated fields), B. impatiens and B. occidentalis workers took 42.6% longer to access artificial flowers while foraging, and foraged at ≈75% the rate of bees in the control group (Morandin and Winston 2003). In addition to this, it was more recently determined that bumblebees forced to actively forage a short distance from their nest for the duration of the study period were significantly more sensitive to the impacts of imidacloprid. The no-observed-effect concentration (NOEC) for B. terrestris that did not forage for food was found to be 20 ppb (similar to the above studies) with an LC50 (the exposure at which 50% of exposed bees die) of 69 ppb, but the LC50 for foraging bees was lowered to 20 ppb, impairment of foraging at occurred at 3.7 ppb, and the NOEC was below 2.5 ppb (Mommaerts et al. 2010b). These results suggest that foraging bumblebees with chronic, sublethal exposure to neonicotinoids are significantly more sensitive to the effects of these compounds than previously believed.
As contact pesticides, clothianidin and imidacloprid were found to be highly toxic to B. impatiens, more so than two common foliar pesticides, deltamethrin—a pyrethroid pesticide—and spinosad—a pesticide derived from soil bacteria—(clothianidin 23x more toxic than spinosad), though these were also toxic to bees (Scott-Dupree et al. 2009). Effects on contact exposure were widely variable by bee species and by pesticide, with Megachile rotundata and Osmia lignaria orders of magnitude more sensitive than Bombus impatiens. Bees gathering pollen and nectar in fields are not likely to come into direct contact with neonicotinoid pesticides, however. Thus, although the foliar pesticides were less toxic, bees are more likely to come into contact with them at high concentrations, and suffer acute effects.
B. impatiens larva develop impaired foraging as adults when exposed to low levels (0.8 mg/kg) of spinosad, although adults did not suffer any lethal effects at this dose, and colony health was only marginally impacted (small decrease in worker weight, and a decrease in foraging efficiency of workers; Morandin et al. 2005). B. terrestris also appear to be sensitive to certain strains of toxins produced by transgenic plants carrying the Cry1C and CryID Bacillus thuringiensis genes, leading to mortality and decreased colony performance due to lower egg production (Mommaerts et al. 2010a).
Much work remains to be done in establishing consistent protocols for testing bumblebee sensitivity to pesticides, as well as in determining how low levels of exposure may influence parasitism, success when overwintering, and other important ecological phenomenon.
Many bumblebees are generalists in terms of foraging for pollen and nectar. However, even generalist species may have preferences, and flower species may evolve defenses against pollen theft that require specific physiological adaptations to make use of their pollen. Larva of two closely related generalist Osmia species, O. bicornis and O. cornuta were raised on pollen from one of four flower types (Echium, Ranunculus, Sinapis, and Tanacetum), and each was unable to develop on two of these pollens: Both failed on Tanacetum pollen, O. bicornis failed with Echium pollen, and O. cornuta failed on Ranunculus pollen (Sedivy et al. 2011). That these preferences vary by species of bee, even in closely related species, may make it difficult to determine what a given species needs to be successful.
In agricultural landscapes, the short-lived abundance of food provided by mass flowering crops such as oilseed rape (Brassica napus) can distort normal pollinator-plant interactions, increasing the abundance of some bees over others, depending on their ability to successfully take advantage of resource pulses. With Brassica napus, generalist short-tongued bumblebees are favored over more specialized long-tongued bees, when diverse, semi-natural foraging resources are scarce (Diekötter et al. 2010).
Contrary to this study, in Ireland and the UK, research showed no relationship between dietary breadth and declines in abundance and range of bumblebee species. Rather, these declines are attributed to the loss of habitat due to agricultural land-use changes (earlier cutting of fields for silage, rather than making hay), and development of open space (Fitzpatrick et al. 2007, Williams 2005). These studies are not able to adequately explain why some species of bumblebees are in decline, and others are not, which may be related to dietary preferences. This idea is supported in part by the finding that declining species worldwide tended to have a preference for deep flowers (Williams et al. 2009). The overarching change of agricultural practices from those involving crop rotations and the planting of nitrogen-fixing species such as alfalfa and clovers (to maintain the productivity of agricultural soils) to a dependence on chemical fertilizers may also be playing a role in the loss of many bee species (Goulson et al. 2010, Spivak et al. 2011).
Appropriate nesting sites are also vital for bumblebee survival. Many bumblebee species nest on the ground or in dense clumps of vegetation, and appropriate nest sites also need to be within foraging range of a reliable food supply. Some British bumblebees had increased rates of nest survival if gardens were a part of the landscape within 750-1000 m; others experienced similar effects when grasslands were within 250-500 m (Goulson et al. 2010). The lack of appropriate nesting sites may cause significant congener competition for the few available sites, and reduce survival for some bumblebee species (Lye et al. 2009).
Bee Phenology & Foraging Habits
The timing of emergence for bumblebees appears to be one of the strongest factors influencing species declines, with most declining species tending to emerge later in the season. Late emergers have a shorter period of time to nest and reproduce, and a scarcity of food plants may be reducing their success rate (Williams et al. 2009). They may also be facing competition for nesting sites from earlier emerging bees (Lye et al. 2009), or other as-yet-unknown factors.
The distance a worker or queen is willing to travel to forage for food also appears to have a strong influence on species success. Bumblebees such as B. terrestris frequently forage at distances in excess of 1 km from their nest, making landscape make-up less significant, while other bees prefer to forage at shorter distances from their nests, making them more dependent on local food sources (Diekötter et al. 2010, Walther-Hellwig and Frankl 2000)