Limits...
A meta-analysis of predation risk effects on pollinator behaviour.

Romero GQ, Antiqueira PA, Koricheva J - PLoS ONE (2011)

Bottom Line: The results of our meta-analysis showed that predation risk significantly decreased flower visitation rates (by 36%) and time spent on flowers (by 51%) by pollinators.The strength of the predator effects depended neither on predator taxa and foraging mode (sit-and-wait or active hunters) nor on pollinator lifestyle (social vs. solitary).However, predator effects differed among pollinator taxa: predator presence reduced flower visitation rates and time spent on flowers by Squamata, Lepidoptera and Hymenoptera, but not by Diptera.

View Article: PubMed Central - PubMed

Affiliation: Departamento de Biologia Animal, Instituto de Biologia, Universidade Estadual de Campinas, Campinas, Brazil. gq_romero@yahoo.com.br

ABSTRACT
Flower-visiting animals are constantly under predation risk when foraging and hence might be expected to evolve behavioural adaptations to avoid predators. We reviewed the available published and unpublished data to assess the overall effects of predators on pollinator behaviour and to examine sources of variation in these effects. The results of our meta-analysis showed that predation risk significantly decreased flower visitation rates (by 36%) and time spent on flowers (by 51%) by pollinators. The strength of the predator effects depended neither on predator taxa and foraging mode (sit-and-wait or active hunters) nor on pollinator lifestyle (social vs. solitary). However, predator effects differed among pollinator taxa: predator presence reduced flower visitation rates and time spent on flowers by Squamata, Lepidoptera and Hymenoptera, but not by Diptera. Furthermore, larger pollinators showed weaker responses to predation risk, probably because they are more difficult to capture. Presence of live crab spiders on flowers had weaker effects on pollinator behaviour than presence of dead or artificial crab spiders or other objects (e.g. dead bees, spheres), suggesting that predator crypsis may be effective to some extent. These results add to a growing consensus on the importance of considering both predator and pollinator characteristics from a community perspective.

Show MeSH

Related in: MedlinePlus

Effects (mean ln R and 95% CI) of all live predators, live crab spiders and ants on (a) visitation rate and (b) time spent on flowers by several pollinator taxa.Sample sizes are indicated next to the error bars. SQU = Squamata, LEP = Lepidoptera, HYM = Hymenoptera, DIP = Diptera, COL = Coleoptera. N/A = data not available. Negative effects indicate decrease in visitation rate or time spent on flowers with predators present; effects are considered significant if 95% CI does not include 0.
© Copyright Policy
Related In: Results  -  Collection


getmorefigures.php?uid=PMC3113803&req=5

pone-0020689-g003: Effects (mean ln R and 95% CI) of all live predators, live crab spiders and ants on (a) visitation rate and (b) time spent on flowers by several pollinator taxa.Sample sizes are indicated next to the error bars. SQU = Squamata, LEP = Lepidoptera, HYM = Hymenoptera, DIP = Diptera, COL = Coleoptera. N/A = data not available. Negative effects indicate decrease in visitation rate or time spent on flowers with predators present; effects are considered significant if 95% CI does not include 0.

Mentions: Effects of live predators on pollinator visitation rates differed significantly depending on pollinator order (Qb = 13.52, P = 0.020, df = 3). Visitation rates of geckonid lizards (Squamata) were affected most (85% decrease), followed by Lepidoptera (46% decrease) and Hymenoptera (42% decrease). In contrast, visitation rates by Diptera were not significantly affected (Fig. 3a). Effects of live predators on the time spent on flowers did not differ among pollinator orders (Qb = 9.32, P = 0.122, df = 4) although only Squamata, Lepidoptera, and Hymenoptera showed significantly reduced time spent on flowers in the presence of live predators, whereas Diptera and Coleoptera showed no significant response (Fig. 3b). Since pollinator order was not independent from predator taxa (e.g., all studies on Squamata examined effects of ants only), we ran separate analyses for the two most studied groups of predators, ants and crab spiders. While presence of crab spiders significantly decreased visitation rate by both Diptera and Hymenoptera (Qb<0.001, P = 0.98, df = 1), ants affected visitation rate and time spent on flowers by Hymenoptera, but not by Diptera (Fig. 3a,b; visitation rate: Qb = 9.62, P = 0.041, df = 2; time spent: Qb = 24.2, P = 0.002, df = 2).


A meta-analysis of predation risk effects on pollinator behaviour.

Romero GQ, Antiqueira PA, Koricheva J - PLoS ONE (2011)

Effects (mean ln R and 95% CI) of all live predators, live crab spiders and ants on (a) visitation rate and (b) time spent on flowers by several pollinator taxa.Sample sizes are indicated next to the error bars. SQU = Squamata, LEP = Lepidoptera, HYM = Hymenoptera, DIP = Diptera, COL = Coleoptera. N/A = data not available. Negative effects indicate decrease in visitation rate or time spent on flowers with predators present; effects are considered significant if 95% CI does not include 0.
© Copyright Policy
Related In: Results  -  Collection

Show All Figures
getmorefigures.php?uid=PMC3113803&req=5

pone-0020689-g003: Effects (mean ln R and 95% CI) of all live predators, live crab spiders and ants on (a) visitation rate and (b) time spent on flowers by several pollinator taxa.Sample sizes are indicated next to the error bars. SQU = Squamata, LEP = Lepidoptera, HYM = Hymenoptera, DIP = Diptera, COL = Coleoptera. N/A = data not available. Negative effects indicate decrease in visitation rate or time spent on flowers with predators present; effects are considered significant if 95% CI does not include 0.
Mentions: Effects of live predators on pollinator visitation rates differed significantly depending on pollinator order (Qb = 13.52, P = 0.020, df = 3). Visitation rates of geckonid lizards (Squamata) were affected most (85% decrease), followed by Lepidoptera (46% decrease) and Hymenoptera (42% decrease). In contrast, visitation rates by Diptera were not significantly affected (Fig. 3a). Effects of live predators on the time spent on flowers did not differ among pollinator orders (Qb = 9.32, P = 0.122, df = 4) although only Squamata, Lepidoptera, and Hymenoptera showed significantly reduced time spent on flowers in the presence of live predators, whereas Diptera and Coleoptera showed no significant response (Fig. 3b). Since pollinator order was not independent from predator taxa (e.g., all studies on Squamata examined effects of ants only), we ran separate analyses for the two most studied groups of predators, ants and crab spiders. While presence of crab spiders significantly decreased visitation rate by both Diptera and Hymenoptera (Qb<0.001, P = 0.98, df = 1), ants affected visitation rate and time spent on flowers by Hymenoptera, but not by Diptera (Fig. 3a,b; visitation rate: Qb = 9.62, P = 0.041, df = 2; time spent: Qb = 24.2, P = 0.002, df = 2).

Bottom Line: The results of our meta-analysis showed that predation risk significantly decreased flower visitation rates (by 36%) and time spent on flowers (by 51%) by pollinators.The strength of the predator effects depended neither on predator taxa and foraging mode (sit-and-wait or active hunters) nor on pollinator lifestyle (social vs. solitary).However, predator effects differed among pollinator taxa: predator presence reduced flower visitation rates and time spent on flowers by Squamata, Lepidoptera and Hymenoptera, but not by Diptera.

View Article: PubMed Central - PubMed

Affiliation: Departamento de Biologia Animal, Instituto de Biologia, Universidade Estadual de Campinas, Campinas, Brazil. gq_romero@yahoo.com.br

ABSTRACT
Flower-visiting animals are constantly under predation risk when foraging and hence might be expected to evolve behavioural adaptations to avoid predators. We reviewed the available published and unpublished data to assess the overall effects of predators on pollinator behaviour and to examine sources of variation in these effects. The results of our meta-analysis showed that predation risk significantly decreased flower visitation rates (by 36%) and time spent on flowers (by 51%) by pollinators. The strength of the predator effects depended neither on predator taxa and foraging mode (sit-and-wait or active hunters) nor on pollinator lifestyle (social vs. solitary). However, predator effects differed among pollinator taxa: predator presence reduced flower visitation rates and time spent on flowers by Squamata, Lepidoptera and Hymenoptera, but not by Diptera. Furthermore, larger pollinators showed weaker responses to predation risk, probably because they are more difficult to capture. Presence of live crab spiders on flowers had weaker effects on pollinator behaviour than presence of dead or artificial crab spiders or other objects (e.g. dead bees, spheres), suggesting that predator crypsis may be effective to some extent. These results add to a growing consensus on the importance of considering both predator and pollinator characteristics from a community perspective.

Show MeSH
Related in: MedlinePlus