ArticlePDF Available

Temporally dependent pollinator competition and facilitation with mass flowering crops affects yield in co-blooming crops


Temporally dependent pollinator competition and facilitation with mass flowering crops affects yield in co-blooming crops

Abstract and Figures

One of the greatest challenges in sustainable agricultural production is managing ecosystem services, such as pollination, in ways that maximize crop yields. Most efforts to increase services by wild pollinators focus on management of natural habitats surrounding farms or non-crop habitats within farms. However, mass flowering crops create resource pulses that may be important determinants of pollinator dynamics. Mass bloom attracts pollinators and it is unclear how this affects the pollination and yields of other co-blooming crops. We investigated the effects of mass flowering apple on the pollinator community and yield of co-blooming strawberry on farms spanning a gradient in cover of apple orchards in the landscape. The effect of mass flowering apple on strawberry was dependent on the stage of apple bloom. During early and peak apple bloom, pollinator abundance and yield were reduced in landscapes with high cover of apple orchards. Following peak apple bloom, pollinator abundance was greater on farms with high apple cover and corresponded with increased yields on these farms. Spatial and temporal overlap between mass flowering and co-blooming crops alters the strength and direction of these dynamics and suggests that yields can be optimized by designing agricultural systems that avoid competition while maximizing facilitation.
Content may be subject to copyright.
Scientific RepoRts | 7:45296 | DOI: 10.1038/srep45296
Temporally dependent pollinator
competition and facilitation with
mass owering crops aects yield in
co-blooming crops
Heather Grab1,2, Eleanor J. Blitzer3, Bryan Danforth1, Greg Loeb2 & Katja Poveda1
One of the greatest challenges in sustainable agricultural production is managing ecosystem services,
such as pollination, in ways that maximize crop yields. Most eorts to increase services by wild
pollinators focus on management of natural habitats surrounding farms or non-crop habitats within
farms. However, mass owering crops create resource pulses that may be important determinants of
pollinator dynamics. Mass bloom attracts pollinators and it is unclear how this aects the pollination
and yields of other co-blooming crops. We investigated the eects of mass owering apple on the
pollinator community and yield of co-blooming strawberry on farms spanning a gradient in cover of
apple orchards in the landscape. The eect of mass owering apple on strawberry was dependent on the
stage of apple bloom. During early and peak apple bloom, pollinator abundance and yield were reduced
in landscapes with high cover of apple orchards. Following peak apple bloom, pollinator abundance was
greater on farms with high apple cover and corresponded with increased yields on these farms. Spatial
and temporal overlap between mass owering and co-blooming crops alters the strength and direction
of these dynamics and suggests that yields can be optimized by designing agricultural systems that
avoid competition while maximizing facilitation.
Increasing consumption driven by a growing global population and demands for more varied and resource inten-
sive diets has placed unparalleled strain on our agricultural production systems and natural resources. Current
agricultural practices rely on fossil fuels, agrochemicals and conversion of new agricultural lands. Yet, yield gains
produced through these practices have plateaued in recent years1 and have come at the cost of increasing green-
house gas emissions, degradation of water quality, widespread pollution, pesticide resistance and unprecedented
biodiversity loss. An alternative solution to meet the planet’s growing needs is ecological intensication, increas-
ing production on existing farmlands in ways that causes less harm on the environment through the replace-
ment of anthropogenic inputs with ecosystem services management2,3. Manipulating and regulating the biotic
interactions underpinning the provisioning of ecosystem services by increasing the structural diversity of agro-
ecosystems had been demonstrated to improve crop yields4–6. In order to implement diversication strategies
successfully, it is critical to understand whether agricultural habitats themselves may act as sources of ecosystem
services or whether diversication may lead to competition for services among crops. Certain crops may have a
disproportionate eect on the ow of ecosystem services due to the large pulse of resources they provide7,8, and
it is essential to understand the eects of these crops on ecosystem service dynamics in order to develop eective
management strategies that can be directly implemented by land managers.
Crops that are grown on large scales and bloom en masse create large pulses of resources that have substantial
impacts on communities of ecosystem service providers. ese dynamics are particularly relevant for pollinator
dependent crops given the dramatic increase in the area planted to these crops9 and their importance for human
nutrition10,11. Pulses in oral resources associated with mass blooming of crops are known to alter pollinator
abundances and visitation rates in nearby crops and natural habitats12–15, which are likely to have direct impacts
on crop yields16. Mass owering crops can increase pollinator ospring production17 and pollinator densities
1Department of Entomology, Cornell University, Ithaca, New York 14853, United States. 2Department of
Entomology, New York State Agricultural Experiment Station, Cornell University, Geneva, New York 14456, United
States. 3Department of Biology, Carroll College, 295 S. Harrison Helena, MT 56901, USA. Correspondence and
requests for materials should be addressed to H.G. (email:
Received: 18 October 2016
Accepted: 23 February 2017
Published: 27 March 2017
Scientific RepoRts | 7:45296 | DOI: 10.1038/srep45296
following mass bloom18,19, particularly for solitary, univoltine bees20 for which the bloom of a single crop may
represent nearly the entire span of their adult foraging activity. erefore, mass owering crops may facilitate
pollination of co-blooming crops when pollinators attracted and supported by the mass blooming crop spill over
into the co-blooming crop, augmenting oral visitation and crop pollination21 (Fig.1). Alternatively, when polli-
nators are limited, as is common in simplied agricultural systems22, plants with high overlap in their pollinator
community may compete for visits from shared pollinators23 (Fig.1).
Little is known about competitive or facilitative interactions between pollinator dependent crops with respect
to pollinators and pollinator services; however, we expect that these interactions are ubiquitous in agricultural
landscapes. ey only require that crops have overlap in their pollinator community though they may bloom in
dierent seasons21 or even in dierent years20. e likelihood of the interaction resulting in either facilitation or
competition depends not only on the degree of overlap in the pollinator community15 but also on the temporal
overlap in bloom between crops13,21,24. Indeed, the two studies available on the eects of mass owering crops on
wild plants have found that mass owering crops can either reduce25 or enhance pollination13 in co-blooming
plants in nearby natural habitats.
At the landscape scale, greater abundance and diversity of bees associated with natural and semi-natural habi-
tats22 may buer against local competition or facilitation eects. For example, in landscapes with high amounts of
natural habitat, competition between co-blooming crops may be lower than expected14. In this case, although bees
are drawn to the mass blooming crop, the number of bees moving from natural habitats into the co-blooming
crop may still be sucient to provide adequate pollination services7. Alternatively, proximity to natural habitats
may reduce facilitation when bees move from mass blooming crops to alternative forage in natural habitats rather
than the co-blooming crop.
Despite the potential importance of pulsed resource dynamics for crop pollination and associated yield, we
are not aware of any studies that have evaluated the eects of mass owering crops on the yield of another crop.
Greater understanding of spatial and temporal factors that shi the balance between competition and facilitation
will allow for management practices that maximize crop yields under the pulsed resource dynamics characteristic
of agroecosystems.
In this study we investigate the potential for pollinator mediated competition or facilitation in two econom-
ically important crops: apple (Malus domestica), a mass owering crop, and strawberry (Fragaria x ananassa
Duch.) in central New York, USA. In this region apple and strawberry have a staggered but overlapping bloom
period. e two crops are both members of the family Rosacea, and thus expected to have high overlap in their
pollinator faunas26,27. Furthermore, the community of bees visiting both apple and strawberry is dominated by
early spring, ground-nesting, univoltine bees in the genus Andrena28–30. e high potential for community over-
lap in pollinators between apple and strawberry make these two crops an ideal study system in which to under-
stand the potential for pollinator-mediated interactions among crops.
We hypothesized that the impact of apple on strawberry pollination may vary temporally, with facilitation and
competition occurring in the same system but at dierent stages of apple bloom. Additionally, we hypothesized
that proximity to natural habitats would moderate these eects and predicted that sites in close proximity to nat-
ural habitat would have greater bee abundance and experience both reduced competitive and facilitative eects.
Both apple and strawberry are economically important crops in the United States, with total apple production
at 327,000 acres and strawberry production at 61,000 acres (USDA NASS, 2013). In New York State, the second
largest apple-producing region in the US (USDA NASS, 2013), it is common for farms to grow apples plus a range
of other small fruit crops including strawberry.
Study Sites. Studies were carried out in the spring of 2013 in the Finger Lakes Region (42°26 N, 76°30 W)
of New York, USA. e study area is characterized by a diversity of agricultural uses, including dairy, row crop,
tree fruits and vegetables with natural and semi-natural areas of deciduous forest, small woodlots and old eld
Figure 1. A simple conceptual model for the consequences of pollinator sharing between a mass owering
and co-blooming crop. (A) Pollinator spillover from co-blooming crops to mass owering crops during mass
owering results in competition for pollinators and a decrease in co-blooming crop yields. (B) Pollinator
spillover into co-blooming crops following bloom of mass owering crops results in facilitation of pollinator
visitation to co-blooming crops and an increase in co-blooming crop yields.
Scientific RepoRts | 7:45296 | DOI: 10.1038/srep45296
dispersed throughout. A total of 35 farms growing apple, strawberry or both were identied. All farms were
used to estimate pollinator community similarity and a subset of 13 farms, all growing strawberry but with a
gradient in the proportion of apple orchard cover in the surrounding landscape (0–37%), were used in further
experiments. Focal strawberry elds on each of these 13 farms were selected. e landscape surrounding the focal
strawberry eld was characterized within a 1 km radius using the 2013 National Agricultural Statistics Service
Cropland Data Layer for New York31 in ArcGIS 10.1. Using these maps we estimated the proportion of land
in agricultural uses (all crop categories including forage and pasture), natural and semi-natural areas (forest,
wetlands, shrub lands, meadows, and fallow), and apple orchards. Apple orchards owered between May 3 and
June 5, 2013, with bloom initiation and duration varying across farms depending on local microclimate and
apple variety. Most farms cultivate a number of apple varieties to ensure pollination success, as apple is varietally
self-incompatible. Measurements occurred between May 6 and May 9 for early apple bloom, between May 14 and
May 17 for peak apple bloom and between May 31 and June 3 for late apple bloom. Apple owering phenology in
2013 would be described as “typical” based on historical data on apple owering in upstate NY32. In the early stage
of apple bloom the percentage owers open of total owers present, estimated by counting the number of open
owers per cluster on randomly selected trees, averaged 26.6% (± 5.4 SE). During the same period, strawberry
bloom had initiated only at four sites (with 16.9% ± 11.5 SE owers open). At peak apple bloom, owering inten-
sity averaged 54.8% (± 5.8 SE) compared to strawberry bloom at 23.2% (± 7.3 SE). Apple bloom intensity during
the late owering stage averaged only 10.9% (± 3.5 SE) while strawberry bloom was 37.5% (± 6.7 SE).
We quantied apple mass owering using a mass owering index. e index describes the total amount of
apple owering within the surrounding landscape and is calculated as the percent apple cover in a 1 km radius
around the focal strawberry eld multiplied by the percent of open apple owers in adjacent orchards (if present).
us, the highest values of the mass owering index indicate high abundance of apple owers locally and within
the landscape.
Pollinator Community. To estimate similarity in the apple and strawberry pollinator communities, bees
were collected using sweep netting during four 15-minute surveys along 100 m transects in apple orchards and
strawberry elds during the peak bloom of each crop. Surveys were conducted between 10:00 and 15:30 on sunny
days with temperature above 16 °C. Bees were identied to species using published revisions33–38 and online keys
( as well as expertly identied reference materials maintained in the Cornell University Insect
Collection (
In order to understand how mass owering apple impacts bee visitation to strawberry, we estimated the abun-
dance and diversity of bees visiting strawberries over the course of the apple bloom within each focal straw-
berry eld and adjacent to the nearest natural or semi-natural habitats. Distances between strawberry elds and
semi-natural habitats on a farm ranged from 20 to 300 m (mean = 46 ± 9 m). Arrays of four white pan traps were
placed at 3 m intervals on transects 2 m from the edge of each focal strawberry eld and semi-natural habitat
during three sampling periods corresponding to the early, peak and late stages of apple bloom. White pan traps
were used as they collect a greater number of bees but maintain a similar community composition to sweep net
sampling compared to other trap colors (H. Grab, unpublished data). Sampling periods were approximately one
week apart, varying based on local microclimatic conditions, beginning on May 6th and ending June 3rd 2013. e
intensity of strawberry and apple bloom was recorded when the arrays were set out and when they were collected.
Bloom intensity was estimated as the percentage owers open of total owers present including senesced owers
and owers in bud stage in the orchards or elds. ese data were then averaged in order to estimate bloom inten-
sity during each stage. Pan traps were collected aer 72 hours and the contents were sorted and identied to spe-
cies. Sampling rarefaction curves for species richness are available in the SupplementaryInformation (FigsS1–3).
Strawberry pollination. To assess the eects of apple mass owering at the landscape scale on strawberry
pollination and fruit set, we measured the pollination rates of sentinel strawberry plants placed within the focal
strawberry eld and adjacent to semi-natural habitat at each farm. Use of sentinel plants allowed us to maintain
consistent strawberry bloom density during each stage of apple bloom and to control for abiotic factors, including
soil and microclimate that could aect yield. During the three periods corresponding to the early, peak and late
stages of apple bloom, we placed 10 individually potted strawberry plants (variety Evie 2) in the same transects
used for pollinator sampling described above. Strawberry plants have one primary ower, two secondary owers
and up to four tertiary owers per inorescence. e number of achenes is greatest on primary fruit and decreases
in subsequent owers. Only primary and secondary fruits were used to estimate yield, as they are the only fruits
usually considered marketable. ese owers were exposed to visitors for 72 hours but on half of the plants, ow-
ers received supplementary pollen applied with a paintbrush to the stigmas. ese hand-pollinated fruits, when
compared to open-pollinated fruits, allowed us to estimate the relative contribution of the pollinator community
to yield while controlling for environmental factors such as microclimate, which may have varied slightly across
the study region. We collected the sentinel plants aer 72 hours and maintained them in a greenhouse chamber
while the fruits developed. Fruits were harvested daily when ripe and weighed. In strawberries, fruit weight can
provide an accurate estimate of pollination rate, as strawberries are an aggregate accessory fruit comprised of as
many as 300 individual achenes39. Each achene must be fertilized in order for the surrounding tissue to develop.
Hence, the weight of a fruit is highly correlated with the number of pollinated achenes40 and an average of four
pollinator visits per ower is required to achieve full pollination and maximum fruit weight41. Only fruits with
a high percentage of fertilized achenes will develop without major malformations that reduce overall yield and
Statistical Analyses. e eect of apple mass owering on bees was assessed using mixed eects models
in the R package “nlme42 with either the dependent variable of bee abundance (total number of bees collected
Scientific RepoRts | 7:45296 | DOI: 10.1038/srep45296
during a sampling stage at each site) or bee diversity (number of bee species collected during a sampling stage
at each site) in separate models. In both models the xed eects included natural habitat proximity (adjacent
or distant from nearest natural area), apple mass owering index, the percentage of strawberry bloom, apple
bloom stage (early, peak, or late), and all possible interactions between the xed eects. Mass owering index
was log(x + 1) transformed to meet distributional assumptions. Farm was included as a random eect in the
model describing bee abundance. In the nal model describing bee diversity, natural habitat proximity nested
within farm was included as a random eect because proximity was removed as a xed eect following model
We used linear mixed eects models to assess the relationship between bee abundance and diversity and the
average weight of strawberry fruits. Models were t separately for bee abundance and diversity as xed eects
along with pollination treatment, apple bloom stage and location. Following model simplication the nal models
retained only the main eects of abundance or diversity. To account for non-independence of samples and the
nested experimental design structure, random eects in the nal model included the nested eects of stage within
pollination treatment within the natural habitat proximity variable within farm.
In order to determine the indirect eects of the apple mass owering index on strawberry fruit weight during
each of the apple bloom stages, we used separate mixed eects models with fruit weight as the response variable.
e predictor variables included pollination treatment, the mass owering index, and all possible interactions
between the xed eects. Fruit order (primary or secondary) nested within the natural habitat proximity variable
nested within farm was included as a random eect in each model to account for the nested sampling design. In
the model describing the eects during peak bloom, weight was log transformed to meet distributional normality
All models were computed in R v. 3.2.343. Minimum adequate models were selected using backwards-stepwise
selection, eliminating predictor variable with p values < 0.1. Once minimum adequate models were identied, the
anova function was used to assess signicance of each factor by obtaining F and p values. In all models apple mass
owering index values were log10(x + 1) transformed to account for overdispersion due to some farms having very
high percentages of apple cover.
Community Similarity. Using bees collected in sweep-net transects in apple orchards (n = 18 orchards,
abundance = 776, species = 51) and strawberry elds (n = 17 elds, abundance = 994, spe cies = 60) during
peak bloom of each crop, we compared the overlap in pollinator communities of each crop. We found that apple
and strawberry share 31 of the 79 pollinator species collected including the most abundant pollinators of each
(Fig.2). In this region, honey bees, Apis mellifera, are oen brought into orchards for apple pollination but not
Figure 2. Pollinator communities of apple and strawberry in the Finger Lakes region of New York State.
Node size indicates total abundance and edge widths represent relative abundance in each crop. Yellow = shared,
Blue = unshared.
Scientific RepoRts | 7:45296 | DOI: 10.1038/srep45296
for strawberry. ese managed honey bee colonies are moved out of apple orchards following peak apple bloom;
therefore, we present honey bee abundance separately from the wild pollinator community. In apple orchards,
honey bees comprised 48% of the pollinator community; while in strawberry, honey bees comprised only 1.3%
of the bees collected. Because our estimates of community overlap are based on collections from geographically
separated locations, they represent a conservative measure of the overlap in apple and strawberry pollinators that
is likely to occur within a farm.
Bee Response to Mass Flowering Apple. There was a significant effect of apple mass flowering on
the abundance and diversity of bees collected in strawberry elds and adjacent semi-natural habitats that was
dependent on the bloom stage (Table1 and Fig.3). When further exploring the interaction between stage and
the mass owering index (Table1) we found that abundance and diversity of bees collected near the sentinel
plants were negatively aected by mass owering during both early and peak apple bloom and positively aected
by mass flowering during late apple bloom (Fig.3, TableS1). Bee community composition was marginally
eected by the stage of apple bloom (FigureS4). As expected, bee abundance was marginally higher adjacent to
semi-natural areas (mean = 16.14, SE = 2.47) compared to strawberry elds with no adjacent semi-natural habi-
tats (mean = 9.01, SE = 2.21 Table1). However, natural habitat proximity did not interact with either stage or the
mass owering index suggesting that the proximity to natural habitat did not alter the impact of mass owering
on the pollinator community. Although species richness was not dierent between strawberry elds and natural
habitats, the composition of bee communities diered between locations (FigureS5). e local intensity of straw-
berry bloom did not impact bee abundance or diversity at any stage, and was therefore removed from all models.
Var i a bl e df F P
Bee Abundance
Stage 2,48 2.10 0.133
Natural Habitat Proximity 1,48 3.86 0.055
Mass Flowering Index 1,48 9.15 0.004
Stage X Index 2,48 8.41 0.001
Bee Species Richness
Stage 2,49 0.94 0.394
Mass Flowering Index 1,49 2.14 0.149
Stage X Index 2,49 6.80 0.003
Table 1. Minimum adequate models describing local and landscape scale eects on abundance and species
richness of bees in strawberry elds sampled during early, peak and late apple bloom from sites located
adjacent or distant from natural areas.
Figure 3. Average bee abundance and species richness during early, peak and late apple bloom in relation
to the mass owering index which describes the total amount of apple owering within the surrounding
Scientific RepoRts | 7:45296 | DOI: 10.1038/srep45296
Strawberry Yield. e average weight of strawberry fruits from sentinel plants increased with both greater
bee abundance (F1,13 = 5.72 p = 0.03) and diversity (F1,13 = 24.22 p = <0.001) (Fig.4). Although pollinator abun-
dances were greater near to natural habitats, fruit yield did not vary with natural habitat proximity. Similar to the
eects observed on the pollinator community, we found the eects of apple mass owering on strawberry fruit
weight diered with the stage of apple bloom (Table2). During both early and peak apple bloom, an interaction
between pollination treatment and mass owering impacted strawberry fruit weight (Table2). In accordance with
the competition hypothesis, mass owering of apple decreased the weight of open pollinated strawberry fruits but
not hand pollinated fruits (Fig.5, TableS2). Conversely, during late apple bloom the mass owering index was
positively associated with fruit weight (Fig.5, TableS2) suggesting facilitation during this stage.
Figure 4. Averaged weight of strawberry fruits per farm relative to (A) bee abundance (B) bee species richness.
Regression lines indicate signicant relationships (p < 0.05).
Var i a bl e df F P
Pollination Treatment 1,876 0.728 0.393
Flowering Stage 2,876 7.757 0.001
Mass Flowering Index 1,876 0.004 0.946
Poll. Trt. X Index 1,876 7.224 0.007
Stage X Index 2,876 8.322 0.001
Table 2. ANOVA table output of minimum adequate models describing landscape scale eects of apple
mass owering on the weights of hand-pollinated and open-pollinated of sentinel strawberry fruits
sampled during early, peak and late apple bloom.
Figure 5. Average weight of hand-pollinated and open-pollinated strawberry fruits during early, peak and
late apple bloom relative to the mass owering index (calculated as the percentage of apple in the landscape
multiplied by the intensity of apple bloom for each sampling period). Regression lines indicate signicant
relationships (p < 0.05).
Scientific RepoRts | 7:45296 | DOI: 10.1038/srep45296
Resource pulses are a common feature of agricultural systems; however, the impact of mass owering crops on
the pollination and yield of co-blooming crops is currently unknown. Here we examined the spatial and temporal
eects of a mass owering crop on bee communities and subsequently on yield in a co-blooming crop species. We
predicted that changes in pollinator abundance over the course of mass owering would lead to either competi-
tion or facilitation at dierent stages, and indeed we found that apple mass owering rst decreased strawberry
pollination and then increased strawberry pollination with corresponding eects on yield.
e mass owering of apple negatively aected bee abundance and diversity in co-blooming strawberry dur-
ing the early and peak stage of apple bloom. However, during the late bloom stage, increasing apple mass ow-
ering was associated with greater bee abundance and diversity in strawberry. ese results indicate that bees are
responding to local changes in resource availability resulting in a dilution of bees when oral resources are plen-
tiful during early and peak apple bloom followed by a spillover of bees from apple orchards to nearby strawberry
elds as apple owering decreases. In natural systems, similar eects have been observed in mixtures of ower-
ing Cirsium and Raphanus plants where the balance between pollinator mediated competition and facilitation
was dependent on the relative densities of Cirsium owers44. ese patterns may be explained by changes in the
foraging preferences of the pollinator community, but population level responses to oral resources pulses may
support overall greater abundances of pollinators in landscapes with high cover of mass blooming crops18,19,21.
Our ndings indicate that both density and timing of owering are important predictors of the outcome of these
We predicted that both abundance and diversity of bees would be greater at sites adjacent to natural habi-
tats. Although bee abundance was greater at sites adjacent to natural habitats, bee species richness did not dier
between sampling locations. is result is likely due to a greater density of nesting sites in less disturbed natu-
ral areas for the ground-nesting bees that dominated the pollinator community45. While the distance between
strawberry elds and semi-natural habitats within a farm was not greater than the ight distance of the average
strawberry pollinator46, it is possible that fewer individuals traveled that distance. Despite overall greater pollina-
tor abundances in sites adjacent to natural habitats, lack of a signicant interaction between the natural habitat
proximity variable and mass owering indicated that proximity to natural habitat did not alter the impact of mass
owering on the pollinator community. Furthermore, the inuence of mass owering apple on the abundance
and diversity of bees was greater than the inuence of proximity to natural habitats. Similar results were reported
by Westphal et al.18, who found that bumble bee densities were positively related to the availability of oilseed rape
and not natural habitats within the landscape. Our ndings reveal that these eects extend to a much broader pol-
linator community. ese ndings also suggest that the eects of agricultural habitats on pollinator communities
has thus far been underestimated and likely represents a common phenomenon among crops with overlapping
pollinator communities.
Mass owering of apple at the landscape scale was negatively associated with the weight of open pollinated
strawberry fruits during early and peak apple bloom and positively associated with fruit weight during late apple
bloom. We hypothesize that these results are due to the parallel changes observed in the abundance and diver-
sity of pollinators, as both measures were highly correlated with the weight of open pollinated strawberry fruit;
however, the decrease in fruit weight associated with early and peak apple bloom may also be due in part to
increased rates of heterospecic pollen transfer47 from apple to strawberry. In the late sampling period, the pos-
itive response of hand-pollinated fruit to the mass owering index may have been caused by incomplete eec-
tiveness of the hand pollination treatment due to the greater storage time of the supplemental pollen at this stage.
e competitive interactions observed between apple and strawberry likely represent a conservative estimate
of the potential magnitude of indirect interactions mediated by shared pollinators. In this case, competitive eects
are moderated by the relatively diverse pollinator community of strawberry48 and the ability of strawberry to
self-pollinate41. erefore, the negative eects of mass owering may be stronger in crops that are more pollinator
dependent or share a greater proportion of their pollinator community with a mass owering crop.
In natural systems, pollinator-mediated facilitation in plant communities is thought to occur through sev-
eral mechanisms. First, coexisting plants may attract greater numbers of shared pollinators by providing aggre-
gate oral displays greater than a single species alone44. Facilitation may also occur when species with staggered
blooming periods support pollinator populations by reducing spatial and temporal variation in oral resource
availability49. In this case, the consecutive bloom of plant species increases the duration of oral resource avail-
ability within years or the reliability of oral resources across years50. ese same dynamics may be particularly
important for pollinator communities in agricultural systems where crop rotation or extreme weather events can
lead to high variability in oral resource abundance among seasons and years. If the greater abundance, diversity
and duration of oral resources can be achieved through complementarity of owering crops, later blooming
crops such as strawberry may even support the pollination services of earlier mass owering crops in the follow-
ing year51.
In agricultural systems, our ndings reveal that crop habitats can act as a source of ecosystem services to other
crops and represent an area of underexploited potential for ecological intensication practices. Studies of spillover
of pollinators between mass owering crops have also reported that prevalence of early mass owering crops in
the landscape can mitigate pollinator dilution in another mass owering crops blooming in a later season21. Our
results advance these ndings by demonstrating that changes in the abundance of pollinators mediated by the
bloom of mass owering crops has consequences for the yield of nearby pollinator dependent crops. Importantly,
our results highlight the importance of timing in determining the outcome of interactions among pollinator
dependent crops and suggest ecological intensication strategies that may be employed to reduce competition
and enhance facilitation among crops that have a signicant number of shared pollinators. By selecting crops
and varieties that bloom sequentially with shared pollinator communities, growers can minimize competition
while maximizing facilitative eects, thereby improving the sustainability of crop pollination. However, when
Scientific RepoRts | 7:45296 | DOI: 10.1038/srep45296
agronomic or other factors constrain variety selection, management strategies may focus on locating co-blooming
crops at distances greater than the average foraging range of their shared pollinator community.
Our results clearly indicate that the timing of owering in co-occurring crops can have consequences for the
yield of pollinator dependent plants. When one crop co-blooms with another, mass owering crop, competition
for pollinators is likely to reduce yield, while owering aer the owering event facilitates pollination leading to
higher yields. We show that the temporal resource pulses associated with mass owering crops are an important
driver of pollinator community dynamics and pollination services at local and landscape scales. Greater under-
standing of these eects will allow for improvements in designing agroecosystems in order to maximize the pro-
visioning of ecosystem services and promote crop yields through ecological intensication.
1. Cassman, . G., Grassini, P. & van Wart, J. Handboo of climate change and agroecosystems Vol 1. (eds Hillel, D. & osenzweig, C.)
Ch. 3, 37–51 (Imperial College Press, 2010).
2. Bommarco, ., leijn, D. & Potts, S. G. Ecological intensication: Harnessing ecosystem services for food security. Trends Ecol. Evol.
28, 230–238 (2013).
3. Garnett, T. et al. Sustainable Intensication in Agriculture: Premises and Policies. Science (80-). 341, 33–34 (2013).
4. Pywell, . F. et al. Wildlife-friendly farming increases crop yield: evidence for ecological intensication. Proc. . Soc. B Biol. Sci. 282,
20151740- (2015).
5. Blaauw, B. . & Isaacs, . Flower plantings increase wild bee abundance and the pollination services provided to a pollination-
dependent crop. J. Appl. Ecol. 51, 890–898 (2014).
6. Iverson, A. L. et al. Do polycultures promote win-wins or trade-os in agricultural ecosystem services? A meta-analysis. J. Appl. Ecol.
51, 1593–1602 (2014).
7. Holzschuh, A. et al. Mass-owering crops dilute pollinator abundance in agricultural landscapes across Europe. Ecol. Lett. 19,
1228–1236 (2016).
8. Geslin, B. et al. Massively Introduced Managed Species and eir Consequences for Plant–Pollinator Interactions. Adv. Ecol. es. 57,
147–199 (2017).
9. Aizen, M. A. & Harder, L. D. The global stoc of domesticated honey bees is growing slower than agricultural demand for
pollination. Curr. Biol. 19, 915–918 (2009).
10. Ellis, A. M., Myers, S. S. & icetts, T. H. Do pollinators contribute to nutritional health? PLoS One 10, 1–17 (2015).
11. Smith, M. ., Singh, G. M., Mozaarian, D. & Myers, S. S. Eects of decreases of animal pollinators on human nutrition and global
health: A modelling analysis. Lancet 386, 1964–1972 (2015).
12. Hanley, M. E. et al. Increased bumblebee abundance along the margins of a mass owering crop: Evidence for pollinator spill-over.
Oios 120, 1618–1624 (2011).
13. ovacs-Hostyanszi, A. et al. Contrasting eects of mass-owering crops on bee pollination of hedge plants at dierent spatial and
temporal scales. Ecol. Appl. 23, 1938–1946 (2013).
14. Dieötter, T., adoya, T., Peter, F., Wolters, V. & Jauer, F. Oilseed rape crops distort plant-pollinator interactions. J. Appl. Ecol. 47,
209–214 (2010).
15. Stanley, D. A. & Stout, J. C. Pollinator sharing between mass-owering oilseed rape and co-owering wild plants: implications for
wild plant pollination. Plant Ecol. 1–11, doi: 10.1007/s11258-014-0301-7 (2014).
16. Boreux, V., ushalappa, C. G., Vaast, P. & Ghazoul, J. Interactive eects among ecosystem services and management practices on
crop production: pollination in coee agroforestry systems. Proc. Natl. Acad. Sci. USA 110, 8387–92 (2013).
17. Jauer, F., Peter, F., Wolters, V. & Dieötter, T. Early reproductive benets of mass-owering crops to the solitary bee Osmia rufa
outbalance post-owering disadvantages. Basic Appl. Ecol. 13, 268–276 (2012).
18. Westphal, C., Stean-Dewenter, I. & Tscharnte, T. Mass owering crops enhance pollinator densities at a landscape scale. Ecol. Lett.
6, 961–965 (2003).
19. Holzschuh, A., Dormann, C. F., Tscharnte, T. & Stean-Dewenter, I. Mass-owering crops enhance wild bee abundance. Oecologia
172, 477–484 (2013).
20. iedinger, V. et al. Annual dynamics of wild bee densities: Attractiveness and productivity eects of oilseed rape. Ecology 96,
1351–1360 (2015).
21. iedinger, V., enner, M., undlöf, M., Stean-Dewenter, I. & Holzschuh, A. Early mass-owering crops mitigate pollinator dilution
in late-owering crops. Landsc. Ecol. 29, 425–435 (2014).
22. icetts, T. H. et al. Landscape eects on crop pollination services: Are there general patterns? Ecol. Lett. 11, 499–515 (2008).
23. Schuett, E. M. & Vamosi, J. C. Phylogenetic community context inuences pollen delivery to Allium cernuum. Evol. Biol. 37, 19–28
24. Cussans, J. et al. Two bee-pollinated plant species show higher seed production when grown in gardens compared to arable
farmland. PLoS One 5 (2010).
25. Holzschuh, A., Dormann, C. F., Tscharnte, T. & Steffan-Dewenter, I. Expansion of mass-flowering crops leads to transient
pollinator dilution and reduced wild plant pollination. Proc. Biol. Sci. 278, 3444–51 (2011).
26. ezende, E. L., Jordano, P. & Bascompte, J. Eects of phenotypic complementarity and phylogeny on the nested structure of
mutualistic networs. Oios 116, 1919–1929 (2007).
27. ezende, E. L., Lavabre, J. E., Guimaraes, P. . Jr. Jordano, P. & Bascompte, J. Non-random coextinctions in phylogenetically
structured mutualistic networs. Nature 448, 925–U6 (2007).
28. Martins, . T., Gonzalez, A. & Lechowicz, M. J. Pollination services are mediated by bee functional diversity and landscape context.
Agric. Ecosyst. Environ. 200, 12–20 (2015).
29. Connelly, H., Poveda, . & Loeb, G. Landscape simplication decreases wild bee pollination services to strawberry. Agric. Ecosyst.
Environ. 211, 51–56 (2015).
30. usso, L., Par, M., Gibbs, J. & Danforth, B. e challenge of accurately documenting bee species richness in agroecosystems: Bee
diversity in eastern apple orchards. Ecol. Evol. 5, 3531–3540 (2015).
31. USDA-NASS. 2013 USDA National Agricultural Statistics Service Cropland Data Layer (2014).
32. Bartomeus, I. et al. Historical changes in northeastern US bee pollinators related to shared ecological traits. Proc. Natl. Acad. Sci.
USA 110, 4656–60 (2013).
33. Mitchell, T. B. Bees of the Eastern United States: volume I. N. C. Agric. Exp. Sta. Tech. Bull 141, 1–538 (1960).
34. Mitchell, T. B. Bees of the Eastern United States: volume II. N. C. Agric. Exp. Sta. Tech. Bull 152, 1–557 (1962).
35. LaBerge, W. E. A revision of the bees of the genus Andrena of the Western Hemisphere. Part XII. Subgenera Leucandrena,
Ptilandrena, Scoliandrena, and Melandrena. Trans. Am. Entomol. Soc. 112, 191–248 (1987).
36. Gibbs, J. evision of the metallic Lasioglossum (Dialictus) of eastern North America (Hymenoptera: Halictidae: Halictini). Zootaxa
216, 1–216 (2011).
Scientific RepoRts | 7:45296 | DOI: 10.1038/srep45296
37. ehan, S. M. & Sheeld, C. S. Morphological and molecular delineation of a new species in the Ceratina dupla species-group
(Hymenoptera: Apidae: Xylocopinae) of eastern North America. Zootaxa 2873, 35–50 (2011).
38. Gibbs, J., Pacer, L., Dumesh, S. & Danforth, B. N. evision and reclassication of Lasioglossum (Evylaeus), L. (Hemihalictus) and
L. (Sphecodogastra) in eastern North America (Hymenoptera: Apoidea: Halictidae). Zootaxa 3672, 1–117 (2013).
39. Webb, . A., Terblanche, J. H., Pur ves, J. V. & Beech, M. G. Size factors in strawberry fruit. Sci. Hortic. (Amsterdam). 9, 347–356
40. Webb, . A., Purves, J. V. & White, B. A. e components of fruit size in strawberry. Sci. Hortic. (Amsterdam). 2, 165–174 (1974).
41. Chagnon, M., Gingras, J. & de Oliveira, D. Eect of Honey Bee (Hymenoptera: Apidae) Visits on the Pollination ate of Strawberries.
J. Econ. Entomol. 82, 1350–1353 (1989).
42. Pinheiro, J., Bates, D., Deboy, S., Sarar, D. & Team, . C. nlme: Linear and Nonlinear Mixed Eects Models (2016).
43.  Core Team. : A language and environment for statistical computing.  Foundation for Statistical Computing (2015).
44. Ghazoul, J. Floral diversity and the facilitation of pollination. J. Ecol. 94, 295–304 (2006).
45. im, J., Williams, N. & remen, C. Eects of Cultivation and Proximity to Natural Habitat on Ground-nesting Native Bees in
California Sunower Fields. J. ansas Entomol. Soc. 79, 309–320 (2006).
46. Zurbuchen, A. et al. Maximum foraging ranges in solitary bees: only few individuals have the capability to cover long foraging
distances. Biol. Conserv. 143, 669–676 (2010).
47. Galen, C. & Gregory, T. Interspecic pollen transfer as a mechanism of competition: Consequences of foreign pollen contamination
for seed set in the alpine wildower, Polemonium viscosum. Oecologia 81, 120–123 (1989).
48. athce, B. Interactions for pollination among coowering shrubs. Ecology 69, 446–457 (1988).
49. Waser, N. M. & eal, L. a. Eective mutualism between sequentially owering plant species. Nature 281, 670–672 (1979).
50. Moeller, D. A. Facilitative Interactions Among Plants Via Shared Pollinators. Ecology 85, 3289–3301 (2004).
51. Schellhorn, N. A., Gagic, V. & Bommarco, . Time will tell: esource continuity bolsters ecosystem services. Trends Ecol. Evol. 30,
524–530 (2015).
We are grateful to Claudio Gratton, Susan Whitehead, the pollination ecology group at Cornell and anonymous
reviewers for comments that greatly improved the manuscript. is research would not have been possible
without our farmer collaborators and dedicated research assistants, particularly Sally Hartwick and Justin
Cappadonna. is project was supported by Smith Lever and Hatch Funds administered by Cornell University
Agricultural Experiment Station and by a USDA-AFRI grant [USDA 2010–03689, to B.N. Danforth].
Author Contributions
E.B., K.P. and H.G. conceived of, designed and conducted the experiments. B.D. collected eld data. H.G. and K.P.
carried out the data analysis and draed the manuscript. All authors approved the nal manuscript.
Additional Information
Supplementary information accompanies this paper at
Competing Interests: e authors declare no competing nancial interests.
How to cite this article: Grab, H. et al. Temporally dependent pollinator competition and facilitation with mass
owering crops aects yield in co-blooming crops. Sci. Rep. 7, 45296; doi: 10.1038/srep45296 (2017).
Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and
institutional aliations.
is work is licensed under a Creative Commons Attribution 4.0 International License. e images
or other third party material in this article are included in the article’s Creative Commons license,
unless indicated otherwise in the credit line; if the material is not included under the Creative Commons license,
users will need to obtain permission from the license holder to reproduce the material. To view a copy of this
license, visit
© e Author(s) 2017

Supplementary resource (1)

... Recent reviews suggest pollinator habitat enhancement does not consistently increase crop pollination outcomes [5,9,12]. This highlights the critical need to understand competitive and facilitative pollination interactions between crops and co-flowering plants, which may include other crops [71,72]. These interactions arise from the integration of crops into surrounding plant-pollinator networks, which shape crop pollination outcomes by influencing the behaviour, diversity and demography of shared pollinators [72]. ...
... For solitary bees and non-bee crop pollinators, floral resource continuity is important, but its role requires further study [90][91][92]. Floral resource continuity may be enhanced by the availability of sequentially flowering crops [71] or through diverse floral resource plantings, although commercial pollinator seed mixes may not provide an adequate range of flowering phenology [93]. In tropical climates, planting non-crop plant species with steady-state flowering characteristics can increase bee species abundances around crop fields, but positive effects on crop pollination may be limited, especially when crop floral abundance is low [94]. ...
... Intensive agricultural practices and landscape simplification impact pollinator abundance, a major driver of crop yields [23]. Mass-flowering crops can increase the abundance of some pollinators [95] and in some cases, facilitate pollination of co-flowering crops [71], yet can also negatively impact surrounding crops by increasing heterospecific pollen deposition and interfering with pollination [96]. Extensive crop fields and narrow flowering phenology means supporting sufficient abundances of wild pollinators to provide crop pollination services in mass-flowering systems remains a major challenge [3]. ...
Bee and non-bee insect pollinators play an integral role in the quantity and quality of production for many food crops, yet there is growing evidence that nutritional challenges to pollinators in agricultural landscapes are an important factor in the reduction of pollinator populations worldwide. Schemes to enhance crop pollinator health have historically focused on floral resource plantings aimed at increasing pollinator abundance and diversity by providing more foraging opportunities for bees. These efforts have demonstrated that improvements in bee diversity and abundance are achievable; however, goals of increasing crop pollination outcomes via these interventions are not consistently met. To support pollinator health and crop pollination outcomes in tandem, habitat enhancements must be tailored to meet the life-history needs of specific crop pollinators, including non-bees. This will require greater understanding of the nutritional demands of these taxa together with the supply of floral and non-floral food resources and how these interact in cropping environments. Understanding the mechanisms underlying crop pollination and pollinator health in unison across a range of taxa is clearly a win–win for industry and conservation, yet achievement of these goals will require new knowledge and novel, targeted methods. This article is part of the theme issue ‘Natural processes influencing pollinator health: from chemistry to landscapes’.
... This conflicts from most previous studies, which have detected such a dilution effect, finding that increasing the amount of mass-flowering crops in the landscape decreases large pollinator abundance, e.g. honeybees and bumblebees (Holzschuh et al. 2016;Bänsch et al. 2020), hence decreasing pollination efficiency in these crops (Shaw et al. 2020) or adjacent crops (Grab et al. 2017) and grasslands (Holzschuh et al. 2011). The gradients of %OSR used in the present study (from 0 to 26% in a buffer of 747 m outside OSR fields) were similar to those used in other studies (e.g. from 0 to 30% in a buffer of 1000 m; Holzschuh et al. 2011), thus other factors must explain this discrepancy. ...
... One possible reason may be the plant phytometer itself: dilution was found when plant phytometers used was not the blooming crop (e.g. strawberry plants with the blooming of apple flowers in Grab et al., 2017), or Primula veris with the blooming of OSR flowers (Holzschuh et al. 2011). Two other possible factors may also be involved in explaining the difference with the present study. ...
Full-text available
Context Recognized as a critical ecosystem service in farmland, pollination is threatened by the decline of pollinators, notably due the homogenization of the landscape and the decline of floral resources. However, there is still a limited understanding of the interplay between landscape features and the pulses of floral resources provided by mass-flowering crops. Objective The goals of this study were to (i) determine how pollination efficiency varies with the amount of floral resources at field and landscape scales through the oilseed rape (OSR) flowering period and (ii) quantify the magnitude of the pollination processes involved. Methods Pollination efficiency (fruiting success) was measured using OSR plant phytometers placed in grasslands, cereals and OSR fields varying in quantity of floral resources at both field and landscape scales. The individual contributions of different processes to pollination were determined using a bagging experiment on plant phytometers. Results Pollination efficiency was enhanced during both the temporal period and in landscapes with a high amount of OSR flowers, and semi-natural habitats as a result of higher pollinator presence. The bagging experiment also supported a complementarity between habitats for pollinators, as insect-pollination in grasslands and cereals was higher after OSR flowering, especially in OSR-rich landscapes, in regard to large-insect-pollination. Conclusions The floral resource availability drives insect-pollination through attraction, spillover, and spatial and temporal complementarities between habitats. These results suggest that maximizing pollination efficiency in farmland landscapes partly consisting of OSR fields should include a combination of habitats that provide continuous floral resources.
... These relationships are easily noticeable. For example, if floral margins are grown on a field to attract pollinators, pollination of crops could be improved, and crop production enhanced (Campbell et al., 2017;Grab et al., 2017). In contrast, studies reporting the relationship between agricultural practices and cultural ES are scarce (Calvet-Mir et al., 2012; Table 4 GLM models analysing the differences between agroecological and conventional farms (explanatory variables) in the indicators of regulating ES (response variables). ...
Agricultural intensification has strongly impacted ecosystems and accelerated the process of global change. Consequently, agroecological practices are being increasingly adopted. Agroecological practices are biodiversity-based solutions that aim to generate sustainable and resilient agroecosystems, which could enhance the supply of ecosystem services. This study compared agroecological and conventional horticultural farms in terms of agroecological practices and ecosystem services supply. We conducted biophysical samplings and interviews on 24 agroecological and conventional farms over two summers in the Madrid Region (Spain). We used multiple indicators as proxies of the supply of 12 ecosystem services, and we identified the agricultural practices applied at each farm. We found that agroecological farmers applied more agroecological practices compared to conventional farmers, and agroecological farms had a higher potential to supply regulating, provisioning, and cultural services. Some agroecological practices, such as crop diversification, light tillage, and the use of organic pesticides, were associated with enhancing soil fertility, pest control, and pollination services. Our study provided empirical evidence that agroecological practices enhance ecosystem services at horticultural farms, which is extremely relevant to upscaling agroecology in the current context of ongoing European policy reforms.
... 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, 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). ...
Full-text available
Numerous studies show that semi-natural habitats within agricultural landscapes benefit native pollinating insects and increase resultant crop pollination services. More recently, evidence is emerging that agricultural diversification techniques on farms, as well as increased compositional and configurational heterogeneity within the cropped portion of landscapes, enhance pollinator communities. However, to date, only a few studies have investigated how diversifying the crops within the farm field itself (i.e., polyculture) influences wild pollinator communities and crop pollination services. In the Central Coast of California, we investigate how local crop diversification within fields, crossed with the proportion of natural habitat in the surrounding landscape, jointly affect pollinator communities and services to strawberry. On 16 organic farms varying in farm type (monoculture vs. polyculture) and proportion of natural land cover, we find that both factors enhance pollinator abundance and richness, although neither affect honey bee abundance. Further, natural cover has a stronger effect on pollinator richness on monoculture (vs. polyculture) farms. Although strawberry can self-pollinate, we document experimentally that pollinator exclusion doubles the probability of berry malformation, while excluding both pollinators and wind triples malformation, with corresponding effects on berry marketability. Finally, in post-hoc tests, we find that berry malformation is significantly higher with greater visitation by honey bees, and observed a trend that this reduction was mitigated by increased native bee richness. These results suggest that both polyculture and semi-natural habitat cover support more abundant and diverse pollinator communities, and that ambient levels of pollinator visitation to strawberry provide an important crop pollination service by improving berry marketability (i.e., by reducing berry malformation). Although further confirmation would be needed, our work suggests that honey bees alone do not provide sufficient pollination services. Prior work has shown that honey bees tend to visit only the top of the strawberry flower receptacle, while other native bees often crawl around the flower base, leading to more complete pollination of the achenes and, consequently, better formed berries. If honey bee visits reduced native bee visitation in our system, this could explain the unexpected correlation between increased honey bee visits and malformation.
... However, the flowering of other plant species both on the farm and in the wider vegetation matrix may also coincide with this crop flowering period. The impact of one flowering species on another can vary along a continuum, from facilitation (positive) to competition (negative) [9], with the strength and direction of the interaction often being driven by temporal and spatial overlap in flowering between plant species [10] and the degree of pollinator sharing [11]. Facilitation occurs when a flowering species increases the abundance or diversity of pollinators in the local area and co-flowering neighbors benefit through spillover of pollinator activity [12]. ...
Full-text available
Co-flowering plants can experience an array of interactions, ranging from facilitation to competition, the direction and strength of which are often dependent on the relative abundance and diversity of the plant species involved and the foraging behavior of their pollinators. Understanding interactions between plant–pollinator networks and how they change over time is particularly important within agricultural systems, such as apples, that flower en masse and that also contain non-crop co-flowering species both within the farm and the surrounding landscape. We determined the degree of overlap between pollinator networks on two varieties of apple (Granny Smith and Pink Lady) and co-flowering plant species within orchards and the wider vegetation matrix in two apple-growing regions (Orange and Bilpin) in Australia. We surveyed plant–pollinator interactions at key stages of the cropping cycle: before mass flowering; during king, peak and late blooms; and, finally, once apple flowering had finished. Overall, we found considerable overlap in the flower visitor assemblage on apples and co-flowering species within the orchard. The introduced honeybee (Apis mellifera) was the most frequent flower visitor to all three vegetation types at all times in Orange. However, in Bilpin, both a native stingless bee (Tetragonula carbonaria) and A. mellifera were highly frequent visitors, both on- and off-crop. Numerous native bees, flies and Lepidoptera also commonly visited apple and co-flowering species within orchards in both locations. We found that native-bee and honeybee visitation to apple flowers was positively correlated with co-flowering species richness (within the orchard and the wider matrix); however, visitation by native bees decreased as the area of co-flowering species in the surrounding landscape increased. Our study highlights the importance of maintaining diverse co-flowering plant communities within the local landscape to increase and support a wide variety of pollinators in horticultural production systems.
... Furthermore, even experimentally induced changes in resource use may not necessarily result in reduced fecundity if alternate floral resources are available ( Paini, 2004 ). In addition, bee foraging behaviour may fluctuate within a day, season, or between years ( Grab et al., 2017 ;Wratt, 1968 ). As a result, visitation may only reflect local conditions for that survey time. ...
Full-text available
Worldwide, the use of managed bees for crop pollination and honey production has increased dramatically. Concerns about the pressures of these increases on native ecosystems has resulted in a recent expansion in the literature on this subject. To collate and update current knowledge, we performed a systematic review of the literature on the effects of managed and introduced bees on native ecosystems, focusing on the effects on wild bees. To enable comparison over time, we used the same search terms and focused on the same impacts as earlier reviews. This review covers: (a) interference and resource competition between introduced or managed bees and native bees; (b) effects of introduced or managed bees on pollination of native plants and weeds; and (c) transmission and infectivity of pathogens; and classifies effects into positive, negative, or neutral. Compared to a 2017 review, we found that the number of papers on this issue has increased by 47%. The highest increase was seen in papers on pathogen spill-over, but in the last five years considerable additional information about competition between managed and wild bees has also become available. Records of negative effects have increased from 53% of papers reporting negative effects in 2017 to 66% at present. The majority of these studies investigated effects on visitation and foraging behaviour. While only a few studies experimentally assessed impacts on wild bee reproductive output, 78% of these demonstrated negative effects. Plant composition and pollination was negatively affected in 7% of studies, and 79% of studies on pathogens reported potential negative effects of managed or introduced bees on wild bees. Taken together, the evidence increasingly suggests that managed and introduced bees negatively affect wild bees, and this knowledge should inform actions to prevent further harm to native ecosystems.
... However, with a few notable exceptions, (e.g. Grab et al., 2017;Kov acs-Hosty anszki et al., 2013), relatively few studies have sought to evaluate the extent to which pollinator networks overlap across the crop flowering season (although see Simba et al., 2018) and the similarity of networks of co-flowering species outside of this period (see Martins et al., 2018;Simba et al., 2018). Furthermore, to our knowledge, very few have focussed on evaluating networks within sweet cherry and co-flowering plants within both the orchard and the broader landscape matrix, however, see Eeraerts et al. (2021b). ...
Many food crops depend on animal pollination to set fruit. In light of pollinator declines there is growing recognition of the need for agro-ecosystems that can sustain wild pollinator populations, ensuring fruit production and pollinator conservation into the future. One method of supporting resident wild pollinator populations within agricultural landscapes is to encourage and maintain floral diversity. However, pollinator visitation to crop plants can be affected either positively (facilitation) or negatively (competition) by the presence of co-flowering plants. The strength and direction of the facilitative/competitive relationship is driven by multiple factors, including floral abundance and the degree of overlap in pollinator visitation networks. We sought to determine how plant-pollinator networks, within and surrounding sweet cherry (Prunus avium) orchards, change across key time points during the cherry flowering season, in three growing regions in Australia. We found significant overlap in the suite of flower visitors, with seven taxa (including native bees, flies, hoverflies and introduced honey bees, Apis mellifera) observed visiting cherry and other co-flowering species within the orchard and/or the wider surrounding matrix. We found evidence of pollinator facilitation with significantly more total cherry flower visits with increasing percent cover of co-flowering plants within the wider landscape matrix and increased visitation to cherry by honey bees with increasing co-flowering plant richness within the orchard. During the cherry flowering period there was a significant positive relationship between pollinator richness on cherry and pollinator richness on co-flowering plants within the orchard and the area of native vegetation surrounding orchards. Outside of the crop flowering season, co-flowering plants within the orchard and wider landscape matrix supported the same pollinator taxa that were recorded visiting cherry when the crop was flowering. This shows wild plants help support the pollinators important to crop pollination, outside of the crop flowering season, highlighting the role of co-flowering plants within pollinator-dependent cropping systems.
... There is clear evidence that pollinators benefit from semi-natural habitats (SNHs) such as grassland or forest, which provide food, shelter and nesting sites (Holland et al., 2017). The stability of floral resources over the season, through organic farming or continuity between early and late mass-flowering crops, is also essential (Grab et al., 2017;Holzschuh et al., 2008). However, high amounts of massflowering crops in the surrounding landscape can also dilute pollinators, resulting in their lower abundance at field scale (Holzschuh et al., 2016). ...
Pollination and pest control are two major ecological functions sustaining crop yield. In insect‐pollinated crops, previous studies have revealed that an increase of resources and habitats in landscapes can increase pest control by natural enemies as well as insect pollination by pollinators. However, data has been lacking that simultaneously considers the effects of landscape on both pollinators and pests, and the direct and indirect effects on yields of farming practices interacting with landscape, bees and pests. This study aimed to fill this gap by focusing on oilseed rape (OSR), an insect‐pollinated crop of high economic value. We first quantified the effects of landscape and farming practices on both bee and pest abundance caught in OSR blooming season in 124 farmed fields over a six‐year study (~20 fields sampled per year), and then used structural equation modelling to assess the direct and indirect links between bees, pests, farming practices and landscape on yield. The results showed that landscape had a stronger effect on bee and pest abundance than agrochemical farming practices. Bees and pests decreased with the amount of OSR in the landscape surrounding the focal field, and showed contrasted effects with the amount of meadow and organic farming: positive for bees and negative for pests. Bee abundance also increased with the amount of sunflower in the landscape the preceding year, and decreased with increasing field size. While agrochemicals surprisingly had barely any effect on bees and pests, their use improved OSR yield, though at a similar magnitude as bee and pest abundances. Synthesis and application. This study, conducted in commercial crop fields, underlines the important contribution of sustainable landscape management for enhancing OSR yield. Despite agrochemicals’ ability to improve or maintain OSR yields, their unconditional use is unsustainable due to negative externalities. Therefore, alternative options such as those highlighted in our study – such as reducing field size, increasing the amount of organic farming in the landscape, or sowing OSR in landscapes rich in sunflowers the preceding year – appear to be relevant tools to promote ecosystem services, maintain yield and conserve biodiversity. These findings support the potential of nature‐based solutions to foster more sustainable agriculture.
Pollination is essential for many crops since 70% of the world's cultivated plants depend on pollinators for their production. Floral resources within cultivated areas, especially those produced by flowering crops such as oilseed rape, are known to have a positive effect on wild pollinators. Nevertheless, little is known about the contribution of other floral resources, such as weeds within cultivated areas, in supporting wild pollinator communities and subsequent pollination services. Here, we investigate the extent to which oilseed rape pollination benefits from floral resources produced within cultivated areas, either crops or associated weeds. Based on the Müller index, we analyzed, during four pairs of consecutive years, the potential for inter-annual indirect effects received by oilseed rape through shared wild pollinators from major crops, and their associated weeds, in a typical French intensive agricultural landscape. Our results show that most of the support for oilseed rape pollinating fauna came from alternative types of floral resources than itself. We also find that weeds support oilseed rape pollination as much as flowering crops. Finally, we show that weeds growing within cereal fields have a major contribution to the support of oilseed rape pollination, exceeding the contribution of other floral resources, except oilseed rape. Our results underline that oilseed rape pollination benefits from floral resources present within cultivated fields, whatever the type of crops, including those that do not depend on pollinators for their pollination. Management practices like herbicide reduction in non-pollinator-dependent crops such as cereals are thus likely to impact the pollination of pollinator-dependent crops.
Full-text available
Context Flowering plants can enhance wild insect populations and their pollination services to crops in agricultural landscapes, especially when they flower before the focal crop. However, characterizing the temporal availability of specific floral resources is a challenge. Objectives Developing an index for the availability of floral resources at the landscape scale according to the specific use by a pollinator. Investigating whether detailed and temporally-resolved floral resource maps predict pollination success of broad bean better than land cover maps. Methods We mapped plant species used as pollen source by bumblebees in 24 agricultural landscapes and developed an index of floral resource availability for different times of the flowering season. To measure pollination success, patches of broad bean ( Vicia faba ), a plant typically pollinated by bumblebees, were exposed in the center of selected landscapes. Results Higher floral resource availability before bean flowering led to enhanced seed set. Floral resource availability synchronous to broad bean flowering had no effect. Seed set was somewhat better explained by land cover maps than by floral resource availability, increasing with urban area and declining with the cover of arable land. Conclusions The timing of alternative floral resource availability is important for crop pollination. The higher explanation of pollination success by land cover maps than by floral resource availability indicates that additional factors such as habitat disturbance and nesting sites play a role in pollination. Enhancing non-crop woody plants in agricultural landscapes as pollen sources may ensure higher levels of crop pollination by wild pollinators such as bumblebees.
Full-text available
Since the rise of agriculture, human populations have domesticated plant and animal species to fulfil their needs. With modern agriculture, a limited number of these species has been massively produced over large areas at high local densities. Like invasive species, these Massively Introduced Managed Species (MIMS) integrate local communities and can trigger cascading effects on the structure and functioning of ecosystems. Here, we focus on plant and insect MIMS in the context of plant–pollinator systems. Several crop species such as mass flowering crops (e.g. Brassica napus) and domesticated pollinating insects (e.g. Apis mellifera, Bombus terrestris) have been increasingly introduced worldwide and their impact on natural communities is addressed by an increasing number of scientific studies.
Full-text available
The use of computer-generated (CG) stimuli in face processing research is proliferating due to the ease with which faces can be generated, standardised and manipulated. However there has been surprisingly little research into whether CG faces are processed in the same way as photographs of real faces. The present study assessed how well CG faces tap face identity expertise by investigating whether two indicators of face expertise are reduced for CG faces when compared to face photographs. These indicators were accuracy for identification of own-race faces and the other-race effect (ORE)-the well-established finding that own-race faces are recognised more accurately than other-race faces. In Experiment 1 Caucasian and Asian participants completed a recognition memory task for own- and other-race real and CG faces. Overall accuracy for own-race faces was dramatically reduced for CG compared to real faces and the ORE was significantly and substantially attenuated for CG faces. Experiment 2 investigated perceptual discrimination for own- and other-race real and CG faces with Caucasian and Asian participants. Here again, accuracy for own-race faces was significantly reduced for CG compared to real faces. However the ORE was not affected by format. Together these results signal that CG faces of the type tested here do not fully tap face expertise. Technological advancement may, in the future, produce CG faces that are equivalent to real photographs. Until then caution is advised when interpreting results obtained using CG faces.
Full-text available
Ecological intensification has been promoted as a means to achieve environmentally sustainable increases in crop yields by enhancing ecosystem functions that regulate and support production. There is, however, little direct evidence of yield benefits from ecological intensification on commercial farms growing globally important foodstuffs (grains, oilseeds and pulses). We replicated two treatments removing 3 or 8% of land at the field edge from production to create wildlife habitat in 50–60 ha patches over a 900 ha commercial arable farm in central England, and compared these to a business as usual control (no land removed). In the control fields, crop yields were reduced by as much as 38% at the field edge. Habitat creation in these lower yielding areas led to increased yield in the cropped areas of the fields, and this positive effect became more pronounced over 6 years. As a consequence, yields at the field scale were maintained—and, indeed, enhanced for some crops—despite the loss of cropland for habitat creation. These results suggested that over a 5-year crop rotation, there would be no adverse impact on overall yield in terms of monetary value or nutritional energy. This study provides a clear demonstration that wildlife-friendly management which supports ecosystem services is compatible with, and can even increase, crop yields.
Mass-flowering crops (MFCs) are increasingly cultivated and might influence pollinator communities in MFC fields and nearby semi-natural habitats (SNHs). Across six European regions and 2 years, we assessed how landscape-scale cover of MFCs affected pollinator densities in 408 MFC fields and adjacent SNHs. In MFC fields, densities of bumblebees, solitary bees, managed honeybees and hoverflies were negatively related to the cover of MFCs in the landscape. In SNHs, densities of bumblebees declined with increasing cover of MFCs but densities of honeybees increased. The densities of all pollinators were generally unrelated to the cover of SNHs in the landscape. Although MFC fields apparently attracted pollinators from SNHs, in landscapes with large areas of MFCs they became diluted. The resulting lower densities might negatively affect yields of pollinator-dependent crops and the reproductive success of wild plants. An expansion of MFCs needs to be accompanied by pollinator-supporting practices in agricultural landscapes.
Although agricultural habitats can provide enormous amounts of food resources for pollinator species, links between agricultural and (semi-)natural habitats through dispersal and foraging movements have hardly been studied. In 67 study sites, we assessed the interactions between mass-flowering oilseed rape fields and semi-natural grasslands at different spatial scales, and their effects on the number of brood cells of a solitary cavity-nesting bee. The probability that the bee Osmia bicornis colonized trap nests in oilseed rape fields increased from 12 to 59 % when grassland was nearby, compared to fields isolated from grassland. In grasslands, the number of brood cells of O. bicornis in trap nests was 55 % higher when adjacent to oilseed rape compared to isolated grasslands. The percentage of oilseed rape pollen in the larval food was higher in oilseed rape fields and grasslands adjacent to oilseed rape than in isolated grasslands. In both oilseed rape fields and grasslands, the number of brood cells was positively correlated with the percentage of oilseed rape pollen in the larval food. We show that mass-flowering agricultural habitats—even when they are intensively managed—can strongly enhance the abundance of a solitary bee species nesting in nearby semi-natural habitats. Our results suggest that positive effects of agricultural habitats have been underestimated and might be very common (at least) for generalist species in landscapes consisting of a mixture of agricultural and semi-natural habitats. These effects might also have—so far overlooked—implications for interspecific competition and mutualistic interactions in semi-natural habitats.
Pollination by bees and other animals increases the size, quality, or stability of harvests for 70% of leading global crops. Because native species pollinate many of these crops effectively, conserving habitats for wild pollinators within agricultural landscapes can help maintain pollination services. Using hierarchical Bayesian techniques, we synthesize the results of 23 studies – representing 16 crops on five continents – to estimate the general relationship between pollination services and distance from natural or semi-natural habitats. We find strong exponential declines in both pollinator richness and native visitation rate. Visitation rate declines more steeply, dropping to half of its maximum at 0.6 km from natural habitat, compared to 1.5 km for richness. Evidence of general decline in fruit and seed set – variables that directly affect yields – is less clear. Visitation rate drops more steeply in tropical compared with temperate regions, and slightly more steeply for social compared with solitary bees. Tropical crops pollinated primarily by social bees may therefore be most susceptible to pollination failure from habitat loss. Quantifying these general relationships can help predict consequences of land use change on pollinator communities and crop productivity, and can inform landscape conservation efforts that balance the needs of native species and people.