t h e Div e r s it y o f po l l in a t o r s o n g r e e n r o o f s
Danilo Be v k
National Institute of Biology, Department of Organisms 
and e cosystems Research, v ečna pot 111, SI-1000 Ljubljana, Slovenia, 
e-mail: danilo.bevk@nib.si
a bstract – The aim of this research was to survey green roofs for pollinator abundance
and diversity. The research was conducted in 2019 on green roofs in Ljubljana and Škofja
Loka (Slovenia). During the whole flowering season, we counted the number of honeybees,
bumblebees, other wild bees, hoverflies, and butterflies. Results showed that various pol-
linators forage on blooming green roofs. The structure of flowering plants and pollinator
communities also changed greatly during the season. During June and July, wild bees
dominated, while in August and September, honeybees had the highest numbers. The
results show that green roofs are important not only for domesticated honeybee or bee-
keeping, but also for biodiversity conservation. Based on the results, we also suggest im-
provements to make these roofs an even better food source for pollinators.
k e y w ORDS: green roofs, green infrastructure, pollinators, honeybee, wild bees, bum-
blebees, hoverflies, urban ecology, biodiversity
izvleček – Pe STROST OPRAŠe v ALCe v NA Ze Le NIH STRe HAH
Cilj raziskave je bilo ugotoviti številčnost in pestrost opraševalcev na zelenih stre-
hah. Raziskava je potekala leta 2019 na strehah v Ljubljani in Škofji Loki. v času
cvetenja smo spremljali število medonosnih čebel, čmrljev, drugih divjih čebel, muh
trepetavk in metuljev. Rezultati so pokazali, da se na strehah hranijo različni opraše-
valci. Cvetenje streh in združbe opraševalcev se preko sezone tudi precej spreminjajo.
Junija in julija so prevladovale divje čebele, avgusta in septembra pa medonosna
čebela. Rezultati kažejo, da zelene strehe niso pomembe samo za medonosno čebelo
in čebelarstvo, ampak tudi za varovanje biodiverzitete. Na podlagi rezultatov predla-
gamo tudi izboljšave, da bi bile strehe za opraševalce še boljši vir hrane. 
k LJu č Ne Be Se De: zelene strehe, zelena infrastruktura, opraševalci, medonosna čebela,
divje čebele, čmrlji, muhe trepetavke, urbana ekologija, biodiverziteta 
5
ACTA ENTOMOLOGICA SLOVENICA
LJUBLJANA, JUNIJ 2021                      Vol. 29, øt. 1: 5–14

introduction
Pollinators provide one of the key ecosystem services – pollination, important for
biodiversity and agriculture. Among insects, the main pollinators are bees, flies, but-
terflies, moths, wasps, beetles and thrips (Potts et al. 2016). The net worth of insect
pollination is estimated to be around 10% of the total worth of agricultural production.
This adds up to around €153 billion globally (Gallai et al. 2009), and up to €22
billion in e urope every year (Potts et al. 2015).
In recent decades, the populations and diversity of pollinators has declined rapidly
(Potts et al. 2010, Hallmann et al. 2017, w agner 2020). For instance, 9.2% of bees in
e urope are threatened with extinction. However, for 56.7% of bees, there is not
enough scientific information to evaluate their risk of extinction (Nieto et al., 2014).
The main drivers for this decline are changes in land use, pesticide use, disease, and
climate change. One of the biggest threats to pollinators is lack of food due to intensive
farming, urbanisation, and climate change (Nieto et al., 2014, Gogala, 2014, Bevk et
al. 2016). Increasing development of urban environments also puts pressure on green
spaces, having a potential negative impact on biodiversity and ecosystem services
(Braaker et. al 2017).
This problem can be partially alleviated by green infrastructure such as green
roofs. A variety of native bee species can use green roofs as foraging and/or nesting
sites (Colla et al., 2009; k siazek et al., 2012; Maclvor et al., 2015). Although
pollinators are present on green roofs in lower abundances and species richness than
in natural habitats and ground-level sites, roofs are potentially valuable sites for pol-
linator conservation in urban areas, particularly if planted with a diverse range of
native plants (Tonietto et al., 2011; Braaker et. al 2017). Subsequently, the integration
of green roofs in cities has great potential to enable higher connectivity among green
spaces (Braaker et. al 2014). Green roofs become even more important under changing
temperature conditions, as they can mitigate the negative effects of increasing tem-
peratures on wild bees (Papanikolaou et al., 2017).
The aim of this research was to survey u rbanscape green roofs for pollinator
abundance and diversity, and to determine their potential as food resource for polli-
nators in urban areas.
Methods
Investigation of pollinator foraging activity was conducted in 2019 on two u rban-
scape green roofs. The first roof (700 m
2
) was in Ljubljana, near the city centre on
residential building y (7th floor). There were three apiaries present within a radius of
500m, and another three within 1000m. The second roof (100 m
2
) was in Škofja
Loka, on the edge of the industrial zone on k nauf Insulation e xperience Center (3rd
floor). There was one apiary within a radius of 500m, and another five within 1000m.
w e counted the number of honeybees, bumblebees, other wild bees, hoverflies
and butterflies. Pollinator activity was monitored during the whole flowering season
(June – September), three to five-times a month on each location, solely in sunny
Acta entomologica slovenica, 29 (1), 2021
6

weather when pollinators are most active. Firstly, we looked at which plants were
blooming and estimated the percentage of flowering roof area. w e then defined five
sampling sites. One sampling site represented one square meter of flowering plants.
Pollinator activity was monitored on sampling sites for two hours between 8 a.m. and
11 a.m. Twenty monitoring counts on each sampling site were carried out per day
(totalling 100 across the five sampling sites). After the last count was carried out, up
Danilo Bevk: The diversity of pollinators on green roofs
7
f igure 1: Green roof in Ljubljana.
f igure 2: Green roof in Škofja Loka.

to three specimens that could not be determined in the field were caught with nets.
e ffort was made to catch samples of morphologically different specimens. These
were prepared for later idenfication, to determine the species. In total in whole season,
we caught 19 bees and one hoverfly in Ljubljana and 12 bees and three hoverflies in
Škofja Loka.
Based on the obtained data, we calculated the structure of pollinator communities,
density of pollinators and the dynamic of flowering on the roof across the season
(composition of flower communities).
r esults 
In Ljubljana, the flowering period (established when at least part of the plants on
the roof started to come into bloom, until the end of the blooming period) lasted from
the middle of June to the end of September. During this period, 14 observation days
were conducted and 869 pollinators were counted altogether. The most abundant pol-
linator observed was the honeybee (463, Figure 3), followed by “other wild bees”,
i.e. wild bees other than bumblebees (392). There were also some hoverflies recorded
(6), butterflies (4) and bumblebees (4). In total, seven species of wild bees and one
species of hoverfly were found (Table 1).
In Škofja Loka, the flowering period lasted only from the middle of June to the
beginning of July. Therefore, only three observation days were conducted altogether,
and 403 pollinators were counted. The most numerous pollinators by far were “other
wild bees” (377). There were also some honeybees (19), hoverflies (6), one butterfly
Acta entomologica slovenica, 29 (1), 2021
8
f igure 3: The structure of pollinator communities recorded on each green roof.

recorded, and no bumblebees. Together, five species of wild bee (Table 1) and three
species of hoverfly were found. Two species of bee were found across both green
roofs in Škofja Loka and Ljubljana
t able 1: List of bee and hoverfly species found on each green roof.
l jubljana Škofja l oka
Bees
Andrena subopaca Andrena subopaca
Lasioglossum politum Lasioglossum politum
Hylaeus punctatus Lasioglossum laticeps
Megachile willughbiella Lasioglossum lineare
Megachile pilidens Lasioglossum malachurum
Anthidium oblongatum
Bombus terrestris/lucorum
h overflies 
Eristalis tenax Helophilus trivittatus
Sphaerophoria sp.
Eupeodes corollae
Danilo Bevk: The diversity of pollinators on green roofs
9
f igure 4: The structure of pollinator communities recorded by month (both loca-
tions together).

The structure of the pollinator communities changed greatly during the season.
w ild bees dominated in June and July, while in August and September, honeybees
dominated (Figure 4). Accordingly, the density of honeybees increased, and the
density of wild bees decreased during the season. The average density in Ljubljana
was 1,02 pollinators per square meter of flowering plants and 1,34 in Škofja Loka.
The density of solitary bees was highest in June and July, in contrast to honeybees
that had the highest density in August and September (Figure 5 and 6). 
The flowering of Sedum spp. was very dynamic. In Ljubljana – where flowering
lasted three and a half months, there were three peaks of blooming: in the middle of
June (20% of roof), middle of July (20% of roof) and in the second half of August
(30%, Figure 5). On the contrary, in Škofja Loka the flowering period lasted only
one month, with peak of blooming (20%) at the end of June (Figure 6). 
There were also differences in the species composition of flowering plants. w e
found nine different Sedum species on roofs. In Ljubljana (Figure 5), Sedum sexan-
gulare (yellow) dominated in the beginning, then Sedum album 'Coral carpet' (white),
followed by Sedum spurium 'John Creech' (red). In Škofja Loka (Figure 6), two
Acta entomologica slovenica, 29 (1), 2021
10
f igure 5: Pollinator density and flowering of the green roof in Ljubljana. The
percentage of flowering green roof is presented on the secondary y-axis. The colour
of the flowers is in line with the colour of the graphs (yellow, red, and white). Lines
represent average density of pollinators (primary y-axis).

species of white-flowered Sedum dominated (Sedum album 'Coral carpet' and Sedum
hispanicum).
Discussion
Results show that green roofs can provide additional or alternative food sources
for pollinators during the blooming period. Considering the lack of food being one of
the main causes for pollinator decline, flowering green roofs can contribute to solving
this problem.
Contrary to our expectations, with Slovenia having one of the highest densities of
honeybee colonies in the e uropean u nion (Chausat et al. 2013), wild pollinators
were more numerous than honeybees. This shows that green roofs are important not
only for domesticated honeybees or beekeeping, but also in the aspect of biodiversity
conservation.
Among wild pollinators, solitary bees were by far the most numerous. Their den-
sities were surprisingly high. Hoverflies, butterflies, and bumblebees were rare. Ac-
cording to lepidopterists (Barbara Zakšek, pers. comm.), the low diversity of butterflies
were expected, however, we expected much more bumblebees. The low number of
bumblebees can at least partly be explained by smaller populations of bumblebees
being present in 2019, due to a very rainy spring. As a result of the small number of
Danilo Bevk: The diversity of pollinators on green roofs
11
f igure 6: Pollinator density and flowering of the green roof in Škofja Loka.

bees sampled (31) as well as hoverflies (4), species diversity is probably underesti-
mated.
The structure of flowering plants and pollinator communities changed greatly
during the season. In June and July, wild bees dominated, while in August and
September honeybees had the highest numbers. A decrease in the number of solitary
bees later in the season was expected, as most of these species are active in spring
and early summer. As most Slovenian beekeepers have stationary apiaries, changes
in the number of honeybees are probably not a consequence of moving bee colonies.
The increase in the number of honeybees in late summer are more likely to be a con-
sequence of decreases in nectar and pollen sources at this time of year. In June, hon-
eybees probably prefer to forage on lime (Tilia spp.) and chestnut (Castanea sativa)
trees and turn to alternative sources on green roofs when these are no longer in
bloom. However, to obtain better understanding, longer research on more locations is
needed.
According to our results, green roofs can play an important role in developing
more pollinator and biodiversity friendly cities in general. Of course, biodiversity on
rooftops cannot be compared to biodiversity of flowering meadows, but green roofs
are one of the solutions for biodiversity and more environmentally friendly urbaniza-
tion.
Based on the results, we suggest some improvements to make roofs an even better
food source for pollinators. w e recommend planting a selection of plants that provide
good sources of nectar and/or pollen, and a combination of plants that provide food
during the whole summer. w e also recommend using a higher diversity of plants and
the use of native plants.
a cknowledgments
Thanks to Mojca Pibernik for help in field work, Blaž k oderman and dr. Andrej
Gogala for determination of bees, dr. Maarten de Groot for determination of hoverflies
and e mily Quinn for the improvement of the language. Thanks also to anonymous
reviewer for all constructive comments and suggestions. The research was funded by
k nauf Insulation d.o.o and Slovenian Research Agency (research core funding No.
P1-0255).
r eferences
Bevk D. 2016: Pestrost divjih čebel in njihov pomen za kmetijstvo in naravo. Zbornik
referatov, 2. znanstveno posvetovanje o čebelah in čebelarstvu, p. 7 –13.
Braaker s ., g hazoul J., o brist K. M., Moretti M. 2014: Habitat connectivity
shapes urban arthropod communities: the key role of green roofs. Ecology, 95:
1010 –2020
Braaker s ., o brist K. M., g hazoul J.,  Moretti M. 2017: Habitat connectivity and
local conditions shape taxonomic and functional diversity of arthropods on green
roofs. Jornal of Animal Ecology. 86: 521–53.
Acta entomologica slovenica, 29 (1), 2021
12

Chauzat M-p, Cauquil l ., r oy l ., f ranco s ., h endrikx p., Chabert M.r . 2013:
Demographics of the e uropean Apicultural Industry, PLoS ONE, 8 (11):
e79018.doi:10.1371/journal.pone.0079018 
Colla s .r ., Willis e ., packer l . 2009: Can green roofs provide habitat for urban
bees (Hymenoptera: Apidae)? Cities and the Environment, 2 (1):  1 –12
g allai n ., s alles J.M., s ettele J., v aissiere B.e . 2009: e conomic valuation of the
vulnerability of world agriculture confronted with pollinator decline. Ecological
economics, 68: 810 –821.
h allmann C.a ., s org M., Jongejans e ., s iepel h ., h ofland n ., s chwan h ., s ten-
mans W., Müler a ., s umser h ., h örent ., g olulson D., de Kroon h . 2017:
More than 75 percent decline over 27 years in total flying insect biomass in pro-
tected areas. PLoS ONE, 12 (10): e0185809
g ogala a . 2014: č ebele Slovenije. Ljubljana, Založba ZRC, 180 str.
Ksiazek K., f ant J., s kogen K. 2012: An asssement of pollen limitation on Chicago
green roofs. Landscape and Urban Planning, 107: 401 –408.
Macivor J.s ., r uttan r .,  s alehi B. 2015: e xotics on exotics: Pollen analysis of
urban bees visiting Sedum on a green roof. Urban Ecosystems,  18: 419–430.
n ieto a ., r oberts s .p.M., Kemp J., r asmont p., Kuhlmann M., Criado M.g ,
Biesmeijer J.C., Bogusch p., Dathe h .h ., De la r úa, p., De Meulemeester
t ., Dehon M., Dewulf a ., o rtiz-s ánchez f .J., l homme p., pauly a ., potts
s .g ., praz C., Quaranta M., r adchenko v .g ., s cheuchl e ., s mit J., s traka
J., t erzo t ., t omozii B., Window J., Michez D. 2014: e uropean red list of
bees. Luxembourg: Publication Office of the e uropean u nion. 84 str. PLoS One,
8, 6: e66665.
papanikolaou a .D., Kühn i., f renzel M., s chweiger o . 2017: Semi-natural habitats
mitigate the effects of temperature rise on wild bees. Journal of Applied Ecology,
54: 527–536.
potts s ., Biesmeijer J.C., Kremen C., n eumann p., s chweiger o ., Kunin W.e .
2010: Global pollinator decline: trends, impacts and drivers. Trends in
Ecology&Evolution, 25 (6): 345–353.
potts s ., Biesmeijer K., Bommarco r ., Breeze t ., Carvalheiro l ., f ranzen M.,
g onzalez-v aro J.p., h olzschuh a ., Kleijn D., Klein a .-M., Kunin B., l ecocq
t ., l undin o ., Michez D., n eumann p., n ieto a ., penev l ., r asmont p.,
r atamaki o ., r iedinger v ., r oberts s .p.M., r undlof M., s cheper J.,
s orensen p., s teffan-Dewenter i., s toev p., v ila M., s chweiger o . 2015:
Status and trends of e uropean pollinators. Pensoft Publishers, Sofia, 72 p.
potts s .g ., imperatriz-f onseca v .l ., n go h .t ., Biesmeijer J.C., Breeze t .D.,
Dicks l .v ., g aribaldi l . a ., h ill r ., s ettele J., v anbergen a . J., a izen M.
a ., Cunningham s .a ., e ardley C., Bf reitas. M., g allai n ., Kevan p. g . ,
Kovács-h ostyánszki a ., Kwapong p. K., J. l i, l i X., Martins D. J., n ates-
parra g ., pettis J. s . , r ader r ., v iana B.f . 2016: The assessment report on
Pollinators, pollination and food production. Secretariat of the Intergovernmental
Science-Policy Platform on Biodiversity and e cosystem Services, Bonn, Ger-
many. 36 pages.
Danilo Bevk: The diversity of pollinators on green roofs
13

t onietto r , f ant J., a scher J., e llis K, l arkin D. 2011. A comparison of bee com-
munities of Chicago green roofs, parks and prairies. Landscape and Urban Plan-
ning, 103: 102 –108.
Wagner D. 2020: Insect Declines in the Anthropocene. Annual Review of Entomology,
65:457–80.
Received / Prejeto:  8. 10. 2020
Acta entomologica slovenica, 29 (1), 2021
14