In April 2017, the Regional Society for the Protection of Birds (LBV) and the Department of Horticulture of the city of Munich created eight 1000 m²-large flower strips; besides these eight strips, we included another 1000 m²-strip established by the same group in 2015 (Fig. 1). All strips, which initially were covered by lawn or roadside greenery, were ploughed by machine and then sown with regional seeds from the seed supplier Kirmer (http://www.krimmer-naturnahes-gruen.de), adapted either for nutrient-rich or nutrient-poor sites, and the LBV also provided man-made nesting sites for cavity-breeding bees at the sites. They therefore all started from bare soil. Flowering plants present after the initial sowing were identified in randomly placed plots of one square-meter per strip at the strips Fockensteinstraße (established in 2015), Willy-Brandt-Allee and Rathenaustraße (nutrient-poor sites established in 2017), and Pasing Stadtpark, and Werner-Seelenbinder-Weg (nutrient-rich sites established in 2017). Plant species found on each strip are listed in Suppl. material 1, Table S1, along with information on herbarium voucher specimens deposited in the Munich herbarium (Botanische Staatssammlung, international acronym M).

Figure 1. 

The nine flower strips monitored for this study (modified from https://www.openstreetmap.org, using QGIS 3.8.2 (QGIS Development Team 2019) and Munich’s bee records (sightings and/or specimens) between 1997 and 2017.

From March to August in 2017 and 2018, each flower strip was visited four to five times. Visits were made between 10 a.m. and 4 p.m. on sunny, warm days with little or no wind. Where possible, bee species were identified directly in the field and were documented via macro-photography in a standardized setup: for close-up pictures, the bees were caught with an insect net and cooled down for 10 minutes in an Eppendorf screw-capped plastic vial stored on ice in a cooled box. When they fell into rigor of cold, they were transferred onto scale paper (using a small box lined with millimetre paper on its bottom) and photographed from all sides (SLR camera: Pentax K-x; Lens: Sigma DG 17-70 mm, 1:2.8, macro). Within one to two minutes, bees warmed up again and were released at the location where they had been caught. For species that are difficult to identify by morphology alone, such as species of Sphecodes, Lasioglossum or Halictus, voucher specimens, preferably males (for re-identification by genitalia preparations), were collected and identified morphologically and via DNA barcoding (methods and primers as described in Hofmann et al. 2018). The voucher specimens are deposited in the Zoologische Staatssammlung Munich (ZSM). Photo vouchers are accessible at the Diversity Workbench server (DWB; https://diversityworkbench.net/Portal/Diversity_Workbench), and DNA barcodes at NCBI GenBank (https://www.ncbi.nlm.nih.gov/genbank). Suppl. material 2, Table S2 in the Online Supporting Material shows all GenBank and DWB accession numbers. Additionally, Suppl. material 2, Table S2 shows each species’ Red List status based on Westrich et al. (2011) as well as foraging and nesting preferences based on Scheuchl and Willner (2016).

To investigate the catchment area of each flower strip, we analysed 7589 records with Gauß-Krüger coordinates made between 1997 and 2017 of single bees or populations and documented either by specimens stored at the Zoological Collections in Munich and/or by taxonomic assessments in Munich’s red lists. We focused on the area within a radius of 500, 1000, and 1500 m from each strip using QGIS 3.8.2 (QGIS Development Team 2019). For species that were recorded on a flower strip, but not within the 1500 m radius from the strip, we measured the distance from the strip to the nearest sighting of the respective species (Suppl. material 3, Table S3). For Fockensteinstraße and Willy-Brandt-Allee, we increased the radius to 1600 m, as there were too few records within the 1500 m radius, while a 1600 m radius yielded comparable numbers of records to those of the other sites.

On the nine 1000 m²-large flower strips, we found 83 species of flowering plants, 35 of them coming from the regional seed mix (Materials and Methods) and 17 self-sown at Fockensteinstraße (Fig. 2), 27 from the seed mix and 28 self-sown at Rathenaustraße and Willy-Brandt-Allee, and all 23 from the seed mix at Pasinger Stadtpark and Werner-Seelenbinder-Weg (see Suppl. material 1, Table S1 for species lists for each site). The flowers of these plants were visited by honey bees and 68 species of wild bees, that is 21% of the 324 species recorded for Munich since 1795 and 29% of the 232 re-observed or newly observed species over the last twenty years (1997–2017).

Figure 2. 

The flower strip at Fockensteinstraße as an example of the urban context of the flower strips studied here.

Of the 68 species, 62 (91%) have the Red List category ‘not threatened,’ three (4%) are listed on the pre-warning-list, and three are ‘threatened’ (Suppl. material 2, Table S2). The respective percentages for the 324-species-pool are 54% (n = 174) not threatened, 11% (n = 35) on the pre-warning list, and 27% (n = 89) threatened. Twenty-two of the 324 species are not Red-listed due to a lack of data, and three are considered locally extinct. There are thus significantly more non-threatened species on the flower strips than in Munich overall (chi-square test with 2 df, χ² = 26.4, P = 1.8 × 10^-6). Of the 232-species-pool recorded for 1997–2017, 156 (67%) species are non-threatened, 29 (13%) on the pre-warning list, 38 (16%) threatened, and 9 (4%) of unknown status. With these numbers, too, the flower strip fauna includes a disproportionate number of non-threatened species (chi-square test with 2 df, χ² = 12.5, P = 0.002).

Of the 68 species found on the strips, 63% (n = 43) are polylectic and 15% (n = 10) oligolectic. Some 22% (n = 15) parasitize other bee species (Suppl. material 2, Table S2). The respective percentages for the 324 species pool are 51% (n = 165) polylectic, 22% (n = 72) oligolectic, and 27% (n = 87) parasitic (Hofmann and Renner, in review), while in the 232 species pool of the last 20 years, 50% (n = 118) of species are polylectic, 25% (n = 59) oligolectic, and 24% (n = 55) parasitic. Of the oligolectic flower strip visitors, seven specialized on Asteraceae, two on Campanula (Campanulaceae), one on Echium (Boraginaceae), and one on Fabaceae. There is thus no significant difference in the frequencies of polylectic, oligolectic, or parasitic species among the flower strips and the remainder of Munich either for the larger pool (chi-square test with 2 df, χ² = 3.62, P = 0.164) or the smaller 1997–2017 pool (2df, χ² = 4.19, P = 0.123).

Our quantification of species recorded between 1997 and 2017 within a radius of 500, 1000, or 1500 m around each of the nine flower strips revealed that the strips at Altostraße (400 records of 105 different species) and Pasinger Stadtpark (329 records of 156 species) were richest in bees within a radius of 1500 m around them, while Weitlstraße and Willy-Brandt-Allee (19 records of 15 species each) have the fewest records within 1500/1600 m around them (Fig. 3; Suppl. material 3, Table S3).

Figure 3. 

Numbers of species and individuals recorded between 1997 and 2017 within a radius of 500, 1000, and 1500 m from the centre of the respective flower strip (compare Fig. 1). For details of how past recordings were made see Materials and Methods.

Figure 4. 

The percentage of bee species recorded at each flower strip that is also found in the surrounding area at distances of 500, 1000, and 1500 m.