Alter mowing regimes on greenspaces and road verges
Overall effectiveness category Awaiting assessment
Number of studies: 7
Background information and definitions
Greenspaces and road verges offer potential habitat patches and corridors for butterflies and moths. However, management of these spaces must be carefully planned. Mowing may be required to keep vegetation down, but also reduces the availability of nectar resources, inhibits the growth of host plants used by caterpillars, reduces the structural diversity of the vegetation, and poses a risk from direct mortality, especially to eggs, caterpillars and pupae (Hopwood et al. 2015). Road verges with a higher species richness of plants attract more butterflies, and yet have fewer butterflies killed by traffic than less diverse road verges (Skórka et al. 2013). Therefore, altering mowing regimes to encourage plant growth and diversity may benefit butterfly and moth populations in urban areas.
See also: “Transportation and service corridors – Restore or maintain species-rich grassland along road/railway verges” and “Pollution – Stop using herbicides on pavements and road verges”.
Hopwood J., Black S.H., Lee-Mäder E., Charlap A., Preston R., Mozumder K. & Fleury S. (2015) Literature review: Pollinator habitat enhancement and best management practices in highway rights-of-way. Federal Highway Administration report by The Xerces Society for Invertebrate Conservation & ICF International.
Skórka P., Lenda M., Moroń D., Kalarus K. & Tryjanowski P. (2013) Factors affecting road mortality and the suitability of road verges for butterflies. Biological Conservation, 159, 148–157.
Supporting evidence from individual studies
A study in 2003 along a highway in South Karelia, Finland (Valtonen and Saarinen 2005) found that roadsides mown in had more ringlet butterflies Aphantopus hyperantus than those mown in mid-summer. There were more ringlets along road verges that were mown in late summer (estimated population size = 840–2720; population density = 1160–4360 individuals/ha) than those mown in mid-summer (estimated population size: 220–1500 individuals; population density: 500–1200 individuals/ha). A lower percentage of butterflies first caught along verges mown in late summer (11–31%) moved out of their original patch than those caught along verges mown in mid-summer (43–63%) (these data were not tested for statistical significance). On 30 days between June and August 2003, butterflies were sampled using mark-release-recapture surveys along a 2.2-km long transect along a highway intersection and adjacent roads. The site was divided into 80–280-m sections based on habitat: an intersection between two roads with either late summer mowing and hay removal or with mid-summer (June) mowing without hay removal, and a highway verge with either partial July and full late summer mowing without hay removal (counted as late summer mowing in the study) or with mid-summer (July) mowing without hay removal. The effect of hay removal was not tested.Study and other actions tested
A replicated, randomized, controlled study in 2004–2006 in 25 roadside verge sites in Limburg, the Netherlands (Noordijk et al. 2009) found that butterflies were recorded along verges with no management and mowed verges where hay was removed but no butterflies were recorded along mowed road verges where hay was not removed. No butterflies were recorded along verges mowed either once or twice a year where hay was not removed, but of insects recorded along verges, butterflies represented 14% where mowed once a year with hay removal, 8% where mowed twice a year with hay removal and 7% with no management. In 2004–2006 five treatments (each replicated five times) were conducted in 12 x 15 m plots along a highway: mowing once a year in mid-September with or without hay removal, mowing twice a year in late June and mid-September with or without hay removal, and no management. On four visits in May–August 2006, flower-visiting insects were counted in four 1-m2 quadrats in each plot over a 15-minute period. Butterflies were not identified more precisely than order level and statistical tests did not look at the difference between the number of mowings or hay removal for butterflies alone.Study and other actions tested
A replicated, site comparison study in 2010 on 60 road verges in southern Poland (Skórka et al. 2013) found that less frequently or later mown road verges, which provided more suitable habitat, had fewer individuals and a lower species diversity of dead butterflies killed by traffic than more frequently or earlier mown verges. Both the number of individual butterflies and number of species killed by traffic were lower on verges mown less frequently, or later in the summer, than on more frequently or earlier mown verges (data presented as model results). Sixty roads, >2 km apart, with verges of similar width and vegetation on each side, were selected. Between roads, verges differed in the frequency and timing at which they were mown. From April–September 2010, butterflies were surveyed 12 times on two 100-m transects along each side of each road. Dead butterflies were collected from the asphalt and the first metre of verge next to the road.Study and other actions tested
A replicated, paired, controlled study in 2011 in 10 urban greenspaces in Baden-Württemberg, Germany (Kricke et al. 2014) found that less frequently mown areas had a greater diversity of butterflies and burnet moths than regularly mown areas. The species richness and diversity of butterflies and burnet moths in areas mown once in the summer (9 species/site) was higher than in areas mown every 3–4 weeks throughout the summer (4 species/site, diversity data presented as model results). In addition, four species only occurred on grasslands which had been mown once/year for >4 years, which had an average of 11 species/site (statistical significance not assessed). See paper for individual species results. One half of each of 10 public greenspaces (>200 m2) was mown or mulched once in July or August 2011, while the other half was mown or mulched once every 3–4 weeks from April–August 2011. Five additional sites had only been mown once/year for >4 years. From April–early August 2011, butterflies were surveyed five times in each site, by walking with nets in large loops until no new species was found for 20 minutes.Study and other actions tested
A replicated, paired, controlled study in 2013 in a public park in Sussex, UK (Garbuzov et al. 2015) found that unmown areas had higher abundance of butterflies and moths, and higher species richness of butterflies, than mown areas. The total abundance of butterflies (123 individuals) and moths (261 individuals) was higher in unmown strips than in strips mown every two weeks throughout the summer (butterflies: 32 individuals; moths: 23 individuals). In addition, the total abundance of butterflies (271 individuals) and moths (391 individuals), and the species richness of butterflies (8 species), were higher on the unmown half of the park than in the mown half (butterfly abundance: 6 individuals; moth abundance: 2 individuals; butterfly richness: 2 species) (moths not identified to species). From spring 2013, half of a 6-ha park was left unmown, while the other half continued to be mown every two weeks from spring to autumn. In addition, four blocks (20 × 30 m) of four strips (5 × 30 m) each were established in the unmown half of the park. Within each block, one strip was assigned to each of four mowing treatments: regular mowing every two weeks through the summer, regular mowing until 5 July, regular mowing until 2 June, and unmown. From June–September 2013, foraging and resting butterflies and moths were surveyed weekly by walking down the centre of each 30-m strip five times/visit. From July–September 2013, butterflies and moths were surveyed eight times on two 500-m transects, one around the regularly mown and one around the unmown half of the park.Study and other actions tested
A replicated, site comparison study in 2007–2008 along 52 road verges and power lines (collectively “transmission lines”) in Manitoba, Canada (Leston & Koper 2016) found that unmown transmission lines had more northern pearl crescent Phyciodes morpheus and pearl crescent Phyciodes tharos butterflies than lines mown twice/year, but mowing regime did not affect the abundance or species richness of other butterflies. There were more crescent butterflies on unmown transmission lines (2.7 individuals/visit) than on lines mown twice/year (0.1 individuals/visit). However, the abundance and species richness of other native butterflies was not significantly different between transmission lines which were not mown (abundance: 11 individuals/visit; richness: 32 species), mown once/year and not hayed (11 individuals/visit; 27 species), mown once/year and hayed (14 individuals/visit; 21 species), or mown twice/year and sprayed with herbicide (10 individuals/visit; 21 species). See paper for species results. Fifty-two road verges and power lines (>30 m wide, >400 m long) were managed in one of four ways: 21 were neither mown nor sprayed with herbicide, but some trees were removed; 14 were mown twice/year with cuttings left on site and sprayed frequently with herbicide; 10 were mown once/year with cuttings left on site and sprayed infrequently with herbicide; seven were mown once/year with cuttings baled and removed with no spraying. From 15 June–15 August 2007–2008, butterflies were surveyed on one 400- or 500-m transect at each site 2–4 times/year.Study and other actions tested
A replicated, site comparison study in 2015 in 30 grassland patches around an urban area in Scania, Sweden (Haaland 2017) reported that biodiversity areas mown once/year had a higher occupancy of scarce copper butterflies Lycaena virgaureae than regularly mown public parks, but lower occupancy than unmanaged grasslands. Results were not tested for statistical significance. Six biodiversity areas managed by the local authority were all occupied at least once, and often 2–3 times, by scarce coppers, but no butterflies were seen in nine regularly mown parks. However, 15 unmanaged grasslands were the most frequently occupied areas on all four surveys (data not presented). Thirty grassland habitat patches managed in three ways were studied: six biodiversity areas (cut once/year in mid-August), nine parks (cut several times/year) and 15 unmanaged grasslands (no cutting or grazing). From July–August 2015, butterflies were surveyed four times in each of 30 patches by systematic searching.Study and other actions tested