Restore arable land to permanent grassland
Overall effectiveness category Awaiting assessment
Number of studies: 10
Background information and definitions
Large areas of grassland have been converted to arable land, leading to a loss of grassland-dependent species. Meadows can support a higher abundance and species richness of butterflies than arable land (Luppi et al. 2018). Restoring arable land to permanent grassland may recreate communities of invertebrates similar to those historically found in an area (Denning & Foster 2018).
For studies of short-term or rotational set-aside, see “Provide or retain set-aside areas in farmland”. For studies on the restoration of species-rich grassland from intensively managed grassland, see “Reduce management intensity on permanent grasslands (several interventions at once)”, “Reduce grazing intensity on grassland by reducing stocking density”, “Reduce grazing intensity on grassland by seasonal removal of livestock” and “Reduce cutting frequency on grassland”. For studies on the restoration of abandoned grassland, see “Restore or create species-rich, semi-natural grassland”. For studies on grassland restoration outside of a farmland context, see “Habitat restoration and creation – Restore or create grassland/savannas”.
Denning K.R. & Foster B.L. (2018) Flower visitor communities are similar on remnant and reconstructed tallgrass prairies despite forb community differences. Restoration Ecology, 26, 751–759.
Luppi M., Dondina O., Orioli V. & Bani L. (2018) Local and landscape drivers of butterfly richness and abundance in a human-dominated area. Agriculture, Ecosystems & Environment, 254, 138–148.
Supporting evidence from individual studies
A replicated, site comparison study in 1999 on three mixed farms in central Switzerland (Bosshard & Kuster 2001) reported that 4–5-year-old flower-rich meadows created on set-aside land had a higher species richness of butterflies and burnet moths than intensively managed meadows, pasture or arable land, and similar species richness to traditionally managed meadows. Results were not tested for statistical significance. Recently created meadows had approximately 14 species of butterflies and burnet moths, compared to 5–7 species in intensively managed meadows or pasture, 10–12 species on traditional meadows, and 1 species in arable fields (data presented for only one farm). Authors reported that adult butterfly abundance was positively correlated with the number of flowers, and up to 98% of flower visits were recorded on only five plant species. In 1994–1995, species-rich grassland was created across 2–6% of the farmed area on three mixed farms (10–25 ha). From May–September 1999, butterflies were surveyed seven times along fixed 10-m-long transects through each habitat type on each farm.Study and other actions tested
A before-and-after study in 1994–2006 on a farm in Oxfordshire, UK (Taylor & Morecroft 2009) found that following adoption of the Environmentally Sensitive Areas scheme, including reverting arable land to permanent grassland, the abundance and species richness of large moths and some species of butterfly increased. After Environmentally Sensitive Area management began, the total abundance (1,000–1,450 individuals) and species richness of large moth species was higher than before (800–1,250 individuals, richness data not presented). One of the five most abundant moth species (lunar underwing Omphaloscelis lunosa) and five of 23 butterfly species (meadow brown Maniola jurtina, brown argus Aricia agestis, common blue Polyommatus icarus, small copper Lycaena phlaeas and red admiral Vanessa atalanta) increased in abundance after the change in management. However, two butterfly species became less abundant (green-veined white Pieris napi and large white Pieris brassicae, data presented as model results). Overall butterfly abundance and species richness increased over the entire monitoring period, but the increase did not just happen after the management change. In 2002, the farm entered the Environmentally Sensitive Areas agri-environment scheme, and 102 ha of arable land was reverted to extensive grassland. In addition, fertilizers, herbicides and pesticides were no longer used, and the total number of livestock dropped from 180 cows and 1,000 sheep to 120 cows and 850 sheep. Butterflies were monitored weekly from April–September on a fixed 3.6 km transect divided into 13 sections. Moths were monitored nightly from dusk to dawn using a light trap in a fixed position in the middle of the farm.Study and other actions tested
A replicated, paired, site comparison study in 2008 on 32 farms in central Scotland, UK (Fuentes-Montemayor et al. 2011) found that species-rich grassland created under agri-environment schemes (AES) had a higher abundance and species richness of micro- and macro-moths than conventionally-managed grassland or crop fields. In created AES species-rich grasslands, the abundance (156 individuals) and species richness (24 species) of micro-moths, the species richness of all macro-moths (46 species), and the abundance of declining macro-moths (44 individuals) were all higher than in improved grasslands or crop fields on conventional farms (micro-moths: 43 individuals, 19 species; all macro-moths: 33 species; declining macro-moths: 21 individuals). However, the abundance of all macro-moths (366 individuals) and species richness of declining macro-moths (10 species) on created AES species-rich grasslands was not significantly different from improved grasslands or crop fields (all macro-moths: 271 individuals; declining macro-moths: 9 species). In 2004, sixteen farms enrolled in AES, and were paired with 16 similar but conventionally-managed farms, <8 km away. On AES farms, species-rich grassland was created on former arable or improved grassland fields by sowing a low productivity grass and herb seed mix, and managed with fertilizer and pesticide restrictions, and no summer cutting or grazing. Improved pastures and crop fields on conventional farms had no management restrictions. From June–September 2008, moths were collected for four hours, on one night/farm, using a 6 W heath light trap located in one field on each farm. Paired farms were surveyed on the same night.Study and other actions tested
A replicated, site comparison study (years not given) in 10 grasslands in England, UK (Woodcock et al. 2012) found that grasslands restored from bare soil by seeding developed butterfly communities increasingly similar to existing high-quality grasslands over the first 10 years after establishment, but the number of species present remained similar. The butterfly communities on grasslands restored by arable reversion were more similar to those on existing grasslands 10–21 years after restoration (42–84% similarity) than one year after restoration (0–33% similarity). However, the number of butterfly species recorded each year on arable reversion sites (~12 species/year) remained similar over time. Four grasslands were restored from bare soil by sowing grassland seed mixes. Three of the sites (two former arable fields and one abandoned road covered with top soil) were then managed by sheep-grazing to produce calcareous grassland, while the fourth site (ex-landfill covered with topsoil) was cut annually and grazed by sheep or cattle to produce a lowland hay meadow. Six high-quality grasslands (three calcareous grasslands and three hay meadows) were used for comparison. From April–September each year, butterflies were surveyed weekly on a ~2 km transect at each site for 9–21 years after restoration.Study and other actions tested
A replicated, randomized, paired, controlled study in 2008–2012 on a farm in Berkshire, UK (Woodcock et al. 2014) found that grasslands established with seed mixes containing legumes and other non-woody, broadleaved plants (forbs) had a higher abundance and species richness of pollinators (including butterflies) than grasslands sown only with grasses. Plots sown with a mix of grasses, legumes and forbs had a higher abundance (9–70 individuals/plot) and species richness (3–7 species/plot) of pollinators over four years than plots sown with grasses only (abundance: 0–3 individuals/plot; richness: 0–2 species/plot). In the first year after establishment, plots sown with grasses and legumes but no forbs had the highest abundance (15–91 individuals/plot) and species richness (5–8 species/plot) of pollinators, but this decreased over time (fourth-year abundance: 3–8 individuals/plot; richness: 2–3 species/plot). Grass and legume plots managed by cutting had a higher abundance (6–91 individuals/plot) and species richness (3–8 species/plot) of pollinators than plots managed by grazing (abundance: 3–33 individuals/plot; richness: 2–5 species/plot). Management had less effect on other seed mixes. In spring 2008, ninety-six 875-m2 plots were sown with one of three seed mixes: a “grass only” mix of five species (30 kg/ha, cost: €83/ha); a “grass and legume” mix of five grasses and seven agricultural legumes (34 kg/ha, €120/ha); or a “grass, legume and forb” mix of five grasses, seven legumes and six non-legume forbs (33.5 kg/ha, €190/ha). Half of the plots were grazed with cattle (3 animals/ha) and half were cut to 10 cm once or twice/year. In May, July and August 2009–2012, butterflies, bees (Apidae) and hoverflies (Syrphidae) were surveyed three times/year on two parallel 20 × 2 m transects/plot.Study and other actions tested
A replicated, site comparison study in 2013 in 40 grasslands in southern Finland (Toivonen et al. 2015, same experimental set-up as Toivonen et al. 2016) found that long-term restored grassland fallows had a greater abundance and species richness of butterflies than recently established meadow fallows. In ≥8-year-old sown grasslands, the abundance (55–85 individuals/site) and species richness (7–12 species/site) of butterflies was higher than in 3–4-year-old sown meadows (abundance: 34–52 individuals/site; richness: 7 species/site). Forty fallow grasslands (0.3–5.8 ha) established under the Finnish Environmental Fallow agri-environment scheme were selected. Twenty long-term grassland fallows (≥8 years old) were either former set-aside areas or production grasslands, originally established by sowing conventional, competitive, perennial grassland mixtures. Twenty short-term meadow fallows (3–4 years old) were established by sowing low competitive meadow plants (see paper for details), which required re-establishment every 4–5 years. All sites were mown at least every three years, and no pesticides or fertilizers were applied. From June–July 2013, butterflies were surveyed four times (two weeks apart) along a 200-m transect in each fallow.Study and other actions tested
A replicated, site comparison study in 2013 in 40 grasslands in southern Finland (Toivonen et al. 2016, same experimental set-up as Toivonen et al. 2015) found that long-term restored grassland fallows had more specialist butterflies than recently established meadow fallows. In ≥8-year-old grasslands, the relative abundance of habitat specialist butterflies was higher than in 3–4-year-old meadows (data presented as model results). Four species (large skipper Ochlodes sylvanus, Essex skipper Thymelicus lineola, lesser marbled fritillary Brenthis ino, mazarine blue Polyommatus semiargus) were strongly associated with ≥8-year-old grasslands, while no species was strongly associated with 3–4-year-old meadows. Forty fallow grasslands (0.3–5.8 ha) established under the Finnish Environmental Fallow agri-environment scheme were selected. Twenty long-term grassland fallows (≥8 years old) were either former set-aside areas or production grasslands, originally established by sowing conventional, competitive, perennial grassland mixtures. Twenty short-term meadow fallows (3–4 years old) were established by sowing low competitive meadow plants (see paper for details). All sites were mown at least every three years, and no pesticides or fertilizers were applied. From June–July 2013, butterflies were surveyed four times (two weeks apart) along a 200-m transect in each fallow.Study and other actions tested
A replicated, site comparison study in 2014 on 52 fields in arable reversion in southern England, UK (Woodcock et al. 2016) found that neither the method of restoring arable land to permanent grassland, nor current management of the field, affected adult butterfly and day-flying moth species richness or caterpillar abundance, but species richness of adult butterflies was lower in fields restored longer ago. One to 30 years after arable reversion began, butterfly species richness and caterpillar abundance were similar on fields established by sowing complex or simple seed mixes, or by allowing natural regeneration, and on fields managed by sheep or cattle grazing, and with or without mowing (data not presented). The species richness of adult butterflies was lower on arable reversion fields >20 years old (0–6 species/site) than fields <10 years old (1–8 species/site), but caterpillar abundance was similar (data not presented). Between 1984–2013, restoration of 52 former arable fields (1.0–22.8 ha) to calcareous grasslands began. Fields were restored by natural regeneration, re-seeding with simple grass or complex grass and non-woody broadleaved plant (forb) mixes, or by spreading green hay. Fields were cut every 1–4 years (normally after 15 July) and lightly grazed (typically 1 livestock unit/ha) by sheep or cattle, with some fields ungrazed. From July–August 2014, adult butterflies and day-flying moths were surveyed twice/day on three days, and caterpillars were sampled by 20 sweeps/day of a net, along a 100-m transect at each site.Study and other actions tested
Referenced paperWoodcock B., Ball S., Amy S., Edwards M., Redhead J., Mountford O., Gregory S., Duffield S., Macgregor N. & Pywell R. (2016) Evaluating the relative importance of site and landscape characteristics for invertebrate communities in grasslands restored through agri-environment schemes. Natural England report, DEFRA Report RP01878.
A replicated, paired, site comparison study in 2015 on 22 farms in Berkshire, Hampshire and Wiltshire, UK (Alison et al. 2017) found that restored grassland supported a higher abundance of moths than unrestored arable fields, and was similar to semi-natural grassland sites. Three to 20 years after restoration, the abundance of moths associated with calcareous grassland (6.3 individuals/trap) and other grassland (49.6 individuals/trap) on restored fields were higher than on arable fields (calcareous: 0.8; other: 14.6 individuals/trap), and similar to semi-natural grassland (calcareous: 7.2; other: 38.3 individuals/trap). The abundance of moths associated with other habitats was higher on restored (25.5 individuals/trap) than unrestored fields (15.3 individuals/trap), but lower than on semi-natural grassland sites (57.9 individuals/trap). Results for species occurrence were similar (data not presented). However, neither moth abundance nor occurrence increased with time since restoration (data not presented). Over 3–20 years, 32 former arable fields (2.6–37.5 ha) on 22 farms were restored to species-rich grassland by either natural regeneration or sowing of wildflowers. All were cut or grazed at least once/year. Thirty-two paired, arable fields (2.2–49.3 ha) were unrestored, and eight semi-natural calcareous grasslands were used for comparison. On 21 nights between June–September 2015, moths were surveyed twice/site (2–4 restored-unrestored pairs/night, with a comparison site on >50% of nights) using one 15 W light trap in the centre of each field. Moths were classified as species associated with calcareous grassland, associated with grassland generally, or not associated with grassland.Study and other actions tested
A replicated, paired, site comparison study in 2017 on four farms in Hualien County, Taiwan (Yen et al. 2018) found that former cropland restored by natural regeneration had a higher species richness of butterflies, but a similar total abundance, than cultivated farms. On uncultivated, restored farms, the species richness of butterflies (16 species/farm) was higher than on active, conventional farms (9 species/farm), but the abundance of butterflies was similar between farms (restored: 185 individuals/ha; active: 191 individuals/ha). Within a National Park, 78 ha of restored former farmland had not been cultivated since the Park was established (number of years not given), and 39 ha of farmland remained in production. In each of two areas, one restored and one active farm were selected. Farms were 250–3,200 m apart. From May–September 2017, butterflies were surveyed once/month along 150-m transects at each farm (number not specified).Study and other actions tested