Reduce grazing intensity on grassland by reducing stocking density

How is the evidence assessed?
  • Effectiveness
    not assessed
  • Certainty
    not assessed
  • Harms
    not assessed

Study locations

Key messages

COMMUNITY RESPONSE (8 STUDIES)

  • Richness/diversity (8 studies): Five of eight replicated studies (including two randomized, controlled studies and six site comparison studies) in the UK, Sweden, Germany, the USA and Switzerland found that grasslands grazed with lower stocking densities of sheep, cattle or a mix of sheep, cattle and horses had a greater species richness of adult butterflies, all moths and burnet moths than grassland grazed at higher stocking densities. However, one of these studies also found that butterfly and burnet moth caterpillar species richness was similar at sites with low and high stocking densities. Two of the other studies found that grasslands grazed with lower stocking densities of cattle and horses or unspecified grazing animals had a similar species richness of butterflies and burnet moths to grassland grazed at higher stocking densities. The other study found that, in one of two study years, grasslands grazed with cattle at a low density had lower species richness than grasslands grazed at moderate density.

POPULATION RESPONSE (12 STUDIES)

  • Abundance (11 studies): Eight replicated studies (including four controlled studies and four site comparison studies) in the UK, Germany, the USA and Sweden found that grasslands grazed with lower stocking densities of sheep, cattle or both (in one case combined with a later start to grazing) had a greater abundance of all butterflies, butterflies with grass host plants, all moths, burnet moths, their caterpillars or specific species (in two cases as part of combined invertebrate counts) than grasslands grazed at higher stocking densities. The three studies on caterpillars only found a higher abundance at two out of three sites or in earlier or later sampling periods, and one of the studies found that sites with low and high intensity grazing had a similar abundance of butterfly and burnet moth caterpillars. Two replicated, site comparison studies in Sweden and Switzerland found that grasslands grazed with lower stocking densities of cattle and horses or unspecified grazing animals had a similar abundance of butterflies and burnet moths to grassland grazed at higher stocking densities. One review of studies in Europe reported that reducing grazing intensity benefitted 41 out of 67 butterfly species of conservation concern, but did not distinguish between reducing stocking density and seasonal removal of livestock.
  • Survival (1 study): One site comparison study in Belgium and the Netherlands reported that the survival of Glanville fritillary caterpillar nests was similar between grasslands with low and high stocking density of sheep.
  • Condition (1 study): One site comparison study in Belgium and the Netherlands found that after 6–10 days of sheep grazing, fewer Glanville fritillary caterpillar nests were damaged in a grassland with lower stocking density than in a grassland with higher stocking density, but there was no difference after two months.

BEHAVIOUR (0 STUDIES)

About key messages

Key messages provide a descriptive index to studies we have found that test this intervention.

Studies are not directly comparable or of equal value. When making decisions based on this evidence, you should consider factors such as study size, study design, reported metrics and relevance of the study to your situation, rather than simply counting the number of studies that support a particular interpretation.

Supporting evidence from individual studies

  1. A replicated, paired, site comparison study in 1989–1993 in three upland grasslands in Scotland, UK (Dennis et al. 1998) found that two of three sites with reduced stocking density had a higher abundance of small invertebrates (including caterpillars) than sites with higher stocking density. In two out of three grasslands, the abundance of invertebrates was higher in plots with reduced stocking density (6–125 individuals) and ungrazed plots (70–250 individuals) than in plots with higher stocking density (4–70 individuals). At the third site, there was no significant difference between reduced stocking (17–78 individuals), ungrazed (35–78 individuals) and higher stocking density plots (17–55 individuals). From 1989–1991, at three sites, experimental grazing plots were established where the number of sheep was adjusted weekly in order to maintain different sward heights from May–October each year. At two sites, two 0.3-ha plots had sward kept at each of 3.0, 4.5 (high stocking density) or 6.0 cm (reduced stocking density). At the third site, four 1–3 ha plots had sward kept at each of 4.5 and 6.5 cm, but from June–August six cattle were grazed on half of the plots. Separate plots which had been ungrazed for one, four or 25 years were also monitored at each site. In August 1993, invertebrates (insects and arachnids) were sampled from both tussocks and low sward at each of six randomly selected points/plot using a d-vac suction sampler.

    Study and other actions tested
  2. A replicated, site comparison study in 1997 in 20 pastures in the Uppsala region, Sweden (Söderström et al 2001) found that on pastures with lower grazing intensity, butterfly species richness was higher. In pastures with lower grazing intensity butterfly richness was higher (data presented as model results). Pastures were mostly cattle-grazed annually from May–October. Grazing intensity was measured as the proportion of grass shorter than 5 cm throughout the period. Grass height was estimated three times in May–June 1997 in 1-ha quadrats covering the whole of each of 20 grazed semi-natural pastures. In July 1997, butterflies were surveyed along the borders of a 100 m2 square walking transect in pastures, repeated 18 times/pasture.

    Study and other actions tested
  3. A replicated, site comparison study in 1996 in 18 grasslands in Schleswig-Holstein, Germany (Kruess and Tscharntke, 2002) found that the species richness and abundance of adult butterflies and burnet moths at sites grazed at low intensity was higher than at intensively grazed sites and similar to ungrazed sites, but the responses of caterpillar richness and abundance to grazing intensity were mixed. Butterfly and burnet moth species richness and abundance were higher at low intensity grazing (average richness: 4; average abundance: 8) and ungrazed (average richness: 7; average abundance: 21) sites than at sites which were intensively grazed (average richness: 2; average abundance: 1). Caterpillar species richness was statistically similar at low intensity grazing (average: 4) and intensively grazed sites (average: 2), but lower in both than in ungrazed sites (average: 7). Caterpillar abundance was statistically similar in low intensity grazed sites (average: 9) and both intensively (average: 1) and ungrazed (average: 12) sites, but ungrazed sites had higher abundance than intensively grazed sites. Intensively grazed sites (6) had an average stocking density of 1.4 cattle/ha annually from May–November, low intensity grazed sites (6) had an average density of 5.5 cattle/ha annually from April–October, and ungrazed sites (6) had not been stocked for 5–10 years. From May–September 1996, at each site separate transect walks were conducted once monthly for adult butterflies and burnet moths (45-minute transects) and caterpillars (45-minute transects in grazed sites and 60-minutes in ungrazed sites).

    Study and other actions tested
  4. A replicated, randomized, controlled study in 2003–2005 on an upland grassland in Perthshire, UK (Dennis et al. 2008, same experimental set-up as 3) found that plots grazed with a lower stocking density had a higher abundance of moth caterpillars than commercially grazed plots, but only after >2 years. After 18 months of grazing, there was no significant difference in the number of caterpillars on lightly grazed (1.9–2.4 individuals/plot), commercially grazed (2.3 individuals/plot) or ungrazed plots (2.8 individuals/plot). However, after 30 months, there were more caterpillars in the lightly grazed plots (1.9–2.4 individuals/plot) than in the commercially grazed plots (0.5 individuals/plot), but fewer than in the ungrazed plots (4.9 individuals/plot). From January 2003, three grazing regimes (light grazing: sheep at 0.9 ewes/ha or sheep and cattle equivalent to 0.9 ewes/ha; commercial grazing: sheep at 2.7 ewes/ha) and an ungrazed treatment were replicated six times each in twenty-four 3.3-ha plots (in three pairs of adjacent blocks). Caterpillars were sampled by sweep net in 2003–2005.

    Study and other actions tested
  5. A replicated, randomized, controlled study in 2003–2007 on an upland estate in Scotland, UK (Littlewood 2008, same experimental set-up as 2) found that lightly grazed plots had a higher abundance and species richness of moths than plots grazed at a commercial stocking density. Plots grazed by sheep at low density had a higher abundance (52 individuals/night) and species richness (12.3 species/night) of moths than plots grazed by sheep at commercial densities (abundance: 34 individuals/night; richness: 10.6 species/night), or plots grazed by sheep and cattle at low density (abundance: 42 individuals/night; richness: 11.3 species/night), and were similar to ungrazed plots (abundance: 48 individuals/night; richness: 13.2 species/night). In January 2003, one of four grazing treatments was established on each of 24 plots (3.3 ha each) on a grazed acid grassland upland estate. The treatments were: low density sheep grazing (3 sheep/plot); commercial high density sheep grazing (9 sheep/plot); low density mixed grazing (2 sheep/plot plus two cows and calves for 4 weeks in autumn); ungrazed control. Moths were sampled between June and October 2007 using four 15 W light traps placed randomly within plots of each treatment, for six or seven sample nights/plot.

    Study and other actions tested
  6. A replicated, site comparison study in 2004 in an agricultural region in central Sweden (Sjodin et al. 2008) found that grasslands grazed at low intensity did not have a greater abundance or species richness of butterflies and burnet moths than intensively grazed or abandoned grasslands. On low intensity pasture, the abundance (22.9 individuals/visit) and species richness (9.1 species/visit) of butterflies and burnet moths was not significantly different from either intensively grazed pasture (abundance: 22.8 individuals/visit; richness: 9.4 species/visit) or abandoned grassland (abundance: 29.5 individuals/visit; richness: 10.4 species/visit). Three pastures, >2 km apart, were selected in each of eight sites (>10 km apart). Within a site, one low intensity pasture was managed by cattle or horse grazing, one high intensity pasture was managed by cattle grazing, and one abandoned pasture had been ungrazed for >10 years. From June–August 2004, flower-visiting insects were surveyed four times on four 5 × 5 m plots/pasture. Plots were observed for 10 minutes/visit.

    Study and other actions tested
  7. A replicated, site comparison study in 2003–2004 in 27 oak and conifer woodland and shrubland sites in Oregon, USA (Runquist 2011) found that at sites with no or low grazing intensity there was lower overall butterfly species richness in one of two years and higher abundance of butterflies with grass host plants in both years compared to sites with moderate grazing, but butterfly density and community evenness were not affected by grazing intensity. In 2003, but not 2004, butterfly species richness was lower along transects with low grazing intensity (0–15%) than those with moderate intensity (30–45%) (data presented as log species richness). There was no difference in species richness between any other grazing intensities (15–30%, 45–60% or 60–75%). Grazing intensity did not affect butterfly density or community evenness (data not provided). Butterflies with grass host plants were more abundant along transects with 0–30% grazing intensity than 30–45% intensity, but there were no differences between abundances at any of the other grazing intensity categories (data presented as log abundance). See paper for individual species results. Twenty-seven butterfly transects of varying lengths were conducted in April–September 2003 and 2004 either in oak woodland and shrubland or in meadows in areas dominated by mixed-conifer forest. Sites were cattle-grazed from May–November. Each transect was surveyed for butterflies 0–7 times in 2003 and eight times in 2004. In 2004, at each transect, cattle grazing intensity (0–75%) was estimated by measuring the height of that transect’s dominant cattle-palatable plant at every metre for 50 m, calculating the average height across the transect’s sampling points, and comparing this to the height of a reference ungrazed plant of that species from the same transect to calculate a grazing intensity percentage.

    Study and other actions tested
  8. A site comparison study in 2009–2010 in two calcareous grasslands in Belgium and the Netherlands (van Noordwijk et al. 2012) found that fewer Glanville fritillary Melitaea cinxia caterpillar nests were damaged at less intensively grazed and ungrazed sites than at a more intensively grazed site. After 6–10 days of autumn grazing, fewer caterpillar nests had signs of damage in a lightly grazed (2/25 nests damaged) and ungrazed (2/24 nests damaged) site than nests in a heavily grazed site (15/25 nests damaged). Two months later, the number of nests with signs of damage was similar in lightly grazed (3/25 nests damaged), ungrazed (6/24 nests damaged) and heavily grazed (6/25 nests damaged) areas. All 24 nests in the ungrazed area, and 24/25 nests in the lightly grazed area, survived until spring, compared to 22/25 surviving in the heavily grazed area (statistical significance not assessed). In July–August 2009, a lightly grazed 0.52-ha grassland and a heavily grazed 4-ha grassland were searched three times for caterpillar nests. At the larger site, half of the area with the most nests was fenced to create a 0.15-ha ungrazed site. The 24–25 largest nests (>1 m apart) in each site were selected, and their location marked on GPS. In September 2009, the 0.52-ha grassland was grazed by 26 sheep for six days, and a 1.23-ha area of the larger site was grazed by 114 sheep for 10 days, after which an expanded 1.76-ha area was grazed by 15 sheep for 50 days. In October and December 2009, nests were checked for damage, and in March 2010 the survival of each nest was recorded.

    Study and other actions tested
  9. A replicated, randomized, paired, controlled study in 2002–2011 in a grassland in Lower Saxony, Germany (Jerrentrup et al. 2014) found that grassland grazed at reduced cattle density had a higher species richness and abundance of butterflies than grassland grazed at moderate density, but there was no additional benefit of more recent very low density grazing. After 8–9 years, plots grazed at reduced density had a higher abundance (18.7–30.6 individuals/transect) and species richness (6.0–8.4 species/transect) of butterflies than plots grazed at moderate density (abundance: 8.1–20.0 individuals/transect, richness: 3.3–5.3 species/transect). However, on plots grazed at very low density butterfly abundance (16.3–34.1 individuals/transect) and species richness (5.4–7.8 species/transect) were similar to the reduced density plots. Over nine years, the abundance of marbled white Melanargia galathea increased more on reduced density plots (2002: 0–1 individuals; 2011: 4–12 individuals) than on moderate density plots (2002: 0–1 individuals; 2011: 1–5 individuals). From 2002–2011, three 3-ha paddocks were each divided into three treatments, each grazed annually from April–October: moderate density (3–6 cattle/ha, target sward height 6 cm), reduced density (1–3 cattle/ha, 12 cm sward) and very low density (2 cattle/ha, 18 cm sward, from 2005 only). Target sward heights were maintained by varying numbers of Simmental cattle in each paddock following biweekly sward height measurements. From July–September 2002–2004 and 2010–2011, butterflies were recorded biweekly on three 50-m transects/treatment.

    Study and other actions tested
  10. A review in 2015 of 126 studies in Europe (Bubová et al. 2015) reported that reducing grazing intensity on grassland benefitted 41 out of 67 butterfly species of conservation concern. Results were not tested for statistical significance. The review reported that 44 studies found that reducing grazing intensity benefitted 41 butterfly species, but did not distinguish between reducing stocking density and using seasonal grazing. The optimal grazing intensity was 0.2–0.5 livestock/ha (data not presented). See paper for information on individual species. Meadows were extensively grazed by different livestock and at different times, sometimes with rotational mowing. The review focussed on 67 butterfly species of conservation concern. The available information was biased towards studies in Northern and Western Europe.

    Study and other actions tested
  11. A replicated, paired, controlled study in 2010–2012 in six permanent pasture fields in Devon, UK (RSPB 2016) found that grassland managed with reduced stocking density initially had a higher abundance of invertebrates (including caterpillars) than conventional grassland, but the effect disappeared over three years. In the first year, the abundance of “bird food invertebrates” was higher on grassland grazed at a reduced density (202 individuals/plot) than on grassland grazed conventionally (112 individuals/plot). However, after two and three years, there was no significant difference between reduced (two years: 97; three years: 48 individuals/plot) and conventionally grazed plots (two years: 78; three years: 42 individuals/plot). From April 2010–September 2012, six permanent pasture fields were divided into two 1-ha plots. One plot/pair was grazed by cattle at reduced stocking density, managed to keep sward height between 9–12 cm, and the other was grazed at conventional stocking density, keeping the sward at 6–8 cm. None of the fields were fertilized during the trial. In July 2010–2012, invertebrates were sampled in 10 locations/plot using a Vortis suction sampler (ten 15-second samples over 0.19 m2) and sweep-netting (20 double sweeps with a 46-cm diameter net). Invertebrates <2 mm long were excluded from analysis.

    Study and other actions tested
  12. A replicated, site comparison study in 1984–2015 in 24 grasslands in Blekinge province, Sweden (Johansson et al. 2017) found that grasslands grazed lightly, with fewer animals or later in the year, had a higher abundance of clouded Apollo Parnassius mnemosyne than grasslands grazed heavily, with more animals or earlier in the summer. In grasslands managed by light grazing, the abundance of clouded Apollo (1–169 individuals/grassland/year) was higher than in heavily grazed grasslands (2–22 individuals/grassland/year) or ungrazed grasslands (0–109 individuals/grassland/year). In addition, abundance was higher on larger grasslands, and grasslands which were close together were more likely to be colonized (data presented as model results). From 1984–2015, twenty-four open grasslands (>150 m apart) with >0.5 m2 cover of the host plant Corydalis spp. and the presence of a major nectar plant Lychnis viscaria were assigned annually to one of three management categories: light grazing (grazing with 1–9 animals/hectare commenced after 15 June); heavy grazing (grazing with ≥10 animals/hectare for ≥8 weeks or commenced before 15 June); no grazing. Grazing animals were cattle and sheep. In 1984–1987, 1991 and 2003–2015, butterflies were surveyed ≥6 times/year on each site, by marking and recapturing individuals along irregular routes through each grassland. In 1988–1989 and 1992–2002, grasslands were visited more irregularly and their presence recorded. Surveys were used to estimate the local population size on each grassland each year.

    Study and other actions tested
  13. A replicated, site comparison study in 2014 in 26 grasslands in Germany (Mangels et al. 2017) reported that grasslands managed with lower stocking density (sometimes also mown) supported more moth species than grasslands grazed at higher stocking density. Results were not tested for statistical significance. Grasslands managed with a lower stocking density had more moth species (143 species) than grasslands managed with a higher stocking density (35 species). From 2006, across three regions, nine grasslands were managed by grazing (by cattle, sheep or horses) at low (0–113 livestock units/ha/year) or high density (113–520 livestock units/ha/year), nine were managed by mowing (1–2 cuts/year, often with nitrogen fertilization), and eight were grazed and mown (76–163 livestock units/ha/year; 1–2 cuts/year). Moths were collected once/month from nine grasslands in each of two regions (May–August 2014), and from eight grasslands in one region (June–July 2014). Each night, a 12 V actinic and black-light trap were placed in the centre of each of three grasslands for 138–317 minutes/night.

    Study and other actions tested
  14. A replicated, site comparison study in 2009–2011 in 133 mixed farms in the Central Plateau, Switzerland (Stoeckli et al. 2017) found that farms with a lower livestock density had a similar abundance and species richness of butterflies to farms with higher livestock density. Both the abundance and species richness of butterflies on farms with a lower density of livestock was similar to farms with higher livestock densities (data presented as model results). A total of 133 farms (17–34 ha, 13–91% arable crops) were managed with “Ecological Compensation Areas” under agri-environment schemes, and pastures were stocked at a range of livestock densities. From May–September 2009–2011, butterflies were surveyed six times on 10–38 transects/farm, totalling 2,500 m/farm. Each transect ran diagonally through a single crop or habitat type, with all available crops and habitats represented. All visits to a farm were completed in a single year, and the species richness was summed across all visits. Total abundance of butterflies was calculated from the number recorded in each habitat, and the availability of each habitat across the farm.

    Study and other actions tested
Please cite as:

Bladon A.J., Bladon, E. K., Smith R.K. & Sutherland W.J. (2023) Butterfly and Moth Conservation: Global Evidence for the Effects of Interventions for butterflies and moths. Conservation Evidence Series Synopsis. University of Cambridge, Cambridge, UK.

Where has this evidence come from?

List of journals searched by synopsis

All the journals searched for all synopses

Butterfly and Moth Conservation

This Action forms part of the Action Synopsis:

Butterfly and Moth Conservation
Butterfly and Moth Conservation

Butterfly and Moth Conservation - Published 2023

Butterfly and Moth Synopsis

What Works 2021 cover

What Works in Conservation

What Works in Conservation provides expert assessments of the effectiveness of actions, based on summarised evidence, in synopses. Subjects covered so far include amphibians, birds, mammals, forests, peatland and control of freshwater invasive species. More are in progress.

More about What Works in Conservation

Download free PDF or purchase
The Conservation Evidence Journal

The Conservation Evidence Journal

An online, free to publish in, open-access journal publishing results from research and projects that test the effectiveness of conservation actions.

Read the latest volume: Volume 21

Go to the CE Journal

Discover more on our blog

Our blog contains the latest news and updates from the Conservation Evidence team, the Conservation Evidence Journal, and our global partners in evidence-based conservation.


Who uses Conservation Evidence?

Meet some of the evidence champions

Endangered Landscape ProgrammeRed List Champion - Arc Kent Wildlife Trust The Rufford Foundation Save the Frogs - Ghana Mauritian Wildlife Supporting Conservation Leaders
Sustainability Dashboard National Biodiversity Network Frog Life The international journey of Conservation - Oryx Cool Farm Alliance UNEP AWFA Bat Conservation InternationalPeople trust for endangered species Vincet Wildlife Trust