Reduce intensity of livestock grazing: freshwater marshes
Overall effectiveness category Likely to be beneficial
Number of studies: 3
Background information and definitions
Domestic livestock can directly consume vegetation, destroy vegetation by trampling, create bare patches of ground (e.g. repeatedly used tracks), affect water infiltration and flows by compacting soils, affect nutrient balance through excretion of waste products, and import seeds of undesirable plants (Morris & Reich 2013). Reducing grazing intensity might allow grazing-sensitive species to recover. However, maintaining some grazing may sustain a mosaic of short and tall vegetation patches, each favouring different plant species (Nolte et al. 2014). The effects of this action might depend on site conditions such as productivity (determined by soil moisture and nutrient levels; Berney et al. 2014).
To be summarized as evidence for this action, studies must have compared different grazing intensities, without completely removing livestock. Grazing intensity could be reduced by altering grazing duration (e.g. allowing livestock to graze for fewer days) or pressure (e.g. letting fewer animals graze, providing supplementary food as an alternative to living plants, encouraging grazing away from focal areas by feeding stations or shelter elsewhere). Comparisons must involve the same type of livestock and at least some overlap in the timing of grazing.
When interpreting the evidence, remember that the overall grazing intensity for a site does not necessarily reflect the local grazing intensity in wetland patches or in different vegetation types. Also note that “low”, “moderate” and “high” are relative terms within each study: they do not always refer to the same absolute intensity across studies.
Related actions: Use barriers to keep livestock off ungrazed freshwater marshes; Exclude or remove livestock from historically grazed freshwater marshes; Use grazing to maintain or restore disturbance; Use grazing to control problematic plants; Modify livestock farming practices in watershed.
Berney P.J., Wilson G.G., Ryder D.S., Whalley R.D.B., Duggin J. & McCosker R. (2014) Divergent responses to long-term grazing exclusion among three plant communities in a flood pulsing wetland in eastern Australia. Pacific Conservation Biology, 20, 237–251.
Morris K. & Reich P. (2013) Understanding the Relationship Between Livestock Grazing and Wetland Condition. Arthur Rylah Institute for Environmental Research, Technical Report Series No. 252.
Nolte S., Esselink P., Smit C. & Bakker J.P. (2014) Herbivore species and density affect vegetation-structure patchiness in salt marshes. Agriculture, Ecosystems & Environment, 185, 41–47.
Supporting evidence from individual studies
A replicated, randomized, paired, controlled, before-and-after study in 1992–1997 of springs and creeks in California, USA (Allen-Diaz & Jackson 2000) found that lightly and moderately grazed areas had statistically similar herbaceous vegetation cover. This was true in five of five years, both in spring wetlands (data not reported) and in downstream wetlands alongside creeks (lightly grazed: 46–47%; heavily grazed: 35–80%). Before intervention, the creek plots had 58–80% herb cover. Methods: Three pairs of pastures were selected for the study. All contained springs and had been moderately grazed by cattle since 1960 (800–1,000 kg/ha Residual Dry Matter: the amount of herbaceous material present left after grazing). From 1992/1993, one random pasture/pair remained moderately grazed (1,100–1,800 kg/ha RDM) and one was lightly grazed (1,200–3,800 kg/ha RDM). Grazing occurred in November and February–May. Vegetation cover was monitored in late May 1992–1997, along four 5–10 m transects/pasture: two in wetlands near the spring source, and two in wetlands along the resulting creek.Study and other actions tested
A replicated, randomized, paired, controlled study in 2000–2003 of ephemeral pools in a grassland in California, USA (Marty 2005) found that seasonally grazed pools typically had similar relative cover of grasses and native plants to continuously grazed pools, and experienced similar changes in native plant richness. In six of six comparisons over three years, seasonally and continuously grazed pools had similar cover of grasses relative to forbs (seasonal: grass cover 46–55% of forb cover; continuous: 34–48%). In three of six comparisons, seasonally and continuously grazed pools had similar relative cover of native plants relative to non-natives. In the other three comparisons, seasonally grazed pools had lower relative cover of native plants than continuously grazed pools (data not reported). Finally, over the three years, seasonally and continuously grazed pools experienced statistically similar changes in native plant species richness (seasonal: 0.6 fewer to 1.2 more species/0.25 m2; continuous: 0.7–1.8 more species/0.25 m2). Methods: In 2000, eighteen plots were established (in six sets of three) on a ranch grazed for >100 years. In each set, one plot was grazed in the dry season, one was grazed in the wet season, and one was grazed throughout both seasons. Plots were grazed by 1 cow-calf pair/2.4 ha. Access to the seasonally grazed plots was controlled by electric fences. Each spring between 2001 and 2003, vegetation was surveyed in three pools/plot and in adjacent upland. Pools were 70–1,130 m2 and dry when surveyed.Study and other actions tested
A site comparison study in 2001 of three wet meadows around an ephemeral lake in Ireland (Ryder et al. 2005) found that lightly and heavily grazed meadows had a similar plant community composition, species richness and overall cover, but that lightly grazed meadows contained taller vegetation. Both lightly and heavily grazed wet meadows had a statistically similar mix of plant species (data reported as a similarity index) and statistically similar plant species richness (lightly grazed: 18 species/150m2 and 15 species/m2; heavily grazed: 17–21 species/150m2 and 15–16 species/m2). Overall vegetation cover was 99% in both lightly grazed and heavily grazed meadows. However, the lightly grazed meadow had greater cover of black sedge Carex nigra cover and lower cover of creeping bent Agrostis stolonifera (data reported as abundance classes; see original paper for data on other species). Statistical significance of cover results was not assessed. The lightly grazed meadow had significantly taller vegetation on average (35 cm) than the heavily-grazed meadows (17 cm). Methods: In 2001, wet meadow vegetation was surveyed in three fields with different cattle grazing intensities. One field was lightly grazed (0.01 cows/ha/day, averaged across the summer) and two were heavily-grazed (0.67–0.76 cows/ha/day, averaged across the summer). In July, vegetation height was recorded at 72 points/field. In September, plant species and the area of bare ground/rock were recorded in six 1m2 quadrats/field.Study and other actions tested