Exclude grazers from semi-natural habitats

Supporting evidence from individual studies

1. A controlled study in 1981-1983 at a semi-desert grassland site in Arizona, USA (Bock et al. 1984), found that bird communities differed between an area from which cattle had been excluded since 1968 and one that had been continuously grazed. Total bird numbers were higher on grazed plots than ungrazed in summer, with no difference in winter (summer: 193 birds counted in ungrazed sites vs. 270 in grazed; winter: 242 birds in grazed vs. 247 in ungrazed). Open-ground foraging species were significantly more abundant in the grazed area, whilst species that prefer grass and shrub cover were the most abundant birds in protected sites, but absent on grazed pasture. The authors argue that the bird communities prevalent in grazed areas were more typical of lower elevations and dry habitats, and may be an indication of desertification of intensively grazed semi-desert and plains grasslands.

   Study and other actions tested

   Referenced paper

   Bock C.E., Bock J.H., Kenney W.R. & Hawthorne V.M. (1984) Responses of birds, rodents, and vegetation to livestock exclosure in a semidesert grassland site. Journal of Range Management, 37, 239-242.

2. In a 1994 site comparison study in Little Valley, Nevada, USA (Ammon & Stacey 1997), the nesting success rates of riparian bird species were found to be lower in an area grazed by cattle than an adjacent area rested from grazing for 30 years (grazed area: 83% of six above-ground nests successful and 67% of 12 ground nests predated; rested area 36% of 14 above-ground nests successful and 43% of seven ground nests predated). Experimental data from placing artificial nests baited with a Japanese quail Coturnix japonica egg and one painted plasticine egg in both areas showed a similar trend (daily survival rates of 55-95% of 120 eggs in grazed area vs. 77-98% of 120 in rested area). The authors suggest that grazing may facilitate nesting predation through changes in predator assemblage or increasing nest detectability.

   Study and other actions tested

   Referenced paper

   Ammon E.M. & Stacey P.B. (1997) Avian nest success in relation to past grazing regimes in a montane riparian system. The Condor, 99, 7-13.

3. A replicated, controlled study in 1991-1994 in semi-arid riparian habitats in Oregon, USA (Dobkin et al. 1998), found that bird species richness and relative abundance were significantly higher on three ungrazed 1.5 ha plots, compared to three grazed 1.5 ha plots (approximately 10-12 species/plot for ungrazed plots vs. 7-10 species/plot for grazed plots). In addition, ten species associated with riparian and wetlands habitats were found only on exclosure plots, and five species associated with uplands habitats only on open plots. Ungrazed plots had not been grazed for 30 years, whilst grazed plots were grazed until 1990. In the final year of study, four years after grazing had been stopped, key wet-meadow species (sora Prozana carolina, Wilson’s phalarope Phalaropus tricolor, green-winged teal Anas crecca, and gadwall A. strepera) were found on open plots. Throughout the study, sedge cover, forb cover and foliage height diversity of herbs were greater within the exclosure; bare ground, litter cover, shrub cover and shrub foliage height diversity were greater on open plots.

   Study and other actions tested

   Referenced paper

   Dobkin D.S., Rich A.C. & Pyle W.H. (1998) Habitat and avifaunal recovery from livestock grazing in a riparian meadow system of the northwestern Great Basin. Conservation Biology, 12, 209-221.

4. A study from 1992-5 in New Mexico, USA (Goguen & Mathews 1998), found no significant differences in songbird abundance or species richness between pinyon-juniper woodland sites that were actively grazed and sites from which livestock grazing had been excluded for 20 years (39 species on ungrazed sites, 36 on grazed).  However, the authors argue that the slow growing woodland may not have had time to recover over the study period. One species, the western scrub-jay Aphelocoma californica, was more common on ungrazed sites.  The authors note that over 75% of blue-gray gnatcatcher Polioptila caerulea, solitary vireo Vireo solitarius and western tanager  Piranger ludoviciana nests were parasitised by brown-headed cowbirds Molothrus ater, raising concern that pinyon-juniper woodland habitat close to grazed areas could act as a population sink for songbirds due to cowbird parasitism.

   Study and other actions tested

   Referenced paper

   Goguen C.B. & Mathews N.E. (1998) Songbird community composition and nesting success in grazed and ungrazed pinyon-juniper woodlands. Journal of Wildlife Management, 62, 474-484.

5. A site comparison study from December-March in 1996-8 in oak savanna in Arizona, USA (Bock & Bock 1999), found that 19 seed-eating birds were 270% more abundant in a livestock exclosure (former cattle ranch, ungrazed since 1968) than on a ‘holistically managed’ ranch, where 60 paddocks (covering 3,238 ha) were grazed intensively on a short rotation. Twenty-four other species (predators, fruit-eaters and insect-eaters) made up a smaller proportion of total bird abundance and did not differ in abundance between grazed and ungrazed sites. Grasses in the ungrazed area were significantly taller (4.4 times) and had higher basal-area ground cover (2.5 times) and higher overall canopy (2.2 times). The study sites were separated by a 7 km boundary fence, which was divided into 1 km sampling transects.

   Study and other actions tested

   Referenced paper

   Bock C.E. & Bock J.H. (1999) Response of winter birds to drought and short-duration grazing in southeastern Arizona. Conservation Biology, 13, 1117-1123.

6. A replicated study in 1994-1995 in the Mojave Desert, California, USA (Brooks 1999), found that bird abundance and species richness were higher inside two 2.25 ha sites protected from sheep grazing and off-highway vehicles (OHV) since 1978, compared to adjacent sites that were grazed and driven over by OHVs. Significant differences were observed for sage sparrow Amphispizia belli, Le Conte’s thrasher Toxostoma lecontei, loggerhead shrike Lanius ludovicianus, verdin Auriparus flaviceps and ash-throated flycatcher Myiarchus cinerascens.  The authors suggest the increased abundance of bird species within the protected area is linked to a greater food supply.

   Study and other actions tested

   Referenced paper

   Brooks M. (1999) Effects of protective fencing on birds, lizards, and black-tailed hares in the western Mojave Desert. Environmental Management, 23, 387-400.

7. A small controlled study from May-July in 1992-4 in river islands in Quebec, Canada (Lapointe et al. 2000), found that, in 1993, more duck nests than expected by an even distribution were found in idle fields, from which cattle were excluded, whilst fewer than expected were found on improved or unimproved pasture. In 1994, unimproved pasture held more than expected as well, but improved pasture held fewer. Nests on improved pasture had significantly lower success than those in other habitats (15% success of 39 nests vs. 47-82% elsewhere), with 33% being trampled. Nesting densities were no higher on idle areas than other habitat types.

   Study and other actions tested

   Referenced paper

   Lapointe S., Giroux J.F.O. & Belanger L. (2000) Benefits of rotational grazing and dense nesting cover for island-nesting waterfowl in southern Quebec. Agriculture, Ecosystems & Environment, 78, 261-272.

8. A before-and-after study from 1986-1990 (Krueper et al. 2003) found that more birds were detected in an area of riparian, mesquite and Chihuahuan desert-scrub in Arizona, USA, after the removal of cattle and the onset of a grazing moratorium in 1988 (average of 221 birds detected/km of transect in 1990 vs. 103 birds/km for 1986). Detections increased for 42 species, 26 significantly, and decreased for 19 species, eight significantly.  Only four species in the study showed similar trends in regional Breeding Bird Surveys.  Insectivores, granivores, midstory species, upperstory species and riparian species were most likely to increase, and migrants tended to show greater increases than residents. Chihuahuan desert-scrub species showed the smallest increases and were most likely to decline, possibly because the Chihuahuan scrub changed the least with the grazing moratorium. Surveys were conducted three times a month, every month over the study period.

   Study and other actions tested

   Referenced paper

   Krueper D., Bart J. & Rich T.D. (2003) Response of vegetation and breeding birds to the removal of cattle on the San Pedro River, Arizona, USA. Conservation Biology, 17, 607-615.

9. A study in May-July of 2000 and 2001 in Kaibab National Forest, Arizona, USA (Walsberg 2005), observed significantly higher fledging success rates of ground-nesting dark-eyed Juncos Junco hyemalis breeding in areas not grazed by cattle (48% of 21 nests) than in immediately adjacent, grazed areas (12% of 17 nests).  The authors suggest that reduced nest cover may expose nests to more extreme climatic conditions and make them more conspicuous to predators.

   Study and other actions tested

   Referenced paper

   Walsberg G.E. (2005) Cattle grazing in a national forest greatly reduces nesting success in a ground-nesting sparrow. The Condor, 107, 714-716.

10. A replicated, controlled study from December 2002 to March 2003 in 46 sampling transects (300 m long, 60 m wide, 1.8 ha, 2-40 km apart) across eight vegetation units and two grazing regimes (6 transects/vegetation unit; 3/grazing regime) in woodland, grassland and rocky habitats in the Córdoba Mountains, Argentina (Garcia et al. 2008) found that bird species richness and abundance was significantly lower in livestock-excluded areas. Livestock exclusion reduced bird density and species richness across all vegetation units for all species and for endemic subspecies alone. Similarly, species richness was higher in grazed sites than in livestock-excluded areas for both insect-eating birds (5.0 compared to 3.8) and seed-eating birds (1.8 compared to 1.6 species / 1.8 ha). Community composition was different between vegetation units, but not between grazing regimes. Traditional livestock management stocking rates ranged from 0.4 – 1.5 cattle equivalents / ha. Livestock exclusion areas were without cattle since 1998.

    Study and other actions tested

    Referenced paper

    Garcia C., Renison D., Cingolani A.M. & Fernandez-Juricic E. (2008) Avifaunal changes as a consequence of large-scale livestock exclusion in the mountains of Central Argentina. Journal of Applied Ecology, 45, 351-360.

11. A study in northern Hawaii, USA (Camp et al. 2010), found that seven species in an open koa Acacia koa forest from which feral grazers were excluded showed long-term population stability or growth, but only two were increasing in a closed forest with grazers excluded. This study is discussed in ‘Threat: Invasive and other problematic species - Reduce adverse habitat alterations by excluding problematic species’ and ‘Habitat restoration and creation – Forest restoration’.

    Study and other actions tested

    Referenced paper

    Camp R.J., Pratt T.K., Gorresen P.M., Jeffrey J.J. & Woodworth B.L. (2010) Population trends of forest birds at Hakalau Forest National Wildlife Refuge, Hawai'i. The Condor, 112, 196-212.