Appendix A. Summary of Included Articles
ArticleCitation / Study Location / Pullin &
Knight
Classification / Primary Outcome(s) / Restoration Technique(s) / Brief Summary of Research/Monitoring Methods / Brief Summary of Results
1. Allen-Diaz 1991 [21] / California; Sierra Nevada; Sagehen Creek Field Station / II-1 / Hydrology
Vegetation / Observational study of water table-plant species relationship in two meadows, one that has been ungrazed for 35 years and one that was subject to about two weeks of grazing by sheep in late August. / Hydrology: 60 piezometers were placed along 5 randomly located transects perpendicular to the creek. Measurements were taken twice per month during the growing season and once per month during the winter from June 1987 to December 1989.
Vegetation: 50 cm x 50 cm plots were located adjacent to each piezometer. Species composition was determined using a 10pt frame placed within each. Aboveground production (at peak standing crop in July 1988) was clipped in 25 cm x 25 cm square quadrats at random locations within 1 m of each piezometer. / Hydrology: Four distinctive water table patterns were identified. The wettest sites exhibited relatively little fluctuation and generally maintained water table levels within 20 cm of the surface. Much greater seasonal variation was observed for the other three patterns/sites.
Vegetation: Five plant community types were identified, each of which was related to specific water table patterns.
Results suggest that changes in meadow species may be predicted by monitoring water table patterns (and that revegetation practices should carefully match species to the expected water table pattern).
2. Ammon and Stacey 1997 [6]
(Ammon and Stacey 1997 cont…) / Western Nevada; Carson Range; Franktown Creek/Little Valley / II-2 / Vegetation
Biodiversity
(nesting/nest success) / Study was conducted on a 70-80 ha meadow; about half was subject to long-term summer grazing and the other half had been protected from grazing for 30 years. / Vegetation: Estimated cover of vegetation and bare ground within 0-50m of the stream that passed through the meadow. Also categorized height of vegetation. A total of 95
1 m2 quadrats were measured on both the grazed and rested side of the meadow, with samples at 0, 10, 20 and 50m from the stream.
Biodiversity: Nest searches were conducted throughout the summer of 1994, starting in early May. Nests were monitored during visits every 3-5 days until they either failed or the young fledged. 240 artificial nests, including ground and above-ground nests, were placed in various locations in relation to the stream and fence. Artificial and real nests in which eggs or young were damaged or disappeared were considered “preyed upon”. / Vegetation: Ground vegetation was more abundant, willows (Salix spp.) were less abundant, and vertical vegetational diversity was lower on the grazed side of the meadow.
Biodiversity: Success of both ground and above-ground real nests was lower on the grazed side. Of the above ground nests, 83% were preyed upon compared to 36% on the rested site. 67% of the ground nests were lost on grazed site compared to 43% on the ungrazed site. Riparian-dependent bird species that nest in willows might be the most affected by grazing.
3. Anderson et al. 2003 [28]
(Anderson et al. 2003 cont…) / Northern Arizona; Coconino National Forest; Clover Spring / II-3 / Geomorphology
Hydrology
Vegetation / Exclosure (livestock and wild ungulate)
Re-channelization Seeding (hydromulching with four grass species)
Transplanting of three wet meadow species / Geomorphology: pre- and post-project longitudinal profiles and channel cross-sections were used to document the status of the channel.
Hydrology: Periodic monitoring of the discharge from Spring Creek was conducted and temperature data loggers were used to determine the presence or absence of flow in the stream. A temporarily placed piezometer was used to assess groundwater.
Vegetation: A total of 18 1 m2 were used to assess species composition and percent cover of vegetation, litter, rocks and bare soil. / Geomorphology: Project met goals of change in existing channel morphology resulting in a sinuosity ratio of 1.22. Stream gradient goals were also met; the pretreatment gradient was 1.52% while the post treatment stream gradient was 1.13%. Average entrenchment ratio before treatment was 1.1 and average entrenchment ratio post treatment is 2.3.
Hydrology: Water discharging from the spring appears to perch on shallow alluvium, with no downward movement into an aquifer. Streamflow in the portion of the meadow upstream from the spring was ephemeral and flows were not recorded beyond early June.
Vegetation: Plant species composition after final monitoring consisted of 15 species. Carex nebrascensis and Juncus balticus, both of which were planted as plugs, established in the lower 1/3 of the project area and were successfully regenerating. Cheat grass (Bromus tectorum) still occured on approximately 25% of plots. Average percent basal vegetation was at 1%, and average percent canopy cover at 10%. Average bare ground was at 64% and average percent litter at 22%
4. Beck and Peek 2004 [60] / Northeastern Nevada; Jarbidge Mountains;
Black Spring, Monument Spring, and Sagehen Springs / I / Vegetation (effects of grazing) / Exclosures, including cattle only and both cattle and wild ungulate (“wildlife”) exclosures / Randomized complete block design with repeated measures for 2 years over 3 seasons. 3 blocks and 3 treatments – control (both cattle and wildlife grazing excluded); wildlife only grazing (wildlife treatment); and wildlife and cattle grazing (cattle treatment).
Ocular estimate of standing crop for each species in 15 randomly selected 20 x 50 cm quadrats at each 225 m2 treatment plot and weighing by species of clipped material in 3-5 quadrats per treatment.
Clipping treatments to simulate effects of various levels of grazing. / No differences in graminoid standing crop were found among treatments in 1999, but in 2000 there were significantly less graminoids in the cattle treatment. Forb availability declined throughout the summer, especially in the cattle treatment.
Species richness was slightly greater in the wildlife treatment than in the cattle treatment in two of three meadows.
Implications for grazing management in mountain meadows were discussed.
5. Blank et al. 2006 [44] / Northern California / II-2 / Soils (effects of grazing) / Pasture fencing and controlled levels of grazing / Using 9 0.5 ha fenced pastures, 3 grazing treatments (no cattle grazing, low grazing and moderate grazing) and 3 landscape positions (stream edge, mid-floodplain and forest edge) were evaluated.
Grazing treatments were short-term (1-14 days) at a high stocking rate. Treatments were initiated in 1990 and continued through 1993.
Note: The study only reported on the no grazing and moderate grazing treatments. / The moderate grazing treatment had greater lysimeter-extractable Na+, Ca+2, Mg+2 and NO-3, higher pH, and less K+ and NH4+ than the exclosure treatment, although effects were mainly significant at the forest edge. Moderate grazing did not result in more anoxic soil conditions and did not have a significant effect on root-length density.
6. Borgmann et al. 2007 [56]
(Borgmann et al. 2007 cont…) / California; Sierra Nevada/Lake Tahoe Region; Cookhouse Meadow / II-2 / Hydrology
Vegetation
Biodiversity / Pond-and-plug
Rechannelization
Willow stakes and mattresses
Sod transplants to the new channel banks / A Before-After-Control-Impact design was employed, with three years of pre-project data and one year of post-project data collected.
Hydrology: Observations were made of standing water and whether soils were wet or dry along transects through June and July.
Vegetation: Willow cover, height and clumping were assessed by a combination of visual estimates and GPS.
Biodiversity: A variety of techniques were employed to assess the species richness and abundance of butterflies, songbirds, owls, small mammals, and bats. / Hydrology: By late June of the post-project year, the proportion of the meadow with standing water or moist soil had fallen below 15%, which was below project goals. However, the year (2007) was drier than average.
Vegetation: Willow cover and clumping data was collected but the time since the project was implemented was not sufficient to draw conclusions about project effects.
Biodiversity: Butterfly richness and abundance increased one year post-project, but similar increases were noted at the two control sites. Songbird species richness increased by 38% but songbird abundance did not increase. Owl species richness was slightly higher post-project, but sampling methods make this difference difficult to fully substantiate. Bat species richness was not different post-project and frequency was slightly lower. In general, richness and abundance of desired condition small mammal species did not increase one
year post-restoration.
Overall, it was recommended that post-project monitoring be continued to allow for more definitive conclusions to be drawn. Both the short duration of post-project monitoring and the unusually dry post-project year affected the results obtained to date.
7. Castelli et al. 2000 [9]
(Castelli et al. 2000 cont…) / Central Nevada; Humboldt-Toiyabe National Forest; Big Creek and Corral Canyon Drainages / II-1 / Hydrology
Soils
Vegetation / Observational study of the relationship between hydrologic regime, soil characteristics and plant species across a topographic gradient within two relatively intact meadows. No restoration treatment was applied. / Hydrology: Piezometer nests were used to measure the depth to groundwater along each transect beginning in the fall of 1997 and throughout most of 1998.
Soils: Soil pits and cores were used to characterize soil profiles and soil texture. Redox potential, soil moisture, depth to saturation, and soil temperature were sampled every 3 weeks from June-October 1998.
Vegetation: Species frequency and cover (of vegetation, litter, bare ground rock and gravel) were sampled along a total of 20 transects located perpendicular to the stream at each site. / Hydrology: Water table depth differed significantly across the 4 vegetation zones (see vegetation below) as a whole. In one meadow, however, there was relatively little difference between the water table of the wet and mesic meadow zones. Wet meadow zones at both sites had the shallowest water tables and the least variation in depth.
Soils: All zones were characterized as Mollisols, with characteristics of productive fluvial ecosystems. The wet meadow zones had the most anaerobic conditions, but both mesic and dry meadows had occasional periods of anaerobic conditions early in the growing season.
Vegetation: Variables most strongly related to plant species occurrence were those associated with depth to water table. Multivariate approaches clearly classified the vegetation into four zones and indicator species were identified for each.
8. Chambers and Lesh 2007
[unpublished manuscript, Northern Arizona University School of Forestry] / Northern Arizona, Coconino National Forest / II-1 / Biodiversity
Vegetation / Grazing exclusion / Three pairs of wet meadows with similar livestock grazing histories were compared; each pair included 1 meadow with and one meadow without ungulate exclosures. Meadows with exclosures included exclosures that excluded all ungulates and exclosures that excluded only cattle.
Biodiversity: Small mammal traps were set out in each meadow during February 2000 and in early-mid June in 2001, 2002 and 2003.
Vegetation: Downed wood was measured in 0.015-ha plots near each trapping station. Vegetation in two 10-cm2 plots near each trapping station was measured for height and then clipped and sorted into 3 categories for biomass estimation: grass and sedge, forb, and other (mostly pine needles and other litter). Vegetation was clipped in February 2000 and June 2003. / Biodiversity: A total of 244 small mammals were trapped, representing 7 species. Mogollon voles (Microtus mogollonensis) and deer mice (Peromyscus maniculatus) were the most common species, with 58% and 37% of all captures, respectively. Capture rates for Mogollon voles were highest in areas with the highest amount of vegetation biomass, which generally were sites within total ungulate exclosures. While Mogollon voles appeared to be highly sensitive to grazing, deer mice did not.
Vegetation: Mean biomass varied from a low of 300 kg ha-1 at one partial grazing site (excluded cattle but not native ungulates) to 4,200 kg ha-1 at one site with total ungulate exclusion. Ungrazed sites generally had higher biomass and greater average vegetation height, although one meadow open to grazing also had high vegetation biomass.
9. Flenniken et al. 2001 [43] / Northcentral Colorado; Roosevelt National Forest, Sheep Creek / II / Geomorphology (on micro scale)
Hydrology
Soils
Vegetation / Grazing management / Applied 4 treatments to a total of 16 3 X 10m plots: (1) grazing plus trampling, (2) trampling only, (3) mowing only, and (4) control. Also conducted rainfall simulations prior to and shortly after the other treatments were applied.
Measured microtopography, soil (bulk density, organic matter and soil moisture), vegetation (stem density and aboverground biomass), and microchannel (sinuosity, drainage density and flow depth) responses. / Grazing (the most relevant treatment in this context) significantly reduced stem density and aboveground biomass, decreased microchannel sinuosity and drainage density, and increased flow depths and total runoff. The grazing treatment did not significantly affect soil organic matter or bulk density, which may be due the short-term nature of the treatment.
10. Godwin 2004 [36]
Note: This was primarily a hydro- geology modeling study, but only the vegetation component is reported here. / Northern Arizona; Coconino National Forest; Hoxworth Springs / II-2 / Vegetation / Rechannelization
Rock walls
Cabled log wall
Cattle and ungulate exclosures / A total of 27 transects were established, with 9 each in the total grazing, wild ungulate-only grazing, and total exclosure treatments. Vegetation and other (e.g., rock, soil) cover was noted at points along each transect towards the end of the 1997, 1998 and 1999 growing season. / No significant differences were found for mean percent plant cover among the grazing treatments. A visual difference in the amount of biomass (“plant height and fullness”) was noted, with more in the total exclosure, but this was strictly a qualitative observation.
11. Hammersmark et al. 2008 [39]
Note: Although not included in this appendix, the report by Mount and Hammersmark [60] contains additional information on this study. / Northern California; Fall River Valley; Bear Creek Meadow
Note: At 1010 m, this is a lower-elevation meadow than most being considered in this review. / II-2 / Hydrology / Pond-and-plug, with streamflow being routed into a combination remnant/created channel. / Two hydrologic models (pre- and post-project) were developed using the MIKE-SHE modeling system. The post-project model was calibrated using 2005 water year data and validated using 2006 water year data, including data from 28 shallow piezometers. There was insufficient data on water table and streamflow prior to project construction to calibrate and validate the pre-project model. / The post-project model successfully simulated observed conditions. Groundwater levels and groundwater storage were higher in the restored scenario throughout much of the meadow system. Overbank flow was more frequent in the restored scenario. Baseflow duration was about two weeks less and total annual runoff was 1-2% lower in the restored scenario. Evapotranspiration was higher in the restored scenario.
12. Hammersmark et al. 2009
[40]
Note: Although not included in this appendix, the report by Mount and Hammersmark (2007) contains additional information on this study. / Northern California; Fall River Valley; Bear Creek Meadow
Note: At 1010 m, this is a lower-elevation meadow than most being considered in this review. / II-1 / Vegetation / Pond-and-plug, with streamflow being routed into a combination remnant/created channel. / Plant species composition and cover (of plants, litter, dung and wood) were estimated using 128 4 m2 plots along 15 transects.
Plant data was collected approximately six years after restoration of the stream channel.
Plant data was analyzed in relation to simulated water table patterns generated using a hydrologic model (see Hammershank et al. 2008). / Four distinct plant communities were identified. In general the community types were distributed along the hydrologic gradient in positions similar to those reported in other articles. The primary exception was with a single species, Carex nebrascensis, which occurred at deeper water table levels than previously reported. Although not quantified in this paper (but see Hammersmark et al. 2010), a greater proportion of the meadow was reportedly dominated by species adapted to wetter sites.
13. Hammersmark et al. 2010
[41]
Note: Although not included in this appendix, the report by Mount and Hammersmark (2007) contains additional information on this study. / Northern California; Fall River Valley; Bear Creek Meadow
Note: At 1010 m, this is a lower-elevation meadow than most being considered in this review. / II-2 / Vegetation / Pond-and-plug, with streamflow being routed into a combination remnant/created channel. / This study built on the previous two (Hammersmark et al. 2008, 2009) by developing habitat-suitability models for 11 herbaceous species. These models were then coupled with the pre- and post-project hydrologic models to simulate the change in distribution of the 11 species as a result of the restoration and related changes in water table depth. / The results indicated an increase in the area of the meadow that is suitable habitat for mesic/wet site plant species and a concomitant decrease in the area of suitable habitat for xeric species as a result of the channel restoration.