Grazing regime is the most influential grazing-related factor that influences the outcome of prescribed grazing in sandplain grasslands and calibration to current and antecedent site-specific conditions is essential to reach desired management outcomes. Targeted grazing is defined as “the application of a specific kind of livestock at a determined season, duration, and intensity to accomplish defined vegetation or landscape goals” (Launchbaugh and Walker 2006).    In sandplain grasslands, type of animal (goat, sheep, or cattle), season, and grazing intensity are important variables that influence the outcome of grazing in sandplain grasslands, and are largely depend on pre-treatment vegetation.

Livestock are capable of modifying plant biomass, structure, and floral composition by removing vegetation through consumption and disturbing soil and ground cover with their movement patterns. These actions can create bare ground that facilitates colonization of grass and forb vegetation through seed germination (Bullock et al. 1994, 1995; Silvertown et al. 1988). Therefore, while grazers may consume grass biomass in the short term (Fig. 2), they may create soil conditions that promote grass survival and reproduction over the long term.

Animal type largely determines the effects to sandplain grassland ecology. Large-scale historic livestock grazing in New England was primarily done by sheep. Sheep are intermediate feeders capable of both grazing on herbaceous vegetation and browsing on woody vegetation. In contrast, cattle are grazers that predominantly consume grass, while goats are browsers that feed predominantly on the branches and leaves of woody plants (Burritt and Frost 2006). C. Neill (Interview) studied the effects of cattle grazing on grassland, edge, and shrub habitats and found that cattle did not decrease shrub and tree cover. Therefore, the choice of livestock suitable for ecological management should be based on pre-treatment vegetation at the site (Table 1).

Table 1. Choice of livestock based on pre-treatment vegetation.

Pre-Existing Vegetation	Type of Animal
Woody vegetation (shrubs and vines)	Goat
Mix of woody and herbaceous species	Sheep
High proportion of graminoids	Cattle
Mix of shrubs, graminoids and forbs	Combination treatments

 

Sheep and cows can be effectively contained using a solar-powered fence system with fiberglass posts supporting several strands of electrified wire, which is cost effective and easily deployed. In contrast, goats are jumpers and can be aggressive and territorial, making them difficult to contain. Goats typically require tightly woven electric net fencing, which is significantly more expensive than electric wire fencing systems (Kott et al. 2006). Pre-treatment vegetation, overall goals, and cost are important factors for deciding which animal to use in any potential management.

Seasonality also impacts the effects grazing could have on sandplain grasslands. Seasonality influences the amount of time, care and cost needed to maintain livestock. Because of relatively short growing seasons in the northeast U.S., grazing cannot typically be sustained without supplemental feed. Therefore, animals must be overwintered and fed grain and hay, sold at the end of each season, or transported for winter. Hendrickson and Olson (2006) suggest that grazing effectiveness relies on target plant species phenology. To reach maximum success in reducing woody growth, grazing should coincide with the period between bud and flower to increase stress on target plant species. Because most plants (including woody plants) are typically most palatable during leaf-out stage, animals that might otherwise avoid these species can be encouraged to graze on them during spring (Kott et al. 2006). Increasing concentrations of toxic secondary compounds, such as tannins in oaks (Quercus spp.) that increase with age, contribute to a decline in palatability (Makkar et al. 1991, Makkar 2003). Timing grazing to the time of maximum palatability when grazing can have the greatest grassland-promoting benefits is a management challenge.

Winter season grazing is a potential option at grassland management sites where multiple target species could be negatively affected by growing season treatments. Experimentation with this type of management is currently under consideration for property owned by The Trustees of Reservations on Martha’s Vineyard, MA (R. Hopping, Interview). During dormant seasons, forb species are dormant while apical meristems of graminoid species are at or below the soil surface and less likely to be grazed. In contrast, the axillary and apical buds and stems of shrubs are exposed and vulnerable to grazing (Hendrickson and Olson 2006). However, one issue that needs to be carefully considered for winter grazing is the need for supplemental feed and the origin of any supplemental feed. Grain and hay can be sources of seeds of non-native invasive pasture weeds that could potentially do more harm than good. Therefore, to reach optimal results, it is recommended to introduce livestock immediately after spring leaf-out or soon after re-sprouting occurs in response to grazing or other treatment (R. Hopping, Interview).

The intensity of grazing is also important to consider, and depends on the size of treatment area, the number of animals per unit area (stocking rate), daily and seasonal weather patterns, and pre-treatment vegetation composition. Historic grazing patterns were likely very different than what can be replicated today in a managed setting because of the difficulty in providing large, continuous free-range areas without introduced predators like coyotes and dogs. Today, fencing is required to keep predators out, and livestock are confined to treatment areas (Fig. 3). Stocking rates, livestock residency period (rotation), and treatment area should be calibrated to balance degradation of grasslands by overgrazing with aims to reduce the dominance of a small number of species. For example, increased grazing intensity due to trampling and consumption could result in bare ground conditions that open niches for native forbs or graminoids. However, there is a threshold at which intensity reaches a point when the area and condition of bare ground stops being beneficial to disturbance-dependent target species, and even might favor non-native invasive species or cause erosion.

Today, sandplain grasslands are the result of many decades of intense grazing and other disturbances, followed by a long release period. To mimic a historic grazing regime, one option could be to graze a small flock long-term within a large area, resulting in less intense grazing pressure as animals can graze across a large area and select preferred forage. Therefore, they are less likely to deplete most of the available vegetation. Though, at the same time, grazing in this manner may not be effective at removing less palatable species such as woody shrubs. As a result, this type of grazing may be more suitable for grassland maintenance rather than restoration. Low intensity grazing can result in a mosaic of grazed and un-grazed vegetation that is beneficial to the persistence of rare plants and insects. It also requires less set-up, management and staff time.

Little information on key aspects of life history exists for many of the infrequent species target for sandplain grassland conservation. There is almost no information on how many of these species respond to grazing. Some evidence suggests that a legacy of historic grazing could promote long-term grass and forb target species following a release from grazing. Dunwiddie (1997) studied patterns in land-use history of historically grazed pasture on Nantucket and found a high abundance of graminoid and target forb cover compared to adjacent ungrazed land. Ungrazed plots contained a high cover or frequency of non-target species such as small bayberry (Morella caroliniensis), while grazed plots included little bluestem (Schizachyrium scoparium), sweet-fern (Comptonia peregrina), and the rare forb, bushy frostweed (Crocanthemum dumosum). These findings indicated that historic grazing benefited some infrequent sandplain grassland species.

Further, evidence suggests grazing could negatively affect sandplain grassland vegetation in the short-term (Fig. 4) and that such a release from grazing over some period of time can be favorable. On Naushon Island, C. Neill (Interview), studied the response of vegetation to cattle grazing and found that native graminoid and forb cover decreased in the short term. Forbes (2011) also found that graminoid and forb cover decreased following grazing, but no long-term monitoring data exists on trends during subsequent post-treatment recovery growing seasons.

When using sheep or cattle, the timing of grazing treatments in an existing grassland is important because both desirable and undesirable plant species can be equally targeted by grazers. Early season grazing would have minimal effects on late-season maturing target species such as northern blazing-star (Liatris novae-angliae) and little bluestem grass. The reverse would be true for desirable early season bloomers such as sandplain blue-eyed grass (Sisyrinchium fuscatum) and bushy frostweed. Similarly, management can be timed to maximize negative impacts to undesirable non-native invasive species such as velvet-grass (Holcus lanatus) and cypress-spurge (Euphorbia cyparissias) by grazing just prior to flowering and seed set. Therefore, it is important to thoroughly inventory the management area prior to treatment to identify potential forage and make decisions about the timing and effect of graze treatments on both desirable and undesirable target species.

A concern about grazing management is that intensive grazing may eradicate existing rare native plants. For example, lion’s-foot rattlesnake-root (Nabalus serpentarius) is highly selected by deer (Patterson III et al. 2005) and is therefore likely to be selectively grazed by livestock. But studies have shown that other rare plant species persist despite grazing, and that undesirable consequences could be mitigated by understanding and applying grazing variables correctly. Dunwiddie et al. (1986) found that grazing had no negative effects on desirable plants and no introduction of non-native species. Karberg and Beattie (2009) found healthy populations of rare plant species after years of grazing, including St. Andrew’s cross (Hypericum stragulum), bushy frostweed, and purple needle-grass (Aristida purpurescens). Further, several species of rare plants now absent from Nantucket were common on the island in the early 1900s, including purple cudweed (Gamochaeta purpurea) and sandplain gerardia (Agalinis acuta). This timeframe is concurrent with the cessation of historic sheep grazing, indicating that some rare species thrived in release periods after large-scale grazing occurred (P. Dunwiddie, Interview). Modern vegetation studies show that rare plants that are not preferentially grazed thrive under grazing management, and non-native invasive plants can be targeted (Fig. 5). Needle-grass indirectly benefitted from grazing because it is likely unpalatable to sheep on Martha’s Vineyard, MA (Patterson III et al. 2005). Sandplain flax (Linum intercursum) may also be unpalatable to grazing animals, as all members of the genus are known to produce toxic or acidic compounds (Zaremba 2003; Patterson et al. 2005). Therefore, grazing impacts each rare plant species differently but many of the rare sandplain plants likely survived and thrived after high intensity historic livestock grazing.

There is some evidence that combinations of grazing and other management practices could reach desired goals. The Nantucket Conservation Foundation conducted four seasons of sheep grazing and mowing management research at Squam Farm on Nantucket starting in 2005 and found that grazing combined with mowing significantly reduced the growth of woody species. Forbes (2011) compared mowing with grazing on Naushon Island, MA and found that grazing had a significantly greater impact than mowing on reducing woody growth and varied cover with gaps and increased areas of bare soil. Dunwiddie et al. (1986) found that in heathland habitat overgrown by scrub oak (Quercus ilicifolia) and dwarf chinkapin oak (Q. prinoides) treated with brush-cutting prior to grazing, all species (both desirable and undesirable) showed a four- to six-fold decrease in cover, indicating possible over-stocking of sheep. Brush cutting alone appeared to be more effective at increasing desirable heath species and was almost as effective at reducing oak species as brush-cutting combined with grazing. These results were based on only one growing season.

Applying grazing during particular times of stress (e.g., droughts or outbreaks of herbivorous insects) could increase the effects of grazing and reduce the frequency of grazing required to obtain similar vegetation responses. However, being able to take advantage of these events by having animals available in particular places will likely be challenging.

The effects of grazing on the response of fauna are understudied and the short- and long-term impacts to target species in sandplain grasslands need further research. Most of the information available on how grazing affects rare sandplain grassland associated fauna is anecdotal and theoretical, as there have been no detailed studies focusing on fauna in this region. Concerns for wildlife include physical crushing or exclusion by the presence of large grazing livestock. This type of disturbance is likely less of a concern in grazing management than disruptive burning and mowing treatments. For example, nesting birds may not be negatively impacted by the presence of grazing livestock during the growing season, and might actually benefit in some ways (Fig. 6). At Squam Farm on Nantucket, the Nantucket Conservation Foundation observed several spotted turtles (Clemmys guttata)—including one that was radio-tracked—traversing areas actively grazed by sheep with no observable negative impacts. Another concern about introducing livestock to grassland habitats is the potential impacts on animal communities throughout the food web and their interactions. Mesopredators might increase in the presence of livestock, which would detrimentally impact small mammals and birds, although evidence for this comes from outside the northeast U.S. (Coates et al. 2016). Effects of grazing on fauna are understudied, and should be the focus of future research.