Recovery Strategy for the Hill's Thistle in Canada [Final Version] 2011: Background

Date of Assessment: November, 2004

Common Name: Hill's Thistle

Scientific Name: Cirsium hillii (Canby) Fern.

COSEWIC Status: Threatened

Reason for Designation: This is a perennial herb restricted to the northern midwestern states and adjacent Great Lakes that is found in open habitats on shallow soils over limestone bedrock. In Ontario, it is found at 64 extant sites but in relatively low numbers of mature flowering plants that are estimated to consist of fewer than 500 individuals. Some populations are protected in national and provincial parks, however, the largest population is at risk from aggregate extraction. On-going risks are present from shoreline development, ATV use, and successional processes resulting from fire suppression within its habitat.

Canadian Occurrence: Ontario

COSEWIC Status History: Designated Threatened in November 2004. Assessment based on a new status report.

Hill's Thistle is listed as Threatened and is on Schedule 1 of the federal Species at Risk Act (SARA). In Ontario it is listed as Threatened on the Species at Risk in Ontario (SARO) List under the Endangered Species Act, 2007 (ESA). The global rank of Hill's Thistle is G3 or Vulnerable (NatureServe 2009). The species is federally listed as a Species of Concern in the United States. It is currently listed as S1 or Critically Imperiled in Illinois, Indiana, and Iowa, and S3 or Vulnerable in Ontario, Michigan, Wisconsin, and Minnesota (NatureServe 2009). The range of Hill's Thistle is completely restricted to the Great Lakes Region, and the Canadian range of Hill's Thistle may account for 50% or more of the global population. See Section 2.1 Population and Distribution Context.

Hill's Thistle (Cirsium hillii (Canby) Fern.) is a perennial thistle with a single deep, hollow tap root or a cluster of roots with tuberous swellings. Spines are present along the undulating leaf margins and at the tips of the scales (involucral bracts) under the flower head. Hill's Thistle plants live as sterile rosettes of leaves for several years (maximum five years) up until the final year when they produce an upright stem (25-60 cm) with a single, large flower head (3.5-5 cm in height) (NatureServe 2010). Flowering occurs from mid-June to mid-September with the main peak in July. Mature flowers are a rich mauve colour. Hill's Thistle also reproduces vegetatively with buds along lateral roots (Higman and Penskar 1999). After flowering and setting seed, the plants and the primary tap root usually die, although new rosettes produced from adventitious buds may continue to grow. Unlike some weedy thistle species, Hill's Thistle does not spread by rhizomes.

Hill's Thistle can be distinguished from other thistles by the stem, which is sparsely hairy to wooly and lacking wings or spines. As well, the leaves of Hill's Thistle are only shallowly lobed to wavy-margined and have fewer spines than those of other thistle species. The spines that are present on both the leaves and flower heads of Hill's Thistle tend to be shorter and finer than those of other thistles (COSEWIC 2004; Higman and Penskar 1999).

In The Flora of North America (Kell 2006), Cirsium hillii is not considered a distinct species but is treated as Cirsium pumilum (Nuttall) Sprengel var. hillii (Canby) B. Boivin. According to NatureServe (2009), Cirsium hillii is apparently very similar in appearance to C. pumilum, but C. hillii differs in being a monocarpic perennial species (living a variable number of years as a rosette before flowering, setting seed, and dying) possessing shallowly-lobed stem leaves with short prickles, and a single hollow, tuberous root. C. pumilum, in contrast, is a biennial that possesses a solid tap root and deeply lobed stem leaves with numerous prickles.

Little information exists in the literature on the biology of Hill's Thistle other than what is already presented in the species description above. Additional background on Hill's Thistle compared to other thistles can be found in Moore and Frankton (1974).

Hill's Thistle requires habitat that is dry and open with little canopy cover (Figure 1a). The species is typically found in patches of open ground growing with low grasses, especially Poverty Oat Grass (Danthonia spicata), reindeer lichens (Cladina rangiferina and C. mitis), and scattered shrubs (Figure 1b). It is not found in dense vegetation or in situations where it is overtopped or crowded by other plants (Jones 1995-2008; COSEWIC 2004; Jalava 2004-2008; Janke et al. 2006; White 2007a). The tree canopy, if present, is predominantly coniferous and very open in a savanna or woodland situation.

The open and grassy habitat required by Hill's Thistle can be part of several different vegetation types, including prairies, sand barrens, oak and jack pine savannas, some types of alvars, in open woodlands, and at the backs of dunes (both current and relict) (Voss 1996; Penskar 2001; NatureServe 2009). Hill's Thistle has been considered by many to be an alvar species (Catling 1995; Brownell and Riley 2000); however it can occur in other vegetation types, provided the correct conditions are present. Many different vegetation community types that support Hill's Thistle have been documented.

Typical associates are native graminoids such as Poverty Oat Grass, Little Bluestem (Schizachyrium scoparium), Ebony Sedge (Carex eburnea), and Richardson's Sedge (Carex richardsonii), as well as reindeer lichens, Common Juniper, Bearberry (Arctostaphylos uva-ursi), Creeping Juniper, and Dwarf Lake Iris (Iris lacustris) (COSEWIC 2004; Jalava 2004-2008; Jones 1995-2008).

Soils range from sandy near the Lake Huron shore to silty and slightly alkaline on alvars. Soils are often shallow or may consist of no more than a mound of sand on top of flat limestone or dolostone bedrock.

Fire is probably required to create or maintain the habitat of Hill's Thistle. Many vegetation types in which the species occurs are considered "fire-prone" or "fire-dependent" (COSEWIC 2004; Penskar 2001; Higman & Penskar 1999) since fire prevents the accumulation of shrubs and trees. However, it is probably more accurate to say these vegetation types were created by fire. Jones (2000) showed that nearly all oak savannas in one region of Manitoulin Island were deciduous forests prior to a historic fire and were created in a single event, but almost none had burned a second time. A large number of alvars on the Bruce Peninsula and Manitoulin Island were burned in the past, but most of the burn evidence appears to be very old with no recent burning (in the last 50-70 years). In addition, a great deal of Manitoulin Island was burned prior to the first land surveys in the 1870s, but there have been very few fires since then (Jones and Reschke 2005; Jones 2000; Reschke et al. 1999). Finally, suitable habitat still exists now, more than 100-150 years after most major fires occurred in the habitat, so fire may be needed only on long cycles of time, perhaps every 100-200 years or more.

There is little evidence that frequent, low-level fire maintained the habitat historically. However, given the need for fire suppression to protect human life and property, it is possible that low-level controlled burning may be required to maintain habitat in the absence of canopy-reducing fires. The results of a controlled burn at Wasaga Beach Provincial Park in 2004 are inconclusive as to whether there was a benefit to Hill's Thistle (White 2007a; Korol pers. comm. 2007). Insights may be gained through ongoing monitoring at this site. As well, Jones (unpublished field notes 2007) observed five locations on Manitoulin Island that had undergone burning in the last five to 30 years near to extant Hill's Thistle populations. None of these burns resulted in the creation of vegetation similar to that in which Hill's Thistle is currently found.

Hill's Thistle often occurs in areas of very old disturbance such as sites of old burning or historic-era logging. Hill (1910) observed the species in 1910 south and west of Chicago "in railway enclosures fenced off from the surrounding prairie before the land has been touched by the plow." He also noted that the species was able to spread into pastures and fallow agricultural fields. Nonetheless, these areas were likely not as disturbed, nor as weedy, as they are today and they probably still contained a significant component of native flora. Today in Canada, Hill's Thistle is never found in heavily disturbed areas or in fallow agricultural fields (Jones pers. obs.; Jalava pers. obs.; habitat data on file in NHIC database, NHIC 2009).

Figure 1. A. Typical Hill's Thistle habitat with open grassy ground.

Figure 1. B. Basal rosettes of Hill's Thistle (centre) with its typical associates of Poverty Oat Grass (throughout background) and Bearberry (small, round shiny leaves, at centre top and bottom right).

Whether or not Hill's Thistle requires disturbance may depend on the quality of the existing habitat. For habitat that is becoming densely vegetated, it has been suggested that light disturbances, such as lightly used trails, may help keep small patches of ground open, thus creating or maintaining habitat for Hill's Thistle (COSEWIC 2004). Indeed, there are several observations of Hill's Thistle growing along trails (TNC 1990 cited in COSEWIC 2004; Jones 1995; Jalava pers. comm. 2009). In some cases where the vegetation is closing in, the trail is the only open ground remaining.

On the other hand, in large areas of good quality, open, grassy habitat, even light anthropogenic disturbance (such as occasional ATV use on a designated trail) can cause considerable damage, by bringing in weeds, creating ruts, disrupting soil, thus causing a general degradation of habitat (Jones unpublished field notes 2007). Therefore, while light disturbance may be useful in marginal habitat, in good quality habitat it may be detrimental. Furthermore, such anthropogenic disturbance would be very difficult to control and is not recommended as a tool to maintain habitat. Threshold levels at which disturbance becomes harmful have not been determined. Further study of techniques and processes to keep habitat open is recommended (see Knowledge Gaps, Table 2).

Loss of suitable habitat from filling in of habitat due to fire suppression or from development are the primary threats to Hill's Thistle. These and other threats are presented in Table 1.

Table 1. Current threats affecting Hill's Thistle by core area. Intensity of threats is: high (H), medium (M), low (L), or nil (N/A).

Lack of suitable habitat has been identified as the primary threat to Hill's Thistle (Jones 1995-2008; COSEWIC 2004). The species is very restricted in its requirements for natural, dry, open, grassy patches. Suitable habitat was formerly much more widespread after extensive historic fires on Manitoulin Island and the Bruce Peninsula (Jones and Reschke 2005). Now, after more than 100 years of human suppression of the natural fire regime, filling-in of the vegetation has resulted in small isolated habitat patches, unsuitably low light levels, and higher competition for ground space and nutrients. Lack of wild fire also allows leaf litter to build up on the ground, which results in poor seedling establishment (Higman and Penskar 1999). Although habitat may close in at a very slow rate, much of the remaining extant habitat is now at the point of becoming unsuitable due to the density of vegetation and the habitat patch size (<1/2 ha).

Most Hill's Thistle occurrences are near the Lake Huron and Georgian Bay shoreline in areas which are prime real estate for development. Even away from the shorelines, for example in the centre of Manitoulin Island, open grassy areas are frequently chosen as places to build cottages, hunt camps, and other structures because the open ground does not need to be cleared. Building and road construction destroy both habitat and individual plants.

Driving heavy machinery in Hill's Thistle habitat destroys individual plants and compacts or displaces shallow soils leaving huge ruts. It also introduces weed species, which may reduce or eliminate the native species. Heavy machinery is used to remove erratic glacial boulders (which have economic value to the landscaping industry) from the habitat. Machinery for logging operations in forests adjacent to alvars frequently ends up crossing open habitat or being parked there. In addition, log landing areas are frequently located in open habitats adjacent to forests.

Several populations in protected areas are located on hiking trails and can threatened by trampling, which can destroy the plants, displace soil, and bring in weeds. However, there are many situations where Hill's Thistle grows on trails that are maintained in a suitable state by a low level of human usage. This is especially true when the trail provides the only remaining suitably open ground. Therefore, managing intensity of usage to achieve the correct balance is needed, and detecting the point where usage becomes impactful is a key issue.

As with trampling, ATV use is a threat to Hill's Thistle when it occurs in sufficient intensity to damage or destroy plants, displace soil, cause ruts, or introduce weed species that reduce native species presence. Moreover, unlike larger vehicles, ATVs are not limited to trails, so the damage is often much more widespread. All-terrain vehicle use has probably caused the extirpation of at least one Hill's Thistle population (COSEWIC 2004). However, there are situations where Hill's Thistle is found along the edges of ATV trails because the trail is the only remaining open ground in an encroaching forest (Jalava pers. com. 2009; and field data on file in NHIC database). Whether a low level of ATV use may maintain habitat or damage it may also depend on the location or vegetation type. Dunes and alvar grasslands are particularly vulnerable to damage, and therefore may not withstand even light ATV use. Again, monitoring the level of usage is essential. See the discussion on disturbance in Section 1.3.2 for further explanation.

The Bruce Peninsula is a prime location for the quarrying of limestone, and alvar habitats where there is little cover of bedrock are preferred sites for this type of development. Hill's Thistle plants located during the environmental work for the approvals process must be protected, but development of new quarries can cause loss of habitat. In the Manitoulin Region, Hill's Thistle occurs within two already-licenced quarries (International Alvar Conservation Initiative field notes 1996; COSEWIC 2002), but the current status of these populations is unknown. Development of new aggregates sites in the Manitoulin Region now requires a natural environment technical study as the region has been designated under the provincial Aggregates Act. Thus, Hill's Thistle plants should now receive more protection in that region.

On Manitoulin Island, stems of Hill's Thistles with the flower heads eaten off have been observed (Jones 1996-2009 unpublished observations). For small populations of Hill's Thistle where only a few individuals may flower, sometimes after a period of many years, loss of flower heads to browsing may be a serious threat. Deer are abundant in the Manitoulin Region and damage to vegetation is frequently observed (Selinger pers. comm. 2010).

Low seed viability and low seed germination rates may be limiting factors for this species, and seedlings may be poor competitors for light and space (NatureServe 2010). However, the primary problems affecting Hill's Thistle are threats, not intrinsic limitations (Jones 2004-2009; Jalava 2004a, 2005, 2007, 2008a, b).

In order to plan recovery of Hill's Thistle, it is important to see the work that has already been done to avoid duplication of efforts. Much work to protect alvars and increase awareness of their significance pre-dated this recovery strategy. Many of these actions have directly protected or otherwise benefited Hills Thistle populations. Some of the major accomplishments include:

This bi-national, range-wide study of alvars produced detailed, standardized field inventories of the majority of significant alvar sites in Ontario, Michigan, New York, and Ohio (Reschke et al. 1999). Fieldwork included botanical surveys, vegetation community inventory, classification and mapping, and specific studies on a number of ecological processes including fire history and natural succession (Schaefer 1996, Schaefer and Larson 1997, Jones 2000, Jones and Reschke 2005). Information on Hill's Thistle was collected at many major sites as part of this survey. As well, several major alvar sites that support Hill's Thistle (including Quarry Bay, Belanger Bay, and Burnt Island Harbour) were protected as a result of this project. Stewardship packages were distributed to alvar landowners to raise awareness of the uniqueness of the alvar ecosystem and its rare species (including Hill's Thistle) (Jalava 1998; Jones 1998).

This ecological study of Ontario alvars ranked significant alvars on a regional basis (Brownell and Riley 2000). Presence of Hill's Thistle was one of the special features upon which the ranking was based.

This park has had a monitoring program for Hill's Thistle in place since 1996 (White 2007a, b) and conducted a controlled burn in its habitat in 2004 (Jackson 2004). The results of these efforts will be useful as background information for the design of range-wide monitoring and habitat management plans for Hill's Thistle.

A number of alvars have been protected in the last 10 years as a result of conservation work for that ecosystem (Parks Canada Agency 2010). Many of these alvar sites support populations of Hill's Thistle. See Section 2.7 Habitat Conservation, for a list of the protected areas which support Hill's Thistle.

At Bruce Peninsula National Park (BPNP), Misery Bay Provincial Nature Reserve, Queen Elizabeth-Queen Mother M'nidoo M'nissing Provincial Park, and private nature reserves such as the Quarry Bay Nature Reserve, management is focusing on maintaining ecological integrity of habitats, including many areas where Hill's Thistle is present.

Table 2 summarizes important knowledge gaps for Hill's Thistle in Ontario. Filling these gaps will provide information that can be used to reduce threats or to better manage habitat. As well, a better understanding of species biology may clarify which threats are serious impacts and which are not.

Table 2. Summary of knowledge gaps.