David Giblin Center for Urban Horticulture University of 3501 NE 41st Street Seattle, Washington 98195

Aster Curtus: Current Knowledge of Its Biology and Threats to Its Survival

David Giblin

Abstract

Aster curtus Cronq. , a taxon in the family , is a rhizomatous, herbaceous perennial endemic to the remnant prairies of , Washington and Vancouver Island, British Columbia. The current population center is located on Fort Lewis Military Base in Pierce and Thurston counties of Washington State. A “species of concern” under the Endangered Species Act, its current listing in Washington is “Sensitive”. Aster curtus cover is highest on prairies with low exotic species invasion and competes poorly with associated native and exotic species at the seedling stage. Under field conditions, recruitment by seed appears to be low, and persistence of this species is achieved primarily through clonal growth. Aster curtus is self-compatible, though putative out crossing appears to result in more filled seeds. Industrial, residential and agricultural developments are the largest threats to its survival. Considering the degree of habitat degradation that has already occurred throughout the range of A. curtus, the long-term survival of this species will depend on how well we understand its autecology and how well we protect its habitat. This paper summarizes past and present information that can be applied to the conservation efforts of this species.

Introduction range, as evidenced by the lack of current records for populations between Aster curtus is endemic to the Pierce/Thurston counties, Washington Willamette-Puget lowlands of the Pacific and Vancouver Island, British Columbia. Northwest. Its present range extends Additionally, no populations are from lower Vancouver Island, British presently known to exist between these Columbia to Lane County, Oregon, with two Washington counties and Clackamas the largest populations found on Fort County, Oregon (Gamon and Salstrom; Lewis Military Base in Pierce and 1992). The reasons for this patchy Thurston counties, Washington (Gamon distribution are unknown, but conversion and Salstrom; 1992). Over the past 20 of grasslands to pasture and commercial/ years, more than 70 new populations of residential development have likely been A. curtus have been reported, however major factors. Despite much attention these discoveries have largely occurred give to A. curtus over the past 15 years in counties where the species was (Alverson, 1983; Clampitt, 1984; Giblin, already known to be extant (Gamon and 1997) many questions remain regarding Salstrom; 1992). The distribution of A. the reasons for its rarity. curtus is not continuous throughout its

D. Giblin Aster curtus 93 Taxonomic Status As a federal level, A. curtus is listed as a Aster curtus Cronq., was originally species of concern (USFWS, 1996). This described as rigidus classification is given to those species Lindl.in 1834 on the basis of collections for which more information is needed made by Scouler along the Columbia before formal listing as endangered or River. Cronquist changed S. rigidus to threatened (Thomas and Carey, 1996). Aster curtus (Hitchcock et al., 1955), The Washington Natural Heritage and more contemporary taxonomic work Program downgraded A. curtus from its has placed Sericocarpus as a subgenus original listings as “Threatened” to of Aster (Jones, 1980; Semple and “Sensitive” (Washington Natural Brouillet, 1980). This subgenus is Heritage Program, 1990), undoubtedly comprised of the western U.S. species A. the result of more populations having curtus and A. oregonensis, the eastern been located since earlier versions of the U.S. species A. paternus, A. list. The state of Oregon lists A. curtus as solidagineus, and A. tortifolius, and the “Threatened”, and this guarantees legal Asian species A. baccharoides. Only A. protection for this species on all state- curtus is a grassland native with the owned or state-leased property (Oregon remaining species typically inhabiting Natural Heritage Plan, 1993). British the forest under story. The common Columbia lists A. curtus as “critically name of A. curtus, white-topped aster, imperiled” because of extreme rarity derives from the effect produced by the (Argus and Pryer, 1990). pappus which expands just prior to seed dispersal. Habitat Characteristics

The ranges of A. curtus and A. A. curtus inhabits the glacial outwash oregonensis overlap from southern prairies and grasslands of the Puget- Washington through central Oregon, Willamette lowlands that were formed however the two species are separated during the Pleistocene glaciations on the basis of habitat preference as well 15,000 years ago. Historically, high as reproductive and vegetative frequency, low intensity fires and morphology. Most notable, the periodically dry soils have kept the woodland species A. oregonensis has bunchgrass-dominated prairie system four ray and over winters as a free from encroachment by forest stout caudex, whereas the grassland vegetation (Hansen, 1947). A. curtus is dweller A. curtus, typically has two ray typically found in open locations where flowers and is strongly rhizomatous cover of native grasses and forbs (Hitchcock et al., 1955). The other five exceeds 50% (Thomas and Carey, 1996), Aster species of the subgenus but will also grow in Quercus garryana Sericocarpus have a haploid (Garry Oak) woodlands where light chromosome number n=9, and this is the levels are high enough to support other presumed condition for A. curtus open prairie vegetation (personal (Clampitt, 1984; Jones, 1980; Semple observation). Native species strongly and Brouillet, 1980). associated with A. curtus include ( fescue), Legal Status lanatum (Woolly daisy),

D. Giblin Aster curtus 94 Hieracium cynoglossoides ( during anthesis (pers. Houndstongue hawkweed), and obs.). A. curtus is self compatible Solidago spathulata var. neomexicana (Giblin, 1997), however pollinators (Thomas and Carey, 1996; Crawford et increase the amount of endosperm-filled al., 1994). seeds produced 4-fold (Giblin, 1997; Clampitt, 1984). The prairie soils of Pierce and Thurston counties, where the majority of A. curtus seed viability is low but not populations are found, typically belong atypically so for species Asteraceae to the Spanaway series (Anderson et al, (Clampitt, 1984). Clampitt (1984) found 1955; Ness, 1958). These are well- that 20% of the achenes produced per drained, gravelly sandy loams which were filled with endosperm. Of become excessively dry between July these filled seeds, only 50% proved to be and September. Low moisture viable through tetra-zolium testing availability during the summer months (Clampitt, 1984). Greenhouse studies results in the dormancy of many spring showed that seed germination was most flowering forbs. favorable after 5 weeks stratification at 5 degrees C, and high light/warm Life History temperature (30 degree C) conditions (Clampitt, 1984). A. curtus is a rhizomatous, herbaceous perennial which produces erect stems 1- Ecology 3 dm in height from slender rhizomes (Hitchcock et al, 1955). New shoots Rabinowitz (1981) developed a emerge in March and become prominent classification system of rarity on the in late spring (pers. obs.). Flowering basis of geographic distribution, habitat occurs from late July through early specificity, and population size. Within September with individual florets being this system, A. curtus is described as a strongly protandrous. Pollen is presented regional endemic due to its narrow in the morning and usually absent by distribution, narrow habitat specificity days’ end as a result of removal by bees, and large population sizes. Indeed, when disruption of the by wind, encountered in the field, A. curtus is and perhaps desiccation. Stigmas appear observed to be locally abundant over a to be receptive for up to 3 days (pers. small geographical area. obs). On average there are 13 disk flowers and 3 ray flowers per capitulum are divided into three groups on (Clampitt, 1984; Hitchcock et al, 1955). the basis of whether their survival Only 10-30% of the ramets within a strategies are ruderal, stress tolerant, or patch bear inflorescences (Gamon and competitive (Grime, 1977). Due to its Salstrom, 1992). Most seeds ripen and slow growth rate, ability to persist under are wind dispersed by late September. severe competition, low stature, and Wind dispersal is achieved via the bristly allocation of resources toward clonal pappus attached to the mature achene. expansion prior to flowering, A. curtus is considered to be a stress tolerant species Effective pollination is achieved by 3-4 (Clampitt, 1984). bee species, though butterflies do visit

D. Giblin Aster curtus 95 On account of its rhizomatous nature, A. strongly rhizomatous Agrostis tenuis curtus forms patches where it occurs. (Bentgrass) and the fall germinating Patch size is extremely variable both in Hypochaeris radicata (Cat’s ear) are two area and number of ramets, with native exotic species which can quickly species abundance strongly influencing colonize microsites favorable for these factors. ON Fort Lewis prairies recruitment of A. curtus seedlings. where native species dominated, Thomas Carey (1996) found patch area to Current Research average 10m2, though they observed patches up to 40m2. They also report In the summer of 1996 I began ramet numbers ranging from 16 to researching the reproductive biology of 10,000 per patch, with a few patches A. curtus, and the influence of different containing up to 200,000 (Thomas and disturbance regimes on seed germination Carey, 1996). Seedlings are seldom and seedling establishment. Greenhouse observed in the field (pers. obs.), and studies are being conducted at the persistence of existing populations is University of Washington’s Center for achieved through clonal expansion Urban Horticulture, and field studies at (Clampitt, 1984). 13th Division Prairie on Fort Lewis.

A. curtus shows a preference for Preliminary results from greenhouse mounded areas in the prairie. Clampitt studies suggest that A. curtus is self- (1984) showed that A. curtus is most compatible. Self and cross-pollinated likely to occur in the middle regions of stigmas were harvested and examined prairie mounds. My own personal using fluorescent microscopy observations on 13th Division and Upper techniques. Pollen grains from nearly all Weir prairies have confirmed this cross-pollinations have shown high association. A. curtus does grow in areas germination percentages on the stigmatic between mounds, but not with the same surface and pollen tube penetration into degree of predictability. The explanation the stylar region. Pollen grains from for this microsite preference likely many of the self-pollinations showed involves the ability of A. curtus to little or no adhesion to the stigmatic tolerate the droughty conditions of this surface, low germination percentages, location during the summer months. and little stylar penetration. Nevertheless, a modest number of self- Where exotic species presence is high, pollinations produced high germination A. curtus cover tends to be low. This percentages and stylar penetration. negative correlation is likely due to the These results suggest that some genets of ability of exotic species to out compete A. curtus are self-compatible. A. curtus for nutrients. Clampitt (1987) found that A. curtus seedlings competed Field pollination trials were conducted in poorly in greenhouse studies when the summer of 1996, and the results grown with associated native and exotic confirmed the preliminary findings of prairie species. These results suggest that the greenhouse trials. Four pollination the recruitment of A. curtus requires the treatments (within-patch between-patch, exclusion of other species during the exclusion, control) were applied to 20 seedling establishment phase. The patches of A. curtus. Using mesh bags to

D. Giblin Aster curtus 96 prevent unwanted pollinations, treatments. Significant differences were inflorescences were hand-pollinated detected between the two foremost and using forceps. Inflorescences used for the two lattermost treatments (Table 1). the exclusion treatment were bagged and received no pollinations. The control Field pollination data suggest that A. inflorescences ere covered with mesh curtus is not pollinator limited and that bags after blooming. The number of self-compatibility is present. These endosperm filled seeds in each results are important in the context of inflorescence was counted, and a number why recruitment of A. curtus by seed is by the total number of seeds per low. One hypothesis is that low inflorescence. Mean dilled seed pollination rates result in low seed percentages for the four pollination production, but results from between- treatments were 39% for open, 35% for patch pollinations discount this. A between patch, 17% for within patch and second hypothesis is that seed 9% for pollination exclusion. These data production is low due to the presence of were arcsine transformed and analyzed a self-incompatibility mechanism. using mixed model two-way ANOVA. Because A. curtus is clonal, most Results showed a significant difference pollinations occur within a patch. Self- between treatment values at the .05 level incompatibility would inhibit seed (F=3.14; p=<.001). Multiple production from such pollinations. The comparisons were run using the least results suggest that seeds, though lower significant difference test (LSD). No in quantity, can be produced from significant difference existed between within-patch pollinations. open and between-patch treatments, or between within patch and exclusion

Table 1. Comparison of four pollination treatments applied to 20 patches of Aster curtus growing at 13th Division Prairie, Fort Lewis, WA. Superscript values show significant differences at alpha= .05 using LSD test.

Treatment Arcsine Values (SE)

Open 1 .66 (.06) Between Patch 1 .61 (.06) Within Patch 2 .37 (.06) Exclusion 2 .21 (.05)

Table 2. Percent of filled seeds from 20 capitula (1 capitulum per plant) of four species growing on 13th Division Prairie, Fort Lewis, WA.

Species % Filled Seeds

Hypochaeris radicata 82 32 Aster curtus 26 Solidago spathulata 21

D. Giblin Aster curtus 97 In separate experiment, filled seed (RAPDs) analysis will be used to percentage of A. curtus was compared to examine genetic variation of the A. associated native and exotic species of curtus patches of the research area at summer-blooming composites. Twenty Fort Lewis. Results from this analysis ripened heads were collected from A. will provide a very general estimate of curtus, H. radicata, Chrysanthemum genetic similarity between these patches. leucanthemum, E. lanatum, and S. spathulata var. neomexicana, and the ratio of filled seeds to total seeds was Threats to Survival of A. curtus calculated for each species. Hypochaeris radicata showed substantially higher Habitat destruction is the biggest threat mean filled seed percentage (82%) than to the long-term survival of A. curtus. all the other species. Filled seed Conversion of prairie to agricultural percentage for A. curtus was comparable production at the turn of the century has to those of the two native species tested. been replaced by industrial and Final results for C. leucanthemum have residential development. As the Pacific not been determined, though preliminary Northwest population continues to grow, numbers suggest a mean percentage of greater pressures are placed on the 50-60%. These results suggest that seed remaining pieces of this unique set for A. curtus is similar to associated ecosystem. Several sources estimate that exotic species (Table 2). 95% of the original prairie system has been destroyed. In this context, the In September of 1996 I installed a future of A. curtus is precarious. Large randomized block experiment testing for tracts of prairie are contained within the the effect of disturbance on recruitment boundaries of Fort Lewis Military Base, of A. curtus. Five treatments (fire, so these areas are free from the pressures mowing, compaction, gophering, of development. Nevertheless, military control) were replicated five times. Fifty training on these prairies can result in filled seeds were sown into each disturbances detrimental to the health of treatment cell and their location marked A. curtus populations. Clampitt (1993) using a quadrat frame. The fate of these found A. curtus to decline where 1250 seeds will be followed over the vehicular disturbance is high, and course of one growing season to personal observations support this determine if establishment rates differ finding. between disturbance types. Invasive exotics pose a serious threat to Ongoing research also includes the long-term survival of A. curtus. comparing germination rates of the seeds Cytissus scoparius (Scotch broom) harvested from the field pollination Agrostis tenuis (bentgrass), H. radicata trials. Difference in germination rates (Cat’s ear), and C. leucanthemum (Ox- between within and between patch seeds eye daisy) are all aggressive colonizers might suggest the presence of inbreeding of open sites on the prairie. Each of depression. The occurrence of these species have proven to be inbreeding depression within A. curtus is outstanding competitors in this water currently unknown. Additionally, stressed ecosystem. Recruitment and randomly amplified polymorphic DNA persistence of these species at both small

D. Giblin Aster curtus 98 and large spatial scales likely influence lack of limitations on availability of the success of A. curtus. Control of these experimental material. invasives should be an integral part of long-term management plans for A. Conclusion curtus. Efforts need to be directed now to the Finally, alteration of disturbance regimes conservation of A. curtus. Because of its has influenced the community structure apparent abundance in the field, it is of the prairie. Removal of fire has led to often perceived as a species whose encroachment by P. menziesii which in existence is not imminently threatened. turns excludes prairie grasses and forbs. Belying this perception is the fact that A. Decline of pocket gopher and other curtus is found in an extremely limited fossorial mammal populations has led to area subject to frequent anthropogenic the loss of burrowing mounds which disturbances and intense development provided favorable microsites for the pressures. Also troubling is the fact that recruitment of native species. New it does not readily colonize new sites. disturbances such as soil compaction have led to increases in exotic species Too often protection is given to rare invasion, and as a result, declines in species only as they approach extinction, native species populations. and such protection is often expensive and difficult. Rather than ignoring A. Future Research curtus on account of its abundance, conservation efforts should focus on Demography studies using stage-based how to best exploit this condition to projection matrices need to be conducted ensure that long-term persistence of this on patches of A. curtus across several species. A. curtus, a regionally rare prairie sites. It is currently unknown as species in a regionally rare habitat, to whether patches are increasing or represents a conservation opportunity decreasing in size over time. Because A. with a high likelihood of long-term curtus persists almost exclusively success if appropriate actions are through rhizome production, it is critical undertaken now. that we understand the current growth rates of these patches. Literature Cited:

Additionally, research is needed to Alverson, Ed. 1983. Observations on the determine the best means of Occurrence and Status of Aster transplanting and establishing A. curtus Curtus in Washington State. in the field. Successful prairie restoration Unpublished Report. Washington efforts will create opportunities for Natural Heritage Program files, establishing A. curtus in historic Olympia, WA. 13p. plus figures and locations. Presently we do not know appendices. whether transplanted rhizomes or patches of rhizomes will survive and Anderson, W.W., A.O. Ness, and A.C. expand. Because local populations of A. Anderson. 1955. Soil Survey of curtus are currently large, research Pierce County, Washington. USDA. conducted now would benefit from the Soil Survey Report, ser. 1939 #7.

D. Giblin Aster curtus 99 Curtus (Cronq.), a Pacific Northwest Argus, G. W., and K.M. Pryer. 1990. endemic. M.S. Thesis,, University of Rare Vascular Plants in Canada/Our Washington , Seattle. Natural Heritage. Rare and Endangered Plants Project, Botany Grime, J.P. 1997. Evidence for the Division, Canadian Museum of Existence of Three Primary Nature, Ottawa. 191p. plus maps. Strategies in Plants and its Relevance to Ecological and Evolutionary Clampitt, C. A. 1984. The Ecological Lif Theory. American Naturalist History of Aster Curtus, a Grassland 111:1169-1174. Endemis in a Forested Region. M.S. thesis, University of Washington, Hansen, H.P. 1947. Climate Versus Fire Seattle. 81p. plus appendices. and Soil as Factors in Post-Glacial Forest Succession in the Puget Clampitt, C.A. 1987. The Reproductive Sound Lowland of Washington. Biology of Aster Curtus American Journal of Science (Asteraceae), a Pacific Northwest 245(5):365-286. Endemic. American Journal of Botany 74(6):941-946. Hitchcock, C.L., M. Ownbey, and J.W. Thompson. 1955. Vascular Plants of Clampitt, C.A. 1993. Effects of Human the Pacific Northwest. Part 5: Disturbance on Prairies and the Compositae. University of Regional Endemic Aster Curtus in Washington Press. Seattle, WA. Western Washington. Northwest Science. 67:163-169. Hitchcock, C.L., and A. Cronquist. 1973. Flora of the Pacific Northwest. Crawford, R.C., C. Chappell, B. University of Washington Press. Stephens, C. Soper, and D. Rolph. Seattle, WA. 1994. Inventory and Mapping of Jones, A.G. 1980. A Classification of the Endangered Native Ecosystems on New World Species Aster Fort Lewis; Draft Final Report. (Asteraceae). Brittonia 32(2):230- Washington Department of Natural 239. Resources, Division of Forest Resources, Natural Heritage Ness, A.O. Soil Survey of Thurston Program, and The Nature County, Washington. USDA. Soil Conservancy of Washington. Conservation Service. Soil Survey Number 6. Gamon, J. and D. Salstrom. 1992. Report on the status of Aster Curtus. Norton, H.H. 1979. The Association Washington Natural Heritage between Anthropogenic Prairies and Program, Department of Natural Important Food Plants in Western Resources. Olympia, WA. Washington. NW Anthrop. Res. Notes 13(2):175-200. Giblin, D.E. 1997. The Relationship of Reproductive Biology and Oregon Natural Heritage Plan. 1993. Disturbance to the Rarity of Aster Natural Heritage Advisory Council

D. Giblin Aster curtus 100 to the State Land Board. Salem, OR. 158p.

Rabinowitz, D. 1981. Seven Forms of Rarity. In H. Synge, ed. The Biological Aspects of Rare Plant Conservation. John Wiley and Sons. .

Semple, J., and L. Brouillet. 1980. A Synopsis of North American Asters: the subgenera, sections, and subsections of Aster and Lasellea.

Thomas, T.B. and A.B. Carey. Endangered, Threatened, and Sensitive Plants of Fort Lewis, Washington: Distribution, Mapping, and Management Recommendations for Species Conservation. Northwest Science 70(2):148-163.

United States Fish and Wildlife Service. 1996. Endangered and Threatened Wildlife and Plants; Review of Plants and Animal Taxa that are Candidates for Listing as Endangered or Threatened Species. U.S. Code of Federal Regulations 50, Part 17.

Washington Natural Heritage Program. 1990. Endangered, Threatened and Sensitive Vascular Plants of Washington. Washington Department of Natural Resources, Olympia, WA.

Washington Natural Heritage Program. 1994. Endangered, Threatened and Sensitive Vascular Plants of Washington. Washington Department of Natural Resources, Olympia, WA.

D. Giblin Aster curtus 101