• E-mail
• Print
• Comment
• Font Size
• Digg
• del.icio.us
• Discuss article

The Habitat and Plant Associates of Eriocaulon Decangulare L. In Three Southern Appalachian Wetlands

Posted on: Sunday, 24 July 2005, 03:01 CDT

ABSTRACT

Eriocaulon decangulare is a wetland species that is rare in the North Carolina mountains. The mountain populations are disjunct from those in the Coastal Plain, where the species is more common. We investigated the habitat and plant associates of E. decangulare in three mountain wetlands during summer 2002. We centered 40, 0.25- m^sup 2^ quadrats around randomly selected E. decangulare plants (8- 20 quadrats per wetland), and estimated the percent cover occupied by all species within the quadrat. We also measured photosynthetically active radiation in each quadrat and determined the soil pH at each site. For seven plant types, only the coverage of woody plants differed significantly among sites. Most of the woody plants occurred as seedlings, small sprouts, or trailing stems. Overall, E. decangulare appeared to favor open, acidic, sunny conditions with abundant Sphagnum mosses. We documented four other unique and uncommon wetland plant species occurring with E. decangulare, and recommend continued management of all three wetlands to suppress woody species and increase the sunlight available to smaller herbaceous plants.

INTRODUCTION

Eriocaulon decangulare L. (common names include ten-angled pipewort, hat pin, and bog button) is a monoecious perennial that occurs in wetlands throughout the coastal plain of North Carolina. It is absent from the Piedmont, but is represented by a few disjunct populations in the western mountains (Radford et al. 1968). Its growth form includes a basal rosette of leaves, with white, button- like flowering heads that are visible from June through October. The flowers are probably wind-pollinated, and the plants can reproduce vegetatively by lateral offshoots, and by rhizomes (Kral 1966).

Eriocaulon is the only genus in the Eriocaulaceae that occurs north of Florida (Kral 1966). All 10 species in the genus have been described taxonomically, but little information is available on their ecology. Published ecological information on Eriocaulon is primarily in the form of general observations, such as Kral's (1966) observation that species in this genus tend to be weedy heliophytes that colonize disturbed wet soils quickly, and are crowded out in later succession. Kral also noted that although most species of Eriocaulon indicate boggy, wet soils, E. decangulare may also occur in relatively dry pine forests. Watson et al. (1994) studied another species in this genus, E. kornickianum Van Heurck & Muell. (dwarf or small-headed pipewort) in the Interior Highlands. They reported that the habitat requirements, population structure, and ecological relationships of this species were virtually unknown, and that the species is declining rapidly in parts of its range.

The North Carolina Natural Heritage Program (NCNHP) has classified Eriocaulon decangulare as regionally rare in the mountains (Amoroso and Finnegan 2002), due largely to the loss of suitable habitat. In the southern Appalachians, the acreage and quality of non-alluvial wetlands have declined due to development, agriculture, drainage, and other land use activities (Moorhead and Rossell 1998). As such, E. decangulare represents just one of many rare and uncommon species that are unique to the dwindling acreage of southern Appalachian wetlands. A decade ago, Murdock (1994) reported that non-alluvial mountain wetlands in North Carolina contained almost 20% of 724 rare plant species monitored by the NCNHP.

Our objective was to examine three populations of Eriocaulon decangulare in the mountains of western North Carolina, in order to gain a better understanding of the habitat and plant associates of this uncommon wetland species. Specifically, our goals were to 1) document the plant species closely associated with E. decangulare, and 2) quantify the ranges of soil pH and sunlight in habitats where this species occurred.

STUDY AREAS

We located three populations of Eriocaulon decangulare in western North Carolina during the summer of 2002. Our study sites were located in Clay, Graham, and Yancey Counties (Figure 1). These sites were chosen based on their known populations of Eriocaulon decangulare, and the logistics of obtaining landowner permission to visit the sites. All three of the wetlands fit the description of southern mountain fens (Moorhead and Rossell 1998).

The Clay County site (elevation 600 m) contains a fen (approx. 4 ha) in which the center is dominated by herbaceous plants such as sedges, Eupatorium spp., and Panicum dichotomum L, and by trailing stems of Rubus hispidus L. The edges of the fen are dominated by shrubs such as Aronia arbutifolia (L.) Ell., Aronia melanocarpa (Michx.) Ell., and Alnus serrulata (Ait.) Willd., and trees such as Acer rubrum L., Liriodendron tulipifera L., and Pinus echinata Miller. This site is burned periodically to manage woody plants. We estimated that the population of E. decangulare consisted of <50 plants.

The Graham County site (elevation 790 m) contains a small fen (1.3 ha) in which open areas are dominated by herbaceous plants such as Carex stricta Lam., Juncus effusus L., and Panicum dichotomum, along with Rubus hispidus. Eriocaulon decangulare was very abundant in open areas of the fen in the mid-1990's (I. Rossell, pers. obs.). However, during the past 10 years, natural succession has increased the dominance of trees and shrubs such as Acer rubrum, Alnus serrulata, Aronia arbutifolia, Aronia melanocarpa, and Rosa palustris Marsh. Eriocaulon is now limited to the edges of a powerline right-of-way, where woody vegetation is controlled with occasional cutting and herbicides. We estimated the population of E. decangulare at this site to be between 50-100 plants.

The Yancey County site (elevation 862 m) contains a small fen (0.4 ha) that is dominated by herbaceous plants such as Rhynchospora spp., Juncus spp. and Solidago spp., and woody plants including Rhododendron maximum L., Acer rubrum, and Alnus serrulata. Woody vegetation at this site is managed by periodic cutting. At the time of our study, this site supported the largest population of E. decangulare (estimated >200 plants).

METHODS

We collected data in Graham and Yancey Counties in July 2002, and in Clay County in early August 2002. At each site, we centered a 0.25-m^sup 2^ quadrat frame around one randomly selected Eriocaulon decangulare plant (no quadrats overlapped). The number of quadrats at each site varied with our estimate of the size of the E. decangulare population: 20 quadrats at the Yancey County site, 12 at the Graham County site, and 8 at the Clay County site (total of 40 quadrats). In each quadrat, we identified all plants growing within the quadrat, and visually estimated the percent cover occupied by each species. For E. decangulare, we estimated the cover of the central E. decangulare plant, along with any other E. decangulare in the quadrat. Voucher specimens from the Graham and Yancey County sites were collected and deposited at the University of North Carolina Herbarium. Plant nomenclature follows Wofford (1989). Although we had landowner permission to visit the Clay County site, we did not have permission to collect plants there.

We used a Li-Cor LI-1905A quantum sensor to take three measurements of photosynthetically active radiation (PAR) in each plot. We took one reading at the top of the Eriocaulon flower closest to the center of the focal plant, and one reading just above the basal leaves of this plant. The third measurement was taken in full sun, so that readings could be expressed as percent of full sunlight. All readings were taken between 1100 and 1400 hours.

We also took two soil samples from each site. Each sample consisted of 20 soil plugs (2.5-cm diameter 20-cm depth) that were mixed thoroughly and air-dried. The pH of each sample was determined from a 1:1 (dry weight:volume) slurry of air-dried soil and distilled water, which had equilibrated for 30 min. The pH was measured with a standard electrode.

We summed the mean percent cover of each of seven plant types (forbs, grasses, sedges, rushes, woody plants, mosses, and ferns) in each quadrat. We used non-parametric Kruskal-Wallis tests (alpha = 0.05) to determine whether the cover of each plant type, the percent of full sunlight received by the flowers, or the percent of full sunlight above the basal leaves differed between the three sites.

We obtained the Region 2 (southeastern United States) wetland indicator status for each plant species by consulting United States Fish and Wildlife Service (1996) and United States Department of Agriculture (2001). Wetland indicator status categories describe wetland affinities as follows: obligate wetland plants (OBL) occur in wetlands >99% of the time, facultative wetland plants (FACW) occur in wetlands 67-99% of the time, facultative plants (FAC) occur in wetlands 34-66% of the time, facultative upland plants (FACU) occur in wetlands 1-33% of the time, and upland plants (UPL) occur in wetlands <1% of the time. Categories are listed in Table 1 for all species for which an indicator status was available.

RESULTS AND DISCUSSION

At all of our sites, Eriocaulon decangulare occurred in acidic conditions. Soil pH ranged from 4.2 in Graham County to 4.5 in Clay County and 5.1 in Yancey County. The plant associates of E. decangulare were very similar across the three sites. In Graham and Yancey Counties, Sphagnum mo\ss covered about 60% of the area inside the quadrats. Quadrats in the Clay County site contained 29% Sphagnum. Although coverage of Sphagnum in Clay County did not differ statistically from that at the other sites (P = 0.494), it is notable that of the three sites, the Clay County site supported the smallest estimated population of E. decangulare. We did not quantify soil moisture or hydrology in the areas where E. decangulare occurred, but the presence of Sphagnum suggests that moisture is an important habitat component. The wetland indicator status of the plants at each site support this observation. Of the species for which we could obtain a wetland indicator status, the percentages that were either OBL or FACW were: 71% in Graham County, 58% in Yancey County, and 61% in Clay County.

Figure 1. Map of North Carolina showing study site locations.

Table 1. Mean percent cover of plant taxa in 0.25-m^sup 2^ quadrats centered around random Eriocaulon decangulare plants in three wetland sites in western North Carolina. Across rows, values followed by different letters differ significantly (P < 0.05)

Table 1. Mean percent cover of plant taxa in 0.25-m^sup 2^ quadrats centered around random Eriocaulon decangulare plants in three wetland sites in western North Carolina. Across rows, values followed by different letters differ significantly (P < 0.05)

Areal coverages of four of the seven plant types (grasses, forbs, sedges, and woody plants) were generally consistent between sites, ranging between 16 and 35%. The only plant type that differed statistically in coverage among the three sites was woody plants (P = 0.023), for which coverage was significantly greater in quadrats at the Yancey County site (30%) than at the other sites (Table 1). However, it should be noted that because all of the sites are managed for woody plant control, our quadrats contained only seedlings, small sprouts, and trailing stems of woody species. At the Clay and Graham County sites, most of the woody coverage was by Rubus hispidus (swamp dewberry), which is a a low, trailing species that occurs close to the ground. In contrast, at the Yancey County site, where woody coverage was the greatest, there was an abundance of small Alnus serrulata (tag alder) seedlings or sprouts. It is likely that differences in management practices (cutting, burning, and herbicides) and in the frequency of management at each site have impacted the species and growth patterns of woody plant regeneration in each wetland.

Ferns and rushes were not important associates of Eriocaulon decangulare at any of the three sites (each occupied <3% of the area inside the quadrats). Interestingly, in the sandy, acidic seeps of Oklahoma, Watson et al. (1994) reported that Eriocaulon kornickianum was positively associated with rushes (Juncus spp.) and bare ground, and negatively associated with mosses (they found only 7% Sphagnum inside their quadrats). The authors concluded that E. kornickianum had poor competitive abilities, and that its continued success required regular removal of the ground cover (through grazing or fire). In Ontario, Wilson and Keddy (1986) determined that a related species, Eriocaulon septangulare With. (seven-angled pipewort) had the lowest competitive ability of seven herbaceous lakeshore species. They found E. septangulare along the most disturbed portions of the lakeshore, and suggested that it was able to succeed only after disturbance had removed the competitive dominants.

Disturbance also seemed to benefit Eriocaulon decangulare in western North Carolina. At our sites, it occurred in open, sunny conditions (>90% full sunlight in Graham and Yancey Counties) (Table 2). Plants in Clay County were growing in only 50% full sunlight, which was significantly less than at the other two sites (P < 0.0001 for the tops of flowers; P = 0.006 for the basal leaves). This site also supported the smallest estimated population of Eriocaulon. The reduced sunlight in Clay County was due to shading by tall herbaceous plants within the quadrats. For example, Sanguisorba canadensis L. and Eupatorium fistulosum Barratt were two plant associates at this site, and both can reach heights of 1 to 2 meters.

Table 2. Percent of full sunlight received by Eriocaulon decangulare plants at three wetland sites in western North Carolina. Light was measured at the top of one flower and just above the basal leaves of randomly selected plants. Within columns, values followed by different letters differ significantly (P < 0.05)

Without regular disturbance or continued management, trees and/ or shrubs will likely dominate all three sites in the future, decreasing the sunlight available to smaller herbaceous plants like Eriocaulon, and lowering the water table through transpiration (Moorhead and Rossell 1998). At our Graham County site, Warren (2002) studied the impacts of woody succession on herbaceous wetland communities between 1994 and 2001. He showed that the number of woody species in the open area of the fen doubled during this seven year period, and that increasing woody basal area was negatively correlated with the coverage of ground-layer plants.

We documented several unique and uncommon wetland plants as associates of Eriocaulon decangulare. All are herbaceous species of relatively small stature. In Yancey County, Cleistes divaricata (L.) Ames (spreading pogonia) occurred in 10% of quadrats, and Drosera rotundifolia L. (round-leaved sundew) occurred in 55% of quadrats. In Graham County, Carex trichocarpa Muhl. ex. Willd. (hairy-fruit sedge) occurred in 42% of quadrats. In Clay County, Sanguisorba canadensis (Canada burnet) occurred in 38% of quadrats. Carex trichocarpa and Sanguisorba canadensis are both classified as significantly rare in North Carolina (Amoroso and Finnegan 2002) and are at the periphery of their range in North Carolina. In fact, the Graham County site contains the southernmost known population of Carex trichocarpa in the United States (Warren et al. 2004).

Clearly, active management of these wetlands to control woody plants and increase sunlight will benefit not just Eriocaulon decangulare, but a suite of unique and uncommon wetland species. Since the distinctive white flowering heads of E. decangulare are easy to recognize and persist throughout much of the summer and fall, we suggest that where it occurs, this species could be used as an indicator species for wetland monitoring and conservation purposes.

ACKNOWLEDGMENTS

We thank J. Dan Pittillo and Robert Warren for assistance with plant identification, Stacey Hatcher, Heidi Losure, and T.R. Russ for field assistance, Kevin Moorhead for helping with the soil analyses, C.R. Rossell, Jr. and two anonymous reviewers for comments on earlier versions of the manuscript, and the landowners of our study sites for allowing us access.

LITERATURE CITED

AMOROSO, J. and J.T. FINNEGAN (eds.). 2002. Natural Heritage Program list of the rare plant species of North Carolina. North Carolina Natural Heritage Program, Division of Parks and Recreation, Department of Environment, Health, and Natural Resources. Raleigh, North Carolina.

KRAL, R. 1966. Eriocaulaceae of continental North America north of Mexico. Sida 2:285-332.

MOORHEAD, K.K. and I. M. ROSSELL. 1998. Southern mountain fens. p. 379-403. In: Messina, M.G. and W.H. Conner (eds.). Southern forested wetlands: ecology and management. Lewis Publishers, Boca Raton, Florida.

MURDOCK, N.A. 1994. Rare and endangered plants and animals of southern Appalachian wetlands. Water Air Soil Pollut. 77:385-405.

RADFORD, A., H. AHLES, and C. BELL. 1968. Vascular flora of the Carolinas. The University of North Carolina Press, Chapel Hill, North Carolina.

UNITED STATES FISH AND WILDLIFE SERVICE. 1996. National list of vascular plant species that occur in wetlands. National Wetlands Inventory, U.S. Fish and Wildlife Service, Washington, D.C.

UNITED STATES DEPARTMENT OF AGRICULTURE. 2001. The PLANTS database, version 3.1. National Plant Data Center, U.S. Department of Agriculture, Baton Rouge, Louisiana(http://plants.usda/gov).

WARREN, R.J., II. 2002. Impact of woody vegetation succession upon herbaceous communities in a southern Appalachian wetland. M.S. thesis, Western Carolina University, Cullowhee, North Carolina.

WARREN, R.J., II, J.D. PITTILLO, and I.M. ROSSELL. In press. Vascular flora of a southern Appalachian fen and floodplain complex. Castanea 69:116-124.

WATSON, L.E., G.E. UNO, N.A. MCCARTY, and A.B. KORNKVEN. 1994. Conservation biology of a rare plant species Eriocaulon kornickianum (Eriocaulaceae). Amer. J. Bot. 81:980-986.

WILSON, S.D. and P.A. KEDDY. 1986. Species competitive ability and position along a natural stress/ disturbance gradient. Ecology 67:1236-1242.

WOFFORD, B.E. 1989. Guide to the vascular plants of the Blue Ridge. University of Georgia Press, Athens, Georgia.

Received November 20, 2003; Accepted June 11, 2004.

IRENE M. ROSSELL* and DAVID A. LOSURE

Environmental Studies Department, University of North Carolina at Asheville, Asheville, North Carolina 28804

* email address: irossell@unca.edu

Copyright Southern Appalachian Botanical Society Jun 2005

Source: Castanea

More News in this Category