Effects of chronic herbivory and historic land use on population structure of a forest perennial, Trillium catesbaei

Applied Vegetation Science, Dec, 2007 by Michael A. Jenkins, Christopher R. Webster, Janet H. Rock

Abstract

Question: How have long-term herbivory and past land use impacted the population structure of Trillium catesbaei, a long-lived rhizomatous herb?

Location: Western Great Smoky Mountains National Park, Tennessee, USA.

Methods: We examined T. catesbaei populations at three sites: (1) Cades Cove (CC), an area of intensive historic land use that has been maintained as open fields and woodlots with a history of chronic deer herbivory, (2) Whiteoak Sink (WOS), a reference area with similar land-use history, geology, and soils that has succeeded to closed-canopy forest with relatively low levels of deer herbivory, and (3) Leadbetter Ridge (LBR), an area of primary forest that has never received significant anthropogenic disturbance. Trillium catesbaei is the most common Trillium species at the three study sites, but smaller in stature, shorter lived, and more of a habitat generalist.

Results: Chronic herbivory in CC has created a highly-truncated age structure with no plants older than 9 years, while plant ages at the other sites were more evenly distributed. Compared to WOS, plants in CC were younger at a given height and more likely to flower when younger. Across all life stages, populations at CC contained 68 x fewer plants than WOS. The age structures of WOS and LBR were similar. Compared to published age estimates for other Trillium species, our results suggest that T. catesbaei is relatively short lived within the genus.

Conclusions: Chronic herbivory had pronounced effects on the population structure of a perennial herb. Other long-lived herbaceous species may exhibit similar truncated age structures and flowering by younger and smaller plants. Habitat generalist species within a genus, such as T. catesbaei, that are able to reproduce more quickly may persist longer under chronic herbivory. However, chronic herbivory has likely caused the loss of herbaceous species from CC and may eventually cause the local extirpation of T. catesbaei populations.

Keywords: Age structure; Deer browse; Demography; Herbaceous vegetation; Population density; Primary forest; Southern Appalachians; White-tailed deer.

Nomenclature: Kartesz (1999).

Abbreviation: CC = Cades Cove; GSMNP = Great Smoky Mountains National Park; LBR = Leadbetter Ridge; WOS = Whiteoak Sink.

Introduction

White-tailed deer (Odocoileus virginianus) are highly adaptable and have flourished in the landscape mosaic of agricultural fields and woodlots that dominates much of the eastern United States and southeastern Canada (Anderson 1997; Rooney 2001; Rooney & Waller 2003). Intensive browsing by over-abundant deer populations has altered the structure, composition, and function of forest ecosystems by causing regeneration failure of woody plants (Horsley & Marquis 1983; Alverson & Wakler 1997; Potvin et al. 2003) and shifting under-story species dominance (Cadenasso & Pickett 2000; Rooney et al. 2004; Taverna et al. 2005; Webster et al. 2005). In addition, intensive browsing by deer threatens the long-term survival of many rare plant species (Miller et al. 1992; Jolls 2003; Furedi & McGraw 2004) and impacts songbird populations by reducing vertical habitat complexity (deCalesta 1994). Because of these spatially widespread and temporally lasting effects on numerous ecological functions, white-tailed deer have been described as a keystone species in forests of eastern North America (Waller & Alverson 1997).

The effects of white-tailed deer on herbaceous plant populations have received considerable study. Research has documented the influence of deer on plant population density (Webster et al. 2005), demography (Augustine et al. 1998; Rooney & Gross 2003), plant height and flowering (Anderson 1994; Augustine & Frelich 1998; Webster & Parker 2000), and seed dispersal (Furedi & McGraw 2004; Myers et al. 2004). Other studies have used height (Anderson 1994; Balgooyen & Waller 1995; Webster & Parker 2000; Webster et al. 2001) and percent reproductive individuals (Hetcher et al. 2001; Augustine & Frelich 1998; Augustine & deCalesta 2003) within populations as indicators of browsing intensity and deer abundance.

The effects of deer herbivory on the age structure of perennial herbs have not been studied. This lack of study largely results from the inherent difficulty of aging herbaceous perennials. However, the plant ages of some genera, including Panax (Anderson et al. 1993) and Trillium (Hanzawa & Kalisz 1993), can be readily determined by counting stem scars on the rhizome. The age structure of Trillium species has probably been studied more than that of any other perennial herb. Studies have shown that disturbance history, forest fragmentation, and forest management influence the age structure of Trillium species (Jules 1998; Jules & Rathcke 1999; Kahmen & Jules 2005; Vellend 2005). Further, Trillium species are a highly preferred food source for white-tailed deer and among the first species to show effects of herbivory (Anderson 1994, Knight 2004, Webster et al. 2005).

Several studies of herbivory effects on Trillium demographic characteristics other than age have shown that deer preferentially browse large flowering plants, leading to stunting and reduced reproductive output (Anderson 1994; Rooney & Gross 2003; Knight 2004). Short-term clipping experiments have shown that while defoliation of flowering plants causes them to regress to a non-flowering form, mortality rates are negligible (Rooney & Waller 2001; Knight 2004). However, the long-term effects of chronic herbivory on plant mortality are unknown. Mortality rates of smaller/younger plants are typically higher than mature plants even without herbivory (Rooney & Gross 2003). In addition, herbivory also reduces the number of reproductive plants in a population, resulting in reduced fecundity and recruitment of new individuals (Rooney & Gross 2003). Consequently, Trillium populations in areas with histories of chronic deer browsing often lack both large and small plants (Augustine & deCalesta 2003). The question remains whether Trillium populations in these areas contain a full range of individual plant ages or a truncated age structure lacking younger and older individuals. In the season following herbivory, individual Trillium plants are often smaller in stature and frequently regress to a non-flowering stage. If this continues to occur under-long-term herbivory, the population should contain plants that are very old, but small in stature and non-reproductive. However, if chronic herbivory causes heavy mortality of these regressed individuals, the population will contain few if any older individuals. In either scenario, a lack of younger plants is likely since deer herbivory negatively affects recruitment.