Aspen age structure in the northern Yellowstone ecosystem: USA

Abstract

Age-structure analysis of aspen (Populus tremuloides) was conducted on Rocky Mountain elk (Cervus elaphus) winter range in the northern Yellowstone area by collecting increment cores from aspen trees in Yellowstone National Park, the Gallatin National Forest, and the Sunlight/Crandall area of the Shoshone National Forest. Our goal was to compare aspen age structure for elk winter range in the park with age structures developed for elk winter range in the national forests. We collected increment cores from aspen in three diameter size classes and three aspen habitat types (xeric, mesic, and scree). A special effort was made to collect increment cores from the relatively rare scree habitat type, since scree forms a “natural exclosure” where browsing pressure on aspen is reduced. The age structure of aspen in Yellowstone was significantly different from the age structures of aspen in either of the national forest areas (P<0.001). The Gallatin and Sunlight/Crandall age structures were not significantly different (P=0.288). Only 6% of aspen stands in Yellowstone contained stems that originated from 1920 to 1989, while 87 and 84% of the stands in the Gallatin and Sunlight/Crandall areas, respectively, contained stems from that period. Within Yellowstone, the age structure of aspen in the scree habitat type differed significantly from the mesic and xeric sites that were available for browsing (P<0.001). Aspen stems originating after 1920 dominated the scree stands, while trees originating between 1870 and 1920 dominated the non-scree stands. Aspen stands have successfully recruited new stems into their overstories in all habitat types from 1880 to 1989 in elk winter range on national forest areas surrounding the park. Within the park, aspen stands recruited new overstory stems between 1860 and 1929 in all habitat types. Since 1930, Yellowstone aspen have recruited overstory stems mostly in scree habitat type stands and other areas of reduced browsing pressure. We concluded that changes in ungulate browsing patterns due to differences in predation risk best explain the spatial and temporal pattern observed.

Introduction

Since the 1920s, researchers on the northern winter range of Rocky Mountain elk (Cervus elaphus) in Yellowstone National Park have documented the failure of existing aspen (Populus tremuloides) clones to regenerate replacement overstory stems (Warren, 1926, Rush, 1932, Grimm, 1939, Barmore, 1965, Kay, 1990, Romme et al., 1995, Houston, 1982, Meagher and Houston, 1998). The decline of overstory aspen is of concern because it is a unique and important species in the park and throughout the greater Yellowstone ecosystem, which includes portions of the states of Idaho, Montana, and Wyoming in the western USA. Aspen is one of the few upland deciduous tree species present in the ecosystem, and it ranks among the highest cover types for aboveground net primary productivity in the area (Hansen et al., 2000). Aspen forests are important for biodiversity; they support a greater variety of plant associations, as well as greater bird species richness and total abundance, than do the surrounding conifer forests (Winternitz, 1980, Turchi et al., 1995, Dieni and Anderson, 1997).

Aspen reproduces principally by vegetative means, where ramets develop from clones thought to be thousands of years old (Barnes, 1966, Mitton and Grant, 1996). Successful aspen reproduction from seed is infrequent, so the maintenance of these existing, self-regenerating clones is critical to the continued presence of aspen in Yellowstone National Park and in other western landscapes (Barnes, 1966, Jelinski and Cheliak, 1992, Mitton and Grant, 1996).

The northern range is located in the valleys of the Yellowstone, Lamar, and Gardiner Rivers. The boundary designation for Yellowstone National Park splits the northern elk winter range in two, with a portion of it inside the park and a portion of it outside, in what is now a mixture of private land and the Gallatin National Forest. On the northern range in Yellowstone National Park, there has been much debate over the possible reasons why aspen has failed to regenerate its overstory throughout most of the 20th century. Even so, researchers agree that there has been little aspen overstory recruitment in Yellowstone National Park since 1920, and that the areal coverage of overstory aspen has declined. Houston (1982) estimated a decline of 40–60% in the park’s aspen canopy coverage during the 20th century. Kay and Wagner (1996) estimated the loss at 95% since 1872, when Yellowstone National Park was established.

From the early- to mid-1900s, scientists attributed the decline of aspen to overbrowsing by ungulates, especially elk (Skinner, 1928, Rush, 1932, Grimm, 1939, Barmore, 1965). Since the 1970s, however, several alternative hypotheses have been proposed. These include the lengthening of the interval between fires (Houston, 1973, Singer et al., 1998), a trend toward a warmer, drier climatic regime (Meagher and Houston, 1998, Singer et al., 1998), the removal of native American environmental influences, including the deliberate setting of fires and overexploitation of elk populations (Kay, 1994, Kay, 1998), and the alteration of trophic cascades relationships among predators, elk, and aspen (Ripple and Larsen, 2000, Larsen and Ripple, 2001, Ripple et al., 2001). Some authors have proposed that an interaction of several factors may best explain the failure of aspen to regenerate overstory stems during the 20th century (Romme et al., 1995, Meagher and Houston, 1998, Yellowstone National Park, 1997).

In a comprehensive study of aspen on Yellowstone National Park’s northern range, Ripple and Larsen (2000) concluded that the current aspen overstory originated primarily between 1860 and 1930, with essentially no overstory regeneration since then on sites available to ungulate browsing. Romme et al. (1995) developed the only other published aspen age structure for the northern range in Yellowstone National Park, sampling only canopy-dominant trees and concluding that the current cohort of large aspen stems originated mostly in the 1870s and 1880s. Warren (1926) collected 31 aspen increment cores from a restricted geographic area (near the park’s Camp Roosevelt) in 1921–1922, dating their origin to the 1870s and 1880s. However, his objective in collecting these cores was to establish an age-diameter growth relationship, not to provide a comprehensive age-structure analysis of aspen on the northern range. Increment cores have also been used to develop aspen stand age structures in the southern portion of the greater Yellowstone ecosystem (Krebill, 1972, Gruell and Loope, 1974, Hessl, 2000).

Most of the aspen research on the northern range has focused on conditions within Yellowstone; the status of aspen on the northern range in the Gallatin National Forest has not been as intensively studied. Kay (1990) conducted a limited aspen age-structure comparison between Yellowstone National Park and the Eagle Creek area of the Gallatin, and found younger aspen stems in the national forest than in the park. St. John (1995) conducted a study of ungulate impacts on aspen on the Gallatin’s portion of the northern range, concluding that current levels of ungulate (cattle and elk) use have resulted in the deterioration of aspen clones.

East of Yellowstone, aspen occur in the Sunlight/Crandall elk wintering area of the Shoshone National Forest. Although there are no published studies dedicated to the status of aspen in the Sunlight/Crandall area, Hyde and Beetle (1964) noted that aspen ramets in this area were heavily browsed and overstory stems had been high-lined by elk during the early 1960s. Internal Forest Service documents also indicate that several Sunlight/Crandall area aspen stands had been cut or burned in 1980–1981, in an attempt to stimulate ramet production in decadent aspen stands (King, personal communication, 1999).

Yellowstone National Park was established in 1872 as the world’s first national park. From 1872 to 1886, great reductions in wildlife populations occurred throughout the Yellowstone area during the “market-hunting era”, when large animals were shots for their hides and their carcasses poisoned to eliminate predators such as the gray wolf (Canis lupus) (Schullery and Whittlesey, 1992). In terms of its effect on aspen, the market-hunting era is thought to have reduced browsing pressure enough to allow a large cohort of aspen overstory stems to be established on the elk’s northern range (Romme et al., 1995, Meagher and Houston, 1998). Elk and other wildlife populations gradually increased, however, as the market-hunting era closed. Yellowstone park management and land-use practices increasingly diverged from non-park areas over time. Hunting was banned within the park in 1883, but remained legal outside park boundaries, including a late-season hunt conducted during the month of January (Haines, 1977). Domestic livestock grazing practices also diverged inside and outside of park boundaries. Limited stock grazing was allowed within the borders of Yellowstone National Park during the early years after establishment, but discontinued over time (Haines, 1977). Outside park boundaries, stock grazing became the dominant use for rangelands (Rush, 1932). In the late 1800s, the federal government began to set aside certain areas outside Yellowstone National Park as “forest reserves”, which were also available for grazing. Shoshone National Forest was established east of the park in 1891, becoming the first national forest in the United States. To the north of Yellowstone Park, the Gallatin National Forest was established in 1899. By 1926, as the need for additional elk winter range was recognized, remaining federal lands on the northern range in the Gallatin National Forest were withdrawn from further human settlement and cattle grazing was reduced (Rush, 1932). During the second-half of the 20th century, additional lands in the Gallatin National Forest have been removed from cattle grazing and are now managed as elk winter range (Lemke et al., 1998).

The goal of this study was to use aspen increment cores to develop and compare aspen age structures for elk winter ranges inside and outside of Yellowstone National Park. The specific objectives of this study were to determine whether there were significant differences in aspen age structure between the portion of the northern range that lies in Yellowstone National Park and that of the Gallatin National Forest, and between the park’s northern range and the elk winter range in the Sunlight/Crandall areas of the Shoshone National Forest. We also sought to interpret any differences in aspen age structures in terms of ecological and/or anthropogenic processes.