Volume 354, 15 October 2015, Pages 254–260
Highlights
- •
- In northern Yellowstone, alder was suppressed by elk browsing.
- •
- Wolf reintroduction in 1995–1996 could cause a trophic cascade benefiting alder.
- •
- Alder recruitment occurred along six streams soon after wolf reintroduction.
- •
- A trophic cascade may have reversed the long-term trend of alder suppression.
Abstract
We explored possible interactions among gray wolves (Canis lupus), Rocky Mountain elk (Cervus elaphus), and thinleaf alder (Alnus incana spp. tenuifoli)
in northern Yellowstone National Park. We developed an alder age
structure based on annual growth rings for plants growing along six
streams in areas accessible to ungulates on the northern range. Alder
stems (n = 412) along the six streams originated only after
wolf reintroduction. By 2013, 80% of the sampled alders along these
streams were taller than 2 m, in contrast with a historical pattern of
height suppression by ungulate herbivory. This pattern of alder
recruitment is consistent with a trophic cascade whereby new alder
growth occurred across all study streams within several years after wolf
reintroduction. Although declines in elk density since wolf
reintroduction likely contributed to the release of alder from
herbivory, the immediate onset of new alder recruitment following wolf
reintroduction indicates that behavioral responses to predation may also
have been an important component in the resulting trophic cascade.
These results suggest that predator conservation could play a role in
the management and ecological restoration of riparian areas.
Keywords
- Wolves;
- Alder trees;
- Elk;
- Yellowstone;
- Trophic cascades;
- Riparian
1. Introduction
The
removal of large carnivores from much of the world has had diverse
ecological effects, often revealed through unexpected and complex
interactions (Terborgh and Estes, 2010 and Ripple et al., 2014b).
One example of predator effects occurs in trophic cascades, where the
effects of predators on prey are translated downward and across food
webs (Estes et al., 2011).
Yellowstone National Park (YNP) has been the focus of recent research
on trophic cascades involving the extirpation and repatriation of gray
wolves (Canis lupus) and represents a large-scale natural
experiment that provides a unique opportunity to examine the interplay
between predators, prey, and plants.
After
wolves were extirpated from YNP in the mid-1920s, park biologists
became concerned about the effects of increased Rocky Mountain elk (Cervus elaphus) browsing on vegetation in the northern and Gallatin ungulate winter ranges ( Skinner, 1928, Rush, 1932, Wright et al., 1933, YNP, 1958, Lovaas, 1970 and Ripple and Beschta, 2006).
Analyses of the annual growth rings of deciduous tree species revealed
that recruitment (i.e., growth of seedlings/sprouts into tall saplings
or trees) occurred regularly in both of these YNP winter ranges when
wolves were present, but declined and became rare after wolf extirpation
( Ripple and Larsen, 2000, Beschta, 2005, Halofsky and Ripple, 2008a and Kauffman et al., 2010).
These tree-ring study results are consistent with reports of YNP
biologists (as cited above) about the decline of woody browse species in
the early and middle 20th century. As a result of park service
biologists’ concerns, a program of elk reductions was initiated in the
1930s and continued through 1968. By the late 1960s, park service
culling had reduced the northern range elk population to less than 5000
individuals ( Fig. 1a), but with no resulting major recovery in recruitment of woody plants ( Houston, 1982, Kay, 1990, Meagher and Houston, 1998, NRC, 2002 and Barmore, 2003). After the elk culling program ended in 1968, elk numbers increased dramatically during the 1970s ( Fig. 1a). In the 1980s and 1990s, elk numbers fluctuated widely due to winter starvation events ( Garrott et al., 2003 and Eberhardt et al., 2007).
During this period of large population size (>19,000 elk in some
years on the northern range), elk were limited by food resources and
consumed relatively unpalatable species such as conifers ( Kay, 1990, Meagher and Houston, 1998 and NRC, 2002).
- Fig. 1.Number of counted (a) elk and (b) wolves on the northern range. Not shown for elk are poor count years of 1977, 1989, 1991, and 2006. The field sites for this study were all located in the eastern and central sectors of the northern range (as defined by Painter et al. (2015)) and the elk counts for these two sectors were summed and shown with circles as a second series in (a). The eastern and central sectors include the portions of the northern range east of and including the Blacktail Creek drainage. Elk data for the eastern and central sectors were not available prior to 1987.
Wolves were reintroduced into YNP during 1995–96 following approximately seven decades of absence (Fig. 1b). Thirty-one wolves were moved from Canada to the northern range of YNP in January 1995 (n = 14) and January 1996 (n = 17). By 1996, five wolf pack territories covered nearly all of the northern range within the park ( Fig. S1 in supplement, Fig. 1 in Phillips and Smith (1997)). During the fall of 1996, each of three different wolf packs on the northern range killed, on average, 1 elk every 2–3 days ( Phillips and Smith, 1997).
Following reintroduction, wolf numbers on the northern range increased
until 2003 and thereafter declined, while the elk population decreased
steadily during this period ( Fig. 1a
and b). The initial decline in elk numbers in the mid-to-late 1990s was
due in part to starvation caused by a degraded winter range and a
severe winter in 1996–97; other factors included predation by wolves,
bears, and continued hunting by humans of elk that left the park ( Eberhardt et al., 2007 and White and Garrott, 2013).
Research
on the effects of predators, ungulates, and other factors on the
establishment and growth of deciduous trees in northern Yellowstone has
focused on aspen (Populus tremuloides) and cottonwood trees (Populus spp.) (reviewed by Ripple and Beschta (2012).
Recruitment of these tree species has increased since the
reintroduction of wolves, although the magnitude of the recovery is
spatially variable ( Beschta and Ripple, 2014 and Painter et al., 2015).
Over the same period, deciduous shrubs in some areas of northern
Yellowstone have increased in height, biomass, or cover including
willows (Salix spp.) ( Beyer et al., 2007, Tercek et al., 2010, Baril et al., 2011 and Marshall et al., 2014) and various berry-producing shrubs ( Beschta and Ripple, 2012 and Ripple et al., 2014a).
Herein we report on the first extensive field study of thinleaf alder (Alnus incana spp. tenuifoli)
in Yellowstone’s northern range. Thinleaf alder, a small tree or tall
shrub, commonly occurs in riparian areas throughout western North
America and can grow up to ∼12 m tall ( Fryer, 2011).
Through its nitrogen fixing properties, it enriches soil and
facilitates the establishment of other native plants. Thinleaf alder
spreads both vegetatively and from small winged seeds, although
vegetative reproduction is thought to be more common. It sprouts
primarily from root crowns, but can also sprout from roots. Dense alder
thickets can provide cover for fish, thermally modify microclimates and
stream temperatures via shading, and protect streams from bank erosion.
Songbirds eat thinleaf alder seeds, squirrels consume catkins, beaver
use stems to build lodges and dams, and various small and large mammals
use alder as cover ( Fryer, 2011).
Thinleaf alder has low palatability as ungulate forage, but it is
consumed by ungulates especially when other forage is limited ( Gaffney, 1941, Nelson and Leege, 1982 and Case and Kauffman, 1997). Northern Yellowstone alders, as well as conifers, were affected by browsing in the 1950s ( Jonas, 1955)
indicating that elk were using low-quality forage even with densities
lower than those of the 1980s–90s. Also, before wolf reintroduction Keigley (1997) observed heavy herbivory effects on various woody browse species on the northern range, including alder.
Because
little is known about Yellowstone’s thinleaf alder, our main objective
was to analyze temporal patterns of thinleaf alder stem establishment on
the northern range of Yellowstone National Park. In light of previous
research showing changes in cottonwood and aspen recruitment following
wolf reintroduction, we hypothesized that thinleaf alder exposed to
ungulate browsing would also increase in recruitment over this same time
period.
2. Methods
This study took place on the northern ungulate winter range, comprising more than 1500 km2
of mountainous terrain and open valleys, approximately two-thirds of
which occurs within the northeastern portion of YNP in Wyoming (NRC, 2002). Much of the winter range is shrub steppe, with patches of intermixed lodgepole pine (Pinus contorta), Douglas fir (Pseudotsuga menziesii), Engelmann spruce (Picea engelmanni),
and aspen. Thinleaf alder, and various species of willow, cottonwood,
and other woody browse plants occur within riparian zones. See Houston, 1982 and NRC, 2002 for a more detailed description of the northern range study area.
We
determined the age structure of alder stems (frequency distribution of
number of stems by year of establishment) growing along small streams,
and then compared the number of alder stems established before wolf
reintroduction versus after wolf reintroduction. This research design
provided two predator treatments: (1) wolves absent (pre-1995) followed
by (2) wolves present (post-1995).
We
located small perennial streams (4th–5th order) on the northern range
within the park that intersected the North Entrance road, the Grand
Loop, and the Northeast Entrance road. We excluded streams where
riparian areas had burned in the large fires of 1988 (i.e., Lava,
Lupine, Elk, Lost, and Tower Creeks). For the remaining streams (Glen,
Blacktail, Oxbow, Geode, Crystal, Rose, Indian, and Pebble Creeks), we
searched for alder 1000 m upstream and downstream of the road. We found
alder along six of these streams: Blacktail, Oxbow, Geode, Crystal,
Rose, and Pebble Creek representing a west-east gradient across the
northern range. All six streams intersect the main west-east road and
were within the central and eastern portions of the northern range
within the park (Fig. 2).
- Fig. 2.Northern ungulate winter range (northern range) study area showing the locations of six riparian study sites.
