- The semi-arid sagebrush steppe ecosystem is one of the largest biomes in
North America. The steppe provides critical habitat and forage for wildlife and is
economically important to recreation and livestock industries. However, the ecosystem
is threatened primarily due to several negative effects associated with expansion of the
exotic annual grass Bromus tectorum (cheatgrass). Because of these changes
rehabilitation of the habitat is extremely difficult and energy intensive. Restoration of
one of the foundation species, Artemisia tridentata (big sagebrush), is a function of
seed availability, seed germination, seedling establishment and mature plant survival.
Many studies have addressed various aspects of A. tridentata seed germination and
mature plant ecophysiology, but gaps in our knowledge include seed bank dynamics
and biotic interactions that may hinder seedling survival.
The prevalence of large fires in B. tectorum-invaded habitat limits availability
of dispersed seeds in the interior of burned areas, leaving pre-existing viable seeds in
the soil seed bank as the primary natural seed source. I investigated seed longevity of
two A. tridentata subspecies over a 2-year period by retrieving seed bags that were
placed at varying depths and sampling soil and litter fractions of the existing seed
bank across six locations in the Great Basin. Artemisia tridentata ssp. wyomingensis
and A. t. ssp. vaseyana exhibited patterns of a steadily decreasing abundance of viable
seeds on the surface and beneath litter, with 100% loss of surface seed viability in 24
months. However, 29-36% (A. t. ssp. wyomingensis) and 30-40% (A. t. ssp. vaseyana)
of buried seeds remained viable throughout the duration of the experiment. Abundance
of naturally occurring seeds varied considerably among locations and between years
across the Great Basin for both subspecies. Loss of viable seeds from the existing seed
bank between post-dispersal and pre-dispersal the following season was higher in
magnitude for A. t. ssp. wyomingensis soil fractions compared to litter fractions, and
higher overall than for A. t. ssp. vaseyana.
Access to resources is critical during early spring when resources are plentiful,
and this access affects the probability of survival through the summer drought period.
Artemisia tridentata resource acquisition may be adversely affected during this time
via root interaction mechanisms that are not mediated through resources (interference
competition). I examined whether and how root interactions affected growth of A. t.
ssp. wyomingensis by forcing its roots to interact with roots of conspecifics, B.
tectorum, Elymus wawawaiensis (Snake River wheatgrass), and Agropyron xhybrid
(cultivar 'Hycrest'). Activated carbon was used to counteract any potentially negative
effects of root exudates. Artemisia tridentata above- and belowground biomass was
not affected when grown with E. wawawaiensis or A. xhybrid compared to control
seedlings, but root growth rate and branching density decreased when grown with B.
tectorum (root growth: p < 0.01; branching density: p = 0.07). These effects did not
occur in potting media amended with activated carbon but may have been the result of
unintentional fertilization. Roots of A. tridentata seedlings changed direction or
stopped growing altogether more often when grown into roots of conspecific seedlings
than when contacting roots of other species (p = 0.08). The odds of this occurring
decreased when seedlings were grown in activated carbon-amended potting medium.
These results suggest that A. tridentata may have a chemical signaling mechanism to
avoid roots of conspecifics.
I also assessed whether root and shoot competition (resource-mediated or
exploitative competition) of the three grass species affected A. t. ssp. wyomingensis
seedling growth and survival, and whether root and shoot competition interacted to
affect growth. Size-asymmetric competition takes place when a resource is preempted
by a larger individual over a smaller individual, and the larger individual receives a
disproportionately larger share of the resource for its relative size. Following wildfires,
B. tectorum cover can increase significantly more than that of other species,
potentially promoting asymmetric aboveground competition between it and A.
tridentata seedlings. Root and shoot competition from B. tectorum lowered A.
tridentata biomass dramatically compared to that of control seedlings, with shoot
competition alone decreasing growth by over 80%. Only full competition from E.
wawawaiensis tended to decrease A. tridentata growth compared to control seedlings,
while A. xhybrid had no significant effect at all seedling growth. Bromus tectorum had
an average of 92% cover and may explain why shoot competition from this species
had such a substantial effect, whereas cover of E. wawawaiensis and A. xhybrid was
71% and 43%, respectively. Root and shoot competition did not interact for any of the
grasses, indicating that there was no mechanism for positive or negative feedbacks
between one form of competition and the other. Competition from shoots is likely not
severe enough for asymmetric light competition to occur.
The first study provides land managers with a bet-hedging application while
the others offer insight into why the seedling life history stage, already vulnerable, has
become more so with B. tectorum invasion. Reseeding techniques promoting burial of
some A. tridentata seeds in the soil seed bank may increase restoration success by
hedging against the potential for failure of establishment in the initial year of seeding.
Selective thinning or removal of potential competitors may be required to benefit
resource status of A. tridentata seedlings before the summer drought period
commences. This is especially important in areas that are dominated by B. tectorum as
competition from the annual reduces A. tridentata root growth substantially and may
impact its ability to take up soil resources.