CALIFORNIA STATE UNIVERSITY, NORTHRIDGE
THE EFFECTS OF INVASIVE PLANTS ON THE ENDANGERED SUNFLOWER,
PENTACHAETA LYONII GRAY
A thesis submitted in partial fulfillment of the requirements for the degree of
Master of Science in Biology
By
Jolene R. Pucci
JANUARY 2007
The Thesis of Jolene R. Pucci is approved:
___________________________________
Christy Brigham, Ph.D.
________________
Date
___________________________________
Paul Wilson, Ph.D.
________________
Date
___________________________________
Paula Schiffman, Ph.D., Chair
________________
Date
CALIFORNIA STATE UNIVERSITY, NORTHRIDGE
ii
ACKNOWLEDGEMENTS
I thank my advisor, Paula Schiffman, for her guidance and support during this
thesis project. Her input on experimental design, procedural logistics, and the writing
process were invaluable to me. I also thank Paul Wilson, my committee member, for his
generous assistance with data analysis and for broadening my view. Thank you to
Christy Brigham, my committee member, for designing a large part of this project, and
for her thoughtful supervision of my work at the National Park Service. Thanks to John
Tiszler for his input and guidance at the National Park Service. Thanks to Tarja Sagar for
inspiring me to be a better botanist. My friends and lab mates were indispensable with
their assistance in the field: Joey Algiers, Kammy Algiers, Taya Cummins, Matthew
Danielczyk, Andrew Ellis, Michael Flores, Camille Franklin, Ray Hernandez, Peter
Holmquist, Alex Li, Robert Pucci, and Rosalyn Son. I am so thankful for the friendship
and moral support of Joanne Moriarty, Ray Hernandez, Kammy Algiers, and Michael
Flores throughout the project. Thank you to Brian Houck and Brenda Kanno for helping
to raise my potted plants. I am grateful to Bobby Espinoza for his encouragement to
enter the Master’s program, to Jennifer Matos for doing field biology with style, and to
Fritz Hertel for sharing his insights on the research process. Special thanks goes to Hedy
Carpenter for her inspiration and encouragement. Funding for this research was provided
by the National Park Service, Santa Monica Mountains National Recreation Area, and
Western National Parks Association.
iii
TABLE OF CONTENTS
Signature………………………………………………………………………….ii
Acknowledgements……………………………………………………….............iii
List of Figures…………………………………………………………………….v
List of Tables……………………………………………………………………..vi
Abstract…………………………………………………………………………...vii
Introduction………………………………………………………………………...1
Methods…………………………………………………………………………….7
Results……………………………………………………………………………..16
Discussion…………………………………………………………………………21
Literature Cited……………………………………………………………………48
iv
LIST OF FIGURES
Figure 1……………………………………………………………………..34
Figure 2……………………………………………………………………..35
Figure 3……………………………………………………………………..36
Figure 4……………………………………………………………………..37
Figure 5……………………………………………………………………..38
Figure 6……………………………………………………………………..39
Figure 7……………………………………………………………………..40
Figure 8……………………………………………………………………..41
v
LIST OF TABLES
Table 1…………………………………………………………………………42
Table 2…………………………………………………………………………43
Table 3…………………………………………………………………………44
Table 4…………………………………………………………………………45
Table 5…………………………………………………………………………46
Table 6…………………………………………………………………………47
vi
ABSTRACT
THE EFFECTS OF INVASIVE PLANTS ON AN ENDANGERED SUNFLOWER,
PENTACHAETA LYONII GRAY
by
Jolene R. Pucci
Master of Science in Biology
Invasive plants threaten native biodiversity and ecosystem function. Non-native
plants can out-compete native plants for resources, reducing population sizes. For rare
species, this can increase the chances of extinction. Pentachaeta lyonii is an endangered,
endemic sunflower, currently ranging entirely within the urbanized Santa Monica
Mountains and Simi Hills. Its former range and number of populations have been
reduced in recent decades due to pressures from urbanization, and the remaining
populations are in decline. This study examined the effects of competition from invasive
plants as a possible cause of declines by evaluating both effects from competition and
effects from community alteration. Three invasive plant groups (annual grasses, Erodium
spp., and Centaurea melitensis) were studied in (1) direct competition experiments in the
field and in pots, (2) observational studies comparing sites where P. lyonii is extant and
extirpated, and (3) manipulative community-level experiments. In the field and pot
competition experiments, all three invasive groups competitively reduced the
reproductive capacity of P. lyonii, and had differing effects on P. lyonii height.
Observational studies showed that the presence of annual grasses and its associated litter
were correlated with extirpation, and retention of bare ground was correlated with P.
lyonii persistence. Restoring P. lyonii habitat to pre-invasion conditions by removing
non-native plants, scraping the soil surface, and adding cryptobiotic crust increased
native species richness and reduced the cover of annual grasses. Seeding P. lyonii
increased its density in existing sites, and was successful in establishing plants in new
sites. Removal of invasive plants and their associated litter in P. lyonii habitat, and
seeding existing and new populations are recommended for restoration and recovery of
the species.
vii
INTRODUCTION
The global spread of invasive species threatens native ecosystem processes,
interactions, and biodiversity (Elton 1958, Gordon 1998, Mack and D'Antonio 1998,
Mack et al. 2000, Wilcove et al. 1998). This results in major environmental damages and
economic losses (Pimentel et al. 2005). The probability and magnitude of an invasion in
an ecosystem depends on biotic and abiotic features, as well as attributes of the invaders
(Blair and Wolfe 2004, Rejmánek 1996, Sans et al. 2004). These features determine an
ecosystem’s invasibility (Dukes 2001, Dukes 2002, Sanz-Elorza et al. 2006). Frequently
cited factors contributing to invasibility are disturbance and the disruption of natural
disturbance regimes (Hobbs and Huenneke 1992). Areas that are heavily populated by
humans are likely to possess both of these features.
California is home to more than 36 million people. Its mediterranean climate is
not only attractive to humans, but also led to the development of a diverse assemblage of
native plants. The high degree of endemism, along with threats posed by human
population pressure, have placed the California Floristic Province on the list of
ecosystems that are considered global biodiversity hotspots (Myers et al. 2000). There
are approximately 6,300 native plants in California (Hobbs and Mooney 1998), and more
than 1,400 of them are endemic (Hickman 1993). Many, by virtue of their limited
ranges, are rare. European occupation initiated the introduction of numerous species of
non-native plants to the region, and today there are more than 1,000 species of introduced
plants that have naturalized in California (Hickman 1993, Hobbs and Mooney 1998,
Rejmánek and Randall 1994). Some of these non-natives have become invasive,
threatening local native plant populations with effects from altered ecosystem processes
1
to altered disturbance regimes and increased competition. These effects can be especially
detrimental to species with small population sizes and/or limited ranges. These two
factors render species more prone to extinction due to stochastic events and can lead to
low genetic variation due to processes such as inbreeding depression that also increase
the probability of extinction. These phenomena might be exacerbated in species that
experience isolation due to habitat fragmentation (Brigham 2003).
The Los Angeles area, including the Santa Monica Mountains and Simi Hills, is a
highly fragmented, urbanized patchwork of developed land and open space. Much of the
open space is privately owned and in danger of being developed, but there are substantial
units of open public land that are managed by various government agencies and nonprofit groups including the National Park Service, California Department of Parks and
Recreation, the Santa Monica Mountains Conservancy, and Mountains Restoration Trust.
Many native communities of plants, some containing rare and endemic species, persist
despite the pressures of urbanization. Chaparral and coastal sage scrub can still be found
in the mountains in significant swaths, but the prairie that once dominated the area’s
flatlands was historically degraded by livestock grazing and agricultural cultivation, and
was subsequently converted to urban sprawl (Schiffman 2005). The remaining patches of
grasslands are heavily invaded by non-native species. The native dominants are annual
wildflowers (Schiffman 2000) associated with perennial bunchgrasses including Nassella
and Poa species. Non-native annual grasses, including Bromus spp., Avena spp., and
Vulpia spp. are naturalized in every southern California grassland site (Heady 1988).
These grasses have direct and indirect competitive effects on native species (Brooks
2000, Dyer and Rice 1997, Dyer and Rice 1999, Suding and Goldberg 1999). Some non-
2
native forbs, including Centaurea spp. and Erodium spp. are also dominant invaders that
can suppress the diversity and abundance of native species (Schutzenhofer and Valone
2006).
Pentachaeta lyonii Gray is a state and federally listed endangered annual in the
family Asteraceae. Its size is variable depending on environmental conditions, and can
range in height from 3 – 48 cm, producing 1 – 36 inflorescences per plant (Hickman
1993, NPS unpublished data). Each inflorescence produces 20 – 40 seeds with 8 – 12
deciduous pappus bristles, indicating dispersal limitations (Fotheringham and Keeley
1998). Flowers are self-incompatible (Fotheringham and Keeley 1998), and pollinated
by generalist insects (Braker and Verhoeven 1998). Extant populations are found on
volcanic clay soils, but garden studies have indicated that it is not edaphically restricted
(NPS unpublished data). It occurs in grasslands on open ridges or hillslopes or in
openings in chaparral and coastal sage scrub. These areas are heavily invaded by nonnative annual species, but native associates include small herbaceous annuals and
cryptobiotic soil crust. Following the extirpation of sites in the southern part of its range,
P. lyonii is now restricted to the Santa Monica Mountains and Simi Hills, living entirely
within the heavily populated and developed area of northern Los Angeles and southern
Ventura counties. Its distribution is comprised of 21 populations on both public and
private lands. Populations have been designated subjectively based on the patchy
distribution. No genetic studies for metapopulation analysis have been done.
Historically, P. lyonii was known to have had a wider distribution in the Los Angeles
basin, Santa Catalina Island, and San Diego (Hickman 1993, Munz and Keck 1959), but
as many as 15 populations have been extirpated within recent decades, and many of the
3
remaining populations appear to be in decline (NPS unpublished data). Although nine
populations are located on public lands with varying degrees of protection, a large
proportion of occurrences are on private property, much of it with great appeal to
developers.
The U.S. Fish and Wildlife Service recovery plan for P. lyonii (1999) identifies
possible causes of the species’ decline as habitat destruction, alteration of habitat structure,
and competition from invasive non-native plants. Habitat destruction and alteration of
habitat structure are undeniable, as a number of population sites have been built upon or
converted to agriculture (NPS unpublished data). Surveys of both P. lyonii numbers and the
presence of invasive species indicated a possible relationship between invasion and declines,
but no competition studies have been done previous to the present work.
Plant species coexist in assemblages despite their close proximity and similar
resource requirements. Because plants are essentially immobile, they can experience
considerable niche overlap, competing for the same resources: light, nutrients, water, and
space. Although the competitive abilities of two coexisting species can be asymmetrical,
when competition is considered in the larger context of population or community
dynamics the outcome is not always predictable (Silander and Pacala 1990). Spatial
(Reynolds et al. 1997) and temporal (Levine and Rees 2004) variability can facilitate
coexistence, as can variation in plant characteristics such as phenology (Dukes 2002),
root placement and depth (Poot and Lambers 2003), and dispersal ability (Levine and
Rees 2002). Furthermore, disturbance can also contribute to long-term species
coexistence, with various species being favored by different kinds and timings of
disturbance (Crawley 2004). Selection in long-term co-occurring plant species can lead
4
to the evolution of increased competitive ability in response to interspecific competitive
pressures (Aarssen 1983).
When non-native invasive plants establish in a new environment, resident species
encounter a competitor that they have not evolved with. Along with the often inferior
competitive response of natives, certain non-native species have aggressive competitive
effects. The result is that these species are invasive. Invasive plant species are non-native
species that produce reproductive offspring without human intervention at numbers and
dispersal distances sufficient to spread over large areas (Richardson et al. 2000). Hypotheses
for why invasive species gain competitive advantage over natives include the enemy release
hypothesis (ERH), which suggests a plastic increase in the size of invasive individuals due to
release from enemies that are present in the native range. Another hypothesis, the evolution
of increased competitive ability hypothesis (EICA), states that in the absence of native
herbivores selection has favored genotypes that allocate resources more to growth and
reproduction and less to herbivore defense (Blossey and Nötzold 1995). The successful
biological control of weeds, along with some experimental evidence support the ERH (Keane
and Crawley 2002). The EICA hypothesis continues to be tested in common garden
experiments with mixed results (Leger and Forister 2005, Leger and Rice 2003, Vilà et al.
2003). Attempts have been made to generalize life history traits that make invasive species
competitively superior, but results have not been consistent and have varied among species
(Gurevitch et al. 2002). Dominance of invasive species in a system is not always due to
competitive superiority (Seabloom et al. 2003b), but evidence to date suggests that
competitive interactions between invasive plants and resident species in a community are
asymmetrical in favor of the invasive species (Vilà and Weiner 2004).
5
Non-native plant presence not only affects the individual performance of native
plants, but can also change ecosystem properties such as hydrology (Sala et al. 1996, Vanlill
et al. 1980), nutrient cycling (Evans et al. 2001, Sperry et al. 2006), natural disturbance
regimes (Hobbs and Huenneke 1992), and biodiversity (Lodge 1993). Ecosystem level
alterations from non-native plants occur through mechanisms such as allelopathy (Bais et al.
2003), changes in the fire regime (D’Antonio and Vitousek 1992, Hobbs and Huenneke
1992, Vilà et al. 2001), and alteration of available mycorrhizal fungi (Hawkes et al. 2006), as
well as competition (Brooks 2000).
In order to assess a range of possible impacts from some of the important invasive,
non-native plants that are present within P. lyonii’s range, this study evaluated both effects
from competition and effects of community alteration. Direct competitive effects were
examined in terms of P. lyonii success when grown with and without non-native competitors
both in the field and in pots. Additive effects from invasive plants on P. lyonii habitat were
evaluated in terms of their impacts on both P. lyonii and its associated native community.
Finally, to try to characterize differences in habitats where P. lyonii has been extirpated and
those where it persists, comparisons were made between environmental features of both
kinds of sites.
The specific goals of this study were (1) to examine the impact of competition from
non-native species on P. lyonii success in the field and in pots, (2) to determine which nonnative species have the greatest competitive effect on P. lyonii, (3) to examine impacts from
non-native species on P. lyonii in the context of its native community, and (4) to evaluate the
environmental conditions that contribute to the displacement of P. lyonii by non-native
6
plants. These goals were formulated to address the request to study competition from
invasive plants in the U.S. Fish and Wildlife Service recovery plan for P. lyonii (1999).
With an understanding of how non-native plants impact P. lyonii, managers will be better
prepared to address the recovery plan goal of achieving stable or increasing populations (U.S.
Fish & Wildlife Service 1999).
METHODS
Non-native competitors
The non-native plants investigated here were (1) annual grasses, including
Bromus madritensis, Bromus hordeaceus, and Vulpia myuros, (2) Erodium spp., namely
E. cicutarium and E. botrys, and (3) Centaurea melitensis. Although other non-native
species were present at the study sites, these three “target” groups were chosen because
of their pervasiveness in P. lyonii habitat and their potential roles as competitors
(DeFalco et al. 2003, Kimball and Schiffman 2003, Schutzenhofer and Valone 2006).
Competitor Removal Experiment
In February 2004 experimental plots were established at two sites: (1) the
Pentachaeta preserve off Tierra Rejada Road in Moorpark, California (“Tierra Rejada,”
34° 15' 54" N, 118° 51' 19" W), and (2) the Westlake Vista/Decker Canyon open space
on Triunfo Canyon Road in Westlake Village, California (“Triunfo,” 34° 07' 09" N, 118°
48' 03" W). Both properties are managed by the Mountains Recreation and Conservation
Authority.
Twenty replicate pairs of 13 x 13-cm plots - competitor removal plots and control
plots - were established for each non-native plant group. Plots with Centaurea and plots
7
with Erodium were established at both Tierra Rejada and Triunfo (10 pairs at each site),
but all 20 annual grass plot pairs were installed at Triunfo because of time constraints
resulting from earlier germination at Tierra Rejada.
The locations of plots within sites were determined by identifying areas
dominated by the target non-native (at least 50% cover) that also contained P. lyonii
seedlings. A no-competition treatment was produced by clipping at the soil surface, all
plants (potential competitors) other than those of P. lyonii in a 26 x 26 cm area centered
over each experimental plot. The treatment areas extended 13 cm outside the plots to
minimize edge effects (Berendse 1983). Clipping was done over a two-week period in
March. At the end of the growing season, P. lyonii heights were measured and
inflorescences were counted. The number of inflorescences was intended as a measure of
how much seed was to be added to the seed bank, whereas height was intended as a
measure of the plastic response to growing among competitors. Height and number of
seeds are not necessarily correlated.
This experiment was replicated in 2004-05 in different plots. The rainy season in
2003-04 was later with much less rainfall (24 cm, Los Angeles Civic Center) than the
2004-05 rainy season (95 cm, Los Angeles Civic Center), consequently the following
adjustments in the methods were made for the second field season. Due to early
germination, plots were established in November, 2004, and competitors were removed
in December. A second removal was done in February, 2005, due to vigorous regrowth
of competitors. All annual grass plots were established at Tierra Rejada instead of
Triunfo because of difficulty in locating appropriate conditions at Triunfo (at least 50%
cover of annual grass with P. lyonii present).
8
To decide whether to evaluate sites separately or pooled, the mean difference in
magnitude of competitive effects from each invasive plant group was compared between
sites using two-sample t-tests. Plots with annual grass as the competitor were not
separated by site within years, and are therefore excluded from the analysis. Dependent
variables were difference in number of inflorescences on P. lyonii plants between
competitor removal plots and control plots [log(competitor removal) – log(control)] and
difference in height of P. lyonii plants between competitor removal plots and control
plots [competitor removal – control]. The data were log transformed to increase
normalcy. Each variable was analyzed separately for each season. Because there were
no significant differences in the numbers of inflorescences or height of P. lyonii between
sites for invasive groups in either year, except for in 2005 in Centaurea plots (Table 1),
sites were pooled in further analyses.
The mean height of P. lyonii plants in each plot and the mean number of
inflorescences per P. lyonii plant in each plot were calculated and then log transformed to
increase normalcy. Paired t-tests were done to test for differences between competitor
removal and control plots. In addition, ANOVA with differences as the dependent
variable was used to compare the size of the competitive effects of annual grasses to
those of Erodium to those of Centaurea.
Pot Competition Experiment
A garden experiment in pots was done to further clarify the field study and test
specific levels of competition. In the winter of 2004-05, P. lyonii seedlings were planted
with seedlings of the invasive non-native groups in 10.16 cm diameter PVC tubes outside
in the garden at California State University, Northridge. Bromus madritensis was used to
9
represent the annual grass group, Erodium cicutarium represented the Erodium group,
and Centaurea melitensis was the Centaurea group. For each invasive group, twenty
replicates of three treatments were attempted: (1) one P. lyonii plant growing alone
(control), (2) one P. lyonii plant growing with 5 non-native plants (low density), and (3)
one P. lyonii plant growing with 20 non-native plants (high density). Low and high
density levels were intended to simulate conditions in sites that are moderately and
heavily infested with non-native plants. Seeds were germinated on filter paper and the
seedlings were transplanted into the pots over three days in December. Dead seedlings
were replaced for up to four weeks. Due to continued seedling mortality after that, final
densities were recategorized at the end of the experiment. Low density pots contained
between one and five non-native plants, and high density pots included more than five
non-native plants. The final numbers of replicates were n = 20 for low density annual
grass, n = 20 for high density annual grass, n = 22 for low density Erodium, n = 18 for
high density Erodium, n = 20 for low density Centaurea, n = 19 for high density
Centaurea, n = 20 for control.
As in the field study, the number of inflorescences and heights of focal P. lyonii
plants were measured at the end of the growing season in June. Number of
inflorescences was log transformed (+1) to increase normalcy, and ANOVAs were
conducted to compare the seven treatments. ANOVAs were followed by Tukey multiple
comparisons.
Comparison of Extant Versus Extirpated Sites
An observational study was done to compare the environmental features of the
sites with existing P. lyonii populations to the features of sites where P. lyonii historically
10
occurred but is now seemingly extirpated. Twelve extant sites and four extirpated sites
were sampled. Ten locations were haphazardly established throughout each site and
sampling points were placed within them. Sampling was done in a 20 x 20 cm plot at
each location. The environmental factors assessed were soil parent material, topography,
soil depth, soil water content, mid-day photosynthetically active radiation (PAR) levels,
percent bare ground, and percents cover of litter, grass, Erodium, Centaurea, other nonnative species, P. lyonii, and other native species. Percent cover was scored by visual
estimation within a 20 x 20 cm plot where a species was present. At the sites where P.
lyonii was extant, numbers of P. lyonii were estimated. In December 2004, soil depth,
soil volumetric water content, PAR, litter depth, and percent cover of bare ground, litter,
and plant cover were scored. In April 2005, measurements were repeated after the plant
community had fully established. Insolation values and fire history for each site were
obtained from the National Park Service (NPS unpublished data). A series of logistic
regressions was done in order to identify which variables were good predictors of the
extirpation of P. lyonii. A nonmetric multidimensional scaling ordination was done using
NTSYS 2.2 to place sites in 2-dimensional space in relation to the environmental factors
that characterized those sites. After standardizing the data by subtracting the minimum
and dividing by the range, Bray-Curtis distances were calculated, and then nonmetric
multidimensional scaling was done. The data were rotated using PC-ORD (MJM
Software Design, Gleneden, OR) to maximally correlate with whether the site was extant
or extirpated. Correlations were overlaid if r2 > 0.2. Those variables with r2 < 2 were not
shown on the graph.
Community Effects Experiment
11
Study sites
In the spring of 2003, experimental plots were established at three study sites
(“Pond,” “Lower,” and “Paramount”) in the Santa Monica Mountains National
Recreation Area on lands managed by the National Park Service. One site supported an
extant population of P. lyonii (Pond), in another site P. lyonii was presumed extirpated
(Lower), and the third site had never had P. lyonii present, but conditions suggested that
it was suitable habitat.
The “Pond” site was located in the Rocky Oaks section of the National Recreation
Area (34° 05' 57" N, 118° 48' 43" W). A small population of P. lyonii persisted, but it
may have been in decline, with only an estimated 450 individuals present in 2003, down
from about 5,500 individuals in 1982. It was heavily infested with non-native invasive
plants. In 2003 before any manipulations, there was approximately 49% cover of nonnative annual grasses and 15% of the ground was bare. The site was disturbed in the
winter of 1996-97 when a drainage ditch was dug through its center.
The “Lower” site was also located in the Rocky Oaks section of the National
Recreation Area, approximately 150 m southwest of the Pond site. By 1995, P. lyonii
was presumed extirpated from the site (NPS unpublished data). The initial non-native
annual grass cover was approximately 4% within the experimental plots in 2003. In
addition, 76% of the ground was bare and a dense stand of Phalaris aquatica (Harding
grass) was encroaching from areas immediately adjacent to the plots. There was
evidence of soil dumping on the site, possibly from the modification of an adjacent
earthen dam in the winter of 1996-97.
12
The “Paramount” site was located in the Paramount Ranch section of the National
Recreation Area (34° 07' 18" N, 118° 45' 38" W). It was considered suitable habitat for
P. lyonii based on soil type, a high pre-treatment percentage of bare ground (53%), a low
initial level of non-native grass invasion (14%), and the presence of species commonly
associated with P. lyonii (Table 2).
Experimental treatments
At each site 20 blocks of four 1 m2 experimental plots were established in the
spring of 2003. Plots with the most similar starting appearance and adjacency were
assigned to the same block. One plot per block received each of four randomly assigned
treatments (1) control, (2) invasive plants (annual grasses, Erodium spp., and Centaurea
melitensis) cut at the soil surface, (3) invasive plants cut and the soil surface scraped, and
(4) invasive plants cut, the soil surface scraped, and fragments of cryptobiotic crust
added. The treatments were intended to address the multiple impacts of invasive plants
on P. lyonii and its associated community. Cutting invasive plants eliminated their
above-ground impacts, such as reduction of available light. Cutting and scraping was
meant to negate the stabilizing effects of the fibrous root systems of annual grasses, and
to remove undecomposed litter. Crust was added to restore a potentially important
component of the native ecosystem (Belnap 2006, Belnap and Harper 1995, Belnap et al.
2001, Bowker et al. 2004, Harper and Belnap 2001). Cryptobiotic crust cover and
diversity is lost or reduced in sites invaded by non-native annual plants (Belnap et al.
2001).
Plots were surveyed in the spring of 2003 (before the treatments) by visual
estimation of percent of bare ground and covers of annual grasses, forbs, Nassella, and
13
shrubs. The cutting treatments were applied in the spring and summer of 2003. The
scraping treatment was done in fall 2003 by abrading the surface 1-2 cm of bare ground
with a trowel. In February of 2004 cryptobiotic crust fragments (approximately 5 cm in
diameter) were collected from local areas, and transferred into the plots to produce
approximately 20% cover. This experimental protocol was repeated in 2004-05, with the
exception of crust addition, since crust fragments were already established within
treatment plots.
In the spring of 2004 percents cover of annual grasses, Erodium, Centaurea,
native annuals, native perennials, native shrubs, and litter were visually estimated. The
amount of bare ground was also determined. Native species densities were determined
by counting individual plants within each plot. In spring 2005, percent cover surveys
were repeated, and individual native plant densities were estimated by subsampling
within 10 x 100 cm quadrats placed through the center of each plot. All P. lyonii
individuals in the plots were counted. Results presented are from final surveys in 2005
because the crust treatment was applied after the 2003-04 growing season was already
underway.
The data were analyzed as a split-plot involving site, blocks within sites,
treatments, and their interactions. Dependent variables were native species richness
(square root transformed to increase normalcy), native species diversity, native species
evenness, density of P. lyonii (square root transformed to increase normalcy), percent
cover of non-native annual grasses (log transformed to increase normalcy), percent cover
of bare ground, and percent cover of native annual forbs. To follow up on site ×
treatment interactions, separate ANOVAs were done for each site with Tukey multiple
14
comparisons. An additional two-way ANOVA was done to analyze site and treatment
effects on P. lyonii density using only plots that contained P. lyonii. This simplification
eliminated the possible confounding factors of initial presence/absence of P. lyonii seed
within plots and of seed addition (see below).
Tests of independence were conducted (replicated by site) to see if the treatments
affected the presence of P. lyonii (and to see if sites were heterogeneous in any such
relationship). Finally, multiple logistic regressions were done to model how the initial
percent cover (pre-treatment, 2003) of annual grass (square root transformed) and/or
amount of bare ground predicted the presence or absence of P. lyonii (after seeding and
treatments, 2005) at each of the sites.
Seed Additions
Because of the potential reduction in native species presence in the seed bank due
to long-term effects of invasive plant presence, a seed mix of species associated with P.
lyonii was added to all blocks at all sites in February 2004 (Table 3). In March 2004, 125
P. lyonii seeds were added to each plot in five randomly selected blocks at the Paramount
site only (not included in Table 3). In October 2004, another mix of associated species,
along with P. lyonii seeds were added to five randomly chosen blocks at all three sites
(Table 3). Locally collected seeds of associated species were obtained from the National
Park Service seed collection maintained at the Rancho Sierra Vista native plant nursery.
P. lyonii seeds were collected from two of the larger populations.
Effects of seeding frequency differences (fifteen blocks at each site seeded only in
February, five blocks at each site seeded in both February and October) were analyzed
15
prior to the above analysis of the main treatment effects. ANOVAs (General Linear
Models) were conducted separately for each treatment (control, cut only, cut and scrape,
cut and add crust), with seeding manipulation (seeded once, seeded twice), site, initial
cover of annual grass, initial percent bare ground, and interactions between seed
treatment and site, seed treatment and annual grass cover, and seed treatment and percent
bare ground as independent variables. Non-significant terms (P > 0.05) were dropped
one at a time to simplify the model. This procedure was carried out for six dependent
variables to seek seeding frequency effects on (1) native species richness (square root
transformed), (2) native species evenness, (3) native species diversity, (4) density of P.
lyonii (square root transformed), (5) percent cover of non-native annual grasses (log
transformed), and (6) percent cover of native annual forbs.
Separate two-sample t-tests were done grouping blocks at each site to compare
density of P. lyonii (square root transformed) in blocks (all treatments) that were seeded
once with associated species and not with P. lyonii to those that were seeded twice with
associated species and also with P. lyonii.
RESULTS
Competitor Removal Experiment
In both the 2004 and 2005 field seasons, competition from all three groups of
invasive species had significant negative impacts on P. lyonii inflorescence number
(paired t-test, P < 0.01 in all cases, Fig. 1). Comparisons of the magnitude of competitive
effects, as indicated by differences in numbers of P. lyonii inflorescences in control
(competition) versus cut (no-competition) plots, showed that Centaurea had a
16
significantly greater effect than did the grasses (Tukey HSD, P < 0.05). The greater
measured effect of Centaurea when compared to Erodium was not significant
(P=0.5688). Grasses and Erodium did not differ significantly in their competitive effects
on P. lyonii (P = 0.1426).
Effects from competition on height of P. lyonii differed among the non-native
groups and between years (Fig. 2). In three of the six cases, there were no significant
differences in height for plants growing in plots with competition from non-native plants
versus those where it was growing in plots without competition. Significant differences
were found in 2004, where plants, released from competition with Erodium spp., grew
taller than those in control plots (paired t-test, P = 0.01, Fig. 2). In 2004, P. lyonii plants
growing without competition from Centaurea were marginally taller than those in control
plots (paired t-test, P = 0.064, Fig. 2). In contrast, in 2005, plants competing with annual
grasses grew taller than those in plots with competitors removed (paired t-test, P < 0.01,
Fig. 2). Comparisons of the magnitude of impacts (differences in heights of P. lyonii
plants in plots with and without competition) showed that annual grasses had a greater
effect on P. lyonii height than Centaurea or Erodium (Tukey HSD, P < 0.05).
Pot Competition Experiment
Under the more controlled research conditions of the pot experiment, competition
from all non-native species had a negative effect on P. lyonii reproductive potential.
Pentachaeta lyonii plants produced significantly fewer inflorescences when grown in
pots with non-native species (annual grasses, Erodium spp., and Centaurea melitensis) at
both low and high densities compared with control plants grown without competition
(one-way ANOVA, P < 0.0001). There were no significant differences between the
17
effects of the three non-native species groups on P. lyonii, or between density levels (Fig.
3). Similarly, P. lyonii plants growing without competition (control) were significantly
taller than plants growing in competition with all three non-native species groups, both at
high and low densities (one-way ANOVA, P < 0.0001, Fig. 3). In all three cases (annual
grasses, Erodium spp., and Centaurea), P. lyonii plants competing in low density pots
were taller than those in high density pots; however, these differences were not
significant within each of the non-native species groups (Fig. 3).
Comparison of Extant versus Extirpated Sites
The logistic regressions suggested that high cover of litter, annual grasses, and
other non-native plants other than annual grasses, Erodium and Centaurea, as well as low
cover of Erodium and bare ground are good predictors of P. lyonii extirpation (P < 0.05
in all cases, Table 4, Fig. 4). Nonmeteric multidimensional scaling yielded a final stress2
of 0.33747. All of the environmental variables correlated with whether the site was
extant or extirpated had r2 > 0.2 except for volumetric water content and percent cover of
Centaurea (Fig. 5). There was a clear separation of extant from extirpated sites (along
the horizontal axis). The factors that were most positively correlated with a site being
extant were percent cover of Erodium, PAR and percent bare ground. Amount of litter
and annual grasses were the most negatively correlated with a site being extant.
Community Effects Experiment
Treatment effects on native species richness differed among sites (there was a
significant site × treatment interaction; Table 5). Overall, the Paramount site generally
had more native species, the Lower site was intermediate, and the Pond site had the
fewest native species. At the Pond all three treatments had significantly higher richness
18
than the controls (Tukey HSD, P < 0.05), but did not differ significantly among
treatments (Tukey HSD, P > 0.05). At Paramount adding crust was associated with an
increase in the number of native species. The cut-scrape-and-add-crust treatment resulted
in significantly higher native species richness than both the control and the cut-only plots
(Tukey HSD, P < 0.05). At Lower the cut-and-scrape treatment had the largest effect,
but was not significantly different from the cut-only or the cut-and-add-crust treatment
(Tukey HSD, P < 0.05; Fig. 6). The sites also differed in native species evenness and
exhibited a pattern that was largely an inverse of the pattern of native species richness.
Pond had the greatest evenness, Lower had a moderate level of evenness, and Paramount
the lowest evenness (Fig. 6). There was no effect of treatment or significant site ×
treatment interaction on native species evenness (Table 5). Native species diversity was
not affected by site, treatment, or site × treatment interaction (Table 5).
The percent cover of non-native annual grasses was significantly affected by a site
× treatment interaction (P = 0.0378, Table 5, Fig. 7). At the Pond and Lower sites, all
three treatments produced significantly lower percents non-native annual grass cover than
the controls (Tukey HSD). At Paramount, the cut only plots and cut with crust addition
plots had significantly lower cover of non-native annual grasses than did the cut and
scraped plots and control plots (Tukey HSD).
The percent cover of native annual species had a significant site × treatment
interaction (P = 0.0008, Table 5, Fig. 7a). At the Pond site, the percent cover of native
annuals in control plots was significantly lower than in the treatment plots (P < 0.05).
However, there was no difference between the effects of the three treatments at that site
(Tukey HSD). At the Lower site, there was significantly greater native annual cover in
19
the cut and crust-addition plots than in the control plots (P < 0.05), but there was no
difference between controls and the other two treatments (Tukey HSD). At Paramount,
none of the treatments had a significant effect (P > 0.05) on the cover of native annuals.
There was no significant effect of treatment on density of P. lyonii at any of the
sites, but the Pond had more P. lyonii individuals than either Lower or Paramount (P =
0.0383, Table 5). The results of the two-way ANOVA that was restricted to plots
containing P. lyonii were consistent with the results of the ANOVA of all plots. The
interaction was not significant, and was dropped. There was no significant effect of
treatment after accounting for site (P = 0.65651). Site had a significant effect on density
of P. lyonii (P = 0.00030), with more individuals per plot at the Pond site than at the
Lower and Paramount sites. Tests of independence for each of three sites, separately and
pooled, showed no effects of the treatments on the presence of P. lyonii and sites were
homogeneous in lacking any effect (P > 0.05 in all cases, Table 6).
The multiple logistic regressions showed that at the Pond, the initial amount of
bare ground was not a significant predictor of P. lyonii presence at the end of the study
(Wald test, P = 0.21430). Therefore, bare ground was dropped from the model. Initial
annual grass cover was a significant predictor of P. lyonii presence at the Pond (Deviance
test, P = 0.00038; ρ2 = 0.136). The Lower site results were similar for bare ground (Wald
test, P = 0.20936) and annual grass (Deviance test, P = 0.01291; ρ2 = 0.06421). At both
the Pond and Lower sites P. lyonii was more likely to be present when there had been low
initial annual grass cover. At Paramount, neither initial annual grass cover (Wald test, P
= 0.59745) nor initial amount of bare ground (Wald test, P = 0.21187) were significant
predictors of P. lyonii presence.
20
Seeding Frequency Effects
The seeding manipulation had no consistent effect on the six dependent variables.
At the Pond site, the number of P. lyonii plants was greatest in all of the treatment plots
that were seeded twice (P < 0.05). In addition, there was a significant interaction
between seed manipulation × log initial bare ground (P < 0.05). There was no significant
effect of the seeding treatment on the native species richness in control plots or in cut and
scrape plots at any of the three sites. However, the cut and crust-addition plots at the
Pond site that had been seeded once had surprisingly higher native species richness than
the plots that had been seeded twice (P = 0.035). In contrast, the cut-only plots at the
Lower site had marginally higher native species richness in plots seeded twice (P = 0.05).
There was no significant effect of seeding manipulation for any of the main treatment
plots on native species evenness, native species diversity, or the percent cover of nonnative annual grasses (P > 0.05 all cases). Because seeding frequency seemed of little
importance, all of the seeding treatments were pooled for the analyses of the main
treatment effects.
When grouped by block, blocks that were seeded twice with associated species
and once with P. lyonii seeds had more P. lyonii plants than blocks that were seeded once
with associated species and no P. lyonii seeds. These differences were significant at both
the Lower site (two-sample t-test, df = 18, P < 0.0001), and also at Paramount (twosample t-test, df = 14, P = 0.0093; Fig. 8). The difference was not significant at the Pond
site (two-sample t-test, df = 18, P = 0.7636). At Paramount, four blocks were eliminated
from the analysis due to inconsistent seeding.
21
DISCUSSION
This study used an integrative approach to understanding the effects of non-native
plants on P. lyonii. Effects at the individual plant level and the multi-species community
level were examined. Four questions were addressed experimentally, and the results
indicated that invasive plants had a negative impact on P. lyonii and its associated native
plant community. There was evidence that non-native plants competitively reduced the
reproductive potential of P. lyonii, and that both direct and indirect effects from nonnative plants may have been significant factors in P. lyonii’s decline. In particular, the
presence of annual grasses not only resulted in direct competitive interactions, but also
impacted the community through the mechanism of the exclusion or reduction of
cryptobiotic crust by filling in the bare spaces that would have been covered by crust in
the pre-invasion conditions.
Does competition from non-native species affect P. lyonii success when grown
in pots and in the field?
The competitor removal experiment took place over two field seasons with very
different environmental conditions (2003-04 and 2004-05), and examined two possible
indicators of competitive effects on P. lyonii: number of inflorescences produced and
plant height. Results of competition on the number of inflorescences remained consistent
over both field seasons for all three invasive plant groups, despite the large difference
between years in available soil moisture. 2004-05 was an exceptionally wet year (56 cm
above average), with four times more rainfall than in 2003-04. Even with an excess of a
potentially limiting resource, competitive interference from non-native plants
significantly reduced the reproductive output of P. lyonii. This indicated that invasive
22
plants have a superior ability to capture other important limiting resources (perhaps
nutrients, space, and/or light). These results were corroborated in the pot competition
study, where growing conditions were more controlled, and water was generously
provided. Since the number of inflorescences a plant produces is an indicator of its total
reproductive capacity, this experiment clearly indicated that at least three non-native
plant groups (Erodium spp., grasses, and Centaurea melitensis) that commonly occur in
P. lyonii habitat negatively impact population inputs. This effect has the potential to
reduce time to extinction (Brook et al. 2002).
The impacts of competition on plant height were less clear. In both seasons P.
lyonii plants growing in the field in competition both with Centaurea and Erodium were
shorter than those in plots where competitors were removed. The opposite was true for
P. lyonii plants growing in competition with grasses. These differing results may reflect
differences in root morphology between grasses and forbs. The forbs, Pentachaeta,
Centaurea and Erodium, possess taproots, whereas the grasses have fibrous root systems.
These rooting differences may cause interspecific variation in the ability of plants to
access below-ground resources and thereby affect aboveground growth and competitive
dynamics. They may also account for the difference in the significances of effects
between wet and dry years. Height differences were greater in the forb groups in 200304, when soil moisture was more limited, than in 2004-05, when it was more abundant.
Additionally, phenology may contribute to differences in height response between
P. lyonii competing with forbs and P. lyonii competing with grasses. Both Centaurea
and non-native grasses typically grow taller than P. lyonii (Centaurea melitensis, up to
1m tall; Bromus madritensis, up to 50 cm tall; Avena fatua, over 1 m tall; Vulpia myuros,
23
around 70 cm tall, versus P. lyonii, up to 48 cm tall; Hickman 1993), and can potentially
reduce available light. However, the grasses grow taller faster than Centaurea, which
matures later in the season. Erodium also grows rapidly, but it is generally of shorter
stature than P. lyonii (10 to 90 cm, but often with prostrate or decumbent stems; Hickman
1993; and generally near the smaller extreme in P. lyonii habitat; personal observation),
and likely does not significantly reduce the light available to P. lyonii plants. The large
height difference (P < 0.001) between taller P. lyonii plants in control plots and shorter
ones in plots from which grass competitors had been removed in 2004-05 may have been
due to the abundance of rainfall that year. Greater moisture availability likely resulted in
exceptionally fast growth rates of grasses, causing P. lyonii plants to elongate to
compensate for early reduction in available light (Gurevitch et al. 2002). Growing tall is
a plastic response that can provide advantages to P. lyonii plants that are competing with
grasses, but this response does not necessarily indicate superior performance. Shorter P.
lyonii plants that were growing without competition from grasses produced more
inflorescences than tall plants growing with competition, indicating a trade-off in
resource allocation (Gurevitch et al. 2002).
Which non-native taxon has the greatest competitive effect on P. lyonii?
In the pot competition experiment, none of the invasive species had a greater
effect on P. lyonii reproduction or height than any of the others. However, there were
differences in impacts in the competitor removal experiment. In the field, although all
three invasive taxa were associated with significantly reduced P. lyonii reproduction,
competition from Centaurea suppressed inflorescence production more than Erodium and
significantly more than annual grasses did. Phenology and morphology may explain this.
24
Centaurea melitensis and P. lyonii are both early-summer-flowering annuals with basal
leaf rosettes and taproots. The bulk of their reproductive efforts occur as, or even after,
annual grasses and Erodium are completing theirs. Two co-occurring plant species with
similar phenologies and morphologies would be expected to compete for, rather than
partition resources (Dukes 2002). The ability of one species to deplete resources more
efficiently than others results in its competitive superiority (Tilman 1990). In semi-arid
southern California, water is a limiting resource after the end of the winter-spring rains.
In addition, another species of Centaurea (C. solstitialis) has been shown to increase lateseason evapotranspiration in the communities it has invaded (Dukes 2001). If Centaurea
melitensis behaves similarly, it may deplete water more efficiently than P. lyonii,
resulting in reduced late-season resource availability and ultimately reduced reproductive
capacity.
Annual grasses had a significantly greater impact on P. lyonii growth in height
than either Erodium or Centaurea in the field. As discussed above, the shorter stature of
Erodium probably meant that it did not compete with P. lyonii for light, and, therefore,
did not induce P. lyonii to allocate resources to increased stem elongation. Although
annual grasses and Centaurea can both grow taller than P. lyonii, the grasses reach their
maximum height much earlier than Centaurea. This means that the grasses can reduce
the light available to growing P. lyonii plants and force stem elongation. Although
elongation increases P. lyonii access to light the costs of this growth is apparently
reflected in reduced resource availability for inflorescences (Gurevitch et al. 2002). The
reduction of inflorescence number combined with increased height in plants competing
25
with annual grasses indicated that increased height did not provide a competitive
advantage to P. lyonii.
Depending on the species composition of the immediate neighborhood of a P.
lyonii plant, competitive pressure could affect an individual plant throughout the entire
above-ground portion of its life cycle. Differences in the phenologies of annual grasses,
Erodium, and Centaurea could potentially submit a P. lyonii individual to competition
early in development (competition with annual grasses and Erodium) through flowering
(competition with Centaurea).
What are the environmental conditions that contribute to the displacement
of P. lyonii by non-native plants?
In the absence of outright destruction of habitat, it is difficult to be certain of the
cause of the local extinction of a species. Comparisons of sites with extant populations
and sites with extirpated populations can be used to identify environmental factors
correlated with extirpation. Logistic regressions indicated that high cover of litter, annual
grasses, non-native plants, low cover of Erodium, and minimal bare ground are good
predictors of P. lyonii extirpation. This can be seen in the ordination of environmental
variables in which extirpated sites had high cover of litter and annual grasses. High litter
cover was the best predictor of extirpation, followed closely by scarcity of bare ground
and high cover of annual grasses. These three variables are related. Scarcity of bare
ground and presence of litter are both effects of annual grass presence. High percent
cover of Erodium in extant sites suggests that it, like Pentachaeta, favors sites without
extensive annual grass presence. The percent cover of Centaurea was not correlated with
either extant or extirpated sites. Considering the magnitude of its effect on P. lyonii
26
reproduction, this result might indicate that its immediate co-occurrence with P. lyonii
may not be as common or may be a much newer phenomenon than the co-occurrence of
P. lyonii with annual grass. Anecdotally, the invasion of Centaurea melitensis seems to
be a recent process, with observed increases in population sizes within the last thirty
years (R. Burgess, personal communication).
It is surprising that extirpation and domination from annual grasses were not
correlated with frequent fires. The invasion of annual grasses can increase the chance of
fire, leading to grass-fire feedbacks (Mack and D'Antonio 1998), which can have
negative direct and indirect effects on the local community and ultimately result in typeconversion to alien-dominated grassland (Keeley 1995a). Within P. lyonii habitat, this
kind of type conversion is a common occurrence. That extirpated sites were characterized
by dominance of annual grass does not necessarily mean that competition from these
annual grasses was the mechanism of P. lyonii extirpation.
What are the impacts from non-native species on P. lyonii in the
context of its native community?
Habitat manipulations (cutting non-native plants, cutting and scraping the soil
surface, and cutting, scraping, and adding cryptobiotic crust) did not affect the presence
of P. lyonii or its density. This result is surprising given the results of the direct
competition experiments showing that non-native plants significantly reduce the number
of inflorescences by competition. This may indicate that reductions in P. lyonii density
due to reduced seed production were not detectable in the two-year period of this study.
The results further indicate that two aspects of life history of P. lyonii, germination and
early-season survival, are not affected by non-native plant impacts such as competition
27
for resources, deposition of litter, alteration of soil stability, and displacement of
cryptobiotic crust. If impacts during early life stages are contributing to population
reductions, then factors other than non-native plant presence such as herbivory, granivory
and/or the availability of safe sites for germination may be important. In addition or
alternatively, below-ground effects of non-native plant presence such as alterations in the
diversity of mycorrhizal fungi available to colonize native plant roots (Hawkes et al.
2006) may have adverse effects on P. lyonii survival.
Manipulations had a positive effect on native species richness at all three sites,
reduced percent cover of annual grasses, and increased the amount of bare ground.
However, these results differed among sites. At the Pond site, neither the cut and scrape
nor the cut and add crust treatment had a greater effect than only cutting non-native
plants. However, at the Lower site, the cut and scrape treatment had a greater effect than
the other two treatments. At Paramount, the cut and add crust treatment was most
effective. Similarly, treatment effects on the percent cover of native annuals differed
among sites. Again, at the Pond site, scraping and adding crust did not increase the
percent cover of native annuals more than simply cutting non-native plants. At the
Lower site, only the cut and add crust treatment had a significant effect. At Paramount,
none of the treatments increased the percent cover of native annual plants. Therefore,
efficient management for native species richness and percent cover is site-specific, and to
reduce costs, potential restoration sites should be tested prior to management plan
preparation.
The increase in native species richness resulting from habitat manipulations is
important because native species diversity is thought to be negatively associated with
28
community invasibility (Dukes 2002, Elton 1958, Naeem et al. 2000), although extrinsic
factors may change this relationship (Levine and D'Antonio 1999). Species-rich
communities tend to use available resources thoroughly, reducing available niche space
for invading species (Tilman 1997). In this study, the treatments increased the number of
native species present in most cases. There is a significant positive correlation between
P. lyonii density and native species richness (NPS unpublished data), which might
suggest that increased native species richness could indirectly benefit P. lyonii by
reducing future invasibility or the impacts from invaders already present (Dukes 2001).
The interaction of effects from competitors and litter has been shown to be more
negative for seedling establishment than the effects of either competition or litter
presence alone (Suding and Goldberg 1999). Although the different treatments had
varying effects on native species richness at the three sites, at least one treatment
involving both cutting (the removal of grass) and scraping (the removal of litter)
increased native species richness compared to control plots at each site. It is not clear
whether the interaction of annual grasses and litter had as great an impact on P. lyonii as
it did for other members of the community because none of the treatments had an effect
on recruitment of P. lyonii.
The reduction in annual grasses is important because competition from annual
grasses significantly reduced reproduction of P. lyonii. Annual grasses were more
abundant at extirpated sites than Centaurea or Erodium and, along with sparse bare
ground, high percent cover of annual grasses was a good predictor of P. lyonii
extirpation. Therefore, significant reduction of annual grass cover and increase in bare
ground should have positive effects on P. lyonii persistence over the longer term.
29
Because neither the additional manipulations of scraping nor adding crust decreased
annual grass cover more than simply cutting non-native plants at any of the three sites,
minimal management of annual grasses would only require their removal.
In California grasslands, the abundance of native annual forbs and perennial
grasses increases with seed addition even in the presence of non-native competitors,
indicating dispersal limitations (Seabloom et al. 2003a, Seabloom et al. 2003b). Plots
seeded with P. lyonii had greater densities of P. lyonii, regardless of treatment.
Treatments did not increase recruitment from the seed bank as did addition of seed,
indicating that P. lyonii is dispersal limited. Although only one site (Pond) had plots
containing P. lyonii before seeding, P. lyonii germinated and survived in seeded plots at
all three field sites. The fact that one of the sites had never supported a population of P.
lyonii is particularly indicative of dispersal limitation, and suggests that conservation
strategies should include establishment of new populations on appropriate protected land,
along with the restoration of existing populations.
Interannual variation in environmental conditions may contribute to the
persistence of small native plant populations when they are competing with non-native
plants, particularly if the natives possess a persistent seed bank and the invasives do not
(Levine and Rees 2004). In southern California, alien annual grasses do not have a
persistent seed bank, but many native annuals do (Pavlik et al. 1993, Rice 1985). The
region’s widely fluctuating interannual rainfall affects population dynamics of species
differently depending on seed dormancy characteristics. In environments dominated by
grasses, a dry year reduces seed production in all species. Neither annual grasses nor
native annuals will produce a large number of seeds. However, due to persistent seed
30
banks, native plant populations are not affected by periodic reductions in seed production
associated with dry years. Therefore, a wet year following a dry year will favor a species
with persistent seed banks, thus interrupting the competitive exclusion process (Levine
and Rees 2004). Although P. lyonii has been thought to lack a persistent seed bank
(Keeley 1995b), extreme interannual population fluctuations and germination studies
(NPS unpublished data) indicate that sites with P. lyonii actually support large dormant
seed pools.
Conclusions and Management Implications
Competition from invasive plants has not been a documented cause of extinction
of any North American native plant species (Davis 2003). However, numerous studies
have shown that competition from invasive plants can have negative effects on rare
native plants (Huenneke and Thomson 1995, Kingston et al. 2004, Miller and Duncan
2003, Thomson 2005, Walck et al. 1999). These effects can reduce population sizes,
increasing the possibility of local extinction (Brook et al. 2002). Hobbs and Mooney
(1998, p. 276) suggest that “the process of species extinction is often simply the endpoint
of a process of population extinctions throughout the former range of a species.” In the
case of a rare endemic such as P. lyonii, which has already lost at least 45% of its
populations in recent decades, and with the remaining patches geographically isolated by
fragmentation, competitive pressures should be expected to contribute to local declines,
and possibly its ultimate extinction.
To forestall extinction, efforts should be made to establish (or reestablish) new
populations. The data indicate that P. lyonii is dispersal limited, and it is a good
candidate for introductions into sites with minimal presence of invasive species, large
31
amounts of bare ground, and presence of associated native species. The continuing
destruction of P. lyonii habitat on private property means that conservation efforts must
be concentrated on protected public lands. A goal stated in the U.S. Fish and Wildlife
Service’s recovery plan is to establish new self-sustaining populations and restore
existing populations to viable numbers (10 populations of 10,000 individuals).
Therefore, suitable sites and sites with extant populations on protected lands should be
seeded and maintained by regular removal of invasive competitors. By placing
restoration sites on managed parklands, existing or developing weed management plans
for the park can be incorporated into a long-term P. lyonii management plan. Wellplanned, holistic approaches to weed management can help to avoid re-invasion, or
subsequent invasion of different non-native species.
Hand-removal of weeds will be necessary because of their co-occurrence with P.
lyonii. This can be labor intensive and therefore expensive. Alternative methods should
be investigated to facilitate the feasibility of large scale, long term restoration efforts.
The use of monocot-specific herbicides early in the season to eliminate competition from
annual grasses should be considered as a possible option, but its effect on the associated
community should be studied before implementation. The use of prescribed burning may
be an alternative. However, little is known about P. lyonii’s response to fire, and
experiments to evaluate the effects of fire frequency, intensity, and seasonality on both P.
lyonii and its associated community should be carried out prior to consideration as a
restoration tool.
Little is known about the metapopulation dynamics of P. lyonii. If seeding new
and existing populations is to be implemented as a restoration tool, pollination and
32
genetic studies should be done to determine population structure throughout the range,
and to uncover potential pollination limitation in population fragments. The remaining
populations are highly fragmented, separated by large urbanized areas, and may be
genetically distinct. Genetic information will better inform decisions about source seed
for restoration projects, and pollination studies will investigate a critical life stage in this
self-incompatible species.
Conservation biology is a crisis discipline (Primack 2002). Because of the
unprecedented rate of species extinctions, often managers are forced to take actions to
attempt to preserve species without thorough investigations. This study provides a basis
to design sound conservation strategies, and serves as a starting point for further
examination of the biological and ecological requirements of an endangered species.
More broadly, the information gained here can be applied to similar conservation
questions in mediterranean grassland ecosystems, which are under increasing pressure
from non-native invaders.
33
Annual grasses 2005
Annual grasses 2004
1
1
0.9
Inflorescences (log)
0.8
0.7
0.6
D=-0.2712
SE=0.0384
N=19
P<0.0001
0.8
0.7
0.6
0.5
0.5
0.4
0.4
0.3
0.3
0.2
0.2
0.1
0.1
0
0
Competition
No Competition
Competition
Erodium 2004
Inflorescences (log)
0.8
0.7
0.6
1
D=-0.2938
SE=0.0417
N=18
P<0.0001
0.8
0.7
0.6
0.5
0.4
0.4
0.3
0.3
0.2
0.2
0.1
0.1
0
0
Competition
Competition
No Competition
Centaurea 2004
Inflorescences (log)
0.7
No Competition
Centaurea 2005
1
0.8
D=-0.4253
SE=0.0688
N=15
P=<0.0001
0.9
0.5
0.9
No Competition
Erodium 2005
1
0.9
D=-0.2107
SE=0.0537
N=17
P=0.0012
0.9
1
D=-0.3541
SE=0.0547
N=15
P=<0.0001
D=-0.4639
SE=0.0822
N=18
P<0.0001
0.9
0.8
0.7
0.6
0.6
0.5
0.5
0.4
0.4
0.3
0.3
0.2
0.2
0.1
0.1
0
0
Competition
No Competition
Competition
No Competition
Figure 1. Mean number (+ SE) of inflorescences (log transformed) on P. lyonii plants in competition and
no-competition plots. D = the mean difference of the logs of the averages of number of inflorescences in
competition and no-competition plots, SE and P values are from paired t-tests.
34
Annual grasses 2005
Annual grasses 2004
250
250
D=4.1774
SE=6.8213
N=19
P=0.5480
Mean height (mm)
200
150
150
100
100
50
50
0
0
Competition
Competition
No Competition
Erodium 2004
250
D=-8.2730
SE=2.9726
N=18
P=0.0128
200
Mean height (mm)
No Competition
Erodium 2005
250
150
D=-90207
SE=12.4442
N=15
P=0.4805
200
150
100
100
50
50
0
0
Competition
No Competition
Competition
Centaurea 2004
No Competition
Centaurea 2005
250
250
D=-13.2260
SE=6.5770
N=15
P=0.0640
200
Mean height (mm)
D=29.7856
SE=5.6971
N=17
P=0.0001
200
150
D=-4.6194
SE=20.1133
N=18
P=0.8211
200
150
100
100
50
50
0
0
Competition
No Competition
Competition
No Competition
Figure 2. Mean height (+ SE) of P. lyonii plants in competition and no-competition plots. D = the mean
difference of the average heights in competition and no-competition plots, SE and P values are from paired
t-tests.
35
Figure 3. Mean (+ SE) number of inflorescences (log transformed) and mean heights of P. lyonii plants
(+SE) grown in pots with high and low levels of competition (densities) from nonnative species. Similar
letters above two bars indicated a non-significant difference between treatments (Tukey HSD).
36
80
70
% Cover
60
50
Extant
40
Extirpated
30
20
10
O
t h l gra
er
ss
no
es
nna
tiv
es
Er
o
di
O
um
th
er
na
tiv
C
en es
ta
ur
ea
P.
lyo
ni
i
un
d
gr
o
nu
a
An
Ba
re
Li
t te
r
0
Figure 4. Differences in composition of sites where P. lyonii is extant and sites where it is extirpated.
37
Stunt
fires since ‘58
Kirsten Lee
%other
non-natives
soil
depth
Tierra
near
Triunfo
near
Malibu Cr
Axis 2
Wildwood Cr
7th
Day
Mulholland
%litter
%annual grasses
%other
natives
Rocky
Oaks
Sidlee
far
insolation
%P. lyonii
Triunfo
far
%Erodium Tierra
far
Wildwood Rd
PAR
%bare ground
Wildwood Stage
Sidlee
near
Cal Lutheran
Axis 1
Figure 5. Nonmetric multidimensional scaling of environmental variables. Ordination rotated so Axis 1
correlates maximally with extant (unfilled triangles) versus extirpated (filled triangles) sites. The lengths
and directions of the vectors indicate the strengths and directions of the correlations with the axes.
38
Native species richness (sqrt)
3.9
Pond
3.7
Lower
3.5
Paramount
3.3
3.1
2.9
2.7
2.5
Control
Cut only
Cut and
scrape
Cut and
add crust
Native species evenness
0.85
0.8
0.75
0.7
0.65
0.6
0.55
0.5
Control
Cut only
Cut and
scrape
Cut and
add crust
Figure 6. Native species richness and native species evenness within sites (mean + SE, n =20 plots per
treatment per site).
39
Percent cover annual grass (log)
1.8
Pond
1.6
Lower
1.4
Paramount
1.2
1
0.8
0.6
0.4
0.2
Control
Cut only
Cut and
scrape
Cut and
add crust
50
Native annuals
45
40
35
30
25
20
15
10
5
0
Control
Cut only
Cut and
scrape
Cut and
add crust
Figure 7. Means (+ SE) of the log of percent cover of nonnative annual grasses (n=20) and percent
cover of native annuals within sites.
40
Number of P. lyonii (sqrt)
7
6
5
*
4
*
3
2
1
0
Pond
Lower
Paramount
Figure 8. Effects of seed addition on number of P. lyonii within blocks. Hatched bars are blocks seeded
once with associated species, and no P. lyonii. Gray bars are blocks seeded twice with associated species,
and once with P. lyonii. * Indicates P < 0.01 (two-sample t-tests).
41
Table 1. Mean difference in magnitude of competitive effects of Erodium spp. and Centaurea melitensis on
P. lyonii between the sites, Tierra Rejada (Tierra) and Triunfo and P-values from two-sample t-tests. Plots
with annual grass as the competitor were established at only one site each year and were, therefore
excluded from this analysis. Each variable was analyzed separately in 2004 and 2005.
Non-native
Site
n
Log(cut)-log(control)
SE
P
2004 inflorescences
Erodium
Tierra
10
0.34396
0.06524
0.18707
Erodium
Triunfo
8
0.23116
0.04080
Centaurea
Centaurea
Tierra
Triunfo
9
6
0.36426
0.33874
0.07772
0.07987
0.82879
Erodium
Erodium
Tierra
Triunfo
10
8
0.13049
0.08262
0.04789
0.05319
0.51349
Centaurea
Centaurea
Tierra
Triunfo
9
6
0.07341
0.12152
0.05905
0.07633
0.62244
Tierra
Triunfo
8
7
0.49001
0.35137
0.11372
0.06880
0.33311
Centaurea
Centaurea
Tierra
Triunfo
9
9
0.44685
0.48114
0.12964
0.10873
0.84192
Erodium
Erodium
Tierra
Triunfo
8
7
0.06618
0.03638
0.05773
0.05831
0.23514
Centaurea
Centaurea
Tierra
Triunfo
9
9
0.13767
0.08071
0.02701
0.09143
0.03591
2004 height
2005 inflorescences
Erodium
Erodium
2005 height
42
Table 2. Some of the native plants associated with Pentachaeta lyonii (NPS unpublished data).
Allium haematochiton
Amsinckia menziesii
Asclepias fascicularis
Bloomeria crocea
Calandrinia ciliata
Calochortus spp.
Camissonia spp.
Castilleja exserta
Chlorogalum pomeridianum
Chorizanthe polygonoides
Chorizanthe staticoides
Clarkia spp.
Crassula connata
Cryptantha spp.
Dichelostemma capitatum
Dodecatheon clevelandii
Dudleya spp.
Eremocarpus setigerus
Erigeron foliosis ssp. foliosis
Fritillaria biflora
Gilia angelensis
Gnaphalium spp.
Hemizonia fasciculata
Lasthenia californica
Lepidium nitidum
Lessingia filaginifolia
Lewisia rediviva
Lomatium dasycarpum
Lomatium lucidum
Lotus purshianus
Lotus salsuginosus
Lotus scoparius
Lotus strigosus
Lupinus bicolor
Navarretia pubescens
Nassella lepida
Nassella pulchra
Pectocarya linearis ssp. ferocula
Phacelia cicutaria var. hispida
Plagiobothrys spp.
Plantago erecta
Pterostegia drymarioides
Salvia columbariae
Psilocarphus tenellus
Selaginella bigelovii
Silene antirrhina
Sisyrinchium bellum
Stylocline gnaphaloides
Thalictrum fendleri ssp. polycarpum
Thysanocarpus spp.
Trichostema lanceolatum
Trifolium willdenovii
43
Table 3. Seed mix applied to plots. In February, 2004, all plots at each site were seeded. In October, 2004,
five randomly selected blocks in each site seeded. *Additional seeds added to one block in each site.
Species
Asclepias fascicularis
Bloomeria crocea
Calochortus catalinae
Calochortus clavatus
Castilleja exserta
Chlorogalum pomeridianum
Chorizanthe staticoides
Clarkia cylindrica
Crassula connata
Cryptantha micromeres
Cryptantha muricata
Dichelostema capitata
Dodecatheon clevelandii
Eremocarpus setigerus
Erigeron foliosis
Fritillaria biflora
Gnapthalium californicum
Hemizonia fasciculata
Lasthenia californica
Lomatium dasycarpum
Lotus salsuginosus
Micropus californicus
Nassella lepida
Nassella pulchra
Pentachaeta lyonii
Phacelia cicutaria
Plantago erecta
Plagiobothrys nothofulvus
Salvia columbariae
Sisyrinchium bellum
Stylocline gnaphalioides
Thalictrum fendleri
Thysanocarpus sp.
Trichostema lanceolatum
#seeds/plot
February 04 October 04
5
6
72*
2
62
3
100
83
5
80
100*
64
84
83*
4
7
6
100
100*
50*
7
7
17
100
100
45
100
16
10
100*
50
100
50
100
680
40
100*
9
25
1
83*
22
56
16
37
60
3
53*
44
Table 4. Comparison of conditions at extant and extirpated sites. Numbers in body of table are means (with
standard deviations in parentheses). P-values are results of deviance tests from separate logistic regressions.
Predictor variable
% Litter
% Bare ground
% Annual grasses
% Non-natives
% Erodium
% Other natives
% Centaurea
PAR
Volumetric water content
Soil depth
Extant sites
(n = 12)
31 (11.7)
21 (7.9)
19 (13.0)
37 (14.7)
4 (2.6)
24 (16.3)
8 (9.6)
969 (152.1)
44 (13.5)
21 (5.2)
Extirpated sites
(n = 4)
73 (14.7)
3 (4.2)
58 (16.9)
72 (16.7)
1 (1.0)
9 (15.8)
2 (3.5)
590 (370.2)
34 (18.6)
24 (7.0)
45
P
0.00002
0.00022
0.00032
0.00116
0.00777
0.06286
0.16733
0.01579
0.22449
0.33519
Table 5. The effects of treatments (cutting, cutting and scraping, and cutting, scraping and adding crust) for
the dependent variables native species richness (sqrt), native species diversity, native species evenness,
number of P. lyonii (sqrt), percent cover of non-native annual grasses (log), percent cover of bare ground
(log), and number of native annual forb species.
Source
Site
Treatment
Site × Treatment
Block(Site)
Treatment × Block(Site)
df
2
3
6
57
171
MS
6.549
0.796
0.199
0.308
0.083
F
21.23326
4.01073
2.39894
P
< 0.0001
0.06975
0.02986
Native species diversity
Site
Treatment
Site × Treatment
Block(Site)
Treatment × Block(Site)
2
3
6
57
171
0.103
0.292
0.126
0.232
0.09
0.44421
2.31162
1.3952
0.6435
0.1759
0.2191
Native species evenness
Site
Treatment
Site × Treatment
Block(Site)
Treatment × Block(Site)
2
3
6
57
171
0.346
0.003
0.009
0.044
0.013
7.82575
0.317647
7.15926
0.000993
0.812737
0.637251
P. lyonii (sqrt)
(all plots)
Site
Treatment
Site × Treatment
Block(Site)
Treatment × Block(Site)
2
3
6
57
171
63.95558
0.58207
1.08939
18.50768
2.62579
3.455624
0.534308
0.414881
0.038325
0.675559
0.868457
Percent cover non-native
annual grasses (log)
Site
Treatment
Site × Treatment
Block(Site)
Treatment × Block(Site)
2
3
6
57
171
3.8475
4.39809
0.24065
0.40964
0.10522
9.392393
18.27588
2.287113
0.000298
0.00202
0.037765
Percent cover
bare ground
Site
Treatment
Site × Treatment
Block(Site)
Treatment × Block(Site)
2
3
6
57
171
436.1625
1187.8111
34.0070
282.3206
82.7633
1.544919
34.92849
0.410894
0.222128
0.00034
0.871117
Percent cover native
annual forbs
Site
Treatment
Site × Treatment
Block(Site)
Treatment × Block(Site)
2
3
6
57
171
600.1625
1586.894
488.2903
481.0408
120.5297
1.247633
3.2499
4.051204
0.294907
0.102085
0.0008
Native species richness
(sqrt)
46
Table 6. Tests of independence for each of three sites separately and pooled. n=20 plots per treatment per
site. Numbers in the body of the table are the percent of plots in which P. lyonii was present.
Pond
Control
40%
Cut only
60%
Cut and
scrape
45%
Cut and
crust added
50%
G2
1.76117
df
3
P
0.62342
Lower
30%
35%
35%
25%
0.65011
3
0.88487
Paramount
30%
40%
35%
40%
0.60152
3
0.89608
3.0128
9
0.96379
1.7363
3
0.6289
1.27654
6
0.97294
Total
Pooled
33%
45%
38%
Heterogeneity
47
38%
LITERATURE CITED
Aarssen, L. W. 1983. Ecological combining ability and competitive combining ability in
plants: toward a general evolutionary theory of coexistence in systems of
competition. The American Naturalist 122: 707-731.
Bais, H. P., R. Vepachedu, S. Gilroy, R. M. Callaway, and J. M. Vivanco. 2003.
Allelopathy and exotic plant invasion: From molecules and genes to species
interactions. Science 301: 1377-1380.
Belnap, J. 2006. The potential roles of biological soil crusts in dryland hydrologic cycles.
Hydrological Processes 20: 3159-3178.
Belnap, J., and K. T. Harper. 1995. Influence of Cryptobiotic Soil Crusts on Elemental
Content of Tissue of 2 Desert Seed Plants. Arid Soil Research and Rehabilitation
9: 107-115.
Belnap, J., R. Rosentreter, S. Leonard, J. Hilty Kaltenecker, J. Williams, and D. Eldridge.
2001. Biological soil crusts: Ecology and management. Bureau of Land
Management, United States Department of the Interior.
Berendse, F. 1983. Interspecific competition and niche differentiation between Plantago
lanceolata and Anthoxanthum odoratum in a natural hayfield. Journal of Ecology
71: 379-390.
Blair, A. C., and L. M. Wolfe. 2004. The evolution of an invasive plant: an experimental
study with Silene latifolia. Ecology 85: 3035-3042.
Blossey, B., and R. Nötzold. 1995. Evolution of increased competitive ability in invasive
nonindigenous plants: a hypothesis. Journal of Ecology 83: 887-889.
Bowker, M. A., J. Belnap, R. Rosentreter, and B. Graham. 2004. Wildfire-resistant
biological soil crusts and fire-induced loss of soil stability in Palouse prairies,
USA. Applied Soil Ecology 26: 41-52.
Braker, E., and L. Verhoeven. 1998. Relative success of flower visitiors in pollination of
the endangered Lyon’s Pentachaeta (Pentachaeta lyonii). Western National Parks
Association.
Brigham, C. A. 2003. Factors affecting persistence in formerly common and historically
rare plants. Pages 59-97 in C. A. Brigham and M. W. Schwartz, eds. Population
viability in plants. Conservation, management, and modeling of rare plants.
Springer-Verlag, Berlin.
Brook, B. W., D. W. Tonkyn, J. J. O'Grady, and R. Frankham. 2002. Contribution of
inbreeding to extinction risk in threatened species. Conservation Ecology 6:
article 16 [online].
Brooks, M. L. 2000. Competition between alien annual grasses and native annual plants
in the Mojave Desert. American Midland Naturalist 144: 92-108.
Crawley, M. J. 2004. Timing of disturbance and coexistence in a species-rich ruderal
plant community. Ecology 85: 3277-3288.
D’Antonio, C. M., and P. M. Vitousek. 1992. Biological invasions by exotic grasses, the
grass-fire cycle, and global change. Annual Review of Ecology and Systematics
23: 63-87.
Davis, M. A. 2003. Biotic globalization: does competition from introduced species
threaten biodiversity? BioScience 53: 481-489.
48
DeFalco, L. A., D. R. Bryla, V. Smith-Longozo, and R. S. Nowak. 2003. Are Mojave
Desert annual species equal? Resource acquisition and allocation for the invasive
grass Bromus madritensis subsp. rubens (Poaceae) and two native species.
American Journal of Botany 90: 1045-1053.
Dukes, J. S. 2001. Biodiversity and invasibility in grassland microcosms. Oecologia 126:
563-568.
—. 2002. Species composition and diversity affect grassland susceptibility and response
to invasion. Ecological Applications 12: 602-617.
Dyer, A. R., and K. J. Rice. 1997. Intraspecific and diffuse competition: the response of
Nassella pulchra in a California grassland. Ecological Applications 7: 484-492.
—. 1999. Effects of competition on resource availability and growth of a California
bunchgrass. Ecology 80: 2697-2710.
Elton, C. S. 1958. The Ecology of Invasions by Animals and Plants. The University of
Chicago Press, Chicago.
Evans, R. D., R. Rimer, L. Sperry, and J. Belnap. 2001. Exotic plant invasion alters
nitrogen dynamics in an arid grassland. Ecological Applications 11: 1301-1310.
Fotheringham, C. J., and J. E. Keeley. 1998. Ecology and distribution of Braunton's
milkvetch (Astragalus brauntonii) and Lyon's pentachaeta (Pentachaeta lyonii).
California Department of Fish and Game, Region 5.
Gordon, D. R. 1998. Effects of invasive, non-indigenous plant species on ecosystem
processes: lessons from Florida. Ecological Applications 8: 975-989.
Gurevitch, J., S. M. Scheiner, and G. Fox. 2002. The ecology of plants. Sinauer
Associates, Inc., Sunderland, MA.
Harper, K. T., and J. Belnap. 2001. The influence of biological soil crusts on mineral
uptake by associated vascular plants. Journal of Arid Environments 47: 347-357.
Hawkes, C. V., J. Belnap, C. M. D'Antonio, and M. K. Firestone. 2006. Arbuscular
mycorrhizal assemblages in native plant roots change in the presence of invasive
exotic grasses. Plant and Soil 281: 369-380.
Hickman, J. C. e. 1993. The Jepson Manual: Higher Plants of California. University of
California Press, Berkeley.
Hobbs, R. J., and L. F. Huenneke. 1992. Disturbance, diversity, and invasion:
implications for conservation. Conservation Biology 6: 324-337.
Hobbs, R. J., and H. A. Mooney. 1998. Broadening the extinction debate: population
deletions and additions in California and western Australia. Conservation Biology
12: 271-283.
Huenneke, L. F., and J. K. Thomson. 1995. Potential interference between a threatened
endemic thistle and an invasive nonnative plant. Conservation Biology 9: 416425.
Keane, R. M., and M. J. Crawley. 2002. Exotic plant invasions and the enemy release
hypothesis. Trends in Ecology & Evolution 17: 164-170.
Keeley, J. E. 1995a. Future of California floristics and systematics: Wildfire threats to the
California flora. Madroño 42: 175-179.
—. 1995b. Seed germination and dormancy of Pentachaeta lyonii. California Department
of Fish and Game, Sacramento, CA.
Kimball, S., and P. M. Schiffman. 2003. Differing effects of cattle grazing on native and
alien plants. Conservation Biology 17: 1681-1693.
49
Kingston, N., S. Waldren, and N. Smyth. 2004. Conservation genetics and ecology of
Angiopteris chauliodonta Copel. (Marattiaceae), a critically endangered fern from
Pitcairn Island, south central Pacific Ocean. Biological Conservation 117: 309319.
Leger, E. A., and M. L. Forister. 2005. Increased resistance to generalist herbivores in
invasive populations of the California poppy (Eschscholzia californica). Diversity
and Distributions 11: 311-317.
Leger, E. A., and K. J. Rice. 2003. Invasive California poppies (Eschscholzia californica
Cham.) grow larger than native individuals under reduced competition. Ecology
Letters 6: 257-264.
Levine, J. M., and C. M. D'Antonio. 1999. Elton revisited: a review of evidence linking
diversity and invasibility. Oikos 87: 15-26.
Levine, J. M., and M. Rees. 2002. Coexistence and relative abundance in annual plant
assemblages: the roles of competition and colonization. The American Naturalist
160: 452-467.
—. 2004. Effects of temporal variability on rare plant persistence in annual systems. The
American Naturalist 164: 350-363.
Lodge, D. 1993. Biological invasions: lessons from ecology. Trends in Ecology and
Evolution 8: 133-137.
Mack, M., and C. M. D'Antonio. 1998. Impacts of biological invasions on disturbance
regimes. Trends in Ecology & Evolution 13: 195-198.
Mack, R. N., D. Simberloff, W. M. Lonsdale, H. Evans, M. Clout, and F. A. Bazzaz.
2000. Biotic invasions: causes, epidemiology, global consequences, and control.
Ecological Applications 10: 689-710.
Miller, A. L., and R. P. Duncan. 2003. Extrinsic and intrinsic controls on the distribution
of the critically endangered cress, Ischnocarpus exilis (Brassicaceae). Biological
Conservation 110: 153-160.
Munz, P. A., and D. D. Keck. 1959. A California Flora. University of California Press,
Berkeley and Los Angeles.
Myers, N., R. A. Mittermeier, C. G. Mittermeier, G. A. B. da Fonseca, and J. Kent. 2000.
Biodiversity hotspots for conservation priorities. Nature 403: 853-858.
Naeem, S., J. M. H. Knops, D. Tilman, K. M. Howe, T. Kennedy, and S. Gale. 2000.
Plant diversity increases resistance to invasion in the absence of covarying
extrinsic factors. Oikos 91: 97-108.
Pavlik, B. M., N. Ferguson, and M. Nelson. 1993. Assessing limitations on the growth of
endangered plant populations, II. Seed production and seed bank dynamics of
Erysimum capitatum ssp. angustatum and Oenothera deltoides ssp. howellii.
Biological conservation 65: 267-278.
Pimentel, D., R. Zuniga, and D. Morrison. 2005. Update on the environmental and
economic costs associated with alien-invasive species in the United States.
Ecological Economics 52: 273-288.
Poot, P., and H. Lambers. 2003. Are trade-offs in allocation pattern and root morphology
related to species abundance? A congeneric comparison between rare and
common species in the south-western Australian flora. Journal of Ecology 91: 5867.
50
Primack, R. B. 2002. Essentials of conservation biology. Sinauer Associates, Sunderland,
MA.
Rejmánek, M. 1996. A theory of seed plant invasiveness: the first sketch. Biological
Conservation 78: 171-181.
Rejmánek, M., and J. M. Randall. 1994. Invasive alien plants in California: 1993
summary and comparison with other areas in North America. Madroño 41: 161177.
Reynolds, H. L., B. A. Hungate, F. S. Chapin III, and C. M. D'Antonio. 1997. Soil
heterogeneity and plant competition in an annual grassland. Ecology 78: 20762090.
Rice, K. J. 1985. Responses of Erodium to varying microsites: the role of germination
cueing. Ecology 66: 1651-1657.
Richardson, D. M., P. Pyšek, M. Rejmánek, M. G. Barbour, F. D. Panetta, and C. J.
West. 2000. Naturalization and invasion of alien plants: concepts and definitions.
Diversity and Distributions 6: 93-107.
Sala, A., S. D. Smith, and D. A. Devitt. 1996. Water use by Tamarix ramosissima and
associated phreatophytes in a Mojave Desert floodplain. Ecological Applications
6: 888-898.
Sans, F. X., H. Garcia-Serrano, and I. Afán. 2004. Life-history traits of alien and native
Senecio species in the Mediterranean region. Acta Oecologica 26: 167-178.
Sanz-Elorza, M., E. D. Dana, and E. Sobrino. 2006. Invasibility of an inland area in NE
Spain by alien plants. Acta Oecologica 29: 114-122.
Schiffman, P. M. 2000. Mammal burrowing, erratic rainfall and the annual lifestyle in the
California prairie: is it time for a paradigm shift? Pages 153-160 in J. E. Keeley,
M. Baer-Keeley, and C. J. Fotheringham, eds. 2nd Interface between ecology and
land development in California. U.S. Geological Survey Open-File Report 00-62.
—. 2005. The Los Angeles Prairie in W. Deverell and G. Hise, eds. Land of Sunshine: An
Environmental History of Metropolitan Los Angeles. University of Pittsburgh
Press, Pittsburgh.
Schutzenhofer, M. R., and T. J. Valone. 2006. Positive and negative effects of exotic
Erodium cicutarium on an arid ecosystem. Biological Conservation 132: 376-381.
Seabloom, E. W., E. T. Borer, V. L. Boucher, R. S. Burton, K. L. Cottingham, L.
Goldwasser, W. Gram, B. E. Kendall, and F. Micheli. 2003a. Competition, seed
limitation, disturbance, and reestablishment of California native annual forbs.
Ecological Applications 13: 575-592.
Seabloom, E. W., W. S. Harpole, O. J. Reichman, and D. Tilman. 2003b. Invasion,
competitive dominance, and resource use by exotic and native California
grassland species. Proceedings of the National Academy of Sciences 100: 1338413389.
Silander, J. A., and S. W. Pacala. 1990. The application of plant population dynamic
models to understanding plant competition. Academic Press, Inc., San Diego.
Sperry, L. J., J. Belnap, and R. D. Evans. 2006. Bromus tectorum invasion alters nitrogen
dynamics in an undisturbed arid grassland ecosystem. Ecology 87: 603-615.
Suding, K. N., and D. E. Goldberg. 1999. Variation in the effects of vegetation and litter
on recruitment across productivity gradients. The Journal of Ecology 87: 436-449.
51
Thomson, D. 2005. Measuring the effects of invasive species on the demography of a
rare endemic plant. Biological Invasions 7: 615-624.
Tilman, D. 1990. Mechanisms of plant competition for nutrients: the elements of a
predictive theory of competition. Academic Press, Inc., San Diego.
—. 1997. Community invasibility, recruitment limitation, and grassland biodiversity.
Ecology 78: 81-92.
Vanlill, W. S., F. J. Kruger, and D. B. Vanwyk. 1980. The effect of afforestation with
Eucalyptus grandis Hill ex Maiden and Pinus patula Schlecht. et Cham. on
streamflow from experimental catchments at Mokobulaan, Transvaal. Journal of
Hydrology 48: 107-118.
Vilà, M., A. Gómez, and J. L. Maron. 2003. Are alien plants more competitive than their
native conspecifics? A test using Hypericum perforatum L. Oecologia 137: 211215.
Vilà, M., F. Lloret, E. Ogheri, and J. Terradas. 2001. Positive fire-grass feedback in
Mediterranean Basin woodlands. Forest Ecology and Management 147: 3-14.
Vilà, M., and J. Weiner. 2004. Are invasive plant species better competitors than native
plant species? - evidence from pair-wise experiments. Oikos 105: 229-238.
Walck, J. L., J. M. Baskin, and C. C. Baskin. 1999. Effects of competition from
introduced plants on establishment, survival, growth and reproduction of the rare
plant Solidago shortii (Asteraceae). Biological Conservation 88: 213-219.
Wilcove, D. S., D. Rothstein, J. Dubow, A. Phillips, and E. Losos. 1998. Quantifying
threats to imperiled species in the United States. BioScience 48: 607-615.
52