Oecologia (1999) 120:605±612
Ó Springer-Verlag 1999
Melinda D. Smith á Alan K. Knapp
Exotic plant species in a C4-dominated grassland:
invasibility, disturbance, and community structure
Received: 9 February 1999 / Accepted: 12 May 1999
Abstract We used data from a 15-year experiment in a
C4-dominated grassland to address the e€ects of community structure (i.e., plant species richness, dominance)
and disturbance on invasibility, as measured by abundance and richness of exotic species. Our speci®c objectives were to assess the temporal and spatial patterns
of exotic plant species in a native grassland in Kansas
(USA) and to determine the factors that control exotic
species abundance and richness (i.e., invasibility). Exotic
species (90% C3 plants) comprised approximately 10%
of the ¯ora, and their turnover was relatively high (30%)
over the 15-year period. We found that disturbances
signi®cantly a€ected the abundance and richness of exotic species. In particular, long-term annually burned
watersheds had lower cover of exotic species than
unburned watersheds, and ®re reduced exotic species
richness by 80±90%. Exotic and native species richness
were positively correlated across sites subjected to different ®re (r=0.72) and grazing (r=0.67) treatments,
and the number of exotic species was lowest on sites with
the highest productivity of C4 grasses (i.e., high dominance). These results provide strong evidence for the role
of community structure, as a€ected by disturbance, in
determining invasibility of this grassland. Moreover, a
signi®cant positive relationship between exotic and native species richness was observed within a disturbance
regime (annually burned sites, r=0.51; unburned sites,
r=0.59). Thus, invasibility of this C4-dominated grassland can also be directly related to community structure
independent of disturbance.
Key words Community structure á Disturbance á Exotic
species á Grassland á Invasibility
M.D. Smith (&) á A.K. Knapp
Kansas State University,
Division of Biology,
Ackert Hall,
Manhattan, KS 66506, USA
e-mail: mendys@ksu.edu
Introduction
Many factors have been proposed as determinants of
invasibility of plant communities (Robinson et al. 1995;
Wiser et al. 1998), and the search for generalizations has
proven dicult (Stohlgren et al. 1999b). However, from
a number of anecdotal and empirical studies, several
hypotheses have emerged concerning the central role
that community structure (i.e., species richness, diversity, dominance) may have on invasibility. Elton (1958)
®rst proposed that species-poor communities would be
more invasible, because interspeci®c competition is not
as strong as in species-rich communities. In addition,
species-poor communities may have more empty niches
(Crawley 1987), allowing exotic species with characteristics not represented in the native ¯ora to establish
(Richardson and Bond 1991). Several studies have supported this hypothesis by demonstrating a negative relationship between native and exotic species richness
(Fox and Fox 1986; Rejmanek 1989; Woods 1993; Pysek
and Pysek 1995; Tilman 1997). In contrast, a positive
relationship between native and exotic species richness
has been noted as well (Pickard 1984; Knops et al. 1995;
Robinson et al. 1995; Planty-Tabacchi et al. 1996; Wiser
et al. 1998). Species-rich communities may be more
readily invaded than species-poor communities because
those factors that increase the establishment of native
species may also increase the establishment of exotic
species (Crawley 1987). Alternatively, the lack of dominance of a community by one or a few species (Robinson
et al. 1995), or higher spatial/temporal variability in
abiotic or biotic factors associated with diverse communities may increase opportunities for the establishment of exotic species (Tilman 1996; Wiser et al. 1998).
Although community structure may in¯uence invasibility, other factors, such as disturbances, may also be
important (Crawley 1987; Hobbs and Huennuke 1992;
Burke and Grime 1996). Disturbances such as grazing
may directly enhance the success of invasive species by
altering resource availability (Burke and Grime 1996;
606
McNaughton et al. 1997; but see Stohlgren et al. 1999a)
or they may alter biotic interactions and community
structure thereby indirectly a€ecting invasibility (Richardson and Bond 1991). Productivity is also altered by
disturbances and this may in¯uence invasibility, with
highly productive, low-diversity (high-dominance) sites
being more resistant to invasions than less productive
sites (Huston 1994; Burke and Grime 1996). Because
disturbances a€ect community structure in several ways,
it is dicult to separate direct from indirect e€ects on
invasibility.
There have been numerous studies of invasibility, but
most lack adequate replication, are short term, and do
not control for spatial scales (Burke and Grime 1996;
Wiser et al. 1998). Moreover, many have been conducted after invasive species have become substantial
components of communities (Burke and Grime 1996);
these studies are therefore complicated by the invading
species and their e€ects on community characteristics.
Long-term, replicated studies that integrate the e€ect of
disturbance and community structure on the di€erent
stages of the invasion process (e.g., establishment, persistence, and spread; Hobbs 1989) are needed to formulate generalizations regarding the factors that
in¯uence invasiblity of plant communities. Furthermore,
identifying these factors is critical for controlling present
invasions, and preventing or predicting future invasions
(Drake et al. 1989; Lodge 1993).
The general objectives of this study were to (1) assess
patterns of exotic plant species occurrence in a highly
productive C4 grassland and (2) determine the factors
that control invasibility, as measured by abundance and
richness of exotic species. We used data from a 15-year
experiment in a tallgrass prairie landscape that included
di€erent ®re frequency and grazing treatments to address the e€ects of community structure and disturbance
on invasibility. Speci®cally, we hypothesized that the
relationship between native and exotic plant species
richness would be positive for this grassland and that
exotic species richness would be negatively correlated
with productivity. These hypotheses were based on
previous studies that have shown that native plant species richness is lowest in the most productive sites, which
are generally those subjected to high ®re frequencies
(Briggs and Knapp 1995). On these sites, dominance, as
measured by productivity of grasses, is high (Hulbert
1988; Briggs and Knapp 1995); therefore, we expected
the abundance and richness of exotic species to be low
on these sites. In contrast, on unburned or grazed sites
native species richness is high (Collins et al. 1998); thus,
we expected higher exotic species richness on these sites.
Materials and methods
Research was conducted at the Konza Prairie Research Natural
Area (39°05¢ N, 96°35¢ W), a 3487-ha tallgrass prairie site in the
Flint Hills region of eastern Kansas, USA, the largest contiguous
tract of tallgrass prairie remaining in North America (Samson and
Knopf 1994). The ¯ora of Konza Prairie is representative of the
tallgrass prairie biome (see Freeman 1998 for a complete description) and is dominated by warm-season (C4) grasses, including
Andropogon gerardii and Sorghastrum nutans.
Since 1972, a split-plot experimental design incorporating watershed-level ®re frequency treatments (whole-plot treatment) has
been implemented at Konza Prairie. This design includes 60
watersheds (ca 60 ha) subjected to spring ®res (April 15 ‹ 20 days)
at annual, 2-, 4-, 10-, and 20-year intervals, each replicated two to
four times. All watersheds include distinct topographic gradients
(sub-plot treatment), with soil type and depth varying from uplands
to lowlands. Sites designated as the 20-year ®re frequency treatments are considered unburned when compared to other ®re frequency treatments. In fall 1987, 30 native ungulates, Bos bison, were
reintroduced into a 469-ha portion of Konza Prairie. As the size of
the herd increased to 200 bison, the total area grazed was expanded
to a 1012-ha portion of prairie. Replicate annual ®re frequency
treatments were also implemented at this time in the grazed portion
of Konza Prairie. Prior to initiation of the experimental design in
1972, Konza Prairie was grazed by domesticated ungulates (cattle)
and burned frequently (1- to 5-year intervals), which is typical
management of tallgrass prairie sites in the Flint Hills.
Species composition and aboveground biomass data collection
began in 1983 and 1984, respectively, for ungrazed watersheds and
in 1987 for grazed watersheds. Five permanent circular 10-m2 plots
were located adjacent to four randomly placed transects in upland
and lowland sites in each of two annual, 4-, and 20-year ®re frequency watersheds (n=20 plots/topographic position per watershed). For each circular plot, species composition was sampled
twice during the growing season, once in late May and in mid to
late August. Cover of each species was estimated using a modi®ed
Daubenmire (1959) method and maximum cover values for the
growing season were used in subsequent analyses. For estimates of
aboveground net primary production (ANPP), peak aboveground
biomass was harvested from 20 0.1-m2 quadrats, 5 located along
each transect (Briggs and Knapp 1995). Biomass was separated
into live (graminoid and forb) and dead components, dried, and
weighed. Total biomass production was calculated as the sum of
grass, forb, and current-year dead biomass.
To determine if ®re frequency or topographic position a€ected
exotic species composition, we focused our analyses on four ungrazed watersheds, two burned annually (26 and 20 ®res) and two
long-term unburned. Because ®re history varied considerably for
the 4- and 10-year ®re frequency watersheds, these were excluded
from this analysis. Analyses were possible for two time periods,
1983±1987 and 1993±1997, due to sampling and treatment history
constraints. The e€ects of treatments and temporal dynamics of
percent cover and richness of exotic plant species were analyzed
separately for each time period using repeated-measures mixedmodel analysis of variance (PROC MIXED; SAS 1997). Individual
transects (50 m2) were treated as observational units so that four
transects within a topographic position made up the experimental
unit (200-m2 plot). We used least-squares means analysis
(LSMEANS; SAS 1997) for comparisons of treatment means when
appropriate. In addition, we tested for correlative relationships
(i.e., Pearson's correlation analysis) between exotic plant species
richness and native species richness, total annual ANPP, or aboveground grass production for data collected from 1983±1987 and
1993±1997. E€ects of ®re history on exotic plant species richness
were evaluated using data collected in 1997 from ten ungrazed
watersheds with varying ®re frequencies, ranging from 26 ®res to no
®res over a 27-year period. Using transects as experimental units (50m2 plot), this relationship was analyzed using correlation analysis.
Since the current grazing regime on Konza Prairie has only
been in place since 1992 and ®re history for the grazed ®re frequency treatments di€ers considerably compared to the ungrazed
watersheds, we did not analyze grazing e€ects directly. Instead, we
evaluated the e€ects of native species richness, as in¯uenced by ®re
and grazing, on exotic species richness using correlation analysis.
This type of analysis allowed us to detect patterns of exotic species
richness in watersheds that varied considerably in ®re/grazing regimes and in native species richness.
607
Results
Temporal dynamics
There was no signi®cant increase or decrease in cover
or richness of exotic plant species during the 1983±1987
or 1993±1997 time periods. Average cover of exotic
species was 0.013% in annually burned and 22.8% in
unburned watersheds. Exotic species richness followed a
similar pattern with <1 species per 200 m2 in annually
burned sites and 3.8 species in unburned sites. Both
exotic species richness and cover exhibited higher interannual variability in the absence of burning, but remained stable with annual burning (data not shown).
In this C4-dominated grassland, exotic species typically
comprised ~9% of the native ¯ora present over the 15year study period with 17 exotic versus 169 native species in unburned, 18 versus 193 in ungrazed, and 21
versus 230 in grazed sites. In annually burned prairie,
exotic species were least common (9 exotic vs 138 native)
and those exotic species found on annually burned or
ungrazed sites were a subset of those found on unburned
or grazed sites (Table 1). Most of the 22 exotic species
encountered are C3 plants (90%) introduced from either
Europe or Eurasia and are considered to be naturalized,
i.e., introduced species able to reproduce and persist
from year to year (Brooks 1986). Records indicate that
10 of these species were present in the Flint Hills Region
prior to 1892, while the remainder were introduced some
time after 1892 or more recently (after 1930) (Table 1).
These species represent ten families, including the
Poaceae (7 species), Asteraceae (3) and Fabaceae (3)
(Table 1). Exotic species occurred more frequently in
unburned or grazed sites than in annually burned or
ungrazed sites (Table 1). Overall, many of the exotic
species occurred only once or a few times (i.e., <5%
frequency) during the study period.
The total cover of exotic plant species was signi®cantly
higher in unburned than in annually burned watersheds
(mixed-model ANOVA: F=24.11, F=31.08, P < 0.05
for 1983±1987 and 1993±1997, respectively; Fig. 1a).
The increased cover of exotic species did not appear to
be related to decreased cover of native species (Fig. 1a).
Exotic species richness was also higher on unburned
sites, which paralleled higher native species richness on
these sites (mixed-model ANOVA: F=20.91, F=131.91,
P=0.01, respectively; Fig. 1b). For the analysis of ®re
history, the number of ®res over a 27-year period for
ungrazed sites ranged from 0 to 26. Fire frequency was
negatively correlated with the number of exotic species
found in plots in 1997 (Fig. 2). Thus, the cumulative
Table 1 Exotic species found during the 15-year study period in
200-m2 plots located in burned, unburned, ungrazed, and grazed
watersheds at Konza Prairie, Kansas (USA). Grazed and ungrazed
watersheds had variable ®re histories. The percent frequency for
each treatment was determined by recording the number of times
each species was present during the study period and dividing by
the total number of plots sampled a (Life form: A = annual,
B = biennial, A/B = annual or biennial, P = perennial; frequency <5% indicates that a species occurred less than three times
over the study period)
Family
Amaranthaceae
Asteraceae
Brassicaceae
Chenopodiaceae
Convolvulaceae
Fabaceae
Lamiaceae
Poaceae
Polygonaceae
Scrophulariaceae
a
Species
Amaranthus retro¯exusb
Lactuca serriolaa
Taraxacum ocinalea
Tragopogon dubiusc
Lepidium densi¯orumb
Thlapsi arvenseb
Chenopodium albuma
Convolvulus arvensisa
Lespedeza stipulaceac
Melilotus albaa
Melilotus ocinalisb
Lamium amplexicauleb
Bromus inermisb
Bromus japonicusb
Bromus tectorumb
Poa compressac
Poa pratensisa
*Setaria glaucaa
*Setaria viridisa
Rumex crispusa
Verbascum blattariaa
Veronica agrestisc
Fire and ®re history
Life form
A
A
P
B
A/B
A
A
P
A
A/B
A/B
A
P
A
A
P
P
A
A
P
B
A
Frequency (%)
Burned
Unburned
Ungrazed
Grazed
±
24.0
<5
28.0
12.0
±
±
±
±
±
<5
±
20.0
14.0
<5
±
34.0
±
±
±
±
±
<5
90.0
10.0
37.5
32.5
<5
<5
15.0
±
<5
±
±
27.5
25.0
10.0
<5
100.0
±
<5
±
<5
<5
<5
53.3
6.7
32.2
21.1
<5
<5
6.7
±
<5
<5
±
23.3
18.9
5.6
<5
63.3
±
<5
±
<5
<5
±
47.5
<5
45.0
85.0
7.5
22.5
15.0
15.0
±
<5
<5
72.5
72.5
<5
±
87.5
17.5
±
<5
17.5
±
Present in the Flints Hills prior to 1892 (Smyth 1892)
Present in the Flint Hills after 1892 (Barker 1968)
c
Present in the Flint Hills after 1930 (McGregor 1976; Great Plains Flora Association 1986)
* C4 species
b
608
e€ects of ®re signi®cantly reduced the presence of exotic
plant species in this grassland.
Productivity and topography
Exotic species richness was negatively correlated with
total ANPP and total aboveground grass production
(Fig. 3). Richness of exotic plant species was also affected by topographic position. Although not signi®cant
for the 1983±1987 time period, exotic species richness
was higher in upland sites (with lower ANPP; Briggs and
Knapp 1995) compared to lowland sites for the 1993±
1997 time period (mixed-model ANOVA: F=8.24,
P=0.0173).
Native versus exotic species richness
For annually burned and unburned watersheds that
were not grazed, native species richness was positively
correlated with exotic species richness over the 15-year
study period (Fig. 4a). Disproportionate sampling of
sites with either low or high richness did not cause this
relationship. The distribution of richness in the sampled
plots was normal, most of the plots having between 45
and 65 native plant species, which approximates the
average richness for all plots sampled (Fig. 4a inset).
Although fewer plots fell within the low or high ends of
this distribution, these plots consistently had either zero
or the highest number of exotic species. There was also a
Fig. 1 Total cover (%) of native and exotic plant species in annually
burned and unburned grassland at Konza Prairie (a). Total richness
(per 200 m2) of native and exotic plant species in watersheds that
di€ered in ®re frequency (b). For each ®re frequency treatment, data
were averaged for two time periods (1983±1987 and 1993±1997) which
had consistent treatment and sampling history. Error bars represent
+1 SE of the mean and asterisks indicate signi®cant (P < 0.05)
di€erences between ®re treatments within each time period
Fig. 2 Relationship between number of ®res in the last 27 years and
exotic plant species richness (per 200 m2) for ungrazed watersheds at
Konza Prairie. Data are for 1997. Error bars represent ‹1 SE of the
mean
Fig. 3 a Relationship between total aboveground net primary
production and exotic plant species richness (per 200 m2) for burned
and unburned watersheds (all ungrazed) on Konza Prairie over the
15-year study period. b Relationship between grass biomass production and exotic plant species richness (per 200 m2) for burned and
unburned watersheds (all ungrazed). Error bars represent ‹1 SE of
the mean
609
respectively, were annually burned, and exotic species
richness increased approximately twofold in these sites.
Discussion
Fig. 4 a Relationship between native and exotic plant species richness
(per 200 m2) for burned and unburned watersheds (all ungrazed) at
Konza Prairie over the 15-year study period. Inset: distribution of
annually burned and unburned plots in di€erent native species
richness classes. b Relationship between native and exotic plant species
richness (per 200 m2) for burned and unburned watersheds subjected
to di€erent grazing treatments. Inset: distribution of grazed and
ungrazed plots in di€erent native species richness classes. Error bars
represent ‹1 SE of the mean
consistent relationship between ®re frequency and the
number of native and exotic plant species; annually
burned plots had the lowest native and exotic species
richness (native range: 25±65), whereas unburned plots
had the highest native and exotic species richness (native
range: 45±85).
Native species richness was also found to be positively correlated with exotic species richness when
grazing treatments, both burned and unburned, were
included in the analysis (Fig. 4b). The frequency distribution of plots sampled was normal and the highest
number of sampled plots fell within the 55±65 native
species class, which approximates the mean for all plots
sampled (Fig. 4b inset). Grazing increased both native
and exotic species richness in annually burned sites. In
the absence of grazing, there were no annually burned
plots that fell within the 65±75 or 75±85 native species
classes. However, with grazing, 26% and 17% of the
plots sampled in these native species richness classes,
The 22 exotic plant species found in this C4-dominated
grassland comprised less than 10% of the native ¯ora on
the study sites, and approximately 30% of these exotic
species occurred less that three times during the 15-year
study period. These results indicate that while turnover
(i.e., local extinction) of some exotic species was relatively high, the majority of exotics are able to persist
from year to year. As well as being temporally dynamic,
these exotic species were not evenly distributed across
the landscape. Patterns of disturbance strongly a€ected
exotic species cover and richness over the 15-year study
period. In particular, long-term annually burned sites
had low cover and few, if any, exotic species, whereas
richness and cover of exotic species was as much as ®ve
times higher in long-term unburned sites (Figs. 1, 2). In
grasslands, disturbances such as ®re and grazing create
sites with a wide range of productivity and native species
richness, both of which vary temporally and spatially
(McNaughton 1985; Frank et al. 1998; Knapp et al.
1998). This range in productivity and native species
richness was correlated with exotic species richness in
this grassland. The sites with lowest productivity and
highest native species richness had the highest number of
exotic species (Figs. 3, 4). These results indicate that
disturbance in¯uences invasibility, as measured by the
abundance and richness of exotic species.
Fire and grazing by native ungulates (B. bison) separately and together strongly in¯uence tallgrass prairie
community structure (Collins et al. 1998; Knapp et al.
1999). Therefore, it is not surprising that these disturbances a€ected exotic species. Annually burned watersheds had lower cover and richness of exotic species than
unburned watersheds (Figs. 1, 4) and, although the effects of grazing could not be tested directly, higher exotic
species richness was associated with both annually
burned and unburned grazing treatments (Fig. 4). Invasibility of grasslands and other plant communities has
been shown to be disturbance-related (Rejmanek 1989;
Robinson et al. 1995; Kotanen et al. 1998; Stepanian
et al. 1998) with changes in disturbance intensity or
frequency often a prerequisite for establishment of exotic species (Crawley 1987; Richardson and Bond 1991;
Hobbs and Huenneke 1992; Burke and Grime 1996).
Disturbance can enhance invasibility by allowing for
greater coexistence of competitors (Lodge 1993), which
may result from a number of interacting factors, such as
altered productivity, dominance, or increased heterogeneity of resources or microsites.
Fire increases biomass production and abundance of
the dominant C4 grasses and decreases production and
abundance of the subdominant grasses and C3 forbs in
this grassland (Briggs and Knapp 1995; Hartnett and
Fay 1998). Our analysis revealed that ANPP was
610
strongly negatively related to exotic species richness
(Fig. 3). Peart and Foin (1985) found that invasion
success was negatively related to biomass, but not to
richness or composition of native species. We also observed a negative relationship between total aboveground grass biomass and exotic species richness. Total
ANPP is comprised mainly (i.e., >60%) of C4 grass
biomass (Hulbert 1988; Briggs and Knapp 1995).
Therefore, community dominance, measured as production of C4 grasses, appears to negatively a€ect
abundance and richness of exotic plant species.
Dominance has been reported to in¯uence establishment and richness of exotic species in other plant communities as well (DeFerrari and Naiman 1994; Burke
and Grime 1996). Establishment and persistence of
invaders may be prevented because one to a few species
are able to physically exclude invaders or compete more
e€ectively for limiting resources (Robinson et al. 1995).
In this productive C4 grassland, ®re alters the availability of nitrogen, moisture, and light. Because resource
availability is reduced after ®re, interspeci®c competition
for resources is likely higher in burned than in unburned
grassland (Knapp et al. 1998). Moreover, plant density
is much higher in burned than unburned sites (Knapp
1985; Hulbert 1988), increasing competition by the
dominant C4 grasses, and reducing diversity and richness of subdominant grasses and forbs (Collins et al.
1998). Thus, annual burning, by increasing the dominance (i.e., production) of C4 grasses, may indirectly
prevent establishment or persistence of exotic species.
Although these results demonstrate that ®re reduces
the presence of exotic species in this productive grassland, the strong positive relationship between native and
exotic species richness in grazed sites (regardless of ®re
treatment) suggests that grazing may be a more important determinant of invasibility. By reducing the cover
and dominance of the C4 grasses, grazing by B. bison
increases overall richness and diversity of native species
due to competitive release of the subdominant forb
species and increased spatial heterogeneity (Collins et al.
1998; Knapp et al. 1999). These factors may increase the
abundance and richness of exotic species as well because
more species can potentially coexist in sites with greater
microsite diversity and reduced competition (Tilman
1996; Palmer and Maurer 1997; Wiser et al. 1998).
Moreover, grazing, by reducing the dominance of C4
grasses, appears to increase the invasibility of sites that
would otherwise have low abundance and richness of
exotic species. This was demonstrated by an increase in
the number of exotic species in annually burned watersheds that were grazed (Fig. 4).
In contrast to the grazing e€ects observed in this
study, in shortgrass steppe, the abundance of exotic
species was highest on ungrazed sites (Milchunas et al.
1989). Di€erences in community characteristics between
the two grasslands can help explain this inconsistency.
Milchunas et al. (1989) suggested that lack of grazing
was more of a disturbance in shortgrass steppe, which
has evolved under heavy grazing pressures. Grazing in
shortgrass steppe increases productivity, cover, and
dominance of the C4 grasses, thereby reducing abundance of subdominant forb species, whereas grazing in
tallgrass prairie reduces the dominance of the C4 grasses.
These ®ndings emphasize the importance of community
dominance, as measured by production of a few species,
rather than disturbance in controlling invasibility in
di€erent plant communities.
In general, exotic species richness in this C4-dominated grassland was positively related to native species
richness (Fig. 4) supporting the hypothesis that speciesrich communities are more invasible than species-poor
communities (Wiser et al. 1998). Others have demonstrated a similar relationship between invasibility and
community structure for plant communities ranging
from grasslands to forests (Pickard 1984; Knops et al.
1995; Robinson et al. 1995; Planty-Tabacchi et al.
1996). The opposite relationship (i.e., greater invasion
susceptibility of species-poor communities) has been
observed for several shrub and forest communities (Fox
and Fox 1986; Woods 1993), a wide range of habitats in
the Czech Republic (Pysek and Pysek 1995), and oak
savanna (Tilman 1997). The latter relationship may be
due to a negative in¯uence of the invader(s) on native
species richness in the post-invasion community (Wiser
et al. 1998) or may be a consequence of using measures
of invasibility (e.g., exotic species richness expressed as a
proportion of native species richness) that are highly
dependent upon initial native species richness (Rejmanek 1989).
Although the relationship between grassland community structure and invasibility is strong, it is dicult to
separate the direct e€ects of community structure from
the e€ects of disturbances that alter the dominance (i.e.,
production of C4 grasses) and richness of native plant
species. In other words, rather than community structure
itself in¯uencing invasibility, disturbance may directly
a€ect invasibility of plant communities independent of
e€ects on community structure. For example, grazing
leads to increased N availability (McNaughton et al.
1997), whereas ®re reduces available N in grasslands
(Blair 1997). Wedin and Tilman (1996) noted that the
abundance of exotic species in grasslands increased after
N fertilization. This suggests that increased N in grazed
grassland could be an alternative mechanism enhancing
invasibility (Stohlgren et al. 1999a). To eliminate these
e€ects of disturbance, we conducted correlation analyses
to determine if community structure-invasibility patterns
observed among watersheds (and treatments) would be
evident within a disturbance regime. Within annually
burned and unburned watersheds, there was still a strong
positive relationship between native and exotic species
richness (r=0.51 and r=0.59, respectively; both
P < 0.05), supporting the role that community structure
(i.e., native species richness) plays in determining the
invasibility of this grassland.
In conclusion, results from this study support the
hypothesis that species-rich communities are more invasible than species-poor communities. At the landscape
611
level, invasibility can be a€ected by disturbances directly
and indirectly through altered community structure. At
smaller spatial scales (i.e., within watersheds), community structure in¯uences invasibility independent of disturbance. These results should be viewed with some
caution because the exotic species encountered in this
study comprise a minor component of the vegetation
and do not appear to be increasing in abundance over
time, and therefore are not comparable to those exotic
species (i.e., invaders) that establish in an intact community, dominate or displace native vegetation, and
alter ecosystem function (Vitousek et al. 1997). Despite
these limitations, these ®ndings may have important
implications for preventing or predicting future invasions, because they provide insight into factors in¯uencing the persistence of exotic species. For example, we
would predict that more aggressive invaders which
currently threaten these grasslands, such as A. bladhii
(Caucasian bluestem) and Lespedeza sericea (sericea lespedeza), would be more likely to establish and persist in
areas of high diversity or those subjected to grazing.
Furthermore, grasslands and other ecosystems are
threatened by fragmentation as well as increased N deposition (Wedin and Tilman 1996; Vitousek et al. 1997),
both of which may increase invasion by exotic species.
Leach and Givnish (1996) suggested that fragmented
grasslands should be burned frequently to reduce the
risk of invasions. Since frequent burning also can reduce
biodiversity, Knapp et al. (1999) suggested that incorporating a range of natural disturbance regimes (burning and grazing) might prevent the loss of native species.
However, results from this long-term study indicate that
grazing, regardless of ®re frequency treatment, may
increase the risk of invasion of this C4-dominated
grassland by exotic species. Therefore, managing for
maximum diversity (i.e., enhanced ecosystem stability
and processes; Tilman 1996) in these grasslands may
lead to greater establishment and persistence of exotic
species over time.
Acknowledgements We are grateful to Marc Abrams, David
Gibson, Gene Towne, and Scott Collins for species composition
data collection. We also thank Je€rey Pontius, Department of
Statistics, Kansas State University, for assistance with the repeated-measures mixed-model analyses. Research was supported
by the NSF Long Term Ecological Research Program, the Konza
Prairie Research Natural Area, and The Nature Conservancy.
This paper is contribution No. 00-1-J from the Kansas Agricultural Experiment Station, Kansas State University, Manhattan,
KS. We thank Russell Monson and two anonymous reviewers for
providing helpful comments and suggestions on earlier versions of
the manuscript.
References
Barker WT (1968) The ¯ora of the Kansas Flint Hills. PhD dissertation, University of Kansas, Lawrence, KS
Blair JM (1997) Fire, N availability and plant response in grasslands: a test of the transient maxima hypothesis. Ecology
78:2539±2368
Briggs JM, Knapp AK (1995) Interannual variability in primary
production in tallgrass prairie: climate, soil moisture, topographic position and ®re as determinants of aboveground biomass. Am J Bot 82:1024±1030
Brooks RE (1986) Vascular plants of Kansas: a checklist. Kansas
Biological Survey, Lawrence, KS
Burke MJW, Grime JP (1996) An experimental study of plant
community invasibility. Ecology 77:776±790
Collins SL, Knapp AK, Briggs JM, Blair JM, Steinauer EM (1998)
Modulation of diversity by grazing and mowing in native tallgrass prairie. Science 280:745±747
Crawley MJ (1987) What makes a community invasible? In:
Crawley MJ, Edwards PJ, Gray AJ (eds) Colonization, succession and stability. Blackwell, Oxford, pp 429±454
Daubenmire RF (1959) A canopy-coverage method of vegetational
analysis. Northwest Sci 33:43±66
DeFerrari CM, Naiman RJ (1994) A multi-scale assessment of the
occurrence of exotic plants on the Olympic Peninsula, Washington. J Veg Sci 5:247±258
Drake JA, Mooney HA, Castri F di, Groves RH, Kruger FJ,
Rejmanek M, Williamson M (1989) Biological invasions: a
global perspective (Scope 37). Wiley, Chichester
Elton CS (1958) The ecology of invasions by animals and plants.
Methuen, London
Fox MD, Fox BJ (1986) The susceptibility of natural communities
to invasion. In: Groves RH, Burdon JJ (eds) Ecology of biological invasions: an Australian perspective. Australian Academy of Science, Canberra, pp 57±66
Frank DA, McNaughton SJ, Tracy BF (1998) The ecology of the
Earth's grazing ecosystems. BioScience 48:513±521
Freeman CC (1998) The ¯ora of Konza Prairie: a historical review
and contemporary patterns. In: Knapp AK, Briggs JM, Hartnett DC, Collins SC (eds) Grassland dynamics: long-term ecological research in tallgrass prairie. Oxford University Press,
New York, pp 69±80
Great Plains Flora Association (1986) Flora of the Great Plains.
University Press of Kansas, Lawrence, KS
Hartnett DC, Fay PA (1998) Plant populations: patterns and
processes. In: Knapp AK, Briggs JM, Hartnett DC, Collins SC
(eds) Grassland dynamics: long-term ecological research in
tallgrass prairie. Oxford University Press, New York, pp 81±
100
Hobbs RJ (1989) The nature and e€ects of disturbance relative to
invasions. In: Drake JA, Mooney HA, Castri F di, Groves RH,
Kruger FJ, Rejmanek M, Williamson M (eds) Biological invasions: a global perspective (Scope 37). Wiley, Chichester, pp
389±405
Hobbs RJ, Huenneke LF (1992) Disturbance, diversity, and
invasion: implications for conservation. Conserv Biol 6:324±
337
Hulbert LC (1988) Causes of ®re e€ects in tallgrass prairie. Ecology
69:46±58
Huston MA (1994) Biological diversity: the coexistence of species
in changing landscapes. Cambridge University Press, Cambridge, UK
Knapp AK (1985) E€ect of ®re and drought on the ecophysiology
of Andropogon gerardii and Panicum virgatum in a tallgrass
prairie. Ecology 66:1309±1320
Knapp AK, Briggs JM, Blair JM, Turner CL (1998) Patterns and
controls of aboveground net primary production in tallgrass
prairie. In: Knapp AK, Briggs JM, Hartnett DC, Collins SC (eds)
Grassland dynamics: long-term ecological research in tallgrass
prairie. Oxford University Press, New York, pp 193±221
Knapp AK, Blair JM, Briggs JM, Collins SL, Hartnett DC,
Johnson LC, Towne FG (1999) The keystone role of bison in
North American tallgrass prairie. BioScience 49:39±50
Knops JMH, Grin JR, Royalty AC (1995) Introduced and native
plants of the Hastings reservation, central coastal California: a
comparison. Biol Conserv 71:115±123
Kotanen PM, Bergelson J, Hazlett DL (1998) Habitats of native
and exotic plants in Colorado shortgrass steppe: a comparative
approach. Can J Bot 76:664±672
612
Leach MK, Givnish TJ (1996) Ecological determinants of species
loss in remnant prairies. Science 273:1555±1561
Lodge DM (1993) Biological invasions: lessons for ecology. Trends
Ecol Evol 8:133±137
McGregor RL (1976) History of naturalized Kansas plants. Reports
of the State Biological Survey of Kansas (no 7), Lawrence, KS
McNaughton SJ (1985) Ecology of a grazing ecosystem: the
Serengeti. Ecol Monogr 55:259±295
McNaughton SJ, Banyikwa FF, McNaughton MM (1997) Promotion of the cycling of diet-enhancing nutrients by African
grazers. Science 278:1798±1800
Milchunas DG, Lauenroth WK, Chapman PL, Kazempour MK
(1989) E€ects of grazing, topography, and precipitation on the
structure of a semi-arid grassland. Vegetatio 80:11±23
Palmer MW, Maurer T (1997) Does diversity beget diversity? A
case study of crops and weeds. J Veg Sci 8:235±240
Peart DR, Foin TC (1985) Analysis and prediction of population
and community change: a grassland case study. In: White J (ed)
The population structure of vegetation, W. Junk, The Hague,
pp 313±339
Pickard J (1984) Exotic plants on Lord Howe Island: distribution
in space and time, 1853±1981. J Biogeogr 11:181±208
Planty-Tabacchi A, Tabacchi E, Naiman RJ, DeFerrari C,
Decamps H (1996) Invasion by Heracleum mantegazzianum in
di€erent habitats in the Czech Republic. J Veg Sci 4:329±340
Pysek P, Pysek A (1995) Invasion by Heracleum mantegazzianum
in di€erent habitats in the Czech Republic. J Veg Sci 6:711±
718
Rejmanek M (1989) Invasibility of plant communities. In:
Drake JA, Mooney HA, Castri F di, Groves RH, Kruger
FJ, Rejmanek M, Williamson M (eds) Biological invasions:
a global perspective (Scope 37). Wiley, Chichester, pp 369±
388
Richardson DM, Bond WJ (1991) Determinants of plant distribution: evidence from pine invasions. Am Nat 137:639±668
Robinson GR, Quinn JF, Stanton ML (1995) Invasibility of experimental habitat islands in a California winter annual grassland. Ecology 76:786±794
Samson F, Knopf F (1994) Prairie conservation in North America.
BioScience 44:418±421
SAS (1997) SAS/STAT software: changes and enhancements
through release 6.12. SAS Institute, Cary, NC
Smyth BB (1892) Check list of the plants of Kansas. Topeka, KS
Stepanian MA, Sundberg SD, Baumgardner GA, Liston A (1998)
Alien plant species composition and associations with anthropogenic disturbance in North American forests. Plant Ecol
139:49±62
Stohlgren TJ, Schell LD, Heuvel BV (1999a) How grazing and soil
quality a€ect native and exotic plant diversity in rocky mountain grasslands. Ecol Appl 9:45±64
Stohlgren TJ, Binkley D, Chong GW, Kalkhan MA, Schell LD,
Bull KA, Otsuki Y, Newman G, Bashkin M, Son Y (1999b)
Exotic plant species invade hot spots of native plant diversity.
Ecol Monogr 69:25±46
Tilman D (1996) Biodiversity: population versus ecosystem stability. Ecology 77:350±363
Tilman D (1997) Community invasibility, recruitment limitation,
and grassland biodiversity. Ecology 78:81±92
Vitousek PM, D'Antonio CM, Loope LL, Rejmanek M, Westbrooks R (1997) Introduced species: a signi®cant component of
human-caused global change. N Z J Ecol 21:1±16
Wedin DA, Tilman D (1996) In¯uence of nitrogen loading and
species composition on the carbon balance of grasslands. Science 274:1720±1723
Wiser SK, Allen RB, Clinton PW, Platt KH (1998) Community
structure and forest invasion by an exotic herb over 23 years.
Ecology 79:2071±2081
Woods KD (1993) E€ects of invasion by Lonicera tatarica L. on
herbs and tree seedlings in four New England forests. Am Midl
Nat 130:62±74