Many communities are characterized by uneven distributions
of species. Understanding the processes underlying these
patterns and their implications for community dynamics and
ecosystem function is a central endeavor of ecology. Often
in communities only a few species are very common (or dominant),
whereas a majority of species occur at moderate or low abundance
(subordinate or rare). Dominant species generally garner
a disproportionate share of resources, contribute most to
productivity and other ecosystem functions, and are consistently
present in the community over time. Rare and uncommon species,
on the other hand, are collectively the most diverse component
of the community, but generally contribute less to ecosystem
functioning (although there are exceptions, e.g., legumes)
and often experience high levels of turnover.
The loss of biota from communities has prompted ecologists
to examine the consequences of reduced biodiversity for
ecosystem function and stability. Recent experimental and
theoretical studies have demonstrated that random loss of
species leads to declines in key ecosystem functions (productivity,
nutrient retention, resistance to invasion). However, such
patterns of loss do not necessarily reflect those in natural
communities, where major drivers of change are widespread.
These stressors, rather than extinguishing species at random,
most likely result in the loss of rare and uncommon species
first with more common (dominant) species less likely to
be lost by virtue of having larger population sizes. Thus,
the natural pattern of loss is likely to be non-random and
ecosystem responses to this pattern of loss should largely
be influenced by the persistent common and dominant species.
In a recent study in tallgrass prairie (Smith and Knapp
2003), we showed that this scenario of non-random loss based
on the relative abundance of species in the community had
no detectable effect on a key measure of ecosystem function
(productivity). This contrasts sharply with other studies
that have found that less diverse communities have reduced
ecosystem function. Instead, we found that largest impact
was when abundance of dominant species was reduced. Thus,
we show that dominant species, as controllers of ecosystem
function, provide short-term resistance to loss in ecosystem
function with non-random species loss. This suggests that
future studies should focus on the role of dominant species
in community and ecosystem processes.
I am expanding upon this research by examining the role
that dominant species play in the stability of ecosystem
function and the assembly of communities. For the former,
I and a collaborator at the University of California Santa
Barbara are examining the way in which dominant species
influence compensation (i.e., the ability of the system
to resist change and maintain function with disturbance
or other environmental changes) and thus community and ecosystem
stability (Smith and Adler in prep). For the latter,
I plan to revisit the experimental plots in which species
were removed non-randomly to examine patterns of re-colonization
over time. This will provide insights into how dominant
species influence colonization and community assembly over
time.
Research in tallgrass prairie suggests that dominant
species (i.e., a few common C4 grasses) can strongly
influence diversity, and it is through the impacts of disturbances
on their abundance that variation in community attributes
and ecosystem processes in space and time are observed.
Moreover, my research indicates that these dominant species
play a central role in the maintenance of ecosystem function
(Smith and Knapp 2003), and that the dominant species are
important in determining resistance of communities to invasion
(Smith and Knapp 1999, Smith et al. 2004).
Because of the key role dominant species appear to play
as drivers of community dynamics and ecosystem function,
my research seeks to gain a more general understanding their
role in ecosystems by melding genetic and ecological information.
I and collaborators at Colorado State University, Kansas
State University, and University of Kansas are currently
examining how Andropogon gerardii, the dominant grass
species in tallgrass prairie, influences community characteristics
(diversity) and ecosystem function (productivity). Our approach
links genetic variation in A. gerardii to ecological
responses of this species to environmental changes (disturbance,
altered nutrients). In the future, I plan to expand this
research to examine how genetic diversity in A. gerardii
impacts community and ecosystem processes using both observational
(examine patterns of diversity) and experimental approaches
(manipulation of diversity).
back to research main page
2) What are the causes and consequences of exotic plant
species invasions?
Communities vary in space and time as a result of a complex
array of abiotic and biotic factors. Understanding the relative
importance of different factors and how they interact is
the focus of a considerable amount of ecological research.
Colonization by native species or invasion by novel (exotic,
non-native, or alien) species represent key biotic drivers
of community dynamics. Invasions are of particular importance
because few ecosystems remain free from their threat, yet
our understanding of the causes and consequences of invasion
are poorly understood.
One approach to understanding invasion is to focus on characteristics
of the invaded community, particularly of the role of diversity
in influencing invasion. Theory predicts that diverse communities
are more resistant to invasion as a result more complete
use of resources and fewer resource opportunities for invaders.
Recent experiments provide support for more diverse communities
being more resistant to invasion, however a number of observational
studies have found an opposite pattern or lack of relationship
between diversity and invasion.
In a recent experiment (Smith et al. 2004), I and collaborators
sought to resolve these conflicting patterns by manipulating
community richness as well as the abundance of the most
dominant species. In contrast to other experiments where
diversity and presence of dominant species often covary,
we were able to show that when diversity and dominance were
separately manipulated diversity had no effect on invasion.
Instead, dominance was the key community characteristic
determining plant invasions. Depending on the situation,
these highly competitive, space-filling species – in this
case, a few C4 grasses – either created a more
competitive environment or alleviated stressful conditions,
and thus had an overriding effect on invasion. Currently,
I and a collaborator at the University of California Santa
Barbara are developing a conceptual framework for understanding
the role of diversity and resource variability in determining
invasion (Smith and Sax in prep).
back to research main page
3) What are the impacts of climate change (increased
temperature, altered timing in rainfall) on community and
ecosystem structure and function?
Invasion, an important global change in its own right,
will most certainly interact with other aspects of human-caused
global change, such as climate change. For many regions,
climate is predicted to be impacted by human activities
in two ways: through altered timing in precipitation and
increased temperature. These changes may facilitate invasion
by exotic plants by altering resource supply and demand.
Thus, current understanding of invasion will likely be further
complicated by predicted climate changes.
 |
 |
Far left: Aerial
view of the Rainfall Manipulation Plot (RaMPs) experiment
at the Konza Prairie Biological Station.
Near left: View of
RaMPs with IR heating lamps. |
Community theory predicts that invasion should occur only
when there is a niche opportunity. These can result from
an enemy escape or resource opportunity. Resource opportunities
arise in two ways: 1) directly due to an increase in resource
supply beyond that required by resident species, or 2) indirectly
due to a decline in resource use by resident species. An
increase in resource supply or decrease in demand may result
from a variety of mechanisms, including fluctuation in climate,
disturbance or resource enrichment. Changes in the mean
and variability in climate may lead to pulses in water and
N availability that could alter resources opportunities
for invaders at different times of during the growing season
and different stages of the invasion process.
As part of a larger, long-term experiment in tallgrass
prairie (the Rainfall Manipulation Plots, RaMPs, http://www.konza.ksu.edu/ramps)
at the Konza Prairie
Biological Station examining community and ecosystem
responses to experimental warming (+1-2°C)
and more extreme precipitation patterns (30% increase in
timing between rainfall events, but no change in total precipitation),
I am examining the effects of these climate changes on invasion
by several exotic plant species over time. With this experiment,
I hope to gain a more detailed mechanistic understanding
of invasion and climate changes may impact invisibility
of ecosystems. In the future, I hope to expand these studies
to other systems (old fields in New England) as well as
incorporate other trophic levels to examine the impacts
of climate change and trophic interactions on invasions.
back to research main page
4) How do genomic responses (i.e., whole genome responses)
link to community and ecosystem responses to climatic change?
Although there remains uncertainty as to
the rate and magnitude of climate change, it is clear that
human-caused changes in climate are already occurring and
will continue into the next century. Hence there is pressing
need to understand and predict the consequences of present
and projected climate changes on ecological systems. Predicting
the responses of ecosystems to climate change requires scaling
up from key mechanisms, such as photosynthesis or growth
that are best understood at the organism level. These mechanisms
are fundamentally linked to genes, gene networks, and their
interplay with the environment. However, our understanding
of the interplay between genes and ecological processes
at levels beyond the individual organism is nonexistent.
Thus, a comprehensive, mechanistic understanding of ecosystem
responses to climate change requires that responses at the
organism level (i.e., individual plant) be directly related
to responses of the genome, and these in turn be linked
to higher levels of organization.
With current technological capabilities in
genomic science, we can directly assess gene expression
at a genome-wide scale using microarray technology (cDNA).
When related to different experimental treatments, an integrated
and ‘global’ view of organism responses to the environment
is the result. I and collaborators from Kansas State University,
Colorado State University, and the University of Minnesota-Duluth
plan to link genomic and ecological approaches by using
the functional genomic information derived from model organisms
to link genomic responses of two closely related and ecologically
important species in a grassland system to ecological responses
to climate change.
This project will take advantage of an ongoing
climatic change experiment in natural grassland - the Rainfall
Manipulation Plots (the Rainfall Manipulation Plots, RaMPs,
http://www.konza.ksu.edu/ramps)
at the Konza Prairie
Biological Station in northeastern Kansas. The RaMPs
experimental infrastructure examines two key, predicted
environmental changes associated with energy production:
(1) increased temperature (1-2 degrees C warming), and (2)
more variable precipitation regimes, specifically increased
time between and intensity of rainfall events. After 6 years
of experimentally increased rainfall variability in the
experiment, impacts at multiple levels of biological organization
have been observed. Warming treatments initiated in 2003
are expected to exacerbate the effects of precipitation
variability.
Now, within this backdrop of known and predicted
ecological responses, this project is poised to gain a more
detailed mechanistic and predictive understanding by explicitly
linking species-level genomic data to fundamental plant
physiological responses. By focusing efforts on two dominant
C4 grasses, Andropogon gerardii and Sorghastrum
nutans that are closely related to a model genomic organism
(Zea mays), known to respond differentially to altered
precipitation, and whose responses strongly influence community
and ecosystem characteristics, the project will attempt
to directly and relevantly scale genome-level responses
to ecological processes observed at the plant and plant
population levels, as well as to those emergent responses
at the community and ecosystem levels.
Specifically, a flexible tiered sampling scheme
that will involve intensive, highly replicated sampling
to assess independent and interactive effects of the temperature/precipitation
manipulations will be used, whereas more explicit analytical
foci are planned for particular climatic events within and
among seasons. Information on the relative changes in gene
expression determined with microarray and real-time PCR
(polymerase chain reaction) technologies will be collected
concurrently with a suite of physiological variables to
identify genes or gene clusters related to photosynthesis,
water stress, and/or heat tolerance that are consistently
up- or down-regulated in response to the experimental manipulations
in the target species. These then will be scaled to the
emergent community and ecosystem level responses based on
differential responses of the individual species. This research
will bridge a fundamental divide between two disciplines
in biology traditionally focusing on divergent domains of
inference, strengthening both fields by developing and testing
an integrative approach to studying potential effects of
climatic change on the structure and functioning of terrestrial
ecosystems.
back to research main page
5) Do ecosystem and community processes in grasslands
respond to fire and grazing in fundamentally different ways
in South Africa vs. North America?
In ecology, we strive to gain a general understanding
of patterns and processes. Such understanding is critical
for predicting future responses of systems to environmental
changes. There are two ways in which we can generalize:
through synthesis and through the formulation and testing
of ecological rules. Synthesis involves combining diverse
results into coherent generalities, while ecological rules
require recognition of general principles that underpin
and create pattern. In my research, I have used synthesis
to gain a more general understanding of variation in productivity
among biomes in response to precipitation (Knapp and Smith
2001, Huxman et al. 2004), patterns of invasion among biomes
(Smith and Dov in prep), and scale-dependence in
diversity-productivity relationships (Chalcraft et al. in
press).
Currently, I in collaboration with researchers
in North America and South Africa are attempting to gain
a more general understanding of the impacts of key drivers
(fire, grazing and climatic variability) on savanna grassland
community dynamics and ecosystem function by deriving rules
and predictions from studies of NA savanna grasslands and
testing them in South Africa savanna grasslands. These systems
are structurally very similar (dominated by C4
grasses) and are influenced by the same drivers, but differ
in evolutionary history and diversity of megaherbivores.
It is these differences that may represent important contingencies
to our ability to generalize across systems.
To date, using extant datasets, we have shown
that some NA grassland rules do apply to SA savannas, particularly
those pertaining to the effects of fire on forb abundance
(Knapp et al. in press). Across these systems, fire
consistently has a negative effect on forb abundance. In
NA grasslands, the reduction in forb abundance results in
a decline in diversity, and therefore may have a similar
impact in SA savannas. However, additional research is needed
to test this, as data at that resolution are not available.
We have preliminary funding and plan to seek additional
funding for a study that takes advantage of several unique
long-term (20-50+ yrs) experiments at sites in NA (Konza
Prairie) and SA (Kruger National Park, http://www.parks-sa.co.za/parks/kruger;
University of KwaZulu-Natal, Pietermaritzburg) manipulating
fire and grazing to examine the effects of these key drivers
on community characteristics (diversity, dominance) and
ecosystem functions (productivity, nitrogen availability
and cycling, C storage). With this study we hope to gain
a more general understanding of savanna grassland systems
and challenge our ability to generalize across systems that
differ in evolutionary history.
Significance of Research
The research I describe above will contribute to our basic
understanding of the role of dominant species in communities
and ecosystems, the factors influencing community and ecosystem
dynamics, and the impacts of diversity on ecosystem function.
These studies also make important strides in linking genetic
variation to variation observed at the community and ecosystem
level, as well as developing ecological rules that could
be invaluable for predicting responses of grassland systems
to environmental changes. Few ecological studies are making
such attempts. My research also has important implications
for understanding how communities and ecosystems will respond
to a suite of environmental challenges, including invasion
and altered climate, and as such contribute to the knowledge-base
that managers and policy makers rely upon to make sound
decisions.
back to research main page |
top
