Tip:
Highlight text to annotate it
X
Dave Lemarie: Ok, good afternoon everyone from the U.S. Fish and Wildlife Service's
National Conservation Training Center in Shepherdstown, West Virginia. My name is Dave Lemarie, but
I am appearing to you on your screen on the web page as Ashley Fortune Isham. I am using
her account. [laughs] I would like to welcome you to our webinar
series held in partnership with USGS National Climate Change and Wildlife Science Center,
which I will fumble through the abbreviation of NCCWSC several times, I am sure. They are
located in Reston, Virginia. The NCCWSC Climate Change Science and Management
Webinar Series highlights their sponsored science projects related to climate change
impacts and adaptation, and aims to increase awareness and inform participants like you
about potential and predicted climate change impacts on wildlife. At this point, I will
turn things over to Emily Fort, the data and information coordinator from the NCCWSC, to
introduce today's speaker. Emily? Emily Fort: Thanks so much. I want to thank
all of you guys for attending. We really appreciate it.
I'm proud to introduce Jennifer Cartwright. She's a biologist at the Tennessee Water Science
Center in Nashville, Tennessee. Her research interests include soil science, terrestrial
microbioecology, and GIS. She recently defended her doctoral dissertation in biology at Tennessee
State University, and will graduate this May. A portion of this talk will also be presented
by Bill Wolfe, who is a hydrologist and the Assistant Director for SurfaceWater Studies
at the Tennessee Water Science Center. Jennifer, I'll hand it over to you. Thanks
so much. Jennifer Cartwright: Thanks to both of you.
My name is Jennifer Cartwright, and I'll be presenting with Bill Wolfe. Both of us are
from the Tennessee Water Science Center in Nashville, Tennessee, and we will be talking
about climatesensitive, insular ecosystems of the Southeastern US.
First, a few acknowledgements. This research was funded by the Southeast Climate Science
Center, and we collaborated with colleagues at NatureServe and at North Carolina State
University. Some of the fieldwork that I'll be discussing, and Bill will be discussing
at the end of the presentation happened at the Arnold Engineering Development Complex
and at Stones River National Battlefield, which is managed by the National Park Service,
both in Tennessee, and some of the lab work was conducted at Tennessee State University.
To start, I'd like to give you our working definition of insular ecosystems. What do
we mean when we say "insular ecosystem"? We have three criteria for our working definition.
First of all, these are systems that have individual occurrences which are spatially
discrete. In the lower left, you see an aerial image of two Carolina Bays, and you can see
how, metaphorically, we could think of them as islands surrounded by a contrasting ecosystem.
They are spatially discreet rather than continuously distant occurrences. Secondly, these systems
tend to have relatively small geographic footprints. We're typically talking about land area that
is much less than one percent of the total land area of the regions in which these systems
occur. Thirdly, these insular systems have steep, ecological and environmental gradients
at their boundaries. At the lower right, you see an image of a rock outcrop ecosystem at
its boundary with the surrounding pine forest. You can see what I mean by a steep, ecological
or environmental gradient. That's sort of our working definition of what
we mean by insular ecosystem. Our project was that we focused on a set of insular ecosystems
in the Southeast U.S., and we reviewed the existing literature on these systems in order
to first of all evaluate their biodiversity contribution both to global and regional biodiversity.
We also wanted to assess the current threats to ecosystem integrity, which are documented
in the literature on these systems. Finally, we wanted to develop a framework for future
research on the climate change vulnerability of these systems. The insular ecosystems we
reviewed fall into a few categories. We looked at rock outcrop ecosystems, grassland and
woodland systems, and wetland and riparian systems. Some of the rock outcrop ecosystems
we looked at included granite outcrops of the Piedmont.
This is an image from South Carolina, but a lot of these granite outcrops are clustered
in Georgia. As you can see, they have subtle depressions within them that sustain unique
assemblages of plant species. We looked at high elevation Appalachian outcrop. These
are often subjected to cloud cover, and immersion in fog.
We looked at limestone cedar glades, and toward the end of the presentation, I'll be talking
about some of my field research in limestone cedar glades within the Central Basin of Tennessee.
That's another rock outcrop system. We looked at grassland and woodland systems, including
midAppalachian shale barrens. These can have anywhere from an almost open canopy to an
almost totally closed canopy, typically steep slopes, and the soil surface covered with
shale fragments. We looked at xeric limestone prairies. These
are sort of a cousin to limestone cedar glades, but they have some geographic and topographic
differences that I can discuss if anyone's interested. We looked at highelevation Appalachian
balds, both grassy balds you can see an example on the left and heath balds, on the right.
Both of these, at high elevations, gradate into rock outcrop ecosystems.
We looked at some wetland and riparian systems, including Carolina bays. You saw, at the start
of the presentation, that aerial image. Carolina bays are often elliptical depressions that
are geographically isolated wetlands in the Coastal Plain. We looked at karst depression
wetlands, and my colleague, Bill Wolfe, will be discussing some of his research at the
end of the presentation in a karst depression wetland on the Highland Rim of Tennessee known
as Sinking Pond. Finally, we looked at riverscour ecosystems,
which occur along the banks of highgradient reaches of large rivers and are subjected
to and have their ecosystem integrity maintained largely by periodic scouring from large flood
events. Here's a map of the general distribution of
these insular ecosystems that we looked at. You can see that they cover a number of different
physiographic provinces throughout the Southeast, including the Coastal Plain, the Piedmont,
the Interior Low Plateau, the Mid and Southern Appalachian region, and the Ozarks, among
others. I should say that there are a number of other
systems within the Southeast that would meet our criteria for being defined as insular
systems. We didn't try to look at all of them. We tried to pick a representative sample.
Some others that would also meet our definition include peat systems, like bogs and fens.
Cliff ecosystems these may be considered linear ecosystems, but if you zoom out at a larger
geographic scale, they could also be considered insular systems.
Within Peninsular Florida, there are a number of systems that could be considered insular,
including cypress domes. Then, ephemeral ponds in the Northeast, they're often known as vernal
pools. That's just to say that there are a number of other systems within the Southeast
US and beyond that meet our working definition for insular systems.
Why are these insular systems special, or why were we interested in reviewing the literature
on them? First of all, they represent fragmented habitat. We tend to think of habitat fragmentation
as a phenomenon associated with landuse change, but we can think of these insular systems
as a naturally fragmented habitat, and that lends itself to studies of island biogeography.
I won't get into that within this talk, but if anyone has questions, we can discuss that
at the end. An important feature is that the geographic
locations of these insular systems are often geophysically determined and constrained.
For example, if we're thinking about a karst depression wetland, that wetland ecosystem,
its spatial location is determined by the geomorphology of the karst depression in which
it occurs. We could also say that about highelevation outcrops at the tops of mountains, where there's
a geomorphic or geophysical constraint on the location of that ecosystem.
A little later I'll be talking about how these systems are typically maintained by characteristic
stress or disturbance regimes, and how that is important when we start to think about
climate change. But right now I'd like to talk about another reason why these systems
are special, is that they often contribute disproportionately to biodiversity, both regional
biodiversity and global biodiversity. To start with, I wanted to show you this map,
which is courtesy of NatureServe. It's hotspots of rarityweighted richness for G1 and G2 species
in the U.S. This is globally critically imperiled and globally imperiled species. We can see
that there are a few hotspots for rare species richness. One is central California and the
desert southwest of California. Another is the big island of Hawaii, and another is the
southeastern US. If we look within the southeast, we can see
that some areas of the interior low plateau, some areas of the mid and southern Appalachians,
a zone within the coastal plain, and then several regions within the Florida panhandle
and peninsular Florida are all hotspots of rare species richness within the U.S.
This is a map from Estill and Cruzan of centers of plant endemism within the southeastern
U.S., and we see a few similar patterns of plant endemism being clustered in the central
basin of Tennessee and within the southern Appalachian region, the midAtlantic coastal
plain, and then, certainly, within the panhandle and center of Florida.
We wanted to think about in what ways the insular system that we were studying might
be contributing to regional and global biodiversity, and one of those mechanisms is by sheltering
endemic species. Endemics are found only within the particular insular ecosystems in question
or closely related contexts. Here are four examples. The map to the right of the photograph
for each has in green the states in which the particular species occurs and in yellow,
the individual counties where the species occurs.
In the upper left is the pool sprite, which is endemic to Piedmont granite outcrops. In
the upper right, you see the shale barren rock cress, so in some cases just the name
of the species gives you a clue that it is endemic to one of these insular systems, in
this case the shale barren. In the lower left, you see the Cumberland rosemary, which is
endemic to river scour systems on the Cumberland Plateau. On the lower right we see the spreading
avens, which is endemic to highelevation southern Appalachian outcrops.
In all of these cases, we see species that have relatively restricted geographic ranges,
and occur only within these insular ecosystems. They contribute both to regional and to global
biodiversity because they're not found really anywhere else. Another type of biodiversity
contribution comes from biogeographic disjunction. These are cases where a species may be secure
within its range elsewhere, outside of the southeast U.S., but is disjunct to a particular
insular ecosystem, meaning that it is geographically separated from its home range elsewhere.
On the left we see an example in the northern bentgrass, which has ranges in northern New
England and in the Rocky Mountains and is disjunct to highelevation outcrops within
the Appalachian region. On the right we see wright's cliffbrake, which is disjunct from
the arid southwest to granite outcrops of the Piedmont.
This is a form of biodiversity contribution, not necessarily to global biodiversity, but
to regional biodiversity. We tried to do a quantitative assessment of the biodiversity
contribution for this select group of insular ecosystems. Here you see numbers of globally
rare taxa. This is G1 through G3 species. We can see that all of the insular systems
that we looked at harbor at least a few globally rare taxa. Some of them, such as limestone
cedar glades, Piedmont outcrops, xeric limestone praries, highelevation systems, Carolina bays
and river scour systems, all harbor more than 20 globally rare taxa.
We looked at globally rare plant associations, and again we see that all of the systems we
looked at harbor at least a few globally rare associations. We see that xeric limestone
prairies and river scour systems both harbor more than 20 globally rare plant associations.
The conclusions from our assessment of biodiversity contributions first are that all of the insular
ecosystems we examined harbor at least a few rare, endemic, and/or disjunct species, and
many of them sustain rare associations and/or high levels of taxonomic richness. These biodiversity
contributions are really disproportionate to geographic area. We're typically talking
about systems that cover much less than one percent of the land area of the regions in
which these systems occur. There are a number of other insular ecosystems
that we didn't examine in detail which we know also support endemic and disjunct species
as well. These include Appalachian bogs, a number of systems within peninsular Florida
such as sand hill scrub island, sandstone rockhouses of the interior low plateau, coastal
plain pocosins, which can gradate into Carolina bays in some cases, Chattooga basin spray
cliffs, and a number of others. The next assessment we made was of what are
the current threats or currently recognized threats within the literature on these ecosystems?
These range from physical destruction from recreation to invasive species.
In the upper right you see the Chinese privet, and some folks attempting to remove this invasive
species. In the lower left you see an aerial image of a Carolina bay that has been ditched
and drained and converted to agriculture. In the lower right is a photo of Bluestone
Dam on the New River in West Virginia, which has been specifically implicated in ecosystem
degradation of river scour ecosystems downstream, on the New River.
What we did was, we did a literature search, and we looked for any mention we could find
of ecosystem threats for these particular ecosystems. Then, we compiled all of these
threat references by threat category, and we came up with a little more than 450 total
references to threats. This is how they broke down, for all ecosystems
put together, the major categories of references that we found in the literature, to threats
to these ecosystems included: land use change, hydrological alteration, invasive species,
resource extraction, *** encroachment from disturbance suppression, and then pollution
and physical destruction from recreation. You'll see that climate change accounted for
four percent of the references we found. I'll come back to that figure in a bit, but I wanted
to show you the breakdown by ecosystem category. Here we have rock outcrop ecosystems and we
see that the main threat references with the literature include recreation, invasive species,
development, quarrying, and pollution. For wetland systems, direct and indirect hydrologic
alteration, was the major recognized threat within the literature, as well as agriculture,
logging, and development. For grassland and woodland systems, overwhelmingly, we saw ***
encroachment from disturbance depression as the most often sited threat to these systems,
followed by invasive species. For river scour systems, hydrologic alteration
followed by *** encroachment from disturbance depression, invasive species, development
and pollution, were the most often sited threats within the literature. That gets us back to
this question of, "What about references to climate change?" Of all of these threat references
that we compiled, which is about 450, only four percent of those were references to climate
change. Some of these papers were published before
the era in which climate change was recognized as a scientific consensus, so we tried limiting
our analysis only to papers published after 1995. That got it down to about 300 papers,
and still climate change accounted for only six percent of the threat references that
we could find. Furthermore, of those 17 references to climate
change, little more than a third of them presented no new empirical research. These were references
that were only in passing, so the authors might have said, "The threats to this ecosystem
are X, Y, Z, and climate change," and maybe said nothing else about it.
That leaves us with this question, "What accounts for the absence, within the literature, on
these insular ecosystems, the absence to references to climate change as a potential threat?"
Is that because climate change is really only a minor threat to these insular systems? Or
is it possible that it's a major threat, but a hidden one?
We could have a conversation about what might be the reason why climate change might be
a hidden threat, in terms of the difficulty of conclusively demonstrating a causal link
between climate change and ecosystem degradation, especially when it might be mediated through
other, more obvious threats, such as invasive species. That is an open question.
To get to that question, by zooming out a little bit, I'm going to attempt a very brief
review of what we know about the ecological effects of climate change at a global level.
Not just in the Southeast U.S., but really at a global level. I know that I am speaking
to an audience of climate change scientists, so I will be very brief, and not go into much
detail. What do we know at a global level about the
ecological effects of climate change? We know that there are phenological changes happening,
globally. In fact, just a few weeks ago, Jake Weltzin presented with the National Phenology
Network, if we look at metaanalyses of spring phenophases, we see that between two to four
days earlier, per decade, has been documented in a number of different studies of metaanalyses
of phenological change. We know that this can lead to phenological
asynchrony in which species respond to different seasonal cues, or respond differently to the
same cue. Ecological consequences of that can include the weakening of interspecific
interactions, for example, predation, or pollination. In some cases, there have been observed population
decline and even localized extinction attributed to phenological asynchrony.
We also know that range shifting has been documented in most major biomes on earth.
That is range shifting toward the poles or to higher elevations. One metaanalysis of
poleward migration put it at about six kilometers per decade. We know that as these species
shift, they don't necessarily all shift at the same rate, and those differential rates
of range shifting can cause novel species assemblages to arise.
In terms of range shifting, when this has been modeled, in terms of climate change modeling,
and ecological effects, migration ability has been an important determinant of projected
extinction rate. Importantly, in the context of insular systems, those topographic limits
on range shifting can lead to range contractions. A wellrecognized phenomenon is that of contracting
mountain island, where there's really a topographic limit that prevents range shifting.
We think that this may be applicable, not just to high elevation systems, but to a number
of the geomorphically or topographically constrained insular ecosystems that we examined.
The bottom line for insular systems, in terms of ecological effects from climate change,
as far as we can tell, is that the simple climate envelope model of species migration
are probably insufficient. Those are the models in which species simply migrate forward or
up slope to track the changing temperature and precipitation patterns.
Those sorts of simplistic models are probably insufficient for insular ecosystems, in particular,
because geomorphic and topographic limitations are present for that range shifting. There
is a real need for us to examine the stress and disturbance regimes that govern these
insular ecosystems. What do I mean by that? We have conceptualized these insular ecosystems
as islands of elevated stress, and we categorized five different stress regime categories. This
shows you some of the stress regime components and then some example ecosystems.
Edaphic climax stress regimes include thin soil, exposed bedrock, high solar irradiation,
seasonally high soil surface temperatures, so we're just typically talking about rock
outcrop systems there. Topography and elevation can be a component for high elevation systems,
a steep slope, or the importance of aspect. Geochemistry can be a stress regime type,
where you see low levels of certain nutrients, either low or high pH, and in some cases,
toxicity of certain plant species. Hydrology as a stress regime category can include widely
fluctuating water availability or inundation, and highly variable hydroperiods. That's the
case for some of these geographically isolated wetlands, and for microhabitats within some
rock outcrop systems. Then, finally, disturbance has to do with
episodic and physically destructive events, so in grassland and woodland systems that
may be fire or grazing. In the context of river scour system, we're talking about periodic
flood scouring. We developed this conceptual model of stress
regime alteration, to try to get at putting forward some hypotheses for ecological consequences
of climate change. What we see here on the xaxis, is a tradeoff
between competition, on the one hand. So that could be biotic, biological stress, and abiotic,
or physical stress, on the other hand. Then on the vertical axis, we have habitat suitability.
We can see that for mesic species that might be inhabiting the ecosystem surrounding these
insular systems, their habitat suitability is really highest for low levels of abiotic
stress, in which they are able to successfully out compete.
Whereas, for stress for disturbance adapted endemic species that are often contributing
to the biodiversity from these insular systems, their habitat suitability is greater at higher
levels of abiotic stress. They tend to be stress adapted, but relatively poor competitors.
The important thing to notice here is that if you move too far in either direction along
the xaxis, you can see that we could predict that there would be ecological effect.
If you move to the left, toward greater competition, this could be a case where stress adapted
endemic species get replaced, or invaded upon by mesic species. That would be a case of
stress regime relaxation. Whereas, if you move too far to the right, we'd be talking
about stress regime intensification, which might overwhelm even the stressadapted adaptation
of the endemic species. That's our conceptual model. We used this
model to generate hypotheses for the direction of stress regime alteration for some of these
systems, based on three possible climate change components. This included higher regional
temperatures, and then both increased precipitation causing increased flooding, both that could
be frequency or intensity of flooding, and/or seasonally decreased precipitation, which
could increase the frequency and/or the severity of drought.
For those climate change components, we were able to hypothesize the direction of stress
regime alteration for these different stress regime categories like high soil surface temperatures
or low temperatures, xeric soil conditions or saturated and inundated soil conditions,
widely fluctuating hydrology, and then disturbance, either from fire or flood scour.
Just to give you a couple examples of what that hypothesis generation process looks like.
This is sort of a simple example of a stress regime at high elevation that is based on
cold temperatures. If in a climate change scenario we were looking at warmer temperatures
for these high elevation systems, then we would be looking at a stress regime relaxation,
a shift to the left. We could see that that might increase the
habitat suitability for the surrounding mesic species, and that could lead to invasion of
mesic species into the habitat that had traditionally been the domain of the stress adapted endemic
species. Another example in the opposite direction
of stress regime intensification this will lead directly into what my colleague, Bill
Wolfe, will be talking about in a minute in karst depression wetlands. Here's an example
if we had increased precipitation, which is increasing the frequency and severity of inundation,
then this would be stress regime intensification for a system in which the abiotic stress regime
included saturation and inundation. Here we might see climate change pushing this
system beyond a threshold in which even the stress adapted species were able to cope.
This is just a couple examples of how we've been using this conceptual model to try to
generate some hypotheses for how climate change might affect these insular ecosystems through
the mechanism of stress regime alteration. Clearly, there is a need for empirical research.
Our conceptual model can generate hypotheses for the direction, either stress regime relaxation
or intensification, but it really doesn't say anything about the magnitude of that change
or spatial or temporal patterns. To get at that, we really do need better empirical research
on these insular ecosystems to evaluate these hypotheses. Specifically, we need empirical
characterizations of the current and historical stress and disturbance regimes, so that we
have a baseline, knowing what we're working with as climate variables change.
Then secondly we need longitudinal studies that track ecological outcomes and climate
variables simultaneously. That leads into two case studies that we would like to discuss
with you. The first is at Sinking Pond, which is a karst depression wetland on the highland
rim of Tennessee. In a minute, I'll hand this over to Bill Wolfe, who did that research.
Then after he discusses that, I will talk a little bit about my work in Limestone Cedar
Glades in the central basin of Tennessee. Over to you, Bill.
Bill Wolfe: Thank you, Jennifer. This is a study that we did with a lot of
cooperation from the University of the South, John Evans down there and his students. It
was funded by the Air Force about 12 years ago.
The site, Sinking Pond, is about 35 hectares in area, and it's a seasonally flooded karst
depression on the eastern highland rim of Tennessee on Arnold Air Force Base, about
an hour southeast of Nashville on I24. It's significant, because it contains what appears
to be a globally unique assemblage of overcup oak, river birch and resurrection fern, a
coastal plain species that are disjunct in the highland rim of Tennessee.
The maximum water depth during the wet season in the pond is almost four meters. It stays
wet long enough to leave a very distinctive moss line or water mark, showing the typical
wet season high water elevation. A striking visual feature of the pond is the absence
of an understory, and the presence of mature adult trees is evidence that recruitment and
regeneration of these oaks has occurred on these sites in the past. The absence of an
understory now raises the question of whether there's been a major shift in regeneration
and recruitment. The impression that something has changed
is certainly strengthened by the spatial patterns of different life history states in the pond.
In the lower right, you can see the distribution of seedlings in 2002. That year, like most
years, the pond interior was basically carpeted with overcup oak seedlings.
In the opposite corner, the upper left, you can see that the adult trees are concentrated
in the dark blue area in the center of the plot, which corresponds to deep areas, with
water depths greater than a meter. In contrast, the younger life states, the saplings and
seedlings, tend to be highly concentrated on the shallow pond margins, areas with less
than half a meter of flooding. The kind of change that might produce this
pattern could be increased hydroperiod, that makes sites that formerly were hospitable
for recruitment and regeneration of these trees now inhospitable. That was basically
the hypothesis the Air Force asked us to examine. I should mention that they were not looking
for a climate story at all. They were concerned that some activity on the base, some management
or even the industrial activity of the engineering center might be having an effect on the ecology
of this pond. We looked at several possible factors, but in the end, the one that was
the strongest by far was climate change. This is a graph illustrating a very welldocumented
and widely known change in regional rainfall patterns of the eastern United States. Basically,
a stepped increase in rainfall right around the year 1970. The bars on this graph represent
departures from the long term mean, but the top graph is for the U.S. as a whole. The
bottom graph is for the local long term weather station of Tullahoma, Tennessee, about 15
miles from the study site. Looking at whether the regeneration patterns
historically in that pond had a temporal pattern that corresponded with that stepped increase
in rainfall, we cored almost 50 trees, about a 10 percent sample of the 2.3 hectare study
plot I showed in the earlier map. Indeed, they show a pretty strong pattern. Before
1970, these oaks regenerated and recruited successfully across the depth range of the
deep and intermediate sections of the pond. They survived after being germinated fairly
evenly through time. After 1970, successful recruitment to adulthood for these overcup
oaks was restricted to the shallow pond margins with depths of less than half a meter.
To further examine whether rainfall changes really were driving these patterns, we developed
a simple hydrologic model that runs from rainfall and temperature records, and that allowed
us to simulate a time series of daily pond stage going back to the mid1850s.
In turn, having a simulated daily hydrograph for the pond let us examine inundation events
of various durations and elevation. This graph shows from 1854 through 2001 the elevation
of area that was ponded at least 200 days in the given year.
What you see is fairly striking. Before 1970, a 200day ponding event was not rare, but it
was fairly sporadic. It had a probability of about 21 percent. Basically, one year out
of five. After 1970, the likelihood of such events nearly triples to 56 percent. That's
pretty strong evidence that the rainfall change that we've seen across the eastern U.S. has
had a significant discernible effect on the ponding dynamic of Sinking Pond.
When we add the tree ring age and elevation data to that graph, the pattern becomes pretty
striking. We see, again, that these trees have successfully recruited to adulthood and
survived across a range of elevations through time, until the constant exposure to prolonged
flooding lasting more than half a year has forced them into the margin.
This, I think, illustrates the kind of stress response that Jen was talking about, and really
the limitations of an extremophile competitive strategy. These trees are very well adapted
to harsh conditions. I'm sure there are trees in here over 200 years old in a site that
pretty much excluded almost all competitors. But now, because of a climate driven shift
in hydroperiod, they are being driven out of this hospitable site and having to compete
on the pond margins with other species. The last, I guess, two points I'd like to
make about this case study, is first, that there was a 30 year lag between the imposition
of the climate stress, the change of rainfall in 1970 and the accumulation of functional
and structural changes that were sufficient to be visible and provoke a management response.
Secondly, that there are many other ecosystems in the eastern U.S. that live on a stress
margin and that are probably experiencing similar effects right now.
Jen? Jennifer: Thanks, Bill. Now I will tell
you a little bit about some of my field research in Limestone Cedar Glades within the central
basin of middle Tennessee. This work happened at Stones River National Battlefield outside
Murfreesboro, Tennessee, which is managed by the National Park Service.
Limestone Cedar Glades are a type of calcareous rock outcrop system, so they have very shallow
soil depth, and have zonal communities of bryophytes, moss, lichen, cyanobacteria, especially
in the thinnest microhabitat. Also have zonal communities of hydrophytic vegetation in seasonally
wet depressions. There are 19 endemic vascular plant species
in cedar glades, and at least 10 disjunct species, and 4 federally listed species, so
they contribute to that clustering of plant endemism within the central basin of Tennessee
that we saw earlier. The objectives of my research were first,
to characterize the abiotic stress regime of these Limestone Cedar Glades based on empirical
measurements. The reason that that is important is that within the existing literature on
Limestone Cedar Glades, there have been a lot of qualitative discussions of the stress
regime in terms of thin soils, widely fluctuating soil water content levels on a seasonal basis,
and seasonally extremely high soil surface temperatures.
But these components of the stress regime are often discussed qualitatively and repeated,
based on references to previous publications, but rarely have they been accompanied by empirical
measurements or statistical analysis. I wanted to try to do a really an empirical characterization
of these components of the abiotic stress regime to get that baseline in the context
of climate change. Then, secondarily, to evaluate the stress
regime and various soil physical and chemical properties as drivers of a few different ecological
outcomes. I was looking at soil respiration and at soil microbial community profiles.
I'll try to explain in a second what I mean by "microbial community profiles."
Here are just some rough correlations. For all of these, soil depths or depth to bedrock
is on the horizontal axis. We can see that there are correlations with other physical
and chemical soil properties, such as pH, organic matter, vegetation density and particle
size distribution. We have here the silt to clay ratio.
Looking at soil water content distribution seasonally, we see the preponderance of the
dry and xeric soil conditions from April through September, which is the time when our region
experiences reduced precipitation and increased evapotranspiration. During this time, we see
an absence of the saturated soil conditions for volumetric soil water content above about
45. By contrast, in the late fall through the
winter through the early spring, we see a reduction in those xeric and dry soil conditions,
and we see more of the saturated and inundated soil conditions. We did an analysis of canopy
coverage as a factor controlling soil surface temperatures, which is important in a system
that may be experiencing *** encroachment from the perimeter in terms of canopy coverage
change. We did that through digital photography and quantified canopy coverage into five zones.
Then found that for six soil points, canopy coverage in any of those zones was significantly
negatively correlated with temperature, canopy coverage cooling the soil. But those effects
were lessened for medium and thin soil areas to where those effects were really only significant
for canopy coverage toward the south and the west. Those are just some examples of the
types of empirical observations of these stress regimes.
Then looking at ecological effects of these abiotic stress components, we modeled soil
respiration, observed patterns based on soil temperature, of water content, organic matter
content, depth to bedrock and density of vegetation. All of those were variables that had explanatory
power for patterns that we observed of soil respiration in cedar glades.
Then, we looked at determinants of soil microbial metabolic profiles, so I can go into more
depth on these methods if anyone has questions, but this is community level physiological
profiling method in which the entire soil microbial community is inoculated onto specially
designed plates that contain a sole carbon substrate in each of the wells.
Over the course of incubation, a color change response is proportional to the community
metabolism of a particular carbon substrate that's present in that well. That gives you
the ability to compare across soil samples that metabolic profiles of the soil community.
Then, I also did some plating, with a platedilution frequency assay to get at a most probable
number of heterotrophic culturable microbes per gram of soil. Found through these microbial
indicators explanatory power for depth to bedrock, so that's soil depth, soil organic
matter, soil pH, water content and density of vegetation.
The implications from some of my work in the Cedar Glades, first that the empirical characterization
of the stress regime is needed and important, as I mentioned before, to get a baseline as
regional climate may change. We saw that some of these abiotic stress regime components
like soil water content and temperature did affect ecological outcomes such as soil respiration
patterns and soil microbial ecology in Cedar Glades.
To get at some conclusions for this literature review, and our work more broadly on the insular
ecosystems. First, as I showed you, insular systems are important to regional biodiversity
by harboring rare species, endemic, disjunct, globally rare associations, and the like.
Second, there are many and a diverse array of recognized threats to these ecosystems
that are documented in the literature. But, at this point, climate change is still largely
uninvestigated in these systems. We have put forward stress regime alteration
as a framework in order to investigate possible climate driven ecological change. We've presented
to you a couple case studies to illustrate some possible approaches for that work. Here
are some references, and we will now take your questions.
Dave: We have a question already from Steve Young that he would like to ask via the phone.
Steve, if you could hit star six, and go ahead and ask your question.
Steve: Can you hear me? Dave: Yes.
Jennifer: Yes. Steve: OK, good. Very interesting talk,
and your research is really exciting to see. I had a question. Early on, obviously, you're
talking about insular systems. These systems that you're talking about, are
they, and we had kind of a discussion here on the side in the chat box, I don't know
if you're reading that. But on the whole, do they make up a greater proportion of the
ecosystems in general than the normal traditional, riparian forest grasslands? Or is it somewhat
of an issue of terminology? If you look across the landscape, it's easy
for us to kind of generalize different systems that we work in. But I think what you're showing
is that these insular systems are obviously critical, but the differences in them and
the number of them actually are contributing to the traditional or general ecosystems that
we work in. Is that true? Or what are your thoughts on that?
Jennifer: In terms of land area and their sort of geographic scope, NatureServe has
this classification of large patch and matrix ecosystems. In the Eastern United States,
a lot of forest ecosystems might be classified as large patch or matrix, meaning that they
are continuous across hundreds or thousands of square miles, so relatively large geographic
areas. The insular systems that we are looking at,
individual occurrences of them are very small. We could be talking about a few acres for
an individual occurrence of a particular insular system. Even when all of the occurrences are
added together, typically, we're talking about much less than one percent of the land area
of the regions in which they occur. These would be small patch ecosystems, according
to NatureServe classification. Steve: One other question I had was you
showed a figure on the stress that looked like one that you, that's your own, correct?
Jennifer: Yes. Steve: OK. I might want to talk to you at
some point. I see your contact information. I want to contact you and discuss that at
a later date. I'll do that offline. Dave: From Sally Sims for Bill. Can you
give additional examples of other ecosystems that might have been affected by climate change
that you referred to in your last slide? Bill: No. I don't have any concrete examples.
But this notion of insularity, back when we were doing this, we thought about sites like
Sinking Pond as ecological islands. Where, like a geographic island is surrounded by
water, if you lower the water enough, it stops being an island. If you raise the water enough,
it stops being an island. I think wetlands, by definition, have water
adopted extremophile vegetation. Probably, I would look at small isolated wetlands in
the Eastern US as candidates that might be worth looking at with a kind of an approach
that tries to combine some sort of historical tracking, longitudinal tracking, of both plant
regeneration or some similar expression of population with hydrologic reconstruction
or some other reconstruction of the stress regime.
It's not easy to do and we were lucky to have, like, one opportunity to do that kind of study.
But I think if you look hard at isolated wetlands, there may well be something there.
Dave: Thanks, Bill. Mitchell, I got a chat message to him, and he would like to contribute
to the previous discussion with Steve Young. Mitchell, hit the star six and go ahead and
speak. Mitchell: Hi, can you hear me?
Jennifer: Yes. Dave: Yes, I can hear you.
Mitchell: Hey, all right. First, just want to say, more of a comment. I went to the University
of the South Sewanee and studied at Sinking Pond during the hydrology course work, so
was thrilled to see that on there and John Evans' work.
My question is pretty general and it's more playing towards the devil's advocate. When
we present these types of examples as a representation of showing, obviously, the effects on insular
ecosystems, is it a bit of a, how should I word this? A potential painting ourselves
into a corner in terms of speaking to the larger public about the effects of climate
change when we're talking about just an insular system being affected? If that makes sense.
Bill: It does make sense. But I think what I would say is that some of these systems
may be sort of alarm clocks or bellwethers. They may be some of the places where we see
effects the earliest. I've been talking with botanists and biogeographers and there's a
sense that this increase in rainfall in the Eastern United States right now may be having
a broader effect on other, especially hardwoods. I think that these localized studies can be
used as a basis for broader biogeographical analyses.
But the other thing is the point that Jen made is that collectively, these things occupy
very little area and they account for a very large amount of our natural heritage in the
Eastern US, and especially the Southeast. I think from a conservation perspective, they
form a theme that might be useful in just looking at priorities.
Jennifer: Just to add to what Bill was saying, a couple of points, when you're thinking about
the general public and communicating to the general public, like Bill was saying, these
could be the canary in the coal mine ecosystems. Because those of us living in the Southeast,
the climate change affects that are happening, at the poles are well documented, but they're
far away. Seeing ecological change in our own backyard so to speak, might have some
potential for communicating climate change impacts to the public.
The second point I wanted to make is, in terms of, maybe not the general public, but conservation
professionals and managers, the implications of this really have to do with how do we protect
these places that are clearly important to natural heritage, and really necessitating
a rethinking of these traditional conservation paradigms that rely on static boundaries,
and just protecting a confined static geographic area, by putting fence around it, and excluding
say recreation. It's more complicated than that, if we're going to protect these systems
in the context of climate change. Mitchell: Can you hear me?
Jennifer: Yes. Bill: Yeah.
Mitchell: I see both your points. I think they're spot on. I work with Audubon, managing
a 3,000 acre property, called Strawberry Plains, in north Mississippi. It's one of those examples
where we have a very, very rural community, kind of what Jennifer was saying, most of
their understanding or information concerned with climate change, or in many of their opinions,
the lack hereof, is these news blasts at the poles.
The property that we manage has several vernal pools and beaver created ponds. We're at the
headwaters of the Cold Water River, and I just wonder, an example like Sinking Pond,
for someone like me, being a biologist, is so impactful.
But I wonder, are there other examples, that you know of, within the Southeast, that are
these insulated areas that have, for us, our downfall is we don't have that track record
of annual or even seasonal documentation, of the changes that are happening. That's
something we're hoping to begin, but are there other insulated ecosystems that are as well
known, that you guys are aware of, or might be looking to investigate?
Jennifer: We've certainly seen an absence within the literature of that empirical link.
For some systems, the work ahead might be as simple as comparing, pulling together different
strands of data from observed ecological changes over time, and linking those to historical
records of temperature, precipitation. In other cases, those longterm ecological records
might be harder to come by. Bill: We've been talking with Milo Pyne,
and the guys at NatureServe, looking for longterm ecological monitoring records, that we might
be able to link to historical climate records and look for associations. But, if we've been
too subtle, so far, I think one of the main points we want to make, really, is there is
a desperate need to collect some empirical data, both on ecological records, or ecological
state, and stress regimes of these small systems. We're proposing that this concept, this stress
regime concept of insularity might be a useful framework for looking at some of these systems.
Dave: I have a question typed in from David Smith, "Did any of your research actually
evaluate impacts of climate change on insular systems, or was it restricted to developing
techniques for future analysis?" Bill: The whole study, at Sinking Pond,
that I presented, comes as close as anything that we're aware of in actually documenting
an ecological effect of climate change. But the main study that Jen is reporting, is primarily
a literature search. Dave: I've got a quick question here from
Steve Young again, "Is this work focused on extreme climate events, too?"
Jennifer: By extreme climate events we might be talking about droughts or storm events,
is that the nature of the question? In both of those cases, we're sort of getting at the
issue of disturbance as a stress regime type or factor.
If we're talking about say largescale storm events that might cause flooding, in the case
of say river scour systems that are primarily disturbance maintained systems, changes in
the frequency or maybe secondarily, the intensity of those events would constitute a change
in the disturbance regime, which is part of the stress regime.
Then if you're talking about say hydroperiod, for geographically isolated wetlands for example,
then certainly changes in the magnitude, or the frequency of storm events, could have
impacts on the hydroperiod which would be a form of stress regime alteration.
When you look at disturbance from fire, these days, so much of that disturbance regime is
managed by people, in the form of either fire suppression or controlled burns. But to the
degree that that disturbance from fire was natural, then you could imagine that drought
would have an effect on the disturbance regime from fire. Drought, certainly, would have
an impact on stress regimes like in rock outcrop ecosystems that involve seasonally xeric soil
conditions. Does that get to that question? Dave: I'm waiting for a response.
Jennifer: OK. Dave: Yes. Steve says, "Yes. Thank you very
much." Leonard was having problems with his star six. You want to try it one more time,
Leonard? If not, I'll read your question. Sorry about that. From what Leonard says,
he works in the Everglade Tree Islands. Small area, high biodiversity, biogeochemical hotspots,
which is high phosphorus and allotropic wetland, but he didn't actually do a question here.
He's going to try star six again. He's finishing his question via text here. "How is this connected
to the surrounding ecosystem?" Bill: We didn't attempt to do much in Florida, just because there is
so much there, and there's also a lot of work being done, but from the description, those
would be strongly insular systems with high gradients with the surrounding landscape.
Sounds like a very fascinating field area. I'm sure, we can't add much insight I think,
probably to somebody who's working there. Dave: Sorry about the technical glitch there.
Now, we've cycled back to Mitchell. I'm sorry I had to cut you off, Mitchell. We just had
so many questions stacked up here. His last question, "Is there any funding for research
and monitoring for insular ecosystems from USGS or USDA?
Bill: Sporadically, yes. We got funding from the Southeast Climate Science Center
to do this literature survey, but the way science funding works, a lot of times is opportunistically
and incrementally. This office right now has some work with the National Park Service,
looking at ecological flows on rivers that include river scour systems.
We have hopes to expand the hydrologic analysis to try to quantify the flooding regime that's
necessary to keep these boulder burrs and flat rock communities scoured. More broadly,
no, there's no concerted program to fund this kind of work at a National scale at this time.
The main thing that we hope to accomplish by this talk, would be to get the ideas out
there, in the hopes that other workers might find or seek small opportunities to look at
some of these systems through this particular lens or a lens similar to it.
Dave: Bill, then I have a twopart question for you, from Mary Morrison. "Did your work
look at the season of increased rainfall? High rainfall in winter would be a different
effect than high rainfall in the spring." Bill: We were painting with a pretty broad
brush. We have, like I said, daily simulated records, but really, what we were looking
at was at an annual basis, what kinds of inundations durations limited the recruitment and regeneration
of these oaks. Basically, to give you an idea, this pond stays wet about half the year, most
years. It's called "Sinking Pond" for a reason. The water level can rise or fall two or three
meters in two or three days. When it falls, it's closely connected to the ground water
system. Also, Overcup Oak, they're very adapted to
seasonal flooding. The oak leaves drop in the spring or in the summer and they expect
to be submerged and they wait until they're exposed to air to germinate. The oak trees
are basically adapted to varying seasons of flooding.
Dave: Bill, before I go onto Mary's second part of her question, David Smith is asking
if there's reference for your work at Sinking Pond?
Bill: Yes, there is. It's a USGS scientific investigations report. It is published online.
I think it's in Jen's reference list. Jennifer: Here's the cover of it.
Bill: It's Water Resources Investigations Report 034217. If you search for that, you'll
come to a URL that is I guess, as permanent as a URL can be. It's a home for that report.
Dave: I just wanted to get that in there, and then Mary Morrison had a second part for
her question. Also, "Did you look at other factors in your study, or did you focus on
the hydrologic factor?" Bill: We really did focus on the hydrology,
but we looked at that from a bunch of different angles. We recorded trees on the spillway,
and looked for signs that the spillway had been altered. We looked at the internal drain
of the pond and for any signs that it might have been filled in. We did a landuse study
from air photographs going back to the 1940's and showed that the only landuse change had
been an increase in forest cover in the ponds basin, which would have probably had a tendency
to dry it out. There was a deer exclusion study that was
going on at the same time, because the deer population on the Arnold Air Force base natural
area had expanded. We were able to convince ourselves that seedlings inside the deer exclusion
plot had no better chance of living than outside. Other than that, that's about the range of
hypotheses that we examined. Dave: I have no other questions stacked
up, so here's your last chance for anyone to ask any questions, or otherwise we'll wrap
up. We'll give you a few more seconds to chat. Emily: This is Emily. I just want to say,
thanks so much everybody for the presentation, and the great questions. We really appreciate
it. Jennifer: Thank you.