Extreme Climate Events and Species Population Dynamics

Video Transcript
Download Video
Right-click and save to download

Detailed Description

This webinar was held as a part of the Climate Change Science and Management Webinar Series, a partnership between the USGS National Climate Change and Wildlife Science Center and the FWS National Conservation Training Center. Webinar Description: Extreme events (floods, droughts, and fires) have a high public profile and changes in their frequency, magnitude, and duration have been linked to changes in climate. For species populations, these events are often associated with high levels of mortality and major changes in habitat, suggesting a strong influence on population dynamics. At the same time, the life history and reproductive strategy of many species, particularly those associated with highly seasonal and variable climates, may mitigate the long-term effects of extreme events relative to more gradual changes in climate. Given the difficulty of accurately forecasting climate extremes understanding their role in population dynamics is critical for effective management and climate adaptation. In this talk, we review some of the basic determinants of population response to extreme events, using case studies based on long-term data from natural populations in the northeastern region, and present a modeling framework for evaluating the relative impacts of changes in timing, duration, and magnitude. We also consider the potential for human responses to perceived and actual risks from climate extremes to interact with, and in some cases override the direct effects of the events themselves.


Date Taken:

Length: 00:55:28

Location Taken: Reston, VA, US


Shayna Carney:  Good morning or good afternoon
and welcome from the U.S. Fish and Wildlife

Service's National Conservation Training Center
in Shepherdstown, West Virginia.

My name is Shayna Carney and I'd like to welcome
you to our webinar series held in partnership

with the U.S. Geological Survey's National
Climate Change and Wildlife Science Center

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 fish and wildlife.

To start things up, please join me in welcoming
Shawn Carter from the National Climate Change

and Wildlife Science Center who'll be introducing
today's speaker.


Shawn Carter:  Thank you.

It's my pleasure today to have Dr. Keith Nislow
with us.

He's a team leader and a research fisheries
biologist for the USDA Forest Service Northern

Research Station.

He's also Adjunct Associate Professor at the
Department of Environmental Conversation at

UMass Amherst.

Finally, Keith is also acting as a CoPrincipal
Investigator at the Northeast Climate Science

Center which is affiliated with our center

Keith has degrees from University of New Mexico
and Dartmouth, and he's been with the Forest

Service Research and Development for the last
16 years.

Today, Keith brings us his expertise in research
dealing with the relationship of ecosystem

change in aquatic habitat and the distribution
and abundance of fish and aquatic invertebrates.

He's particularly interested in using basic
science to assist restoration, conservation

and management.

It's my pleasure to introduce someone I'm
very fond of, Keith Nislow, for today's webinar.

Take it away, Keith.

Keith Nislow:  Thanks very much, Shawn, and
thanks, everyone, at NCCWSC.

It's my pleasure to talk to you guys today
and everyone on the line about some of the

work we've been doing and some of the issues
that we're interested in.

My very first thought in starting to put this
talk together after having given Holly the

title, seems like many eons ago, was, "Wow!

What a stupid title."

We've all been in this situation where we
send in a title for a talk, it sounds amusing

at the time, and then we just sort of grimace.

I was faced with the choice, do I try and
change it at the last minute, or do I double

down and try and make some sense of it?

Like a fool, I kept the title, and I'll try
and make some sense of it to everyone online.

We're interested in extreme climate events.

Obviously, they have an influence and draw
attention way out of proportion to their actual


In a sense some of the extreme events, particularly
that have hit the Northeastern US in the last

three to four years, have become now the poster
child for potential climate change in the


Going back to my bad title, this actually
was a bit of a controversy back in the early

days of ecological science.

We have an interesting debate between two
general camps.

We've got folks led by Andrewartha and Birch
who stress an overriding influence of climate,

particularly climate extremes in driving population
dynamics and the distribution and abundance

of animals.

In contrast in the no big deal school, although
this is an overstatement, we had folks like

G. Evelyn Hutchinson, Bob MacArthur, and a
group that was really focused on equilibrium

dynamics, carrying capacity, and population
aggregation via density dependent processes.

It's a stretch to say that for these folks
climate extremes and these climate events

are no big deal, but much more focus on how
populations got back to an equilibrium or

carrying capacity and less of a focus on those
extremes and driving dynamics.

Obviously, as we've been moving forward in
the last 50 years, we've lost the science.

We certainly realize that we need to bring
these two perspectives together in order to

understand the effects of these extreme climate
events on population dynamics, a topic that

has become even more important given the nonstationarity
of frequency, timing, and duration of climate

extremes that we expect in a changing regional

This is some data up from the Northeast Climate
Science Center, from Ray Bradley's lab and

his postdoc.

It deals with an aspect of climate that I'll
be focused on a lot today, particularly the

frequency of extremeprecipitation events.

I'm an aquatic ecologist.

I'm really interested in floods, and as I
mentioned, these large floods really have

been a hallmark of climate extremes in this
region over the last four to five years, since

the Climate Science Center has been in existence.

Obviously, there's lots of variability, lots
of uncertainty in these predictions, but it

does seem like it's going to be a blue world
with respect to intense precipitation and

the possibility for extreme climate.

This brings up some really important science
questions from the perspective of fish and

wildlife population dynamics.

First, are there critical thresholds in frequency,
duration, timing, and magnitude that increase

risks to population?

Then, from more of an operational standpoint,
but also from the basicunderstanding standpoint,

what is the relative importance of changes
in extremes versus changes in central tendencies

with respect to climate?

This has particular importance because of
the difficulty in generating robust forecasts

of climateextreme regimes.

From the wildlife and fish perspective, from
the naturalresource perspective, we need to

have a really good sense of what the importance
of these events are if we're going to task

modelers with coming up with better and better

With respect to management implications, some
obvious questions of interest are, how do

these changes in extreme events influence
predictions of distribution abundance?

Maybe even more importantly, does considering
extreme events change the prioritization and

the relative value of specific management

Obviously here, in the Climate Science Center
at NCCWSC, we're interested in actionable

science, and this is our major concern.

Are we going to ask managers to do things
differently if we consider changes in these

extreme events?

The rest of the talk, I just want to give
you a bit of a road map.

I'm going to be talking about population demography,
conservationgenetics, habitats, and then I'm

going to end with human responses.

I am going to talk a lot about fish, which,
that's the way I am and it's hard for me to

get away from, but I hope to achieve some
level of generality using fish and a few other

texts and case studies.

Before I get too far in my talk, I want to

acknowledge that the work I'm going to be
talking about is very, very much a collective

effort, involving a lot of great cooperators,
from the USGS, particularly Ben Letcher, my

longtime cooperator at the Conte Anadromous
Fish Research Center, who pioneered a lot

of this work on brook trout.

Jason Coombs, my postdoctoral researcher,
who's done a lot of the modeling and additional

work, and Andrew Whiteley, who runs the Aquatic
Conservation Genetics Laboratory here at UMass,

which is cosupported by the Forest Service
and UMass, and then, of course, acknowledge

all of the institutions that have funded or
supported this research, including the Northeast

Climate Center, USGS, Nature Conservancy,
UMass, Forest Service, and many others.

Not only am I going to be talking a lot about

I'm going to be talking a lot about a particular
kind of fish, brook trout, in a particular

place, our longterm study site in Westbrook,
in Western Massachusetts.

We've been at it for a long time now, pushing
20 years.

We've got the site very well monitored.

Lots of years of longterm, very intensive,
individualbased data, which, over the past

10 years, we've branched out to include conservationgenetics
perspectives as well.

You'll hear a lot about this site at various
points in the talk.

I'm going to start the talk talking about
demography, and particularly talking about

temporal variation.

One of the things that's really interesting
about a lot of the species that we deal with

is that we do see a lot of climateassociated
variability in population numbers, but we

see populations persist.

Before we get into the potential effects of
climate extremes and changes in climateextreme

regimes, I’m going to take a little time
and talk about how populations and species

that we're interested in deal with the kind
of climate variability that they're experiencing

now and have experienced over most of their
evolutionary history.

A really important concept here to start out
with is the concept of stock and recruitment,

which relates to the numbers of stock, number
of spawners, number of adults, however you

want to describe it, for whatever species
you're interested in, yielding a certain number

of recruits, the number of young that survive
recruitment stage to become potential spawners.

I'm going to focus a little bit here on recruitment,
because it's a really important life stage

for many, many species, and it has an important
intersection with climate and particularly

climate extremes.

We're really interested in recruitment, and
it's often very important, because for many

species, it's the most vulnerable lifehistory

You're dealing with small, inexperienced,
competitively inferior individuals, whether

you're talking about tree seedlings or fledgling
birds, there in the middle of the screen,

or fish larvae that are just rid of their
yolk sac, and oftentimes they're in the process

of transitioning from dependence on maternal
resources to independence.

As a consequence of this vulnerability, we
often see very high mortality during this

stage, so very high variation in survival,
leading to very high annual variability in

recruitment, that is in turn tied to interannual
variation in climate.

This is an example of what a larval brook
trout in Westbrook might be experiencing as

they begin to absorb their yolk sacs and emerge
from the gravel.

What we have here, in the stippled line, the
stipple line shows a decrease, from high flows

in the spring, April, down to base flows in
the later spring, consequent increase in temperature.

This is the period when young larval brook
trout, brook trout frys, recruit from the

fry stage to the young juvenile stage.

One of the things that we see time and time
again in many streamsalmonid populations,

like brook trout, is that highrecruitment
years are linked to successful match between

environmental conditions at recruitment and
the stage the fish are in when they're ready

to recruit.

Conversely, when there is a mismatch between
these fry requirements for example, if they

emerge early, when flows are too high, water
is too cold they can have very, very low survival

and can even completely fail to recruit.

You can have years, depending on particular
flow conditions and particularly associated

with extreme flows that can wipe out an entire
recruit class.

Conversely, there are some species, like these
floodplain trees, that depend on extreme events

during critical recruitment phases to successfully

These silver maples here lining the Connecticut
river, they may go through many years of no

successful recruitment, until they get just
the right flood at just the right time to

allow their seedlings and saplings to recruit
to their population.

Just to follow on this a bit, we see this

high annual variability, and it's variable
to the extent that for many populations, we

have many years where recruitment is zero
or close to zero.

As a further consequence, a lot of these populations
may indeed be recruitmentlimited.

The number of individuals that get through
that stage is going to determine how big a

cohort you're going to have, how abundant
that population is throughout the course of

that cohort's existence.

Just to demonstrate that a bit, this is some
work that I did with a colleague, John Armstrong,

in Scotland.

We were looking at when we would expect to
see recruitment limitation.

This is just a diagram that illustrates that.

The lefthand side of the graph, you've got
two levels of salmon fry recruitment, and

what the graph describes is the change in
numbers on the Yaxis, the log change in numbers

of individuals, and the change in the size
of those individuals.

As in almost all populations, this isn't just
true for streamdwelling salmonids.

It's particularly true for forest trees.

As individuals increase in size, they are
reduced in number.

That reduction, as you see leading from the
recruitment part of the graph, in the lightly

stippled lines, that's just densityindependent
mortality, so there's just some level of mortality

going on.

What we can see is that as long as that mortality
or that survival doesn't exceed the carrying

capacity of the habitat for older juveniles
in the postrecruitment phase, that the number

of individuals coming out of this juvenile
phase, which is the size of the arrows on

the righthand side of the graph, is directly
proportional to the levels of recruitment.

This is an example of where recruitment is
an overriding influence on cohort size or

number of individuals.

We see a lot of recruitment variation and
persistence, as I mentioned, to the extent

that, in many populations, you can have a
number of years of zero recruitment, zero

survival during this stage, and yet these
populations persist.

I'm going to talk about both the ability of
these populations to persist to buffer many

years of low recruitment, and some of the
mechanisms that are involved in that buffering,

in that compensation for low recruitment and
low abundance, and then I'm going to talk

about how that relates to changes in extreme
event regimes.

One important mechanism that helps populations
recover from low density at any stage is density

dependence, and classic density dependence
where we have increased survival and growth

of recruits at low density as well as in some
cases increased survival growth and fecundity

of adults.

Both of those processes help populations recover
from low density.

You can see here this is the graph from a
paper by my colleague, Sigert Hynimaneye,

and he looked at how initial density of salmon
fry influenced the total number of recruits,

of juvenile recruits that we had at the end
of this stage.

We can see two things here.

The dark bars are in a lowflow year.

I'm sorry, the dark circles are...we did this
experiment in a lowflow year, so relatively

benign year, no floods.

The white circles are when we did this experiment
in a year with a pretty major flood during

the recruitment stage.

We see here the process, the outcome, of both
differences in low regimes, in flood regimes,

but also a really strong density dependence
so that in that benign year that the performance,

the survival, of fish at low densities was
great enough to result in potentially the

same output of fry as when we had lots of
fry escaping that recruitment.

Increased performance, increased survival,
growth, and fecundity at that density are

a very powerful mechanism allowing these populations
to respond to low recruitment and low abundance.

The other important mechanism, or another
important mechanism, is a component of what

we call the ecological storage effect.

What this component is is essentially that
in populations that have longlived highly

fecund adults like forest trees, like a large
adult salmon, the storage or the ability of

those adults to persist across multiple bad
years of poor recruitment is a critical component

to persistence.

This is really manifest, or it's possible,
because as bad as recruitment can be it can't

be less than zero.

If you can't recruitment less than zero it
means, again, particularly for these highly

fecund species, which in a good year when
everything goes right, when all the holes

in the Swiss cheese line up, can have just
absolute boons of recruitment and the good

years can be much better than the bad years.

What I have illustrating here is this is the
relationship between egg number...between

length of individual brook trout and the number
of eggs that females have.

What we can see is a very strong size dependence
so that if you get to be two, three years

old, reach 200 millimeters, your potential
reproductive output, your fecundity can be

an order of magnitude greater than fish that
mature at 105, 125 millimeters.22:16

To pull this together a bit one of the ways
that species can persist in the face of lots

of highly variable climates and influential
extreme climate events is to combine these

aspects of compensatory scope.

What I've tried to do here is display this
in three dimensions, and just to review, the

factors that are involved in this recruitment
variation persistence, longlived adults, adults

that live multiple years and are often more
resilient in the fact of climate regimes than

these vulnerable juveniles, these vulnerable
recruits that I talked about.

Not only can adults be longlived, but if they
continue to grow throughout their lives so

that their size is somewhat dependent on age
and their growth is indeterminate, unlike

us, fish, trees, continue to grow and as a
consequence of that growth they can be very

fecund, have a very high potential reproductive
output so high in size dependent fecundity.

Then, finally, as I mentioned a couple slides
earlier, strong density dependent, big increases

in performance at low density.

To illustrate those factors, just draw a contrast
and to get away from fish a little bit, let's

consider your typical, neotropical songbird.

It's the Blackburnian warbler, I think.

Like most songbirds it's not particularly

Its fecundity, or its variation fecundity,
particularly compared to some of the other

species we'll talk about, is tiny, very small
variation in fecundity, so not a lot of scope

for highly fecund adults to compensate for
bad years.

Then, finally, not a lot of consistent evidence
for strong density dependence in a lot of

these songbird species.

All of these factors combine to put species
like Blackburnian warblers and other songbirds

at the very corner of this threedimensional
space that describes the potential for compensatory


As I mentioned you can have orders of magnitude
variation in fecundity as a function of adult

size and similarly forest trees where you
also have that same very high variation in

fecundity particularly for forest trees.

They're longlived.

They continue to grow, continue to get big,
and for both the salmonid fishes that I've

studied and forest trees that folks like Tony
and many others study - very strong density

dependence so very good performance at low

All of these factors giving species that are
in this piece of life history space a lot

of compensatory scope to buffer the effects
of extreme events.

Given this very high variation in recruitment
with simultaneous persistence, what are some

of the management implications for dealing
with changes in extreme event regimes associated

with these kinds of temporal variations in

One important one, or one thing that's really
worth looking at, is can we define recruitment

failure thresholds?

We talked about populations' abilities to
buffer individual bad years, individual recruitment

failure events, but how many climaterelated
bad years are too many, and can we directly

relate those frequencies and magnitudes to
climate predictions?

I want to talk a little bit about some work
that Yoichiro Kanno did when he was a post

doc, working in Ed's lab, working with us
on brook trout, and he used a matching model

to look at brook trout abundance from a really
good longterm population monitoring program

from the Shenandoah National Park so a great
data set.

I think about 25 years of data.

He parameterized a model, environmental model,
based on those data, and then in the model

he changed the frequency of different kinds
of extreme climate events in different seasons

low flows, high flows, winter, summer, fall.

Because he had that kind of data he was able
to do some scenarios and look at how these

populations would respond particularly with
respect to their persistence under different


What he found was that under reasonable levels
of extreme event frequency as we see now we

saw a strong ability of these populations
to exist.

So low flows every five years, high winter
flows every five years.

We really didn't see a change in equilibrium
adult abundance.

As we got into more extreme situations, higher
magnitudes, higher frequencies, and different

combinations, Yoichiro did start to see some
thresholds that beyond which these populations

would decline to very low and very vulnerable

If we've got good population models we can
try and look at the scope for persistence

under different types of extreme climate event
regimes, and that can help to improve our

forecasts, get a better sense of that original
science question I posed, "What is likely

to be more important, changes in extreme events
or changes in mean temps?"

Another interesting, from my perspective,
imagined implication, implication for how

these populations persist under recruitment
uncertainty, is that we tend to focus a lot

on resilient habitats with respect to population

But our understanding of population dynamics
suggests that if we manage to maximize compensatory

scope and storage for those species where
it’s really important we might do a lot

of good, as much good as focusing on habitats.

One way to do this is to focus on individuals
and/or life history strategies with high reproductive

value, for example this giant lake trout that
might have a potential fecundity of 20,000,

30,000 eggs.

We're justified in this focus in a wide range
of research.

Some of our own research in our study system
in Westbrook has really pointed out the value

of these large, potentially highly productive
value individuals in the population.

We have here in the top graphic, the only
graph, is we've got size state on the Xaxis

and the summed elasticity, which is how influential
variation in survival of fish at any of those

size classes is to overall population performance,
in this case measured as lambda, the population

growth rate, one if it's a stable population,
above one, below one if it's declining.

What we found was in this model and this is
work that we published in 2007 in PLoS is

that the influence of these large high reproductive
value individuals, even though they were a

relatively small proportion of the population,
was well out of proportion to their abundance.

It suggested that if we target management
strategies for these large individuals that

we might be lending these populations a lot
of resilience in the face of variability.

What's nice is that we've actually got some
management tools to try and address this.

Almost everyone in fisheries has heard of
slot limits, and this is in contrast to size

limits where fish need to be a certain size
before you take them, that where in contrast

to the size limit in a slot limit you've got
protection both for young fish, for small

fish, that haven't yet recruited to be a spawning
population, which is the classic justification

for the size limit in managing a stable fishery,
but you also have protection for those large

highvalue individuals.

You're not allowed to take small individuals,
but you're also not allowed to take large


Similarly, in forest ecology everyone is familiar
with diameter limit harvest so these are two

wellestablished management techniques that
are done for all kinds of reasons but we suggest

might also contribute to resilience from extreme
events and the population variability associated

with it.

If we get our models right and do this work
correctly we can give managers an idea of

just how much resilience they're getting from
these management techniques.

Exploitation isn't the only thing that selects
against large body size, selects against these

individuals that might be very important in
resilience to extreme events.

One of the other things we found in our modeling
exercise, modeling work at Westbrook, is that

we've got three sites, the red, green, and
blue lines here that are part of a connected

system, so two tributaries and a main stem
where individuals can move freely between

those habitats.

The purple line describes the situation in
one of our tributaries which is isolated by

a waterfall from the rest of the system.

What we found really clearly was that we looked
at probability of survival.

Survival probabilities for these large potentially
highvalue fish were substantially lower in

these isolated tributaries providing evidence
that there's selection, potentially, against

large fish in isolated habitat.

This selection against large highvalue fish
in isolated small stream habitats combined

with access to highgrowth downstream lake
and ocean habitats.

Allowing fish to move to gain these growth
opportunities, which is again probably the

fundamental reason why fish adopt migratory
strategies in the first place, allowing fish

to do that, allowing not just fish but individuals
from any species to maximize growth opportunity

if and when size is related to fecundity and
reproductive value can help to contribute

to resilient populations.

I'm going to move from temporal variation
recruitment associated with environmental

variability to spatial variation, and one
interesting thing to note is that I've talked

about, some temporal mechanisms that can restore
or regulate the numbers of individuals, that

can help regulate population size and keep
them low.

One interesting thing is even though these
mechanisms can help population numbers recover

from low values they can't restore alleles.

They can't restore genetic diversity.

As these populations go through bottlenecks,
even if they're able to respond demographically

in the way that we've talked about, they're
still going to lose genetic diversity with

some potentially important consequences.

This is a really interesting contrast with
spatial compensatory mechanisms where you've

got the potential both to restore numbers,
the mechanisms I'll talk about, and to restore

and conserve genetic diversity.

This has some implications for populations
in the stage of extreme events.

The basic idea here is that instead of good
years compensating for bad years, good locations,

locations where numbers are good for compensating
for bad locations, this is supposedly tied

to metapopulation concepts where we have subpopulations
potentially subsidizing each other and maintaining

the overall population level.

And also more recently to what Dan Schindler
described as portfolio effect where if you've

got variation, environmentally associated
variation in recruitment, you're encompassing

a diversity of conditions that may respond
differently to extreme events, you can enable


That's why this...if you've got this diversity
across the landscape and subpopulations are

connected can be a really important mechanism
for keeping populations resilient.

I said I was going to talk a lot about Westbrook

and brook trout.

Now I'm going to talk a lot about Hurricane
Irene, which was a major flood event that

occurred in 2011, record flooding in western
New England and a flood of record in many,

many rivers, lots of damage, lots of impact,
and along with this, of course, we saw some

immediate reductions in trout numbers.

Oftentimes trout populations are quite resilient,
particularly the adults, to pretty high flows.

They're good at finding refuges.

They're good at persisting under what seems
like pretty intense conditions.

In this case we saw some substantial losses,
particularly of our radiotagged fish where

we had really good records.

They were adults.

They were successful.

It was a really big flood, and you can see
from the picture, the top picture there, you've

got lots of bed movement, lots of scour, lots
of action, lots of power, and that's the thing

that really gets trout.

We were interested, and we actually had the
ability to try and look at how this reduction,

this extreme event, put populations in jeopardy
of reductions of genetic diversity with potential

longerterm consequences.

Again, even though they could potentially
recover demographically, what were the consequences?

Genetically we were in luck because we happened
to have genetic samples from a number of rivers

that were subsequently impacted by Irene,
before Irene and then after Irene.

On the left hand side this is the Millbrook
watershed, which is in western Massachusetts,

and on the right we've got a number of alleles,
measurable allelic diversity in two sites,

which, again I wish I could find a pointer,
but I can't in the Millbrook above and below

a barrier to migration.

While we saw reductions in allelic diversity
at a number of those sites, what was interesting

or the interesting result that seems to be
emerging and these data are still being analyzed

is that we were most concerned about sites
I don't know if you can see this small headwater

sites that were above barriers, which our
thought was they would lose allelic diversity

as a consequence of this reduction in number
in the flood, but it couldn't then be subsidized

from downstream.

Those were our major concerns, but what turned
out is that the sections that seemed to lose

the most allelic diversity were the further
downstream sections more towards the main

stem Millbrook, whereas our headwaters actually
retained allelic diversity quite well.

This bears directly on this concept of the
portfolio effect, because it's potentially

a really good illustration of how extreme
events, extreme climate event disturbances,

if they manifest differently in different
systems can be buffered against by the populations.

What we think we're seeing here is that even
in very large floods, very, very small headwaters

right at the threshold of perenniality may
be less vulnerable because they experience,

depending on the way the flood is generated,
reduced increases in per unit power during

flood generation.

Once you get to the midreaches, once you get
downstream, you're going to see increase in

power, bed movement, overbank floods, and
that jives very well with what we actually

saw on the ground with Irene that the streams
and rivers that really got hammered were midreaches

and not headwaters.

What's interesting, too, is that other kinds
of extreme climate you could see the exact


For example, droughts might be particularly
challenging for headwaters again at this threshold

for perennial flow, which is actually quite
a bit of brook trout habitat.

Maintaining these connections between different
habitats that respond differently to extreme

events could be quite important.

With respect to management implications, well,
I mean, it's like being against apple pie

to be against increasing connectivity.

Everyone wants to do it.

We all know it's important.

It's a major management initiative, but again,
by recognizing some of these effects and establishing

them we can value them with respect to their

More interesting in a sense is that you do
have some potential challenges or some potential

conflicts for species with relatively narrow
habitat requirements.

For example, cold water fishes like brook
trout if you're prioritizing based on thermal

resilience, and you're focusing only on the
headwaters, or you're focusing only on one

longitudinal strata in the network you're
not going to get that portfolio effect of

differential impacts at differential points
in the system helping to make populations

more resilient.

That's where some of these prioritization
schemes, folks like Matt Tibold, in the Midwest

that specifically consider stream order diversity
in terms of prioritizing barrier removals

can be particularly important even for species
with relatively a narrow requirement if that

stream order diversity also results in extreme
events having differential effects in different

parts of the connected network, and, obviously,
incorporating this into model simulations

is key.

Then, finally, you've got some interesting
potentials for directly mitigating the effects

of losses of alleles during loss of genetic
diversity during extreme events, and we're

working with that now with the genetic refuge
experiment where we're bringing fish from

anthropogenically isolated into ontogenetically
isolated populations and trying to restore

that diversity and seeing what effect that
has on this.

I'll go relatively quickly through the rest
of my slides.

In addition to demographic and conservation
genetic effects, obviously this has important

effects of climate regimes on habitat themselves.

The real difference here is that with the

demographic effect we're talking about climate
events moving things up and down with respect

to some carrying capacity.

With respect to the climate regime's effect
on habitats we're looking at changes in the

carrying capacity itself.

A way to demonstrate this is, in contrast
to the top graph that I showed earlier, the

bottom panel you've got a lower carrying capacity
for older juvenile fish, and that's going

to result in a substantially lower ultimate
population size, and that often results in

changes in habitats.

Streams and rivers: extreme events always
had some recognition of their importance,

strong influence with the twoyear flood in
shaping channel planform, bed caliber, flood

plain/channel connectivity, a lot of things
we're interested in.

What about the more extreme flows associated
with things like Hurricane Irene and that

may become more frequent and more intense
in the future?

An interesting point here is that you could
potentially see a shift in some dominant habitatforming

mechanisms, mechanisms associated with importing
wood, sediment, and other materials from terrestrial

environments to streams, changing the relative
dominance of chronic processes like single

tree mortality, bank erosion, to more episodic
mechanisms, for example, extreme winds and

hill slope failure shown in that study, the
picture to the left.

Similarly in forests, direct relationship
with the frequency and magnitude of timing

of extreme events with respect to the relative
importance of climate associated events like

fires, floods, wind flow, drought, versus
the classic competition and successional dynamics.

In terms of management implications I think
that the interesting thing here is to try

and bring extreme event predictions and habitat
goal expectations together and also underscore

the need to combine empirical mechanistic
models, both of which have strengths and weaknesses

with respect to their ability to incorporate
extreme events.

Empirical models, those events are in there.

For example, all the mechanisms that determine
a climate envelope for a given species, frequency

and magnitude of extreme events is implicit,
but mechanistic models can help make them

explicit and help us look at thresholds for
tolerance and perspectives.

This could be particularly useful when we've
got a lot of work.

For example, this is work done by Dave King
in my lab on disturbance dependent birds establishing

these relationships, the ability to tie these
relationships between time sensitive disturbances,

in this case actual manual treatment, but
potentially natural disturbance, could make

these models much more effective.

I'm going to end quickly with respect to some
human responses, and when I'm talking about

human responses to extreme events, I'm talking
about responses to the events themselves and

also responses to the risks, of either the
perceived or actual risks of these events.

With the general point being that in highly
settled regions like the north and northeast

human response has the capacity to really
override natural dynamics and do what I'm

going to call and I realize this is a valueladen
way to describe it very strong influence on

whether or not you catalyze ‘virtuous’
versus ‘vicious’ cycles of response and

impact from the perspective of natural resources.

Just briefly what I mean by that.

Obviously, Hurricane Irene, as you guys know
well, and other hurricanes, that not only

affected habitats and natural populations
but a big impact on people of particularly

limited infrastructure.

One route, one virtuous cycle that could be
catalyzed by all these road failures is recognition

of their importance, real emphasis on right-sizing
road stream crossings with the confidence

and benefits of less damage next time.

We've shown pretty clearly that those road
crossings that were right-sized for our groups,

and Nat Gillespie, and National Park groups,
those folks.

That those were crossings that were right-sized,
that were over the banks or width of the channel

and on that graph in the presentation there,
were much less likely to fail than the large

majority of crossings that were less than
bank-full size.

You've got less infrastructure damage, and
you also have more habitat connectivity and

more resilient populations, what I would call
a virtuous cycle.

In contrast, catastrophic flooding, roads
go out, regulations lifted.

Anything that's at hand gets thrown in the
stream to rebuild the road, and you've got

the possibility for more damage next time,
more habitat fragmentation, and more vulnerable


I wanted to close by referencing a riparian
study by Anita Milman here at UMass, the Department

of Environmental Conservation.

Recognizing how important people's responses
and attitudes are in moving things along different

paths, is currently working on a survey of
riparian landowners in Vermont who have been

affected by Irene.

Their responses are worth listening to and
worth engaging if we want to more fully manage

in the context of these kinds of changes and
extreme events.

One of the things that we can really do with
respect to looking at the multiple dimensions

of resilience extreme events is to put this
in the context of vulnerability and exposure.

This is a graphic that Andrew, Ben, and I
put together and what it shows is how with

increasing exposure to any kind of climate
impact, which is on the Xaxis.

The deviation of that horizontal line, which
is ‘no sensitivity’.

So essentially, you're having no change in
the performance parameter, in this case abundance,

with increasing exposure to climate extremes
or other aspects of climate.

Then, the family of curves below that show
increasing levels of sensitivity to a given

level of exposure and we found this a really
helpful framework within which to put both

the different mechanisms of the resilience
and the different management actions that

can be taken to try and address that resilience.

Shayna:  Shawn, I’ll turn it over to you.

Did you have any closing comments or words
for us?

Emily:  Hi.

This is actually Emily Fort (Shawn had to
duck out) from NCCWSC, but just to say thanks

to Keith and to everyone for attending.

As always, we appreciate it.