What's the Deal with Wisconsin's Walleye?

Video Transcript
Download Video
Right-click and save to download

Detailed Description

This video was created as part of the Climate Change Science and Management Webinar Series hosted by the U.S. Geological Survey's National Climate Change and Wildlife Science Center and the U.S. Fish and Wildlife Service's National Conservation Training Center.

In this webinar, the speakers describe the status of ongoing research examining trends in Wisconsin’s sport fish communities and drivers of those trends. Although walleye populations and recruitment have declined in most lakes, some populations have remained stable and others have increased, suggesting that the relative importance of various drivers differs among lakes. The research team developed a regional model of lake temperatures used to identify thermal characteristics of lakes associated with fish community changes.

This research on fish community changes in Wisconsin is relevant to managers and researchers interested in how environmental change may influence lake communities, and how management responses can be most effective.


Date Taken:

Length: 00:59:22

Location Taken: WI, US


Ashley:  Good afternoon, or good morning,
from the U.S. Fish and Wildlife Service’s

National Conservation Training Center, in
Shepherdstown, West Virginia.

My name is Ashley Fortune and I would like
to welcome you to our webinar series, that

is held in partnership with the U.S. Geological
Survey's National Climate Change and Wildlife

Science Center in Reston, Virginia.

The NCCWSC's Climate Change Science and Management
Webinar Series highlights their sponsored

science projects related to climate change
impacts, and adaptation, and it aims to increase

awareness, and inform participants like you
about potential, and predicted climate change

impacts on fish and wildlife.

I do have the pleasure of announcing our speakers
for today's webinar.

We have three folks with us today that worked
on the project, and two will be presenting,

both Gretchen Hansen, and Dan Isermann, and
then Steve Carpenter is also on the line for


I'm going to introduce all three right up
here at the front of the webinar, and then

we'll get started.

Gretchen Hansen is a research scientist with
the Wisconsin Department of Natural Resources,

where she studies long term changes in lake

She has a Master's degree from Michigan State
University, and a PhD from the University

of Wisconsin, Madison.

Gretchen currently lives in Madison with her
husband and her two daughters.

Dan Isermann is the leader of the Wisconsin
Cooperative Fisheries Research Unit, and director

of the Fisheries Analysis Center at the University
of Wisconsin at Stevens Point.

Dan's research attempts to address fisheries
management issues, with a specific focus on

the population dynamics of walleyes and black

Steve Carpenter served as the director of
the Center of Mammalogy, and the University...excuse

me, at the University of Wisconsin in Madison,
where he was the Steven Alfred Forbes Professor

of Zoology.

Steve, Gretchen, and Dan have been working
together on Wisconsin walleye problems for

the past four years.

I'd like to welcome all three of our speakers,
and turn it over to Gretchen.

Thank you.

Gretchen Hansen:  Thanks Ashley.

I hope my mutes off now.

This is Gretchen Hansen.

I will be speaking first today, and I'll hand
it over to Dan Isermann for a little bit,

and then I will wrap it up, and then all three
of us will take questions.

Thanks everybody for joining us today.

It's exciting for me to be able to talk about
the culmination of four years of research

that we've been working on here in Wisconsin.

Since it's a webinar, and it's a little weird
that you can't see us, I thought I would put

our pictures up here, so you know who you're
dealing with.

There we are on your screen.

I need to figure out how to work this.

I should also, upfront, acknowledge that this
project was funded by the USGS Climate Change

Science Center.

As well as the Sport Fish Restoration Fund
and the Wisconsin DNR.

I want to acknowledge that upfront, and also
acknowledge that this, what I want to talk

about today, is work that has come together
form the hard work of a number of people.

We have this bass, walleye group, that we
call ourselves, that have been working for

the past four years on this research, and
we've really taken a collaborative approach,

and a multifaceted approach to do a lot of
different kinds of research that I'm going

to talk about today.

Just want to thank all these people for their
involvement in this project over the years.

I'm going to start by giving you a little
bit of background of how we got started on

this project.

In the early to mid 2000s, here in Wisconsin,
there were a lot of anecdotal reports coming

in to the Wisconsin DNR on a lake by lake
basis, mainly, of declines in the walleye

population, and largemouth bass population

I should point out right now, I'm going to
talk a lot today about bass, and I might just

say “bass”.

For the most part, I'm referring to largemouth

We do have smallmouth bass also in Wisconsin,
but our data are not as good on smallmouths.

We don't see the same level of trends, so
I'm just not really going to talk about smallmouths


When I say bass, I mean largemouth for the
purpose of this talk.

In the early to mid 2000s there were lot of
anecdotal reports of declines in walleye and

increases of largemouth bass.

This is concerning to managers in the state
of Wisconsin.

Walleye are the most targeted sport fish species
in the state, so anything that leads to walleye

declines is concerning.

There is some belief among some people in
the state that because these two trends are

happening concurrently, that possibly they
were directly related, that largemouth bass

increases were the cause of walleye decline.

So, there was some motivation both within
the DNR, and then among people at universities

to try to investigate this, to try to understand
a little bit more about the magnitude, the

extent, and the cause of these changes.

I should note that there were similar anecdotes
coming in from around the region as well.

It seems like this wasn't necessarily a Wisconsin

In Minnesota, Michigan, and Ontario there
was some inkling that similar trends might

be going on.

As these reports were coming in, there have
been some management responses in Wisconsin

to these declines in walleye, and increases
of bass.

One of the major ones has been what's known
as the Wisconsin Walleye Initiative.

$12 million were devoted to stocking more
walleye, specifically to stocking extended

growth fingerling walleye, so larger walleye
than are normally stocked.

That was one response to declining walleye

At the same time, there have been some restrictions
on harvest, including closure of a very prominent

fishery in Wisconsin to walleye harvests recently,
so certainly management is responding to these

declines in walleye, even when the causes
might be unknown.

At the same time, there has been a liberalization
of bass regulations to try to encourage more

harvest of bass, including some localized
interesting fishing tournament events to promote

the idea that harvesting, and cooking and
eating bass is something the public should

get interested in.

All this management response, while this is
going on we still have a lot of questions

about what the causes are, what's going on,
and causing these trends in these two species.

Like I said, there's some idea that possibly
the increases in bass and declines in walleye

were directly related, but as you can imagine,
there are a number of other factors that we

could hypothesize might be causing changes
in these fish species, and in fact all of

these factors may interact in a very complex
web of interactions, influencing bass and

walleye individually, as well as the way they
might interact with each other.

When we started this research project it became
clear that no single approach to understanding

the system was likely to work, and that we
needed to take a collaborative, and multifaceted

approach to start to disentangle this web.

That's what we did.

I'm going to organize my talk today, talking
about three major areas, some broad areas

of research that we've done over the past
four years.

The first thing that we wanted to do, was
to really quantify the magnitude, and the

extent of these patterns and trends in bass
and walleye.

So try to understand how severe is the problem
of walleye declines?

How widespread is the problem?

And the same questions for bass increases,
how much has bass increased, and how widespread

are those trends?

After we nailed down the magnitude of the
problem, we spent a lot of time generating,

and evaluating hypotheses about what might
be causing them.

Something I want to talk a lot about today
is water temperature, and how that might be

related to these trends.

We also tried to identify other factors we
thought might be associated with the trend,

and also identify knowledge gaps, and areas
where we needed to design some new studies,

and new research to get into some of the mechanisms
of what might be going on.

At the end of the talk I'm going to touch
on some of the ongoing, and future research,

and Dan, Dr Isermann will talk about that
as well.

Finally, these first two areas of research
are certainly related to management.

But we also spent some time focusing on research
that was really directly management oriented,

so really evaluating the role of harvest as
a management tool for largemouth bass to try

to understand, could increased angler harvest
control bass populations in Wisconsin?

We have started an adaptive management study
to try to evaluate our management responses

in the field.

I'll talk a little bit about that, and also
develop some models for prediction and prioritization

of locations to try to maximize the success
of our management actions by targeting them

to places where we think they will work.

I'll talk briefly about that as well.

Let's start at the beginning, identifying
patterns and trends.

This seemed like an important first step to
this project.

Like I said when we started, we had some anecdotal
reports on an individual lake basis, of declines

in walleye, and when we dug into all the data
we had, and we found indeed that state wide

walleye recruitment to age zero was declining.

I'm want to note, I'm going to talk a lot
about walleye recruitment today, and in all

cases I'm talking about recruitment to age
zero, so survival of young walleye to their

first fall.

I know that other states might define recruitment
differently, and certainly recruitment to

age zero is not the same as recruitment to
the fishery, but because we're seeing these

strong declines in recruitment to age zero,
where walleye in many places in Wisconsin

are not surviving past their first summer.

If they can't make it past their first summer,
they certainly can't make it to be a six,

or 10, or 20 year old walleye that can be

We're focusing a lot on why aren't they making
through that first summer.

What we found when we looked state-wide was
an average decline in recruitment of about

6.6 percent per year.

This is statewide average numbers, since 1989.

That was pretty concerning.

Obviously the statewide average doesn't tell
the whole story.

We have tens of thousands of lakes in Wisconsin,
so looking at individual lake trends was also


We looked at lakes where we had enough data,
over the past three decades, to try to identify

a trend.

We found a similar story, that in fact, in
the majority of lakes where we had data, walleye

recruitment was declining.

This histogram here in the upper left corner
shows annual percent change on the X axis.

That's this slope of the line on the long
scale of walleye recruitment over time, then

the percentage of lakes that show that trend.

The red dash line is the zero line.

Everything to the left is zero means walleye
recruitment in those lakes was declining.

Anything up here in the positive range means
recruitment was increasing since 1989.

There are a substantial number of lakes where
we do see increases, but the vast majority,

we see decline.

Those trends are plotted here on the map,
color coated with the blue color showing declines,

and the green showing increases.

What was interesting to note early on was
that there was a large degree of spatial heterogeneity,

so we could have lakes right next door to
each other where you might see strong declines

in recruitment in one lake, and then increases
in recruitment in another.

That spatial heterogeneity told us that this
wasn't just a regional trend that was operating

the same in every lake, that there were some
complexities that we needed to understand

to know why these lakes were responding differently.

That's walleye recruitment.

When we looked at adult walleye, we saw also
the statewide average adult walleye densities

were also declining since 1989, but the rate
of decline was not as large.

Average decline of about two percent per year
in this case.

That makes sense because adult walleye numbers
are really influenced by a large number of

things besides recruitment.

We do a lot of stalking in Wisconsin, as is
the case in many places, and harvest pressure

can also influence adult population.

So it makes sense that the declines would
not be as strong, but we did see declines

in adult density as well.

Again, when we look on a lake specific basis,
we once again see some heterogeneity, with

some lakes showing increases in adult walleye
density, but the majority having decreases.

When we looked at largemouth bass we saw,
for the most part, increases in largemouth

bass throughout the state.

The statewide average rate of increase was
about four percent for year.

In this case, in the majority of lakes, bass
were increasing, and then in a small number

we saw some decreases, and again, some spatial
heterogeneity throughout the state.

When we tried to look at concurrent trends
between large mouth bass and walleye, we found

we had not a huge number of lakes where we
had the ability to quantify trends in both

species, about 30 lakes where we had data
for both species that we could look at the

concurrent trends.

What you see here is a biplot.

On the X axis is the largemouth bass trend,
so anything over zero means largemouth bass

are increasing.

On the Y axis we have the walleye trend.

This is walleye recruitment in this case.

Anything below this zero line would mean walleye
are decreasing.

Perhaps not surprisingly given the trend in
the species individually, we see in most cases

where we have data for both, largemouth bass
are increasing, and walleye are decreasing

in the quadrant here.

Because largemouth bass themselves are most
of the time increasing, and walleye are most

of the time decreasing, we wanted to test
whether the cooccurrence of these trends is

happening more often than you would expect
by chance.

The result was somewhat equivocal, a P value
of 0.06 when we do Chi-squared test here.

I would say there's some moderate maybe, possibly
evidence that these trends are happening at

the same time more often than you would expect
by chance, but certainly nothing really conclusive

came out of this.

It's important to remember the lesson that
all of us have heard probably hundreds of

times but sometimes it's easy to forget, but
correlation doesn't equal causation.

We can see that in a lot of lakes bass are
increasing, walleye are decreasing, but this

doesn't really tell us much about the mechanism
of what might be causing these things.

I'd like to show a slide from this great website
called Spurious Correlations, where you can

find any number of interesting correlations.

This is my personal favorite.

In the U.S., per capita cheese consumption
correlates quite well with the number of people

who died by becoming tangled in their bed

This has an R squared of 0.9.

Most of us would be pretty excited to get
R squares of 0.9 in our analysis.

Maybe you can come up with some post hoc explanation
of why these two things might be related,

but I think all of us can agree that this
is a spurious correlation.

I like to put this up as a reminder that because
you see these trends happening at the same

time, it doesn't mean that they're directly

A big part of our job as researchers here
is to dig a little deeper, and understand

what might be the mechanism.

That was the next step in our research approach,
was to, as I said, dig a little deeper and

look into what else might be changing at the
same time in these lakes that we know could

potentially be associated with these two species,
and then to design some new project together,

some new data, to try to understand more about

A lot of what I have been working on is focused
on water temperature, and the potential role

of water temperature in driving trends in
fish species in lakes in Wisconsin.

Temperature can be thought of as a “master
factor” in ecology.

It controls the rates of pretty much every
process that we might care about, from nutrient

cycling to oxygen concentration, algal dynamics,
zooplankton dynamics, and of course fish.

Temperature controls the distribution, growth,
survival, reproduction, every major rate of

fish population, so it's very important.

We wanted to evaluate how temperature might
be related to the trends that we have seen

in bass and walleye in Wisconsin.

For those of you who are not lake people,
I thought I'd take a minute to talk about

water temperature and lakes because if you
want to know something about what is the water

temperature of this lake, it's not a matter
of knowing a single number.

Water temperature in lakes, the kind of lakes
that we're most interested in, for the most

part in Wisconsin is heterogeneous.

Most of the lakes that we're dealing with
that have bass and walleye in them stratify

in the summer, meaning that the water segregates
based on temperature.

With warm water in the upper region of the
lake known as the epilimnion and cold water

in the deeper waters, known as the hypolimnion.

Those layers don't really mix because of the
density differences.

In water, they're really separated from each
other for most of the summer.

This is important from a fisheries perspective
because fish species have distinct temperature


Largemouth bass are a warm water fish that
are most likely going to prefer the upper

waters of a lake, whereas walleye are a cool
water fish.

Probably more likely to be found in the middle
area of a lake, where the water is a bit cooler

than you find at the surface.

In trying to understand the role of temperature
and explaining the trends that we saw, it

seemed like it was important to understand
temperature on a whole lake basis for our

lakes in Wisconsin.

Unfortunately, we don't have a lot of data
on water temperature in lakes, particularly

over the time scale that we were interested
in, the past three decades, at the resolution

that we might care about.

So knowing something about temperature over
the whole course of a season and certainly

not for knowing temperature across the entire
profile or depth range of a lake.

We just don't have that data for most lakes
in Wisconsin or really in the world I would


The approach that we have taken is to model
temperatures from known conditions and try

to hindcast what we think water temperatures
in lakes were likely to have been in the past

using a mechanistic thermodynamic model.

I'm not going to talk a lot about the details
of this model.

You can find those details in this paper listed
here or contact me or Jordan Read or my other

coauthors here later.

We'd be happy to talk about it.

For the purposes of this talk, I will say
that this model uses air temperature and solar

radiation and wind information from past days,
where we have that information, combined with

lake specific characteristics, like water
clarity and canopy cover, which influences

how wind will affect the lake.

Then, like I said, uses a thermodynamic model
to hindcast daily temperature profiles of


The output of this model is depth specific
daily temperature values.

We did this for about 2,400 lakes from 1979
through 2012 in Wisconsin.

This heat map shows an example of the data
you'd get for one lake for one open water

season, with warmer water at the top in the
warm colors, cold water at the bottom in the

cool colors.

Imagine we had this level of data for 2,400
lakes for 30 plus years.

The model works quite well to hindcast water

We were pretty happy with the result.

But we wanted to distill this vast amount
of data into metrics that were biologically

relevant for the species that we're interested

For example, instead of using daily temperature
profiles, we would calculate metrics such

as growing degree days, which is a measure
of the cumulative water temperature in a lake,

as well as a large number of other temperature

These are the metrics that we then try to
associate with fish populations to see if

temperature could explain the trends that
we were seeing.

But as a side note, one thing that we found
was that water temperatures were quite variable

across Wisconsin, so when we look at this
map of growing degree days, on this scale

it's probably not surprising that you see
lakes in the southern part of the state are

more red.

Meaning higher growing degree days, meaning
warmer water.

Warmer water, higher growing degree days in
the south compared to the north.

That's probably not very surprising.

But if you drill in and zoom in a little bit
closer...and I should note that the color

scale here has changed, but it still represents
a fairly large difference in growing degree


When we zoom in close like this, you can see
that lakes right next to each other can have

very different temperatures.

Like if we circle this little group of three
lakes here, three lakes almost right on top

of each other that span a range of about 500
growing degree days.

They're quite different.

Seeing this small scale heterogeneity in water
temperatures was interesting, given that we

saw small scale heterogeneity in walleye trends
as well.

So this was encouraging as we started our
temperature modeling.

The next step was then to more formally try
to relate water temperature metrics that we

thought might be related to walleye and bass
to the walleye and bass populations that we

had data for across the state.

The way that we did that was using a statistical
model known as a random forest model.

Again, I'm not going to get too deep into
the details of this modeling.

I'm happy to talk about it later with anybody
who's interested.

For the purposes of this talk I will say it
is a tree based method that classifies data...in

our case, what I'm showing here is probability
of walleye recruitment success.

In this case, a yes or no.

Did recruitment happen or did it not happen?

So the random forest, it will look at a large
number of predictor variables, and identify

relationships between those predictor variables
and recruitment success.

Random forest is a great method for our purposes,
because it can identify nonlinear relationships

as well as interactions, which can be really
important in complex systems like this.

What I'm going to show you here is the relationship
between a single variable and probability

of recruitment success for the variables that
we've selected using a model selection technique

as the best predictors of walleye recruitment.

In random forest, the effects of one variable
often depend on the level of another variable

or of all the other variables, so what I'm
going to show in these figures is the median

effect of the variable of interest.

In this case, lake area is what I'm showing.

So the black line is the median effect, and
then the gray bars, the interquartile range,

given all the other values of the other variables.

What we see for walleye recruitment, this
first most important variable was lake area.

We see lakes with larger surface areas have
a higher probability of walleye recruitment.

There were four other variables that came
out as important in predicting walleye recruitment


Three of those were related to water temperature,
so we have the coefficient of variation of

surface water temperatures, both 30 to 60
days posticeoff, and 0 to 30 days posticeoff.

So variability in water temperatures as walleye
are spawning, as walleye are in their egg

stage and then immediately following some
up when they're fry and larval stage.

That's these two CV metrics.

Then growing degree days, which I talked about
before, which is the cumulative measure of

water temperature in a year.

Here we see lower growing degree days means
cooler water.

Walleye recruitment is more likely when degree
days are lower.

Less likely in warmer waters when degree days
are higher.

For the variability metrics, walleye recruitment
is more likely in both cases when variability

is lower in spring water temperature.

Like I said, this random forest technique
can also identify interactions between variables.

The effect of one predictor, say growing degree
days, may depend on the level of another predictor.

In this case, the variability of water temperatures
in that 30 to 60 days after ice goes off of

the lake.

I'm showing you here is a contour plot where
the darker purple colors represent higher

probability of walleye recruitment, and the
lighter pale blue represents lower probability

of walleye recruitment.

Here we see there is this sweet spot for walleye
recruitment, where degree days are less than

about 2,4002,500 and variability of water
temperatures are below around 0.17.

In that sweet spot, walleye recruitment is
more likely.

Outside of that in any direction, the probability
of walleye recruitment goes down.

But because of some of the interactions, you
can see a little bit darker colors in this

region and this region than out here in the

For example, you can still have a decent probability
of walleye recruitment even with high degree

days as long as your CV of water temperatures
is low.

And conversely, you can still have a decent
probability of walleye recruitment at high

levels of variability as long as your degree
days are low.

These counter plots help us to identify interactions
and what kinds of conditions are most conducive

to walleye recruitment.

What was interesting to see, what we then
did was to look at how our walleye lakes in

Wisconsin have changed over time in terms
of these temperature variables.

That's what I'm showing here.

This plots a path of the median of all Wisconsin
walleye lakes in terms of these two water

temperature variables.

So you see a movement from this sweet spot
purple zone out into the not so good blue


Over time, we're moving away from places where
recruitment is most likely.

We did the same thing for bass.

Same techniques to try to predict bass abundance
from variables we thought might be important.

Again, we see in this case only one temperature
variable came out as important.

It was degree days once again.

But in this case, the relationship was opposite.

Largemouth bass abundance was predicted to
be highest when degree days were higher, so

in warmer waters.

Then some other variables really did, to lake
morphometry and landscape position were also


I thought it was interesting to plot the effects
of degree days for both species side by side.

So we can see that they look like almost mirror
images of each other with a threshold at around

2,4002,500 degree days separating high probability
of walleye recruitment from low and high probability

of there being a lot of bass from low.

We found in both cases that the temperature
effect was strongest in small lakes.

So the effect of growing degree days was much
higher in lakes of, say, 100 hectares, shown

here for both species with the black line,
than in bigger lakes, say, of 1,000 hectares,

shown in the blue lines.

It's interesting to find, using totally independent
data sets we found really similar results

for both walleye and for bass, suggesting
that maybe they're both responding to temperature

in some of these lakes.

Again, I have to go back to my favorite slide,
we can do this high level statistical modeling

that provides some really great information
of the types of conditions that are most associated

with good walleye recruitment or high bass

But again, we're still working really with
correlations and we don't know the mechanism

in this case.

When growing degree days gets high it's not
like it's necessarily too warm for walleye

to live.

There's some complex interaction going on
there that we needed to drill a little deeper

to understand the mechanisms.

That's where my colleague, Dr. Isermann, will
talk now.

Dan Isermann:  All right, so as Gretchen
mentioned early in her presentation with some

of her slides, the interactions within these
ecosystems can be pretty complex.

When we think about the relationship between
bass and walleye there are two rather obvious

mechanisms that come to most peoples' mind
that could cause that to occur.

That would be direct predation by bass on
young walleye and then competition for available


To assess the extent of those mechanisms we
went out and collected diets from hundreds

of largemouth bass and walleyes from four
northern Wisconsin lakes with various combinations

of walleye and largemouth bass abundance.

We did this from May to October and, for brevity's
sake, I'm showing you summarized data here.

We also used DNA bar coding to help us reduce
error associated with partially digested diet


What we've found is that largemouth bass rarely
consumed walleyes.

We saw one incident of this in 945 largemouth
bass diets, so fairly rare.

This was true even when relatively strong
year classes of age zero walleyes were present.

Walleyes and bass shared a wide variety of
prey items.

The top five are summarized here.

As you can see the two species where we saw
the highest level of niche overlap, or groups

of species here, was sunfish and then yellow

Of course, our findings don't allow us to
declare that competition is occurring because

these resources would have to be limiting.

But it at least gives us an idea that these
are the two groups where the highest potential

for competition might be.

As Gretchen also demonstrated that there's
been these landscape level changes, probably

in temperature regimes that have provided
for increased bass recruitment over the last

decade or more.

Many previous studies have suggested that
the growth of bass and the length that they

attain in their first year of life can influence
recruitment through size selective mortality


And that these factors can be influenced by
hatch timing.

But there's very limited information on this
for largemouth bass in northern lakes.

What we did is went out to seven lakes across
the state of Wisconsin.

We've been doing this over the last four years.

We've collected age zero largemouth bass of
approximately this size and then we've removed

their otoliths so that we can use the daily
rings pictured here to estimate their hatch


We're collecting these bass at the end of
July, early August, and counting their daily

rings and their otoliths to see.

Are we seeing trends in hatch dates that could
lead maybe to increased size and eventual


I'm going to show you two years here.

2012 was the first year we did this.

This was the year in the upper Midwest where
we had some 80 degree water temperatures at

the end of March.

Fairly early ice out, even early April.

Then in 2013 this was a year where people
were ice fishing in some places on the walleye


So, two very stark years in terms of temperature

You can see that the median hatch date in
2012 was much earlier than the median hatch

date in 2013, although we didn't really see
any major differences in hatching duration

which is depicted in this other graph here.

Now the important part about this is whether
it equated to anything in terms of the size

of the bass at the end of summer.

Certainly when we look at early hatched fish,
middle, and late hatched fish, in 2012 we

do see a trend that these early hatched fish
were slightly larger but the differences are

pretty minor.

We're talking on the order of three to six

Then in 2013 with the late ice out we did
not really observe any, or we didn't observe

any significant differences in the average
size of these fish at the end of their first


We've got two more years of data to add to
this analysis, so we've continued to collect

these fish from these lakes.

One of the other questions Gretchen hinted
at is, we know that we're losing these walleyes

essentially in the first year of life based
on not catching them in age zero electro fishing


One of the primary questions is when exactly
in their development are we losing them?

Then these lakes that have different recruitment
histories, do we see variation in abiotic

and biotic factors such as temperature or
water clarity?

Hadley Baum, who's one of our current grad
students, is working on this project.

We selected four lakes in northern Wisconsin,
two that have a history of sustained walleye

natural reproduction, meaning that it continues
at a variable rate as would be expected.

That's Escanaba and Big Arbor Vitae lakes.

Then two lakes where we've seen declining
natural reproduction of walleyes to the point

that we have not observed any walleyes in
these systems in a couple of years.

Then we essentially went to these systems
and threw everything but the kitchen sink

at them.

In terms of trying to collect walleyes in
their first year of life.

This included egg mats, larval towing and
light traps, staining, micromesh gill nets,

and then our typical age zero electro fishing
that occurs at the end of each summer in the

early fall.

This allowed us to develop a method for assessing
walleyes in these lakes, which includes larval

towing at night in late May, these micromesh
gill nets in midJuly to lateJuly, and then

the typical age zero electro fishing to sample
them at the end of the summer.

What we also found is that in these lakes
where we're seeing sustained natural reproduction,

we always encountered adults, we always captured
eggs on the egg mats and we observed both

larvae and juveniles in both years.

But on the declining NR lakes there were some
adults still present, and we did collect relatively

low numbers of eggs in both years, but we
never encountered a larva walleye or a juvenile

fish in age zero electro fishing.

This makes us think that the bottleneck is
probably at the larval stage or earlier.

We're still processing zooplankton samples,
so the differences in these lakes in terms

of temperature, water clarity, and zooplankton
are yet to be determined.

However, we do know the two declining NR lakes
are generally clearer than our two sustaining,

natural reproducing systems.

With that Gretchen's going to take back over

Gretchen:  It's still you, Dan, for a couple
more slides.

Dan:  Next we're going to talk about this
next phase of the project, which is these

management actions.

Of course, one of the things Gretchen mentioned
is that the DNR has liberalized bass harvest

regulations on some systems in some locations.

To try to potentially alleviate any effects
that might occur between the two species.

Then additionally, for the sake of bass management,
increasing abundance does pose some problems

in terms of density dependent growth and maybe
our ability to provide quality sized fish

for anglers.

On the four lakes where we did the diet work
we also wanted to simulate what effects do

fishing mortality have on bass abundance and
what would it take to reduce bass abundance?

This is in light of the fact that most anglers
release most of the bass they catch.

At best guess, we're talking about exploitation
rates that are probably, on an annual level,

five percent or less.

On these systems, we collected a variety of
data that included mark recapture, population

estimates, otolith based estimates of fishing
mortality and growth trajectories.

We developed some stock recruit relationships
and then put these in to age structured models

to simulate the effects of fishing mortality
under different harvest regulations that were

chosen by the Wisconsin DNR's bass management

One of the first things we learned is that
bass in these northern lakes can live a really

long time.

This is the oldest fish that we captured in
the study.

You'll see 21 annuli here, at least I see
21, and then if you add a year you get a 22yearold


They regularly live beyond 10 years of age.

That means that a strong year class can persist
for a very long time and influence abundance

estimates well into the future.

A lot of these really large fish, say fish
over 18 inches long that anglers really want

to catch, they're generally over 15yearsold.

It's taking them quite a while to get to those
larger sizes.

When we look at the results of our simulations,
I'm going to show you one lake here so you

can get a feel for this.

These white dots here represent a 25 percent
reduction in initial abundance.

You can see, regardless of the harvest regulation,
they require a pretty substantial amount of

fishing mortality relative to what we're seeing
on the landscape now.

In terms of reducing abundance while still
maintaining size structure to some level.

These two middle harvest regulations probably
offered the best case scenario.

Certainly a no minimum length limit was the
best option and required the lowest amount

of fishing mortality to get the 25 percent
reduction in abundance.

A substantial increase in fishing mortality
would be needed in order for this to really

work effectively.

All right, now I'm going to pass it on to

Gretchen:  Thanks Dan.

I'm going to wrap up with a couple more, mostly
ongoing and future projects.

Let me get my little pointer back.

I mentioned in the beginning we have initiated
an adaptive management experiment in collaboration

with many of our biologists throughout the
state who've been willing to work with us

to try to understand what's going on out there
in the landscape.

As I said in the beginning, this whole research
project started because there were a lot of

reports coming in of lake specific declines
in walleye and increases in bass.

There were a lot of places where the stakeholders
and the biologists wanted regulations to be

changed in order to try to reverse those trends.

We designed an adaptive management experiment
to try to do these regulation changes in a

way that will allow us to learn if they work
or if they don't.

What we've done is we have 20 experimental
lakes and 10 reference lakes.

Sorry, let's start with the experimental.

So, we have 20 experimental lakes where we've
seen these declines in walleye and increases

in bass, where the regulations have been changed
in three important ways.

First, increased stocking of those extended
growth, large walleye fingerlings is going


Also, more restrictions on walleye harvest
to try to protect adult walleye populations

and then a liberalization of largemouth bass
regulations to try to encourage more bass


We're really throwing the three major tools
that we might use in the state of Wisconsin

to try to change the trajectories of sport
fish populations.

We're throwing all three at them at once to
see if the trends in these lakes can be reversed.

Importantly, for the prospect of learning
we also have 10 reference lakes.

In these lakes the same trends have been observed
but we're not doing any of those three regulation


So, this will allow us to track overtime things
that we care about, like walleye recruitment,

in both the reference and the treatment lakes
and see if there's any difference in response

in the places where we've done these regulation
changes compared to the places where we haven't.

This is ongoing and hopefully we'll have at
least some preliminary results in the next

couple of years.

Then another thing we're doing is to try to
use the statistical models that we've developed

to help prioritize management actions.

One example of this that's already occurred
is we have used the walleye recruitment model

that predicts the probability of natural walleye
recruitment for really any lake in Wisconsin.

We've used that to help prioritize stocking
under the Wisconsin walleye initiative.

There were a large number of lakes, several
hundred lakes, that were proposed for stocking.

We ran them through our model to say what
is the probability that these lakes could

support natural recruitment?

Places where the probability was high, so
conditions seemed good that walleye natural

recruitment could occur but it wasn't happening,
those were prioritized for stocking with the

idea that stocking should be prioritized in
places where maybe natural recruitment could

be restored.

Ongoing work now is to take our water temperature
model that I described earlier and extend

that out into the future.

So, what I talked about before was hindcast
water temperatures and we also have developed

forecast water temperatures under various
climate models.

In progress now is to use that information
to make projections about future fish populations,

future walleye recruitment, and hopefully
use that information to help prioritize management

to places where it's most likely to be successful.

Identify where we expect resilient walleye
populations to exist and maybe focus our protection

efforts on those lakes and also identify lakes
where they might not have as great of a chance.

That's ongoing.

Now I'm going to sum up.

Like I said in the beginning, we separated
our project into three major areas of research.

We identified patterns and we discovered that
walleye's adult densities and recruitment

have declined over time while largemouth bass
abundance has increased over time.

But that in both cases these trends can be
somewhat specially variable, not going in

the same direction or at the same magnitude
in every lake.

We have looked at a number of hypotheses and
generated some new ones.

We found that walleye recruitment is most
likely in large, cool lakes.

Largemouth bass abundance is highest in small,
warm lakes.

We have some evidence that walleye recruitment
failure is occurring at the fry stage or potentially

even earlier in some lakes.

Also, we found that adult largemouth bass
rarely consume walleye, suggesting that direct

predation is not the mechanism operating here.

We also found that largemouth bass and walleye
share prey and have a substantial degree of

overlap between the prey resources.

But we can't really say if competition is
happening or not because we don't know if

those resources are limiting.

We found that ice out timing influences hatch
timing of largemouth bass, but that hatch

timing seems to have fairly little influence
on the length of those bass at the end of

their first summer.

Finally, in terms of management, we have identified
that substantial increases in angler harvests

are going to be needed in order for angling
to reduce largemouth bass abundance because

most anglers release most of the bass that
they catch and because bass are so long lived

and have low mortality.

Our adaptive management experiment will allow
us to evaluate the effectiveness of the regulation

changes that have already been implemented.

We hope that the predictive modeling we continue
to work on will help to identify locations

where management and success can be most likely.

So, that's it.

Looks like we have about 10 minutes for questions.

Here's the contact information for both me
and Dan.

I should say, I cut off the end of his beautiful
pike here in the first picture, so I thought

I had to put it on in the last one to try
to...And I had to throw a picture of myself

with a fish, too.

Ashley will take over to moderate the question

Ashley:  Yes.

Thank you very much Gretchen and Dan, wonderful

From Henry, it says, "What about northern
pike population trends?"

Gretchen:  Unfortunately, we don't have great
data on northern pike populations over time

in Wisconsin.

We do have some research scientists working
on that on a smaller scale basis, but I haven't

worked much on it, so I can't really say much

Dan:  I would say in the course of our diet
work we did, when we encountered pike, often

look at their diet items.

We did not see any walleye predation during
the study.

Although, we have seen some predation in additional
samples that have come through the office.

So, we certainly know they can be a predator
for walleyes, but whether they're abundance

is up or down is a little more difficult to
tease out of our sampling gears.

Ashley:  We have another question coming
in from Sean.

It says, "You said you're planning to increase
the harvest of largemouth bass, but elsewhere

you said there was no direct predation by
largemouth bass on the walleye and little

competition for food resources.

Just wondering why you think this will help?"

Gretchen:  Yeah, that's a good question,

One answer is that, Dan alluded to this a
little bit, that in addition to having concerns

about potential direct effects of largemouth
bass on walleye.

There are additional concerns about the increased
densities of largemouth bass causing changes

to their size structure that are not desirable.

That by decreasing densities you can improve
the largemouth bass fishery, in terms of its

size structure.

That's one answer.

Another would be that some of these regulation
changes were put in place before those predation

results were attained.

I don't know if Dan has anything more to add
to that?

Dan:  I think a lot of the initial changes
in the regulations were a good example of

a management agency being proactive in responding
to the changes, and also were somewhat experimental

as part of a recovery effort.

If we make these changes to management on
some system, do we garner a response from

the walleyes in the system, and then maybe
sort out the causative mechanisms later because

it does take more time to do that.

There hasn't been any more, to my knowledge,
anymore real recent liberalization attempts.

A lot of these happened very early in this
research where the first thing we was these

trends between the two species.

Ashley:  We have another question from Patrick,
and it's "Was there any insight gained into

the potential effect of zebra muscles on the
walleye recruitment?"

Gretchen:  We don't have a ton of lakes with
zebra mussels in Wisconsin, and that wasn't

something that I looked at in the large scale
analysis, so my gut response is zebra mussels

are not a major mechanism.

That said, clarity does seem to potentially
be a mechanism, so certainly zebra mussels

could be an influence, but it's not something
that has really been focused on here in Wisconsin.

Ashley:  Another question comes out on the
chat from Rick, and it says, "Are lakes with

decreasing recruitment of walleyes the same
as the lakes with decreasing trends of the

walleye adults?"

Gretchen:  Yes, sometimes.

We don't have as much data on adults.

We measure adults doing mark-recapture studies
in most cases, in Northern Wisconsin in particular.

Those are, as you can imagine, pretty resource

We do many more recruitment surveys per year
than we do adult density surveys, so we don't

necessarily have good time series data on
both walleye recruitment, and walleye adults

in all of the lakes, but yes, in places where
we do see declining adults, we generally also

see declining recruitment.

Usually we see recruitment declining first.

Ashley:  Thank you.

From Gregor, it says "Have there ever been
efforts to establish walleye populations south

of their historical range?"

Dan:  Certainly that would be true within
the state of Wisconsin, and across North America.

They've been stocked probably, I'm going to
say in most of the 48 contiguous states, so

certainly in southern reservoirs, including
places like Tennessee and elsewhere, where

they've been stocked outside of their...

They may have occurred there, but they've
been stocked into systems where they did not

previously occur.

Ashley:  A further comment on that.

It says, "Such efforts might be analogous
to efforts to retain walleye in the southern

parts of their historical range, as temperatures
warm, and so might provide useful information

for efforts to resist the effects of warming."

Gretchen:  I would say that natural recruitment
in those southern populations is probably

not occurring, at least in small inland lakes,
like we're talking about here, so that is

a useful system to look at.

Ashley:  I have two more questions, one from

It says, "Are the efforts of the Walleyes
for tomorrow aiding walleye reproduction?"

Dan:  One of the additional studies that
we have, that's been ongoing, is looking at

availability of spawning habitat in relation
to recruitment.

This, I know, is something that Gretchen's
working towards as well.

A lot of these groups are either stocking,
or making some effort to improve spawning


With regards to the spawning habitat end of
it, we've had one initial study where we looked

at 16 lakes in Northern Wisconsin, and were
really unable to find any strong evidence

that the amount and the quality of spawning
habitat was really influencing walleye recruitment

in those 16 lakes.

That data's really lacking at a large scale
because collecting habitat data in the past

has been a time consuming task, so we are
working to develop side scan sonar to do the

substrate mapping.

Ashley:  Thank you.

Our last question comes from Heath.

It says, "During the past two decades WDNR
has been stocking two inch walleye in large

numbers in the lakes with declining walleye
abundance with limited recruitment success.

Any thoughts on why these fish would not survive
if the recruitment bottleneck is at the fry


Dan:  I would say that there could be multiple
bottlenecks, and the only one we're seeing

in those two lakes we're working on is that
they're just not making it to the fry stage.

We've discussed how to address Heath's question
with an additional exercise.

Our hope is we're going to expand the work
that we have been doing, and do sampling on

more lakes.

It would be interesting to see if possibly
you injected fry into a system, do they make

it beyond that stage, or if two inch fingerlings
are going into a system, are they making it

beyond that phase as well?

Gretchen:  That's really the million dollar
question, of why they're not surviving.

That's what we're continuing to work on.

Ashley:  Excellent.

Thank you.

Did Dan, Steve or Gretchen have any closing

Dan:  No.

Thanks for everybody attending.

I was impressed.

Gretchen:  Thank you very much for your attention,
and your time.