2018 South Haven Fishery Workshop — Matthew Kornis

2018 South Haven Fishery Workshop — Matthew Kornis


Alright, hello everybody. It’s a beautiful evening to talk about the
fishery. Thank you all for coming. So, as Dan mentioned, I’m with the Great Lakes
Mass Marking Program, which is a multi-agency effort coordinated by the U.S. Fish and Wildlife
Service. And I am happy to be here today to share some
of the results we have. A very quick outline for you today. For those
who are, perhaps, unfamiliar with the program, I will briefly go over a review, and then
focus on results from a study on chinook salmon, as well as lake trout. Then I’ll close today with some results from a
salmon and trout diet study in Lake Michigan that’s been ongoing since 2014. A mass marking program. Great Lakes states and tribal agencies, as
well as the Fish and Wildlife Service, stock millions of salmon and trout in the Great
Lakes every year. That fishery’s worth about $7 billion annually
to local economies. The mass marking program was founded to help
provide better information and foster enhanced understanding of the fishery and to assist
with management of the fishery. It’s a collaboration amongst the federal,
state, and tribal agencies, again coordinated by Fish and Wildlife Service. It’s established to help address questions
for the management objectives for salmon and trout. The program provides tagging, marking, field
data collection, and analytical support services. None of this would actually happen, though,
without the participation of anglers and stakeholders like you. Our field teams get out there and interact
with anglers every summer, and thanks to your support, we have six years-worth of data with
over 118,000 fish from open-water anglers in this data set. That gives us a very powerful basis for addressing
some of these questions with regards to survival, movement, wild reproduction, and contributions
to fisheries. The tagging and marking operation began for
lake trout in 2010, chinook salmon in 2011, and we started steelhead this past year. It happens at the state and federal hatcheries
using these automatic fish tagging trailers. We tag about 10 million fish each year. The tags provide critical information on the
stocking location, the age and the year class, and the genetic strain of these fish. The fin clip allows us to distinguish between
fish that are of hatchery origin and fish that are wild-produced. In order to do this study, we also get out
there, we collect tags from fish that are harvested by anglers as well as other sources. Basically these are the adults returning to
the fishery. Our six field teams are located where these
red stars are on this map, and they sample over 40 ports a year. Those are these blue dots on the map here,
as well. Handling in a full season over 20,000 fish
a year in Lake Michigan and a little over 1,000 in Lake Huron. And these are just salmon and trout species,
so a fairly large sample size. The field survey costs about $250,000 a year
in salary for the staff, and what we get in return for that is about 450 cumulative sampling
days per year from April through September to provide these data. And then to actually get at these tags, back
in the lab, they’re taken out by hand, each one of these tags is only a fraction of an
inch long. Throughout the program, we’ve been able to
recover a little over 86,000 coded wire tags from 91,000 snouts. That’s about a 5 percent tag-loss rate, which
is what we see in the hatcheries and is what we would expect. Because we’re out there in the field and we’re
handling all these fish, we have several studies we’ve been able to assist with. Ray had mentioned the salmon and trout diet
studies, we’ve been collecting muscle tissues for isotope analysis, which provides an indication
of diet. We’ve been doing that since 2014. We’ve been collecting stomachs since 2015. We’ve also helped with a Notre Dame-led study
on mercury biocontamination, studies looking at the origin of wild steelhead and chinook
salmon, chinook salmon growth rates, lamprey wounding on the different species of salmon
and trout, and effects of lamprey attack on host condition. Getting a lot of bang for our buck here with
the mass marking program. The funding outlook for the program 2018 to
2019, you can see the most recent year at the top here. We’re looking at a projection of about $1.5
million in funding. Again, this is to help provide information
on a $7 billion fishery, so that’s a small investment to make for the amount of data
we have to provide. I want to point out the primary funding source
here is the Great Lakes Restoration Initiative, that’s the GLRI. That is not a permanent funding program. It’s not a base funded program. One thing to inform folks here about is that
there actually has been an act that’s been introduced as a bill into the Senate by Senator
Debbie Stabenow of Michigan, with co-sponsors Senator Peters of Michigan, Senator Sherrod
of Ohio, and Senator Schumer of New York. This bill, if it were to make it through the
Senate, would formally establish — sorry, if it were to make it through and be passed
by the Senate and the House and signed — would formally establish this program under the
Fish and Wildlife Service, specifying the collaboration that we have between the states,
tribes, and federal agencies; help make the data available to all agencies, which is something
that’s already ongoing; and would potentially authorize the program for up to $5 million
annually. That would actually allow the mass marking
program to expand to other species and other lakes. As I mentioned, at present, we’re looking
at lake trout, chinook salmon, and steelhead stock in Lakes Huron and Michigan. That’s a brief overview of our program, and
without further ado, I want to get into some of the results. For chinook salmon, we’ve been able to look
at a lot of different things. I’m going to start with wild recruitment. This is something we evaluate annually. The stocked fish have an adipose fin clip,
so that’s a lack of an adipose fin here, whereas the wild fish have no fin clip and no coded
wire tag. We actually looked at well over 100,000 fish
with our quality control checks in the hatchery, and only one half of one percent of those
stocked fish are released without a fin clip, due to error. We also looked at close to 1,200 fish in the
field, looking at fish that had no fin clip, examining for wire tags, and we didn’t see
very many at all. So there’s very little evidence of fin regeneration. So what we conclude is that 99.5 percent of
those unclipped chinook salmon are, in fact, wild chinook salmon. We look at the percentage of chinook salmon
every year, and this is the year class at age 1, and you’ve got to subtract a year,
so in 2017, we’re looking at the 2016 year class that were age 1 in 2017. The percent wild is between 50 and 60 percent
for much of the time series here. That’s really reemphasizing this point that,
at this point, the majority of the chinook salmon fishery in Lake Michigan is comprised
of wild fish. Using this value and the number of fish stocked,
we can do a ratio to look at and estimate the actual number of chinook salmon smolts
in millions from each year class. In this figure, you have the blue bars are
the wild smolts that we estimate, and the orange bars are the stocked smolts that we
know from hatchery release records. This most recent year class, 2016 year class,
we saw an uptick here: about 4.2 million wild smolts estimated to be entering the lake,
and you have the stocked fish on top of that. You’re at about 6.6 million for the 2016 year
class. One key component to note from this figure
is if you look just at these blue bars, this value of 4.2 million is right in line with
the time-series average, if you look at 2006 through 2009 and 2011, this blue bar is right
in line with those values. But we do see a fair bit of variability in
year class strength. There were two very strong year classes for
2010 and 2012, and two rather weak year classes recently for 2013 and 2015. This emphasizes the need to look at this on
an annual basis. To summarize what we’ve learned from wild
recruitment: the majority of chinook salmon in Lake Michigan and Lake Huron are wild fish. Wild recruitment is variable and needs to
be monitored annually. Another thing we’re able to look at is the
survival of chinook salmon. This is a very complicated figure, but it’s
one simple message: basically, every bar you see here is the estimated survival, corrected
for number of fish stocked, based on return rate for each individual tag lot of chinook
salmon stock in the lake. It’s organized by state. Basically, the taller the bar is, the higher
the survival. You can see that our highest survival rates
tend to be from the Wisconsin shoreline, although there are certainly some good producers in
every state. By and large, there’s a larger volume here. So we look at this across year classes, and
we compile that into a map to summarize this. Basically, these blue areas here have consistently
high survival rates, whereas the yellow areas are more average. These purple districts were a little bit strange. Some year classes were really good, in terms
of their survival, and other year classes were really poor. And then MM6 here and Green Bay, these red
ones, those were consistently poor survivals, in terms of the number of returns we see per
unit of fish stock. We don’t know exactly why this is, but there
are some potential factors that could factor into this. We have temperature differences, typically
colder temperatures on the western shore of the lake. Possible differences in food, although if
you look at the maps that Chuck showed in the previous talk, that tends to be variable. We do know that the western shore does have
more rocky substrate, which could be important for invertebrates, which may or may not be
a good food source for very young chinook salmon as they smolt out. Evidence suggests there’s more wild recruits
coming out of the streams here in Michigan, so maybe there’s some competition that’s reducing
the survival of fish stocked here. Another potential factor is predation, and
I highlight Green Bay as an example here, because of that strong walleye fishery there. Again, just some factors that could be contributing
to these patterns. To summarize here for survival: the fish stocked
on the western shore appear to be surviving the best, with poor survival for fish stocked
in Green Bay and MM6. We’re going to look at movement rates for
chinook salmon, based on their coded wire tags. We look at were they were stocked and compare
that to where they were harvested. One example of this: this is a map showing
the locations of fish that were landed in Frankfort, Michigan. Each one of these dots is proportional to
the number of fish landed here in Frankfort, Michigan, a little bit north of here — or
a lot of a bit north of here, I guess. You can see that these fish are coming from
all over Lake Michigan, some coming from Lake Huron. The dot sizes are proportional to the number
of recoveries. So it’s in line with what we see with that
survival pattern, the largest dots are here on the Wisconsin shore, even though Frankfort
is closest to fish stocked in Traverse Bay and MM6. This emphasizes that movement that we see. However, in the fall, we see what we’d expect
to see with chinook salmon: they move back to where they were stocked. This is an example from the 2011 year class. What you’re seeing here on this axis is the
percent of chinook salmon captured in the same management unit or statistical district
where they were stocked. For age 1, 2, and 3 from April to September
or October. And the bottom line here is that from April
through July, less than 10 percent of the chinook salmon are actually captured anywhere
close to where they were stocked. August is a bit of a transitional month. Once again, in September, October, you’re
looking at 70 or 80 percent of those fish being recovered in the same area where they
were stocked. The recap here is we see evidence of high
chinook salmon movement during the summer, where the capture location is not likely to
be the stocking location. But the fall fishery is determined by that
stocking location. Recently, we were able to take a look at growth
rates of chinook salmon. This is a figure of just stocked fish. You see the total average length versus their
age. The main takeaway here is the growth rates
are really not differing all that much, these lines overlap a fair bit. Not a huge difference by stocking region. But that small difference that we did see
actually lines up with the differences in survival. On the right side of this figure here, this
is actually that same survival map I showed earlier with the different colors, except
now it’s a heat map where your higher survival are your darker red areas. If you look at the estimates of growth, those
same areas have higher estimates of growth where we have higher survival. So some of the same factors we discussed earlier
could be contributing to this pattern. Another pattern we observed is that there
appears to be a very tight coupling between alewife density and chinook salmon growth. This is the chinook salmon mean length at
age 1, and that’s this blue line right here. This orange line is the year and older alewife
density in fish per hectare, as estimated by USGS from Warner, et al. You can see these patterns are identical,
and the fact that they’re so tightly correlated really is indicative of that limited food
supply. If food supply was not limited, you’d expect
growth to be dictated by other things, like the climate in any given season or the abundance
of chinook, and we actually see it’s a really tight coupling with the food supply. So, to recap here, growth was similar among
locations. That’s consistent with what we see for movement
of chinook salmon throughout the lake. The growth and survival patterns were similar. I didn’t have time to show the data, but we
also have evidence that, for pretty much every year class, the stocked fish have — they
don’t grow faster than wild fish, I should have rephrased that. They are larger than wild fish at a given
age, because they have a little bit of a leg up in the hatchery, and it appears that the
wild fish never catch up. The lines are actually parallel to one another. So they don’t grow faster, they just have
a little bit of a leg up and keep that leg up throughout their lives. We also see annual variability in growth that’s
linked to annual abundance of alewife, and this is something we wouldn’t expect if alewife
were not limited. So with that, I’m going to move on now to
lake trout. I’ve got a couple of quick things to show
here. We looked at, similar to the chinook salmon,
we looked at the patterns in lake trout wild recruitment. We measure this by looking at the percentage
of wild fish in each of our management units. And as someone mentioned earlier, we see higher
percentages here in the southern part of the lake than the north part of the lake, really
high percentages in Lake Huron. One thing that’s likely happening, a gentleman
earlier asked a question about why we don’t see any small lake trout. It looks like much of that wild reproduction
is probably happening in some of these offshore reefs in the mid-lake complex or at Julian’s
reef. From a different study, we see some evidence
of stocked fish that appear to be coming into the near-shore fishery when they’re a little
bit older, so it’s possible that we see this wild recruitment, but they don’t actually
show up in the fishery until they’re a little bit older. I want to emphasize here that the population
is not rehabilitated, but certainly this is a positive trend in terms of progress. These numbers throughout Lake Michigan are
up about 3 to 19 percent from last year depending on the management unit. Another thing we’re interested in looking
at, we got a lot of feedback from groups just like this, we’re stocking all these lake trout
offshore, while anglers fish near shore. Why are we putting all these fish where anglers
can’t catch them? We were interested in looking at what’s the
contribution of those lake trout that are stocked offshore to the nearshore fishery? And what we’re looking at here in this figure,
this is the number of fish caught per 100,000 fish stocked. Each of these blue bars is an offshore stocking
location. These red bars are the nearshore stocking
locations. What we’re seeing here, correcting for number
of fish stocked, is that, on a lakewide basis, Julian’s reef and the southern refuge, or
the midlake reef complex, have a really high return rate to the nearshore fishery, and
it’s actually much higher on average than fish that are stocked nearshore. Here in Michigan, the differences are a lot
smaller, but still, the highest contributor in terms of the return rate is the southern
refuge. To our surprise, we are actually learning
that these lake trout that are stocked offshore are the ones that are typically being caught
by anglers nearshore. In fact, if you don’t correct for number of
fish stocked and just look at the percentage of an average angler’s creel, lakewide, about
62 percent of stocked lake trout in anglers’ creels are actually coming from offshore stocking
locations. What may be happening here, if you think about
it, these fish that are offshore, they’re stocked in an area that has a minimal exploitation
rate for the early stages of their lives, and by design they’re being stocked in areas
that are supposed to be great trout habitat. So if they’re surviving at a higher rate than
most nearshore stocked fish, even though some of them will stay offshore their whole lives,
those that do move nearshore appear to be providing a higher number to the fishery than
those that are stocked in the nearshore area. So, there’s a positive trajectory for lake
trout recruitment, and lake trout stocked offshore do contribute the most to nearshore
recreational catch. I’m going to conclude today with some information
on salmon and trout diets. Most of what you’ll see is from a collaboration
with Purdue, University of Illinois, SUNY-Rockport, and University of Wisconsin-Milwaukee, looking
at diets from 2015 and 2016 in Lake Michigan. As Chuck Madenjian mentioned earlier, we see
these changes in the forage base, alewife have been trending downward, round gobies
have been trending upward, and a number of other changes. As you might imagine, understanding how the
diet and the potential for competition among the salmon and trout species is a pretty critical
question to be asking in the lake at this point. To jump right into here, this is a figure
showing diet data from stomach contents. These are lakewide summaries, and this is
the percent of diet by weight. In all these figures today, you’re going to
see the same color-code, so we have the percentage here on the y-axis and the five different
species on the x-axis. Right away you see these blue bars are the
largest bar in pretty much every species. Those are the alewife. Everybody’s eating alewife. But we do see some differences among the species. The chinook salmon and also the coho salmon
are the ones that have the highest reliance on alewife in terms of their percent diet. Species like brown trout and lake trout have
a more diverse diet, in particular because of this orange bar — those are gobies we’re
seeing in their diets. And then the steelhead, this big green bar
are the terrestrial invertebrates everybody sees kind of exploding out of their stomachs,
especially in the springtime. In 2016, we see the same pattern with some
other items popping up. This shows that there actually is some variance
or variability from year to year, which is part of the importance of continuing this
study with the [unclear] et al. work that Dan O’Keefe mentioned at the beginning. In terms of similarities, those are these
blue bars, those are alewife, those are sill the majority for each species. We still see a lot of gobies for brown trout
and lake trout. Still see invertebrates here for steelhead. We see more yellow perch, those are the yellow
bars, and more bloater from 2016, the black bars. And then these pink bars are actually spiny
water flea. The percent of the diet may be pretty high
for spiny water flea, but the animals aren’t actually eating a ton of water flea. The way these percentages are calculated,
it looks at the average in a unit time period, and if over that time period the animals didn’t
consume any prey fish, their stomachs were pretty much empty and you get a high percentage
of these water fleas. So that was kind of our first lesson from
the diet, is that most energy are from fish prey even if invertebrates are abundant. We can do this by comparing the percent of
diet with what’s called the “ration.” That is basically just the total mass consumed
from each diet. Here’s our 2016 percent plot again. You see our invertebrates bythotrephes, — I’m
sorry, the spiny water flea — and the terrestrial invertebrates here in green. If we look at this as a ration plot and this
is the mass consumed, those bars pretty much disappear. We still see some invertebrates, but the spiny
water fleas are almost nonexistent. This gives us a more clear picture that, yes,
it’s fish that are really important. 2016, a lot of gobies for brown trout, some
gobies for coho, a moderate amount of gobies for lake trout. Mostly alewife here for chinook salmon. The second thing we learned is that there’s
a lot of seasonal variability in diet. So there are more gobies in the earlier part
of the year, and terrestrial invertebrates in the early part of the year; more bloaters
late, and alewives all the time. To just highlight this, I’m going to show
you one example: what you’re seeing here, this is the proportion of diet by weight comprised
by round goby. The green bars are lake trout, and the grey
bars are burbot. What we see here is in the springtime, there’s
a high amount of goby in the diet for these animals, and the same again in fall, but not
so much in the summer. What I think is happening here is this is
relating to the annual migration that round gobies do. In the fall, round gobies migrate out to deeper
water; it’s a little bit warmer there in the wintertime. In the spring, they migrate back, and in the
summer, they’re in shallow, rocky water, typically under rocks so they can spawn. They’re not very accessible to predators here
in the summertime. But when they’re moving through these deeper
areas, they are more accessible, and we see that uptick in the diets of the lake trout
and burbot at that time. We also see a lot of spatial variability. What you see here in South Haven is not necessarily
what an angler in Sturgeon Bay is going to see in the diet of their fish. Two quick examples: one is chinook salmon,
and I wanted to highlight chinook salmon because, by and large, this is the most consistent
diet you can get. They feed almost exclusively on alewife. But, in 2016, in northeastern Lake Michigan,
so that’s the area around Frankfort, almost half their diet was comprised of bloater chubs. It’s the only time we’ve seen that high of
a composition of anything that wasn’t alewife or chinook. It kinda highlights that spatial variability. We see that spatial variability in lake trout
as well. This is from a publication that came out in
2017 for lake trout diets. In Huron, these green bars are rainbow smelt. If you look in Lake Michigan, the red bars
in these pies are alewife, a lot of alewife on the western shore. These blue bars are gobies, a lot more here
on the southern part of Lake Michigan. A lot of spatial variability in diets as well. Finally, there’s a potential for depth variability. Here is a figure provided by Jory Jonas from
Michigan DNR from two areas of Michigan. Basically, yellow is alewife, red is goby,
and you’re looking at a time series plot from ’96 to 2016. What I’m trying to get at here, these are
from bottoms of gill nets, so the percentage of the diet of lake trout here for gobies
is 70, 80 percent, which is way higher than we see in our angler diets. There’s a seasonal component here too. These are April through June, and our diets
typically are May through July, even in the early period. But there’s likely something going on here
where, if you’re catching a fish on the bottom, it’s probably eating bottom-oriented prey
like gobies. If you’re catching them while you’re out trawling,
it’s probably oriented to open-water prey like alewife. The bottom line is that diet depends on where,
when, and how a fish is caught. How can we determine how much of each prey
type is eaten over the long term? There’s a technique that we’ve used, stable
isotope analysis, that provides a longer term indication of diet. These isotopes reflect diet over the past
4 to 6 months, whereas the stomach contents the last couple of days. To orient you here to what we see in an isotope
plot, these are our salmon community in the colored dots here. This x-axis is providing offshore to nearshore,
and this y-axis is showing both trophic level and depth. To highlight some examples, we have round
gobies in the nearshore versus sculpins and alewife offshore. In terms of our differences in the food chain,
the higher you go here, the higher you are in the food chain. These are our zooplankton and our mussels. Here are our planktivores, the alewife and
smelt. Here are the things that are eating those
alewife and smelt. You can see that clear increase in the food
chain. We also see this depth effect. The sculpins actually were higher in this
nitrogen component than even a lot of the salmon. It’s not that sculpin are eating salmon, it’s
actually this microbial process in the lake that results in this enrichment out deep. How are Lake Michigan salmon and trout adjusting
to this forage base? One of the things we see is that the difference
between the lake trout, on average, and the rest of the communities is about the same
difference between that community and large alewife, which is a pretty substantial difference. It supports this idea that lake trout have
a more diverse diet. To highlight this, there’s a technique we
use to look at the niche overlap — basically a way of looking at competition between the
animals. If we look at this value here, this is the
overlap between lake trout and chinook salmon. We can look at this through ellipses. Basically this plot, which looks a little
complicated at first, all of our salmon and trout species are in these colored ovals. In this case, that difference between levels
of the food chain has been corrected for. If you see an oval overlapping with a prey
item, it means that’s a component of the diet. This 41 percent overlap is this yellow are
here. This green circle here are the lake trout,
and this orangey circle here is the chinook salmon. Of course, that overlap is occurring right
where alewife occur. The rest of this lake trout orb is encircling
bloater, rainbow trout, and goby, and that’s where that diversification shows, and showing
that difference here in the diet as indicated by the isotopes. What do we take from all this? Any given stomach does not indicate a species’
.

Leave a Reply

Your email address will not be published. Required fields are marked *