Tip:
Highlight text to annotate it
X
[Introduction: Margaret Mantor] Hello everyone, thank you so much for coming today. We're
going to go ahead and get started. I'm Margaret Mantor; I work for the CESA program in the
Habitat Conservation Planning Branch. Before we get stared I just wanted to remind everyone
that if you are with CDFW we are giving training credit through OTD so just make sure that
you sign in at the door. If you're listening on WebEx you can also get training credit
so I've sent out sign in sheets. If you haven't gotten one, just contact me, again I'm Margaret
Mantor. Also, to let people online know, you can type in questions for the speaker at any
time and we will make sure that he gets your questions. This is part of a lecture series.
Our next one is going to be on December 17 and it's going to be on the northern spotted
owl. Today we are very happy to have Dr. Brian Halstead presenting his lecture on giant gartersnakes.
Dr. Halstead is a research wildlife biologist in the Western Ecological Research Center
for the USGS in Dixon. He received his PhD from the University of South Florida in 2008
and his research focused on predator prey interactions of snakes and lizards. His current
research is focusing on the ecology and conservation of reptiles and amphibians and today he's
going to be talking about his work on giant gartersnakes. Thank you for being here.
[Dr. Brian Halstead] Thanks Margaret and thank you everybody for coming here. It's not just
my research it's also the research of my co-authors. Glenn Wylie is here today; he got our giant
gartersnake research program started. Shannon Skalos is here as well, she really facilitated
this presentation. And Mike Casazza who has been here since the beginning as well but
he is out east for work. Just to give you a brief outline of how I'm
going to organize my talk, I'm going to talk a little about the history of giant gartersnakes,
just kind of their nomenclature, their historic distribution. Then I'll talk a little bit
about our research methods, including identification, and then I'll get into biology and management.
It's going to be kind of a mix of different topics. Some of our research will be mixed
in with general knowledge and kind of feed together a little bit. So if it seems a little
bit discombobulated, I hope it's not too bad. The giant gartersnake was first described
by Henry Fitch in the 1940s. At that time it was Thamnophis ordinoides gigas. That later
got switched to Thamnophis couchii gigas. In 1987 Rossman and Stewart elevated the giant
gartersnake to a full species. The giant gartersnake has been listed in the state as threatened
since 1971 and by the US Fish and Wildlife Service it was listed in 1993.
The reason it was listed is because of the loss of its habitat. This is the historic
range of the giant gartersnake. Tule marsh habitat in the central valley, it occurs nowhere
else on earth. But with the damming of streams and the conversion to agriculture the giant
gartersnake has lost a lot of its habitat. Most of this habitat in most of its range
down here is no unsuitable for giant gartersnakes. There are a few remnant populations but we've
lost almost 2/3 of the range of the snake. Our research program was established around
1995 and kind of the cornerstone of our research program is detection/non-detection methods.
I say that instead of presence absence because these guys are hard to see, so detection is
a more apt term. And capture-mark-recapture. And the important feature here is that these
snakes are hard to find and we would need to account for that when we are doing our
surveys. I don't know if you can see it in this frame or in this picture but there's
a giant gartersnake there and it occupies most of that photo. So you can see they're
really hard to detect visually. Now in certain situations some people develop
a really good search image for them and they're able to find them and catch them pretty well,
but it involves both finding them and catching them so although some people have great success
with visual surveys and captures, we tend to focus more on trapping techniques.
The two traps shown here, the one on top is just a vinyl coated; the one on the bottom
is galvanized hardware cloth. These are commercially available traps just modified so they float
at the waters' surface so that if a snake gets in there it doesn't drown if the water
level happened to rise. Now because the snakes tend to swim at the surface, we tend to catch
more snakes by floating them at the surface than we would if they were sunk or if they
were out of the water. We aren't perfect at catching snakes, in fact
historically we were trying to get our hands on snakes for radio telemetry, and we did
a good job at that. But when we try to estimate demographic parameters, we need a little more
precision. And so we looked at our historic data to see how different conditions affected
our ability to detect giant gartersnakes. And in this figure you can see this red line
is just a probability of 0.95. And the dark black line is the mean cumulative probability
of detection after sampling for a certain number of consecutive days. So you can see
that as we increase population size (we have a very dense population of giant gartersnakes),
not surprisingly we can detect them easier. There's just more snakes moving around out
there getting caught in the traps. When we increase the number of traps we also increase
our detection probability. Those two things are really not surprising at all. Warmer water
tends to lead to higher capture probability and that's probably because the snakes tend
to spend more time in the water; if the water's cooler they probably have to come out of the
water to bask more frequently so the warmer water leads to more time in the water and
more encounters with traps. And earlier in the year, in the active season anyway, we
also find that snakes are easier to detect. But you can see the dotted line in these figures
represents the lower fifth percentile of detection probability. And you can see that in some
cases, in many cases actually, we can sample for up to two months straight and not be able
to achieve a cumulative probability of detection much above 0.5.
So we set out to improve our traps. This is just a video clip showing how snakes interact
with our traps. You can see the snake at the bottom there. Right here. Hopefully everybody
can see that on the WebEx as well. We videotaped some of our traps to see how the snakes interacted
to give us ideas for how to improve our trapping. Obviously you can see that snake bypassed
that trap opening. That's a pretty common occurrence in our videography. We also would
videotape snakes getting into traps or being in traps and then go check the trap and there
was no snake there so you know that they escape pretty well too. So we tried a series of different
modifications. Including, we looked at both trap type; we
looked at including these little fingers in the traps as kind of a one way valve that
the snake can push through. They're made of just cable ties. The idea is that the snake
can push through that pretty easily but then when it tries to get back out they kind of
fold over each other. It also provides some visual obstruction to the trap entrance. And
we also added these wings or funnel extensions to the wide end of the trap. The idea behind
those is that first, we could articulate the trap more closely with the emergent vegetation
or with the bank but it also, even if we aren't able to do that, it still increases the width
of the funnel so that the snakes are more likely to encounter that. So we tested all
different combinations of these traps and of the trap materials and what we found was
that when we combined the galvanized traps with both of these modifications we had the
highest capture rates. So the bars are just different numbers of captures in different
habitats at two different sites. The black dots and the error bars, that's our best fit
model predictions of how many snakes we would catch. So the model was significant in that
all the trap modifications improved capture rate. So these are the traps that we currently
use. So finally we're now getting our hands on
snakes (at least in the talk anyway). So what we do when we catch the snakes is we mark
them. That's probably the most important thing that we do for our research is marking them.
We do that by two methods. The first method is with a Passage Integrated Transponder or
PIT tag, and the second method is by using a disposable medical cauter unit to give each
snake a unique central code we branded into them. These things get nice and hot and I've
hit myself with it several times. Most snakes don't even flinch when you do it so it's actually
a really nice alternative to the old way of marking them which was scale clipping which
was a lot more intrusive. This seems to be a nice clean and easy technique. So what you
end up with after you mark the snake is a permanent mark something like this. Now these
snakes, they live in an aquatic environment and a lot of the time they have aberrations
on their ventral scoots. So we actually brans up the side of the snake a little bit a couple
of scale rows. So if you look closely, and I don't know if you guys will be able to see
this in this photo -- you can see that the scale rows here are a little bit shrunk from
the brand whereas those over here are not shrunk. This is actually just a natural scar,
some natural scarring that the snake has. So that extra part of the mark is really important
for distinguishing some of that natural damage from our mark.
We also measure the snakes. We measure the snout-vent length, we measure the mass, and
we do scale counts and measure some of the scales on the head of the snake as well, as
you can see there. These things are important for a couple of reasons.
First, if we look at giant gartersnakes, there's a lot of variation in coloration, both between
populations and within populations. So we've got snakes with yellow stripes, snakes with
orange stripes, snakes with no stripes, snakes that are checkered -- all giant gartersnakes.
So if we try to identify them based on pattern alone, that can be problematic.
They also occur with two other species of gartersnakes. On the left is the common or
valley gartersnake. The easiest and quickest way to tell them usually, if you have a clean
snake, is that there is red between the dorsal and lateral stripes on the valley gartersnake.
Giant gartersnakes don't have that. You can also look at scale counts, labial scales on
the common gartersnake, they have seven on each side, the giant gartersnake has eight.
The one that's a little more problematic, especially in areas near the foothills or
areas near the Sutter Buttes is terrestrial or mountain gartersnake, and that one tends
to have more distinct stripes. It tends to have a brighter yellow against that darker
brown color. But probably the best way to tell is to take a close look at the head.
Just by glancing at them you can tell that one of these snakes has more of a blunt nose
and the other one has more of a pointy nose. The terrestrial gartersnake has more of a
blunt nose and that you can see, again I apologize if you can't see it that closely, but these
inter-nasal scales on terrestrial gartersnakes are wider than they are long and they aren't
pointed, whereas on the giant gartersnake, they'll tend to be more triangular, almost
pointed on the end and they'll be about as long or a little bit longer than they are
wide. So that's why we do all those scale measurements -- so that we know for sure when
we're looking at the data whether we're looking at a giant gartersnake or a terrestrial gartersnake.
But for our technicians this is kind of a pain because they can grab a snake and look
at it, and you can tell when it's a giant gartersnake in the valley. It's pretty easy
once you get the feel for it. So how giant are giant gartersnakes? Well
as far as gartersnakes go, they get pretty big. They can get over 1.2 meters total length.
I think if you read the Stebbins Guide, I think it says 1.62 meters is the maximum length
ever recorded. That's total length -- that's snout-vent plus tail. That's a pretty sizable
gartersnake. Most of the snakes we catch are less than a meter snout-vent length, but they
get pretty big. They can also weigh more than a kilogram.
Some of these larger females, especially when they're gravid, they can be pretty hefty snakes.
You almost notice their mass more than their length with some of these snakes.
So what do they eat to get that big? Well, they prey upon fish, tadpoles, and frogs,
which isn't surprising because they are aquatic snakes.
Their growth is pretty typical of reptiles. It tends to slow with size. And males and
females exhibit different patterns of growth. You can see that females, their growth rate
doesn't vary much within the active season. But males, when we looked at during the breeding
season, in early spring, and the rest of the active season, they exhibited a depressed
growth rate in the spring. And that's because they're using their energy to look for mates
and they aren't foraging at all. So they aren't really putting any energy into growth. That,
surprisingly, that probably doesn't lead to the *** size dimorphism, that's probably
and evolutionary thing. Because the larger females can then produce more offspring. So
you can see here, the males are in blue, this is the mean snout-vent length and our uncertainty
associated with measuring that mean for males in blue, and females in red. And you can see
the females are both longer and heavier than the males. They also exhibit different patterns
of body condition through the active season which is kind of an interesting finding. In
the spring the males tend to start a little lower in body condition than the females and
they continue to drop. So this dotted line here is males, just a couple different models
that we fit to the data. The solid lines are females, and then we have a body condition
index on the Y-axis. So in the spring the females are starting to gain body condition
slowly. The males are still dropping and then they accelerate. By fall both sexes are going
into hibernation at about the same condition. And this is just related to the reproductive
activity. The average litter size of giant gartersnakes
is about 17 young. 13 to 21 is about average. They can be larger or smaller. The litters
are usually born in mid-July through mid-September. The young are about 21 cm in snout-vent length
and they weigh about 5 grams. Then length of the young actually varies quite
a bit from litter to litter. This plot on the left shows box plots of neonate size for
each mother and then the mother's snout-vent length is indicated on the bottom. So we just
ordered the mothers from smallest to largest and plotted their baby sizes and you can see
that there's not really a pattern there with the mother's length. As the mothers get larger,
their babies don't get larger. In fact, if we look at a trade-off here on the right,
we have the relative mass of the litter per offspring -- so we basically take the whole
mass of that female's litter, divide it by the number of offspring, and then plot it
against the length of the mother, and we can see that as females get longer, they tend
to invest less in each offspring. So what they're doing is larger females produce more
offspring rather than larger offspring. So that really puts a premium on females getting
larger and producing large clutches so that they have some offspring survive to contribute
to the next generation. In addition to our mark-recapture, detection-non-detection
surveys, we also do some radio telemetry surveys. Right now we aren't doing any but in the past
that was really the cornerstone of our giant gartersnake research.
And we've used these data for a number of different things. One thing we've looked at
is survival of adult females. Why is it just survival of adult females? Well, because females
are the larger sex, we're able to radio track them easier; males tend to be too small to
put a radio transmitter in them without worrying about the transmitter affecting their behavior
or survival. These analyzes are limited mostly to females. Our annual probability of survival
on average throughout the Sacramento Valley is about 61% but there is quite a bit of variation
among sites. So each of these dotted lines here is a survival curve for a different site;
a different population throughout the Sacramento Valley. There's actually even more variation
among years than there is among sites. Again each of the dotted lines represents a survival
curve for a different year of study. We also examined what different variables might affect
survival rates of the female giant gartersnake. And we didn't find much of a relationship
with snout-vent length or with body condition, but when we looked at habitat we found that
the snakes were at lower risk of mortality when they were in terrestrial habitat than
when they were in aquatic habitat, which seems counter intuitive -- these are aquatic snakes,
they're eating aquatic prey -- what's going on here? The reason we think that this occurs
is because when the snakes are in a terrestrial environment, they're almost always underground
in a burrow or under some form of cover. So they're just basically seeking refuge in terrestrial
environments that are near the water. And then they go out into the water to forage
and conduct their daily activities (or whatever). So they're not at much risk when they're in
the terrestrial environment, but when they're out foraging, that's when predators will encounter
them and they're just a little bit more exposed when they're swimming around. We found a really
interesting pattern with linear habitat. And that's that on average linear habitat didn't
affect the daily risk of mortality. Basically how we did this was we looked at for each
location was a snake in a canal or other linear waterway, but essentially just canals where
we do our studies, or was it in a rice field or a wetland. What we found was that at some
sites, when they were in linear habitats they were at a greater risk of mortality, and at
others, they were at less risk of mortality. And this really has to do with the context
of the site. For example, this high peak below one...the dotted line here represents that
there is no difference with that variable and the daily risk of mortality...so this
peak over here means that they're at a lower risk of mortality in linear habitat than in
aerial habitat. And that one happens to be the Natomas Basin before any habitat restoration
occurred, so this is the Natomas Basin in the late 1990s. What happened there was for
much of the gartersnake's active season the habitat was restricted to canals. So essentially,
if they were in the canals they had a better chance of surviving than if they were off
in a cultivated field or a rice field at the wrong time of year, or some other habitat
because that just wasn't very good habitat. In contrast, this curve over here is Gilsizer
Slough just a few years ago. There's been a lot of habitat restoration there, there's
a lot of freshwater marsh. The snakes there, if they're in the marshes, they're actually
at a lower risk of mortality than if they're in the canals. Here, what we think happens
is that when the snakes are in the canals, they're restricted to a relatively narrow
band of habitat that's pretty easily searched by predators. A lot of the mortalities there
are from things like otters, we've also had a couple that were taken by raptors. When
they're in that linear habitat, the snakes don't have...the predators have a much better
chance of finding the snakes. There's just not as much room for them to hide as there
is in a marsh where there's a lot of emergent vegetation. So that's why the context really
matters. If linear habitats are the only thing going, it makes it look like "oh, they do
great there," if you have other habitats available, the linear habitat starts to look like not
quite so good anymore. What are their predators? Raptors, as you
can see here, wading birds will take them, especially the younger ones. Otters have taken
a few of our radio snakes. Bullfrogs -- Mike and Glenn did a study on bullfrog predation
at Colousa National Wildlife Refuge and found that bullfrogs can take up to around 20% of
annual production of giant gartersnake neonate. It's a really interesting dynamic here because
the bullfrogs provide prey for large snakes but they're predators for small snakes. It
would be interesting to see what the net effect of bullfrogs on giant gartersnakes really
is. And large predatory fish probably do play a role as predators, particularly if the habitat
doesn't allow for escape - so if there's not a lot of emergent vegetation and other cover
from the fish. Other sources of mortality include parasites
like these nematodes that in-cyst in the snakes. Here's one right here. If they have a lot
of these and they erupt, they can lead to other infections of just massive trauma to
the snake. Diseases -- we really don't know a lot about diseases. But in some parts of
the US there's an emerging pathogen called snake fungal disease. We haven't found it
to my knowledge in California but it is something to be aware of -- that disease can play a
role in the ecology of these species. Introduced prey can be problematic. This is an introduced
bullhead. They have very sharp stiff spines on their dorsal and pectoral fins and they
can erect those and tear open a snake. And we've actually see snakes with small bullheads
stuck in their mouths and a snake doesn't have a very good way to get those out of their
mouths once those spines are embedded. So introduced prey can be really problematic.
And of course humans both intentionally with a shovel or a car or unintentionally with
a car, or just by our decisions as far as management practices -- we can be a pretty
substantial source of mortality in some situations. Fortunately I think there is a lot we can
do with habitat to ameliorate some of these threats. We mapped habitat suitability for
giant gartersnakes using a couple different factor analysis methods. We found that giant
gartersnakes were more likely to be found in the Sacramento Valley in areas that were
near rice, near water, and had a high density of canals. So this isn't really surprising
since these are their aquatic habitats. We did find too a relationship with wetlands
-- they're found near wetlands. But our sampling also occurred closer to wetlands so we can't
really tease out whether that's our sampling frame or if snakes are actually found closer
to those areas in this particular study. As you can see, those areas occur in the major
basins -- in the Butte Basin, we had a lot of rice right up here just north east of Yuba
City and Marysville, Sutter Basin, Sutter County. But it looks like there is quite a
bit of suitable habitat for giant gartersnakes remaining in the Sacramento Valley.
So we did a follow up study that actually accounts for our inability to detect giant
gartersnakes at some locations even when they're there. So we did an occupancy analysis. With
this analysis we found that the probability of occurrence with giant gartersnakes, the
best predictor of that was the distance to historic tule marsh. So if you're within the
historic tule marsh areas, there's about almost an 80% chance of finding giant gartersnakes.
There's quite a bit of uncertainty around that estimate, but as you extend out to about
25 to 30 km away from that historic tule marsh habitat, the probability that you're going
to find giant gartersnakes there, regardless of what habitat is currently there, drops
down to about 20%. And this is mapped out on the right here. So you can see, if I go
back, this area to the north east of Yuba City and Marysville, if we look at this map,
we sampled there -- we didn't find snakes at either one of these locations and the habitat,
quite honestly, looked great. So what we think this represents is dispersal limitations of
giant gartersnakes. They probably historically occurred out in that tule marsh and out in
the California prairie where you had pockets of tule marshes, but areas that were separated
by wide riparian corridors, like the riparian area around the Feather River for example
cuts off that historic tule marsh habitat from this area up here and we suspect that's
why we don't find, or we haven't found giant gartersnakes there. What's really going on
is probably a combination of both suitability and the probability of occurrence. With more
data we could fit more complex models that account for the present day variables. And
we did that; we tried to fit models that included present day variables of natural habitat,
the microhabitat at the sites we actually sampled where we look at proportion of different
types of habitat and vegetation type, we looked at prey abundance...but of all the variables
we looked at, the probability of occurrence was most strongly affected by how far it was
from historic tule marsh. Radio telemetry has also led to some habitat
selection studies, these are currently in preparation. At the macrohabitat level, we
find that habitat selection is context dependent. If you have permanent marsh habitat available,
giant gartersnakes will select that. That seems to be their preferred habitat. If you
don't have permanent marsh, they will select rice, and in particular they will use the
canals associated with rice because the rice really doesn't provide appropriate habitat
through the full active season. There's a period of the active season where the rice
fields are not yet emergent. They're either bare ground or they're just open water and
that doesn't provide any cover for the snakes. When it becomes emergent in mid-June (about),
some snakes will move into it (not all) and use it. And then when they draw the water
down, snakes that are in there probably really go to town on stranded prey, but then there's
no more water so the snakes come back to the canals. The canals are probably one of the
most important features of rice agriculture for the snakes. And that's a study that we're
initiating to see how important are the rice fields to these snakes, independent of the
canals. If we keep the canals and don't have the rice, will the snakes be OK? That's something
that we're looking into right now. I mentioned that those linear waterways were important,
some open water is also important. The snakes like to be located along the edge of open
water. So if you have like, a bank against open water, that's OK for the snakes. If you
have emergent vegetation next to open water, that's great for the snakes. They really like
to be at the edge of that open water. We looked at microhabitat selection. This
was going out in the field and putting a hula-hoop around the snakes' location and characterizing
the habitat and vegetation within that hula-hoop and then picking a random location within
50 meters in a random direction and doing the same thin, and comparing those. We find
that emergent vegetation is strongly selected. Just for reference, this horizontal line at
1 means no selection for or against that habitat. Terrestrial vegetation, again, is strongly
selected. Again, the terrestrial component of the habitat you kind of have to take with
a little bit of a grain of salt because in a lot of these locations the snakes are underground
and we aren't able to characterize what those burrow characteristics are; we're looking
at the surface. So we know the snake is under the ground and this is what's on the surface.
We don't know exactly how much that reflects an active selection by the snake. Bare ground
and open water are used roughly in proportion to their availability. I should back up just
a second here, the big dots with the error bar, that represents the population average
for selection for those habitat types. The small gray dots represent selection by individual
snakes. So you can see that in response to litter, the snakes are much more variable
in their response to that habitat type than they are to say, water. If you look at water,
we have a really tight relationship right at or centered around 1. There's no selection
for or against it. This is partly because they like the edges. So if you go to the snake
location, you get 30 to 70 percent water a lot of the time and if you go to a random
location you're going to either have more or less or similar to that because the snakes
are sort of on that edge most of the time. That's why I think we don't have much variation
among individuals in that variable. Interestingly, rice came out as somewhat avoided by the snakes.
Some snakes positively selected it and others did not. Part of this is because the rice
really isn't available to them for the whole active season. And some snakes, even when
the rice they stayed in the canals. This study was done at Gilsizer Sloughs. That doesn't
mean that rice doesn't provide benefits to the snake. It probably does. It probably helps
to improve the prey base in the canals, increase the productivity. It just means when we're
thinking about rice as habitat when comparing what the snakes use to what's available. The
rice is hugely available to a lot of these snakes. So in order for them to use it in
proportion to its availability they'd have to be found there almost all the time. And
that just doesn't happen because of the phenology of the rice.
When we look at vegetation selection, again this figure is pretty much the same as the
last one just with vegetation types instead of the habitat types, we can see that tules
are the most strongly selected vegetation type. Cattails are also positively selected
but not nearly as strongly. Grasses and forbs are also positively selected, again with the
caveat that a lot of these snakes are underground. One of the things that's interesting is that
primrose was somewhat positively selected, but the snakes were really variable in their
response. And that probably has to do somewhat with what the environment around them is and
how much primrose there is. If it's really dense primrose and there's not a lot of water-primrose
edge, the snakes are probably going to be found avoiding those situations. But if there's
primrose along the edge and that's basically the cover that they have, that's what they'll
be using. So the primrose is, in some situations, not really all that bad for the snakes. Mosquito
fern and floating vegetation does tend to be avoided somewhat. We see that in even stronger
response when we look at occupancy. At some locations we actually see that there is a
lower probability of occurrence of the snakes in areas of floating vegetation than in other
areas. So this is what their active season habitat
looks like. And if you think about the snake's biology, and history for that matter, you
see why they might prefer tules over cattails, remembering that the cattails are still OK
habitat --they'll still use it. When you look at the tules, there's a lot of horizontal
structure here. The snakes are able to bask in those areas and readily drop through the
tules and escape aerial predators. The tules tend to come up -- the stems come up singly
as opposed to the cattails which come up in this big fan-shaped clump. If your body is
long and thin, you can kind of snake through there and get through there and other things
can't quite as easily. Whereas with the cattails you've got less horizontal structure, you've
got these big fan bases that other animals can get through them as easily as the snakes.
Just evolutionarily, the snakes evolved in these tule marshes so they're well adapted
to those types of habitats. Like I said, we find snakes in canals a lot.
Not all canals are created equal. If there's a lot of emergent vegetation and terrestrial
vegetation along the edges, like this canal is a pretty nice looking canal for snakes.
A few more tules would probably be better but at least there's cover right up to the
water's edge. This canal, you'd probably find snakes if it's in the right location here
as well. It would be a little less pleasant for you with all the starthistle but you can
find snakes in these canals as well. So canals make up an important habitat where there's
nothing else for the snakes. Rice fields, again, you can find snakes in
the rice at certain times of year. When the rice is emergent and dense like this you can
find snakes along the checks and you can find them at the edges here along the rice, but
they don't seem to seek it out, partially because they have to move into it from an
area where they can be in appropriate habitat for all of the active season.
In the winter snakes are usually found along banks. They'll be found along canal banks
or near the marshes that they occurred in during the summer. We can find them also,
even in situations - I don't know if you can see it here -- it's flooded up here; there's
a snake hibernating underneath this clump of grasses. They can tolerate that low sheet
flow. They might be less tolerant of high flow systems like in the bypasses. If there's
a lot of flooding they'll move out farther from the water. This is at Badger Creek and
there was a snake hibernating I think right down in hear near the fence posts. And that
was, if I remember correctly from what Glenn and Mike told me, about 200 meters from the
marsh that you can see in the background there. That's about the farthest that we've seen
snakes go to hibernate. They'll be found in roadsides because in a
lot of cases that's the high ground along the canals. Where you have these little basically
two track trails for equipment along rice fields and along canals, a lot of times snakes
will be found on the edges of those. And again, here's another situation where there are snakes
just along the edge of this road; they came out in the spring just fine, so low level
flooding doesn't seem to be too bad. And also in riprap. Along the railroad bed
or along water control structures where they place riprap, snakes will hibernate in those
areas as well. So we have an idea of what's good and bad
habitat; how does that affect how many snakes we find in an area and what are some of their
population characteristics? In most populations that we've studied, the sex ratio has been
really close to 1, the 0.93 was kind of a Sacramento Valley overall sex ratio for several
different studies done in the late 1990s and early 2000s. Abundance and density do vary
with context. Their lowest, in this case at Colousa National Wildlife Refuge, which at
the time was managed mostly for water fowl - they had a moist soil management program,
so they were growing seed crops during the summer and flooding it during the winter.
So during the summer the snakes really didn't have much habitat because there was no water
and emergent vegetation. Management has switched there and the snakes are doing really well
there, but at the time they weren't doing so well. At the other end, we have a place
like Badger Creek, which again this study was done in the late 1990s, which was a natural
marsh and it had a very high density of snakes regardless of how we measured it. In rice
agriculture, the densities were intermediate. So here we had Gilsizer Slough which had a
remnant of a natural water feature but water was pumped out and drained into it quite a
bit so it wasn't quite as natural as badger creek that was surrounded by rice and other
crops. And then we also had the Natomas Basin which was almost all rice. And what's interesting
is that when we look at body condition, it follows the same pattern of that snakes in
the marshes had highest body condition. So what this means is that where the snakes are
abundant, they're also very robust. They're not prey limited even where they achieve these
higher densities, which is interesting because it means that even at those high densities,
it doesn't seem like their condition is suffering at all.
What this means for management is that water management is extremely important. If you
don't have water during the active season of giant gartersnake, you're not going to
have snakes there. So even though we have what could be a really nice marsh in this
picture, when it's dried down like that, unless there's a pocket of water inside the marsh
that we don't see here, the snakes are going to either stay there and maybe hang on for
a little while, die, or emigrate to some place where there is water. And we have seen some
pretty spectacular migration of snakes at Colousa National Wildlife Refuge when we were
doing that abundance study. We also had some snakes radioed. In that situation where they
were overwintering at Colousa and then they would do the moist soil management and they
would migrate to the neighboring rice field where there was water in that situation. So
these snakes made fairly long movements and had large home ranges. We don't know what
kind of cost that might have with regard to exposure to the environment or predators.
It also has implications for invasive plant control so if you have a situation where you
have primrose that looks like this, you know, you've got it kind of thick on the edge but
you've got some open water still, snakes are probably going to do OK there. If it went
all the way across, then all bets are off in those situations. They'll tend to avoid
it. But what we've seen with patterns of occupancy is with things like these thick mats of mosquito
fern, the snakes tend to avoid those areas. So it's actually probably more of a management
concern where you have a lot of this floating vegetation than in areas where you have some
primrose along the edges. The transmitters that we've implanted in snakes
are temperature sensitive so we've done some work on their thermal ecology and we've found,
not surprisingly, that snakes don't use the thermal environment at random. Instead they
tend to concentrate their body temperatures right around 30 degrees Celsius, just a little
bit above that, which is warmer than what the environment typically is (the environment
being the mean of water and air temperatures in this case). The spoils surfaces can actually
get much hotter. If we included those values it would pull that environmental mean over.
Interestingly, males and females use the thermal environment differently. Males tend to elevate
their body temperature in the late winter and early spring, as you can see here, and
that's probably associated with mating activity. They're out seeking mates; they're going to
bask and try to find those females before those females get up and are moving around
too much. In contrast, the females tend to elevate their body temperature relative to
males and the environment in the summer when they're gravid. So they're trying to thermoregulate
to promote embryonic development at those times.
The thermal ecology of the snake is an important component of their habitat that we don't always
think of as habitat. Things like timing of mowing can be really important when it comes
to the thermal ecology of the snake. If you have a cold overcast day during the inactive
season in, you know, February or January, mowing at that time is probably find because
the snakes are going to be underground in burrows. If it's a warm sunny day at that
time of year, the snakes do come up and bask at the mouths of the burrows that they're
staying in or on riprap, or wherever they happen to be hibernating. Likewise, if it's
a hot afternoon during the active season, the ground temperatures are too warm for the
snakes to be active. So if they have a burrow, they're going to be down in the burrow. Otherwise
they're going to be in the water or very near it. So mowing during hot afternoons during
the active season, as long as you're not starting any fires with sparks, is fine. But what you
really want to avoid is mowing on warm sunny mornings, especially during the spring time.
The sun comes up, the snakes are basking to elevate their body temperature so that they
can go forage or move, or whatever they're going to be doing, that's a time when they're
still king of slow but they're trying to warm up and they're really susceptible to getting
hit with mowers. The other thing with mowing is to leave enough stubble so it leaves some
ground cover for snakes. Elevating the mower with not chopping the snakes but it's more
important for the cover that it leaves. Spoil piles from dredging, you know there's
a lot of guidance regarding not dredging and putting spoils next to canals in the winter
time, but you should remember that there's always a chance that snakes could be using
burrows, particularly burrows that are right on canal banks. The idea here, or the management
recommendation, is that in the winter, most of the snakes are going to be in these burrows.
So if you're piling up soil and wiping out these burrows, you're going to be wiping out
populations. In the summer if you're doing it, you're probably going to be taking a few
individuals, but at least you're not wiping out whole populations. It's just something
to be aware of that these snakes can be in burrows at any time of year.
One of the things that was brought to our attention by Sacramento National Wildlife
Refuge was the importance of debris piles and how you handle those. They had been clearing
out a water control structure of all the cattails and tules and branches that wash up to it
and just piling it up alongside. They came back about two weeks or a month later to get
rid of the pile and it was just loaded with giant gartersnakes. There's two options when
you're clearing out these water control structures. If you can make a pile of the debris and just
leave it, that's great because snakes will take cover there from the heat; they'll likely
hibernate in those locations if they're close enough to the water. But if it's not a location
where you can do that, the best thing to do is just put the stuff in a truck right away
and hauling it away without making a pile. You don't want the snakes to colonize that
created habitat and then be removed when you disturb it.
Just to reiterate, ground disturbing activities during hibernation are always a bad thing.
The closer to the wetlands and canals the worse, but they can move some distance from
those wetlands and canals. Just to summarize, in the past decade or so,
we've really increased our knowledge about giant gartersnake biology, but we have a lot
of information gaps still. One of the things that's still a big gap is the response to
specific management practices. We have these general ideas about how snakes respond to
habitat, what's good snake habitat, but we really don't have good controlled studies
of the snakes' response to a specific practice. And likewise, we have an idea of what the
habitat should look like and how habitat restoration should be practiced. We've got a good idea
of where it should happen now. We know that if we do it too far away from that historic
tule marsh it's probably not going to be colonized, at least not on its own, we might need to
do repatriation. But we don't know what are the most important features when we restore
these habitats and how the snakes respond to specific items that we put into these plans.
The relative value of different habitat types, again we kind of alluded to that earlier,
but we don't really know what is the value of a canal, and is a canal a canal. What about
wetlands? Rice agriculture, the future of rice in California is uncertain. How much
marsh would we need to make up for a loss of rice? And knowing how valuable each of
these different habitats are in terms of population growth rate, carrying capacity and these types
of things would be really beneficial. The effects of invasive species of all types -- prey,
predators, and plants -- is another thing that we think deserves more attention. I alluded
to the bullfrogs earlier. Are they a net benefit or a net cost to giant gartersnakes? Do they
take more young ones out than the food that they provide the adults causes them to produce?
I think that's an interesting question that, I don't know how much we can really manage
to get rid of bullfrogs, but it would be interesting to find out anyway. The other big question
mark is male and juvenile survival. They're too small for radio transmitters. We've now
got some more long-term mark recapture projects initiated that will hopefully help to answer
some of those questions. The radio telemetry is really nice because you are following individuals
and you can tell right when they die to within a couple of days usually. But the mark recapture
method will make up for that, it just takes more time to gather more data. There are a
lot of other questions that we don't know. I've just highlighted a few here.
With that I'd like to acknowledge the US Fish and Wildlife Service and California Department
of Fish and Wildlife for providing access to their refuges, the numerous biological
technicians that have worked on these projects, without them this wouldn't happen and they
really do a great job for us. Landowners that have provided access for us to their lands
have been instrumental to the occupancy study that I talked about in particular. And water
districts have, for the most part, been really cooperative as well. We've received funding
from a number of different groups, federal, state, and non-profits as well. And with that,
I'd be happy to take any questions.