Cues used by Predators to Detect Freshwater Turtle Nests may Persist Late into Incubation

Abstract

Previous studies have found that turtle nest depredation is concentrated immediately post-oviposition, likely because cues alerting predators to nest presence are most obvious during this time. In algonquin Provincial Park, Ontario, we examined the frequency of nest depredation during the incubation period for snapping Turtles (Chelydra serpentina [Linnaeus, 1758]) and Midland Painted Turtles (Chrysemys picta marginata [agassiz, 1857]). Contrary to most past findings, nest depredation occurred throughout the incubation period for both species. In fact, 83% and 86% of depredation interactions with snapping and Painted Turtle nests, respectively, occurred more than a week after oviposition at our study site. Peaks in nest depredation (weeks with ≥10% nest depredation) occurred late in incubation and may have coincided with hatching. Trail cameras deployed at four nesting sites revealed six predator species interacting with nests. The presence of predators at nest sites increased late in the incubation period indicating a persistence or renewal (from hatching) of cues; additional research is necessary to determine the nature of these cues. These findings have implications for both research and turtle conservation. Further research should examine the relationship between temporal changes in predator species' density and patterns of nest depredation. additionally, in areas where protective nest caging is used as a species recovery action, it may be important to ensure that cages remain in place throughout the incubation period until emergence of hatchlings.

Full text

179
Introduction
Freshwater and marine turtles are among the most at-
risk groups of vertebrate animals, and one of the many
threats to turtle populations is unnaturally high levels of
nest predation (Gibbons et al. 2000; Spotila 2011). A
broad range of animals from multiple taxa depredate
turtle nests, including mammals, birds, reptiles, and
in vertebrates (Ernst and Lovich 2009). High levels of
turtle nest depredation commonly occur in human-
disturbed landscapes because food resources associated
with human presence (e.g., trash, agricultural crops)
increase the abundance of subsidized predators of tur-
tle eggs (Christiansen and Gallaway 1984; Garber and
Burger 1995; Mitchell and Klemens 2000; Seburn
2007; Fordham et al. 2008; Kurz et al. 2011). In addi-
tion, the removal of top predators in urban areas has
caused an increase in the number of mesopredators,
lead ing to a concomitant increase in depredation of tur-
tle eggs (Barton 2003; Prugh et al. 2009; Ritchie and
Johnson 2009). In some turtle populations, subsidized
mesopredators annually depredate 100% of nests (e.g.,
Geller 2012), resulting in chronic prevention of recruit-
ment (Spinks et al. 2003; Seburn 2007; Fordham et al.
2008). Substantial repeated reductions in recruitment
(approximately a 50% or more decrease in egg and/or
hatchling survival) perpetuate population declines
(Crouse et al. 1987; Crowder et al. 1994; Tomillo et al.
2008). World-wide and in Canada, turtle populations
are in decline (Gibbons et al. 2000); thus, it is important
to understand all possible threats, including the ecology
and behaviour of nest predators, to increase the effec-
tiveness of management and recovery actions.
Many studies of turtle nest depredation report that
most depredation events occur within the first week
post-oviposition (Tinkle et al. 1981; Christens and Bider
1987; Congdon et al. 1983, 1987; Robinson and Bider
1988; Burke et al. 2005; Strickland et al. 2010; Geller
2012; Wirsing et al. 2012; Holcomb and Carr 2013).
For example, Congdon et al. (1983) found that 87% of
Blanding’s Turtle (Emydoidea blandingii) nest depre-
dation occurred within the first 5 days post-oviposition.
Congdon et al. (1987) found that of depredated Snap-
ping Turtle (Chelydra serpentina) nests, 59% were
depredated within the first 24 h, 73% of these nests were
depredated within the first 6 days, and 100% by the
32nd day post-oviposition. Robinson and Bider (1988)
found that 57% of depredation events occurred within
72 h of nest construction, and 87% occurred within 5
days post-oviposition. Similarly, Desroches and Picard
(2007) found that 80% of nests were depredated in the
first week post-oviposition. Holcomb and Carr (2013)
found that 86% of Alligator Snapping Turtle (Mac -
rochelys temminckii) nests were depredated within the
first 24 h, and, on average, the time until depredation
was 19 h.
The high peak of depredation immediately post-
oviposition is thought to occur because cues alerting
predators to the presence of a nest are most prominent
at this time. These cues might include olfactory stimuli,
such as the scent of oviposition fluid (Legler 1954;
Congdon et al. 1983; Spencer 2002), and visual cues,
such as the presence of a female turtle (Congdon et al.
1987; Eckrich and Owens 1995) and soil disturbance
Cues used by Predators to Detect Freshwater Turtle Nests may Persist
Late into Incubation
JULIA L. RILEY1and JACQUELINE D. LITZGUS1, 2
1Department of Biology, Laurentian University, 935 Ramsey Lake Road, Sudbury, Ontario P3E 2C6 Canada
2Corresponding author: jlitzgus@laurentian.ca
Riley, Julia L., and Jacqueline D. Litzgus. 2014. Cues used by predators to detect freshwater turtle nests may persist late into
incubation. Canadian Field-Naturalist 128(2): 179–188.
Previous studies have found that turtle nest depredation is concentrated immediately post-oviposition, likely because cues
alerting predators to nest presence are most obvious during this time. In Algonquin Provincial Park, Ontario, we examined
the frequency of nest depredation during the incubation period for Snapping Turtles (Chelydra serpentina [Linnaeus, 1758])
and Midland Painted Turtles (Chrysemys picta marginata [Agassiz, 1857]). Contrary to most past findings, nest depredation
occurred throughout the incubation period for both species. In fact, 83% and 86% of depredation interactions with Snapping
and Painted Turtle nests, respectively, occurred more than a week after oviposition at our study site. Peaks in nest depredation
(weeks with ≥10% nest depredation) occurred late in incubation and may have coincided with hatching. Trail cameras deployed
at four nesting sites revealed six predator species interacting with nests. The presence of predators at nest sites increased late in
the incubation period indicating a persistence or renewal (from hatching) of cues; additional research is necessary to determine
the nature of these cues. These findings have implications for both research and turtle conservation. Further research should
examine the relationship between temporal changes in predator species’ density and patterns of nest depredation. Additionally,
in areas where protective nest caging is used as a species recovery action, it may be important to ensure that cages remain in
place throughout the incubation period until emergence of hatchlings.
Key Words: Snapping Turtles; Chelydra serpentina; Midland Painted Turtles; Chrysemys picta marginata; Turtles; Algonquin
Provincial Park; Ontario; ecology; nest depredation; predator detection; predators; species recovery

180 THE CANADIAN FIELD-NATURALIST Vol. 128
(Strickland et al. 2010; Spencer 2002). Acceptance
of this evidence has led some researchers examining
cues for nest depredation (e.g., Marchand and Livaitis
2004; Strickland et al. 2010) to base their methodology
on the premise that depredation of turtle nests is
restricted to the first week post-oviposition. For only
2 weeks, Marchand and Livaitis (2004) monitored fake
nests constructed to investigate depredation, and Strick-
land et al. (2010) monitored nests for 2 days post-
oviposition. Yet, is an early peak the only temporal pat-
tern of turtle nest depredation reported in the literature?
Although most studies of nest depredation report
that it occurs within a short time post-oviposition, a
few, largely neglected, studies document substantial
levels of nest depredation throughout or late in the
incubation period. Snow (1982) found that 55% of
depredated Painted Turtle (Chrysemys picta) nests were
older than 3 days; however, all of the nests were still
depredated within the first month of incubation. Brooks
et al. (1992) found that all Wood Turtle (Glyptemys
insculpta) nest depredation occurred 9 weeks after the
last nest was laid. Burger (1977) found that the risk of
depredation of Diamond-backed Terrapin (Malaclemys
terrapin) nests did not decrease over the course of
incubation; instead, nest depredation was significantly
higher 60–90 days post-oviposition (75% of remain-
ing nests depredated) than 1–30 days and 30–60 days
post-oviposition (27% and 20% of nests within those
periods, respectively). Gillingwater (2002) observed
nest predation throughout the incubation period and,
in some cases, Northern Map Turtle (Graptemys geo-
graphica) nests were depredated the following spring
after hatchlings had overwintered in the nest. Some re -
ports document bimodal predator activity, with peaks
around nesting and hatching. For example, Congdon
et al. (1983) found that a few nests (6%) were depre-
dated at the end of the incubation period during hatch-
ling emergence and, similarly, Desroches and Picard
(2007) found that 5% of freshwater turtle nests were
depredated during hatchling emergence. In these stud-
ies, because of a large peak in nest depredation imme-
diately post-oviposition, there may not have been many
nests left to depredate in the fall. Also, nest debris may
make nest-searching by predators more challenging.
Nevertheless, predators often appear to cue in on
nests later in incubation. During fieldwork for a 41-
year study of the ecology of Snapping Turtles and a
35-year study of Midland Painted Turtles (Chrysemys
picta marginata) based out of the Wildlife Research
Station (WRS, 45º35'N, 78º30'W) in Algonquin Provin-
cial Park, Ontario (R. J. Brooks, University of Guelph
and J. D. Litzgus, Laurentian University), field tech-
nicians observed that nest depredation was occurring
throughout incubation. These observations and our sub-
sequent review of the literature led to our research ques-
tion: do predators detect nest cues that persist long
after oviposition and lead to later occurrences of nest
depredation?
In the long-term Algonquin Park study, eggs are re -
moved from the nest cavity less than 4 h post-oviposi -
tion, measured, and reburied within 24 h (Samson et
al. 2007). This study method is common in turtle re -
search, management, and conservation programs glob-
ally; thus, depredation patterns documented in our
study may be representative of what is occurring under
these circumstances. This method also provides the
unique opportunity to examine the temporal pattern
of nest depredation that occurs when fresh cues left
by the mother are reduced. The aim of our project was
two-fold: first, to quantify the frequency of nest depre-
dation throughout incubation for Snapping and Mid-
land Painted Turtles when nest cues are reduced imme-
diately post-oviposition and, second, to compare this
temporal pattern to patterns reported previously to
determine if nest cues and depredation peaks are pres-
ent later in incubation.
Study Area
The study took place along the Highway 60 corridor
on the west side of Algonquin Provincial Park, Ontario,
Canada. The study area is within the Algonquin–Lake
Nipissing ecoregion, which is a rugged landscape
underlain by Precambrian Shield outcrops (Ontario
Ministry of Natural Resources 1998). Forest cover
domi nates, with predominantly upland mixed forests
of sugar maple (Acer saccharum), yellow birch (Betula
alleghaniensis), and eastern hemlock (Tsuga canaden-
sis) and lowland sites with spruce (Picea spp.) and
tamarack (Larix laricina) (Hughes 2003). Field sam-
pling was concentrated at two sites: the WRS and
Arowhon. The WRS is within the North Madawaska
watershed and nesting sites in this area vary from nat-
ural sand dunes beside lakes to gravel embankments
along access roads and Highway 60. At the Arowhon
site, nesting occurs on a decommissioned rail-line that
is used as a public hiking trail (Mizzy Lake Trail;
Schwarzkopf and Brooks 1985). Elevations on the west
side of Algonquin Park (370 – 570 m above sea level)
are higher than the surrounding landscape and expe-
rience a cool and wet climate (Ontario Ministry of
Natural Resources 1998); this climate is reflective of
the northern range limits of both turtle species.
Methods
From 30 May to 4 July 2011, Snapping and Mid-
land Painted Turtle nesting was monitored daily. Nest
sites were monitored visually by researchers on foot
and using binoculars. Monitoring occurred from dawn
(about 0500) to about 1000, and from just before to
dusk (about 1700) until after dark, as long as nesting
activity was occurring, to capture the time frames in
which Snapping and Painted Turtles experience peak
nesting (Ernst and Lovich 2009). The nesting activity
of each female was recorded as it occurred. Nest loca-
tions were marked with metal stakes and flagging tape.
Nests were carefully excavated, maintaining the eggs

in the same order and orientation in which they were
found (Samson et al. 2007). After the eggs were re -
moved, the nest cavities were filled with soil excavated
during egg removal. The eggs were taken to a field
laboratory to record clutch data for the long-term study
(egg data were not used in the current study). Eggs
were reburied in their original nest chambers within
24 h post-oviposition, before the vitelline membrane
adhered to the inner shell surface and formed a “white
spot” on the top of the egg (Yntema 1964; Rafferty and
Reina 2012), thus ensuring no trauma to developing
embryos (Samson et al. 2007). We cannot discount the
possibility that removal, processing, and reburial of
eggs by researchers in the first 24 h affected depreda-
tion rates, especially at the start of incubation. Although
we present data for the first 72 h post-oviposition, we
are aware they may not be indicative of depredation
rates on nests from which eggs were not removed (Con-
gdon et al. 1983; Tinkle et al. 1981; Christens and Bider
1987; Burke et al. 2005; Strickland et al. 2010; Geller
2012; Wirsing et al. 2012).
Each nest was randomly assigned to a caging treat-
ment (above ground, below ground, or wooden-sided
cage, or uncaged control) and cages were deployed
when the eggs were reburied (69 Midland Painted Tur-
tle and 52 Snapping Turtle nests). Various nest cage
types were used to test their effects on hatchling fit-
ness for another study (Riley and Litzgus 2013); here
we report only the details of predator interactions with
these nests. During the nesting season (June) and the
hatching season (September), the nests were monitored
daily for predation attempts and successful depreda-
tion events. During the remainder of the incubation
period (July and August), the nests were monitored
weekly (Burke et al. 1998; Kolbe and Janzen 2002)
because of logistic constraints. A depredation attempt
was recorded when substrate was cleared away from
around the nest or nest cage and/or the nest cage was
dug up but the predator did not reach the nest cham-
ber (Figure 1A). A successful depredation event was
recorded when eggs were dug up and/or eaten (Fig-
ure 1B). If predator tracks and scat were discernible
2014 RILEY AND LITZGUS: TURTLE NEST PREDATION 181
FIGURE 1. A. An above-ground nest cage with substrate cleared away from the sides indicating a depredation attempt. B. Suc-
cessful depredation of a nest protected by a wooden-sided cage indicated by the excavated nest chamber and eggshell
fragments. C. A trail camera photograph of a Red Fox (Vulpes vulpes) interacting with a nest outfitted with an above-
ground cage. All photographs were taken during summer 2011 in Algonquin Provincial Park, Ontario, Canada.

around a nest, the predator species was identified and
recorded. To gauge the predator assemblage patrolling
the nesting sites and to capture interactions between
predators and nest cages, four trail cameras (119456C,
Bushnell Corporation, Overland Park, Kansas, USA)
were set-up from 1 July to 1 October 2011 at four nest
sites.
The numbers of depredation attempts and success-
ful depredation events were combined for analysis to
create a total number of depredation interactions. This
total did not differ among cage types or from the un -
caged control nests over the incubation period (Riley
and Litzgus 2013). It was assumed that if the nest
cages were not protecting the clutches, a depredation
attempt would have been successful. If a nest was sub -
ject to multiple depredation attempts (i.e., a predator
targeted the same nest multiple times), only the first
attempt was included in the dataset.
We quantified predator interactions with nests in two
ways. First, we used a non-parametric product-limit
survival analysis (Kaplan and Meir 1958; Engeman
et al. 2006). The survival time equaled the time (days)
from oviposition to the first predator interaction. When
eggs hatched, those nests were considered “censored”
after the time of that event, and thus removed from the
analysis. A Mann–Whitney–Wilcoxon test was used to
compare the survival curves between species. Second,
we undertook a more detailed analysis of the specific
timing of the depredation interactions during incuba-
tion. This part of the analysis included only nests that
were subject to depredation interactions during incu-
bation (Snapping Turtle, 29/52 nests; Midland Paint-
ed Turtle, 28/69 nests). For each species, the number
of depredation interactions was totaled for each week
of incubation (29 May to 1 October 2011). This num-
ber was divided by the total number of depredation
interactions over the incubation period and multiplied
by 100 to calculate the percentage (i.e., frequency) of
depredation interactions that occurred weekly. For each
species, this observed distribution of depredation fre-
quency was compared with an expected distribution
based on previous quantified reports of depredation in
the literature, using a Kolmogorov–Smirnov goodness-
of-fit test. From all literature sources that contained
usable data, we averaged the reported data to generate
an expected distribution of depredation interactions by
week post-oviposition (Table 1). Also, using a Kol-
mogorov–Smirnov goodness-of-fit test and our data,
we compared the distribution of weekly depredation
interactions post-oviposition for Snapping Turtles with
that for Midland Painted Turtles to determine whether
there were differences in pattern. Depredation peaks
were identified as weeks with ≥ 10% nest depredation.
A significance level of α = 0.05 was used for all sta-
tistical tests. Statistical analyses were performed using
R (R Foundation for Statistical Computing, Vienna,
Austria).
For predator interactions where tracks or scat were
identified, the number of interactions observed for each
predator species was summarized monthly. Similarly,
the number of photographs of each predator species
taken by the trail cameras was summarized monthly
to elucidate temporal patterns of predator presence at
the nest sites over the course of incubation.
Results
For both turtle species, the probability of nest sur-
vival decreased steadily with number of days post-
oviposition (Figure 2). The survival curves did not dif-
fer between species (W1= 365, P= 0.11). In 2011, the
first Snapping Turtle nest was found on 7 June, and
the last nest was found on 24 June. The observed fre-
quency of weekly predator interactions with Snapping
Turtle nests over the incubation period was different
than the expected distribution of depredation interac-
tions post-oviposition (D = 0.63, P < 0.01; Figure 3A).
For Midland Painted Turtles, the first nest was found
on 2 June and the last was found on 2 July. The ob -
served frequency of weekly predator interactions with
Midland Painted Turtle nests over the incubation period
182 THE CANADIAN FIELD-NATURALIST Vol. 128
TABLE 1. Rates of turtle nest predation over the course of incubation reported in previously-published studies. The means of
these rates were used to create an “expected distribution” of depredation frequencies for comparison with our observed fre-
quencies for Snapping Turtles and Midland Painted Turtles in Algonquin Provincial Park using a Kolmogorov–Smirnov
goodness-of-fit test. For the purposes of the test, the mean depredation frequency estimated for weeks 4+ (3.7%) was split
over weeks 4–16 (0.3% each week).
Depredation frequency (%)
Study Species Week 1 Week 2 Week 3 Week 4+
Congdon et al. 1983 Blanding’s Turtle 87 544
Christens and Bider 1987 Painted Turtle 86 0014
Burke et al. 2005 Diamond-backed Terrapin 100 000
Wirsing et al. 2012 Snapping Turtle 98 101
Painted Turtle 98 101
Geller 2012 Map Turtle spp. 90 70 3
Snow 1982 Painted Turtle 64 21 12 3
Mean expected distribution used
in Kolmogorov–Smirnov test 89 5 2.3 3.7

was also different from the expected distribution (D
=0.63, P < 0.01; Figure 3B).
For Snapping Turtle nests, 17% of predator inter-
actions occurred in the first week post-oviposition.
Another peak of predator interactions (weekly interac-
tions ranging from 10% to 17% of the total depreda-
tion interactions) occurred from weeks 10 to 14 (64 to
91 days) post-oviposition. Predator interactions oc -
curred throughout the incubation period up to 105 days
post-oviposition. For Midland Painted Turtle nests,
14% of predator interactions occurred in the first week
post-oviposition. Another spike in predator interactions
occurred 3 and 4 weeks post-oviposition, when 18%
of total depredation interactions occurred each week.
Another peak in depredation occurred at 11 weeks (71–
77 days) post-oviposition, when 14% of total depre-
dation interactions occurred. We also observed elevat-
ed levels of predation in week 12 (7% of total depre-
dation interactions). Similar to Snapping Turtle nests,
predator interactions with Midland Painted Turtle nests
occurred throughout incubation until 109 days post-
oviposition. The number of predator interactions per
week over the incubation period did not differ between
species (D = 0.31, P = 0.42).
Six predator species were identified interacting with
nests and were present at the study sites throughout
2014 RILEY AND LITZGUS: TURTLE NEST PREDATION 183
FIGURE 2. Survival curves (solid lines) for (A) Snapping Turtle
(Chelydra serpentina) and (B) Midland Painted Turtle
(Chrysemys picta marginata) nests over the incuba-
tion period in Algonquin Provincial Park. Vertical tick
marks indicate when eggs hatched and their nest was
“censored” from the analysis. The grey area represents
the period when hatchlings were emerging from the
nests. Dashed lines indicate standard errors.
FIGURE 3. For (A) Snapping Turtle (Chelydra serpentina)
and (B) Midland Painted Turtle (Chrysemys picta
marginata) nests in Algonquin Provincial Park, ob -
served weekly frequencies of predator interactions
over the incubation period (29 May to 17 September
2011) were significantly different from those expect-
ed based on previous reports (see Table 1 for list of
studies). The grey area represents the period when
hatchlings were emerging from the nests.

incubation: Red Fox (Vulpes vulpes), Eastern Wolf
(Canis lycaon), Raccoon (Procyon lotor), American
Crow (Corvus brachyrhynchos), Common Raven
(Corvus corax), and Wild Turkey (Meleagris gallopa-
vo) (Figure 4). The number of photographs of preda-
tors at nest sites captured by the trail cameras was high -
est in August and September; specifically, the number
of photographs of canid (Red Fox and Eastern Wolf)
predators increased during this time. Of the predator
interactions for which species were identified, Red Fox-
es were the most common predators of nests from July
until the end of incubation.
184 THE CANADIAN FIELD-NATURALIST Vol. 128
FIGURE 4. Assessment of the predator assemblage at Snapping Turtle (Chelydra serpentina) and Midland Painted Turtle
(Chrysemys picta marginata) nesting sites in Algonquin Provincial Park by (A) number of predators identified from
tracks and scat interacting with nests and (B) number of photographs taken by trail cameras of each predator species
by month over the incubation period.

Discussion
We observed nest predation throughout incubation,
and there were peaks in depredation (one for Snap-
ping Turtles and two for Midland Painted Turtles) more
than a week after oviposition. Our observed patterns
of nest predation were atypical compared to those gen-
erally reported in the literature. Only a handful of oth-
er published studies have reported nest predation
throughout or late in incubation (Snow 1982; Burger
1977). Also, the presence of predators at nest sites in
our study, particularly canid species, increased later in
incubation (August and September). These later peaks
in depredation may be a result of a larger predator pop-
ulation at this time of year because canid pups begin
to hunt independently in the fall (C. Callaghan pers.
comm.), and because predators are somehow cueing to
nests at hatching. Wild dogs have been documented to
target Leatherback Sea Turtle (Dermochelys coriacea)
and Green Sea Turtle (Chelonia mydas) nests during
hatching and to dig up nests and prey on hatchlings
be fore their emergence from the nest (Spotila 2011).
Burger (1977) found that crows and gulls depredated
terrapin nests only during oviposition, whereas mam-
mals (raccoons and foxes) depredated nests at a low rate
during oviposition and at a high rate during hatching.
Also, mammalian predators depredated Wood Turtle
(Glyptemys insculpta) nests 9 weeks post-oviposition
(Brooks et al. 1992). In our study, removing eggs from
nests in the first 24 h post-oviposition (which may
have reduced early-incubation nest depredation) pro-
vided an opportunity to investigate depredation in the
later stages of incubation.
It appears that cues to nests exist and can be detected
by predators throughout egg incubation, but what are
these cues? The most obvious answer may be that the
predator-exclusion cages around several of the nests
monitored in our study may have served as visual cues
to the nests’ presence. However, our study also includ-
ed nests with no cages and nests protected by cages
that did not extend above ground (Riley and Litzgus
2013) and, therefore, did not provide a visual cue. In -
deed, predation interactions did not differ among the
nest caging treatments, including controls with no cages
(Riley and Litzgus 2013); thus, it does not appear that
predators were using nest cages as visual cues to a food
source. Other researchers have also found that mark-
ing nests did not increase depredation rates (Burke et
al. 2005; Strickland et al. 2010). In contrast, raccoons,
coyotes, and Corvus sp. have been found to use nest
cages and nest markers, respectively, as visual cues
for nests (Mroziak et al. 2000; Rollinson and Brooks
2007; S. D. Gillingwater, pers. comm.). This is most
likely a learned response (Rollinson and Brooks 2007),
and it appears that, at our study site, predators have not
yet learned to associate nest cages, which were present
for only 2 summers, with the presence of turtle eggs.
A few other studies that examined unprotected turtle
nests also found that nest predation occurred late in
incubation (Snow 1982; Burger 1977; Brooks et al.
1992); thus, a natural cue must be attracting predators
to nests at that time. For Midland Painted Turtles, the
depredation peak at 3–4 weeks post-oviposition could
be associated with predators returning to the nesting
sites when these turtles lay their second clutches (ap -
proximately 2 weeks after the first clutch is deposited
at our study sites; Brooks et al., unpublished data). The
return of nesting female turtles may present a visual cue
for predators. The presence of hatchlings at a nesting
site may also act as a visual cue for predators later on
in incubation and trigger further searching and preda-
tion of nests. At Rondeau Provincial Park and Long
Point, Ontario, raccoons and coyotes have been ob -
served following the tracks of early-emerging turtles
back to their natal nest and consuming the remaining
eggs and young (S. D. Gillingwater, pers. comm.).
How else would predators know where unmarked
turtle nests are located? Embryonic fluids released dur-
ing hatching could serve as olfactory cues to attract
predators to nests in August and September. In addi-
tion, rotting undeveloped or unfertilized eggs may pro-
duce olfactory cues. At our study sites, canid presence
increased in August and September, and this group of
animals is well-known for its outstanding olfactory
abilities (Spotila 2011). Another possibility for nest
detection may be auditory cues. Some turtle hatchlings
have recently been reported to vocalize within the nest
cavity after hatching (Ferrara et al. 2013).
In our study, the later peak in predator interactions
with nests was found at 64–94 and 71–77 days post-
oviposition for Snapping and Painted Turtles, respec-
tively. These peaks of depredation precede the begin-
ning of hatchling emergence for Snapping Turtles (77
days post-emergence; Riley et al. unpublished data)
by 13 days, and for Midland Painted Turtles (74 days
post-emergence; Riley et al. unpublished data) by 3
days. Thus, these peaks could correspond to the release
of embryonic fluids (olfactory cues) and potential
vocalization by hatchlings (auditory cues). The study
by Brooks et al. (1992) lends support to this idea, as
mammalian predators began depredating nests at their
study site 9 weeks after the last nest was completed,
which corresponds to the hatching period and hatch-
ling emergence for Wood Turtles (Ernst and Lovich
2009). Although this is a convincing association, it is
impossible to correlate definitively the later peak in
predator interactions with the hatching period from
our study alone, as it is unknown when the eggs were
hatching. Hatchlings often remain in the nest chamber
after hatching, sometimes emerging 1–9 days later
(Bur ger 1976; Christens 1990) or, in the case of Painted
Turtles, in the spring after overwintering in the nest
(Storey et al. 1988; Costanzo and Lee 1995; Costanzo
et al. 2000; Packard and Packard 2003, 2004). Future
work should examine olfactory, auditory, and visual
cues at nests throughout incubation to determine wheth -
er the presence or types of cues change with time.
2014 RILEY AND LITZGUS: TURTLE NEST PREDATION 185

In our study, given that we potentially changed the
nest cues during the first 24 h post-oviposition, we
were able to measure peaks in nest depredation later
in incubation for two freshwater turtle species. For
Snapping Turtles, there was a later peak in nest depre-
dation 10–14 weeks post-oviposition, and, for Midland
Painted Turtles, there were peaks in nest depredation
during weeks 3, 4, and 11 post-oviposition. There was
also an increase in the number of canid predators at
the study sites during these later peaks in incubation,
which corroborates the findings of other studies in
which mammal predator presence at nest sites increased
before hatchling emergence from nests (Spotila 2011;
Burger 1977). We report that there were comparative-
ly low peaks in nest predation during the first week
post-oviposition: only 17% and 14% of depredation
interactions with Snapping and Midland Painted Tur-
tle nests, respectively, occurred at this time. Some re -
searchers have observed similar trends at other study
sites, with nest predation occurring both during nesting
and hatchling emergence (Burger 1977; Spotila 2011;
Congdon et al. 1983; Desroches and Picard 2007). At
various sites, differences in predator species densities
and interspecific differences in the cues used by pred-
ators to find nests (i.e., avian species using visual cues,
and mammal species using visual and olfactory cues)
may be responsible for geographic variation in depre-
dation frequency over incubation. Depredation frequen-
cy may also depend on individual predator behaviours
and how they vary among sites. More research is need-
ed to examine the effects of predator species and pred-
ator behaviour on temporal patterns of nest predation.
Management of at-risk turtles throughout North
America commonly includes the use of cages to pro-
tect nests from subsidized mesopredators (Seburn
2007; Riley and Litzgus 2013). Some strategies sug-
gest removal of nest cages halfway through incuba-
tion, as it is thought that depredation will not occur dur-
ing the last half of incubation. In a study by Rahman
and Burke (2010), nest protectors were removed from
nests after 21–25 days of incubation, and, in one study
area (out of three), depredation significantly increased
in the 11 nights after nest protector removal. Engeman
et al. (2006) also found that after removing nest cages
from sea turtle nests, depredation rates increased sub-
stantially. Soil disturbance during nest cage removal
may provide a cue to attract predators (Rahman and
Burke 2010). An alternative to removing nest cages
partway through incubation is including an opportu-
nity for turtles to escape from the cage (e.g., escape
hatches or holes). Our study, as well as others that doc-
ument nest depredation before hatchling emergence,
indicates that turtle nests can be at risk of depredation
throughout the full incubation period. Thus, it is impor-
tant to understand and study depredation timing at each
location, as it may be site or predator specific, to gauge
the effectiveness of a particular nest protection method
for that site. In some areas, species recovery may re -
quire that nest protection measures remain in place
throughout incubation and extend to hatchling emer-
gence.
Acknowledgements
All work was carried out in compliance with the
Canadian Council for Animal Care guidelines and
under approved Laurentian University Animal Care
Committee protocol 2008-12-02. All work was author-
ized by permits from the Ontario Ministry of Natural
Resources (OMNR). Financial support for this work
was provided by the Natural Sciences and Engineer-
ing Research Council, the Canadian Wildlife Federa-
tion, the OMNR’s Species at Risk Stewardship Fund,
the Toronto Zoo, and Laurentian University. In-kind
contributions were provided by the Wildlife Research
Station (WRS) in Algonquin Provincial Park and the
University of Guelph. We thank R. J. Brooks for allow-
ing access to his long-term study sites. Thanks also to
the following people who assisted with fieldwork: K.
Hall, M. Keevil, H. McCurdy-Adams, P. Moldowan,
L. Monck-Whipp and staff and volunteers from the
WRS. Thanks to J. Baxter-Gilbert, C. Davy, P. Mol do -
wan, S. Gillingwater, and an anonymous reviewer for
their helpful comments on the manuscript.
Literature Cited
Barton, B. T. 2003. Cascading effects of predator removal
on the ecology of sea turtle nesting beaches. M.Sc. thesis,
University of Central Florida, Orlando, Florida, USA.
Burger, J. 1976. Behavior of hatchling diamondback terra-
pins (Malaclemys terrapin) in the field. Copeia 1976: 742–
748.
Burger, J. 1977. Determinants of hatching success in Dia-
mondback Terrapin, Malaclemys terrapin. American Mid-
land Naturalist 97: 444–464.
Burke, V. J., S. L. Rathbun, J. Bodie, and J. W. Gibbons.
1998. Effect of density on predation rate for turtle nests
in a complex landscape. Oikos 83: 3–11.
Burke, R. L., C. M. Schneider, and M. T. Dolinger. 2005.
Cues used by raccoons to find turtle nests: effects of flags,
human scent and diamond-backed terrapin sign. Journal
of Herpetology 39: 312–315.
Brooks, R. J., C. M. Shilton, G. P. Brown, and N. W. S.
Quinn. 1992. Body size, age distribution, and reproduc-
tion in a northern population of wood turtles (Clemmys
insculpta). Canadian Journal of Zoology 70: 462–469.
Christens, E. 1990. Nest emergence lag in loggerhead sea
turtles. Journal of Herpetology 24: 400–402.
Christens, E., and J. R. Bider. 1987. Nesting activity and
hatching success of the painted turtle (Chrysemys picta
marginata) in Southwestern Quebec. Herpetologica 43:
55–65.
Christiansen, J. L., and B. J. Gallaway. 1984. Raccoon re -
moval, nesting success, and hatchling emergence in Iowa
turtles with special reference to Kinosternon flavescens.
Southwestern Naturalist 29: 343–348.
Congdon, J. D., G. L. Breitenbach, R. C. van Loben Sels,
and D. W. Tinkle. 1987. Reproduction and nesting ecol-
ogy of snapping turtles (Chelydra serpentina) in South-
eastern Michigan. Herpetologica 43: 39–54.
186 THE CANADIAN FIELD-NATURALIST Vol. 128

Congdon, J. D., D. W. Tinkle, G. L. Breitenbach, R. C.
and van Loben Sels. 1983. Nesting ecology and hatch-
ing success in the turtle Emydoidea blandingi. Herpeto-
logica 39: 417–429.
Costanzo, J. P., and R. E. Lee, Jr. 1995. Supercooling and
ice nucleation in vertebrates. Pages 221–237 in Biological
Ice Nucleation and Its Applications. Edited by R. E. Lee,
Jr., G. J. Warren, and L. V. Gusta. American Phytopatho-
logical Society, St. Paul, Minnesota, USA.
Costanzo, J. P., J. D. Litzgus, J. B. Iverson, and R. E. Lee,
Jr. 2000. Seasonal changes in physiology and develop-
ments of cold hardiness in the hatchling painted turtle
Chrysemys picta. Journal of Experimental Biology 203:
3459–3470.
Crouse, D. T., L. B. Crowder, and H. Caswell. 1987. A stage-
based population model for loggerhead sea turtles and
implications for conservation. Ecology 68: 1412–1423.
Crowder, L. B., D. T. Crouse, S. S. Heppell, and T. H.
Martin. 1994. Predicting the impact of turtle excluder
devices on loggerhead sea turtle populations. Ecological
Applications 4: 437–445.
Desroches, J. R., and I., Picard. 2007. Évaluation de l’im-
pact des routes sur les population de tortues en Outaouais,
Québec. Ministère des Transports du Québec, Québec,
Québec, Canada.
Eckrich, C. E., and D. W. Owens. 1995. Solitary versus
arribada nesting in the Olive Ridley Sea Turtles (Lepi-
dochelys olivacea): a test of the predator-satiation hypoth-
esis. Herpetologica 51: 349–354.
Engeman, R. M., R. E. Martins, H. T. Smith, J. Woolard,
and C. K. Crady. 2006. Impact on predation of sea turtle
nests when predator control was removed halfway through
the nesting season. Wildlife Research 33: 187–192.
Ernst, C. H., and J. E. Lovich. 2009. Turtles of the United
States and Canada. Second edition. Johns Hopkins Uni-
versity Press, Baltimore, Maryland, USA.
Ferrara, C. R., R. C. Vogt, and R. S. Sousa-Lima. 2013.
Turtle vocalizations as the first evidence of post-hatching
parental care in chelonians. Journal of Comparative Psy-
chology 127: 24–32.
Fordham, D. A., A. Georges, and B. W. Brook. 2008.
Indigenous harvest, exotic pig predation and location per-
sistence of a long-lived vertebrate: managing a tropical
freshwater turtle for sustainability and conservation. Jour-
nal of Applied Ecology 45: 52–62.
Garber, S. D., and J. Burger. 1995. A 20-yr study docu-
menting the relationship between turtle decline and human
recreation. Ecological Applications 5: 1151–1162.
Geller, G. A. 2012. Notes on the nest predation dynamics of
Graptemys at two Wisconsin sites using trail camera mon-
itoring. Chelonian Conservation and Biology 11: 197–205.
Gibbons, J. W., D. E. Scott, T. J. Ryan, K. A. Buhlmann,
T. D. Tuberville, B. S. Metts, J. L. Greene, T. Mills, Y.
Leiden, S. Poppy, and C. T. Winne. 2000. The global
decline of reptiles, déjà vu amphibians. Bioscience 50:
653–666.
Gillingwater, S. D. 2002. A selective herpetofaunal survey,
inventory and biological research study of Rondeau Provin-
cial Park. Report to Rondeau Provincial Park. Morpeth,
Ontario.
Holcomb, S. R., and J. L. Carr. Mammalian depredation of
artificial alligator snapping turtle (Macrochelys temmin -
ckii) nests in north Louisiana. Southeastern Naturalist 12:
478–491.
Hughes, E. J. 2003. Nest-site selection, hatching success
and hatchling over-winter success in painted turtles (Chry-
semys picta). M.Sc. thesis, University of Guelph, Guelph,
Ontario.
Kaplan, E. L., and P. Meir. 1958. Nonparametric estimation
from incomplete observations. Journal of the American
Statistical Association 53: 457–481.
Kolbe, J. J., and F. J. Janzen. 2002. Spatial and temporal
dynamic of turtle nest predation: edge effects. Oikos 99:
538–544.
Kurz, D. J., K. M. Straley, and B. A. DeGregorio. 2011.
Out-foxing the red fox: how to best protect the nests of
the endangered loggerhead marine turtle Caretta caretta
from mammalian predation? Oryx 2011: 1–6.
Legler, J. M. 1954. Nesting habits of the western painted tur-
tle, Chrysemys picta belli (Gray). Herpetologica 10: 137–
144.
Marchand, M. N., and J. A. Litvaitis. 2004. Effects of
landscape composition, habitat features, and nest distri-
bution on predation rates of simulated turtle nests. Bio-
logical Conservation 117: 243–251.
Mitchell, J. C., and M. W. Klemens. 2000. Primary and
secondary effects of habitat alteration. Pages 5–32 in Turtle
Conservation. Edited by M. W. Klemens. Smithsonian
Institution Press, Washington, DC, USA.
Mroziak, M. I., M. Salmon, and K. Rusenko. 2000. Do
wire cages protect sea turtles from foot traffic and mam-
malian predators? Chelonian Conservation and Biology 3:
693–698.
Ontario Ministry of Natural Resources. 1998. Algonquin
Provincial Park management plan. Ottawa: Queen’s Printer
for Ontario.
Packard, G. C., and M. J. Packard. 2003. Natural freeze-
tolerance in hatchling painted turtles? Comparative Bio-
chemistry and Physiology 134: 233–246.
Packard, G. C., and M. J. Packard. 2004. To freeze or not
to freeze: adaptations for overwintering by hatchlings of
the North American painted turtle. Journal of Experimen-
tal Biology 207: 2897–2906.
Prugh, L. R., C. J. Stoner, C. W. Epps, W. T. Bean, W. J.
Ripple, A. S. Laliberte, and J. S. Brashares. 2009. The
rise of the mesopredator. BioScience 59: 779–791.
Rafferty, A. R., and R. D. Reina. 2012. Arrested embryonic
development: a review of strategies to delay hatching in
egg-laying reptiles. Journal of Comparative Physiology
B: Biochemical, Systemic, and Environmental Physiology
279: 2299–2308.
Rahman, S., and R. L. Burke. 2010. Evaluating nest pro-
tectors for turtle conservation. Pages 1–23 in Final Reports
of the Tibor T. Polgar Fellowship Program, 2009. Edited
by D. J. Yozzo, S. H. Fernald, and H. Andreko. Hudson
River Foundation New York, New York, USA.
Riley, J. L., and J. D. Litzgus. 2013. Evaluation of predator-
exclusion cages used in turtle conservation: cost-analysis,
and effects on nest environment and proxies of hatchling
fitness. Wildlife Research 40: 499–511.
Ritchie, E. G., and C. N. Johnson. 2009. Predator interac-
tions, mesopredator release, and biodiversity conservation.
Ecology Letters 12: 982–998.
Robinson, C., and J. R. Bider. 1988. Nesting synchrony: a
strategy to decrease predation of snapping turtle (Chelydra
serpentina) nests. Journal of Herpetology 22: 470–463.
Rollinson, N., and R. J. Brooks. 2007. Marking nests in -
creases the frequency of nest depredation in a northern
2014 RILEY AND LITZGUS: TURTLE NEST PREDATION 187

pop ulation of painted turtles (Chrysemys picta). Journal
of Herpetology 41: 174–176.
Samson, J., E. J. Hughes, and R. J. Brooks. 2007. Excava-
tion is a nondeleterious method for obtaining fecundity
and morphometric data from small-sized eggs of fresh-
water turtles. Chelonian Conservation and Biology 6: 255–
259.
Schwarzkopf, L., and R. J. Brooks. 1985. Sex determina-
tion in northern painted turtles: effect of incubation at con-
stant and fluctuating temperatures. Canadian Journal of
Zoology 63: 2543–2547.
Seburn, D. 2007. Recovery strategy for species at risk turtles
in Ontario. Ontario Multi-Species Turtles at Risk Recovery
Team, Ottawa, Ontario, Canada, 72 pages.
Snow, J. E. 1982. Predation on painted turtle nests: nest sur-
vival as a function of nest age. Canadian Journal of Zool-
ogy 60: 3290–3292.
Spencer, R. J. 2002. Experimentally testing nest site selection:
fitness trade-offs and predation risk in turtles. Ecology 83:
2136–2144.
Spinks, P. Q., G. B. Pauly, J. J. Crayon, and H. B. Shaffer.
2003. Survival of the western pond turtle (Emys marmora-
ta) in an urban California environment. Biological Con-
servation 113: 257–267.
Spotila, J. R. 2011. Saving Sea Turtles: Extraordinary Stories
from the Battle Against Extinction. Johns Hopkins Uni-
versity Press, Baltimore, Maryland.
Storey, K. B., J. M. Storey, S. P. J. Brooks, T. A. Churchill,
and R. J. Brooks. 1988. Hatchling turtles survive freezing
during winter hibernation. Proceedings of the National
Academy of Sciences 85: 8350–8354.
Strickland, J., P. Colbert, and F. J. Janzen. 2010. Experi-
mental analysis of effects of markers and habitat structure
on predation of turtle nests. Journal of Herpetology 44:
467–470.
Tinkle, D. W., J. D. Congdon, and P. C. Rosen. 1981. Nest-
ing frequency and success: implications for the demogra-
phy of painted turtles. Ecology 62: 1426–1432.
Tomillo, P. S., V. S. Saba, R. Piedra, F. V. Paladino, and
J. R. Spotila. 2008. Effects of illegal harvest of eggs on
the population decline of leatherback turtles in Las Baulas
Marine National Park, Costa Rica. Conservation Biology
22: 1216–1224.
Wirsing, A. J., J. R. Phillips, M. E. Obbard, and D. L.
Murray. 2012. Incidental nest predation in freshwater tur-
tles: inter- and intraspecific differences in vulnerability are
explained by relative crypsis. Oecologia 168: 977–988.
Yntema, C. L. 1964. Procurement and use of turtle embryos
for experimental procedures. The Anatomical Record 149:
577–585.
Received 11 January 2014
Accepted 25 January 2014
188 THE CANADIAN FIELD-NATURALIST Vol. 128