Content uploaded by Rosalind M Rolland
Author content
All content in this area was uploaded by Rosalind M Rolland on Jul 06, 2016
Content may be subject to copyright.
INTRODUCTION
Effective conservation and management of many cetaceans
has been hindered by insufficient non-lethal methods to
acquire data on feeding ecology, reproductive parameters,
individual and population health and the physiological
impacts of environmental and anthropogenic stressors (e.g.
marine biotoxins, contaminants, global climate change).
This has been particularly problematic for large whales,
which are elusive and extremely difficult to live-capture for
sampling of blood or tissues. While remote biopsy darting
provides samples for genetic, contaminant and fatty acid
analyses, the data that can be obtained from skin and
blubber cores are limited.
A significant decline in reproduction and health in the
western North Atlantic right whale population (Eubalaena
glacialis) in the late 1990s raised concern among managers
and researchers (Kraus et al., 2001; Pettis et al., 2004;
Hamilton and Marx, 2005). In response, the International
Whaling Commission (IWC) Workshops on the
Comprehensive Assessment of Right Whales, and Status
and Trends of Western North Atlantic Right Whales (IWC,
2001a; b) gave priority recommendations to develop
methods for assessing health, stress and reproductive
failure. Subsequently, a suite of faecal-based studies were
validated and applied to northern right whales to assess the
reproductive status of individual whales, and to study
factors potentially affecting health and fecundity.
Measurement of faecal metabolites of steroid hormones
has now been used to determine reproductive status of free-
swimming right whales (Rolland et al., 2005). That study
showed that concentrations of reproductive hormone
metabolites were reliable predictors of gender, pregnancy
and lactation in females and sexual maturity in males.
Current extensions to this work involve identifying
individuals by creating genetic profiles using right whale
DNA isolated from their faeces (R. Gillett, unpublished
data) and measuring metabolites of adrenal hormones to
assess relative stress levels (Hunt et al., 2006). Faecal
parasitology studies have shown that right whales have the
highest prevalence of infection with potentially pathogenic
protozoa (Giardia spp. and Cryptosporidium spp.) of any
marine mammal yet examined (Hughes-Hanks et al., 2005).
In that study, over 70% of the faecal samples collected from
right whales were Giardia spp. positive and 24% were
positive for Cryptosporidium spp. Finally, faecal
measurements of the paralytic shellfish poisoning (PSP)
toxins produced by the ‘red tide’ organism Alexandrium
showed that sampled right whales were being exposed to
this potent neurotoxin by feeding. In some cases, toxin
levels reached 0.5mg saxitoxin equivalents g
–1
faeces, near
the levels at which human advisories for shellfish are issued,
although the biological effects on right whales remain
unknown (Doucette et al., 2006). All of these studies were
derived from multiple assays of the same faecal (scat)
samples where the individual whale can frequently be
identified either photographically (by comparison with the
North Atlantic Right Whale Catalogue; Hamilton and
Martin, 1999) or genetically (by comparing scat DNA
profiles to biopsy DNA profiles of known whales).
Preliminary results show that at least 14 whales have been
sampled more than once within a field season and/or in
multiple years (R. Rolland, unpublished data). These studies
represent the foundation of an individual-based profile of
health and reproductive status, that when integrated with
the Right Whale Catalogue, provide insights into
population-based models of reproduction, health, mortality
and trends.
J. CETACEAN RES. MANAGE. 8(2):121–125, 2006
121
Faecal sampling using detection dogs to study reproduction and
health in North Atlantic right whales (Eubalaena glacialis)
ROSALIND M. ROLLAND
*
, PHILIP K. HAMILTON
*
, SCOTT D. KRAUS
*
, BARBARA DAVENPORT
+
, ROXANNE M. GILLETT
†
AND
SAMUEL K. WASSER
‡
Contact e-mail: rrolland@neaq.org
ABSTRACT
Conservation and management of many cetaceans is hindered by the difficulty of acquiring samples from free-swimming individuals to
obtain essential data on health, diet, reproduction and physiological impacts of environmental and anthropogenic stressors. This is
particularly true for large whales, which are logistically difficult to live-capture for sampling. In North Atlantic right whales (Eubalaena
glacialis), a significant decline in reproduction and health in the 1990s led to the application of faecal-based analyses to study stress and
reproductive endocrinology, marine biotoxin exposure and prevalence of disease-causing protozoa. However, this approach was limited by
low sample acquisition rates with opportunistic faecal (scat) collection methods. The work presented here evaluates the relative sampling
efficiency of scent detection dogs trained to locate North Atlantic right whale scat versus opportunistic scat collection during photo-
identification surveys. Three years of sample collection using both detection dogs and opportunistic methods are summarised. Faecal sample
collection rates using detection dogs were over four times higher than opportunistic methods. The use of detection dogs for scat collection
from free-swimming right whales has for the first time provided adequate numbers of samples for statistical analyses. The endocrine,
disease, genetic and biotoxin studies currently being performed on these samples markedly improve the ability to address fundamental
questions vital to effective conservation and management of highly endangered right whales.
KEY WORDS: SAMPLING STRATEGY; NORTH ATLANTIC RIGHT WHALE; HORMONES; REPRODUCTION; GENETICS;
DISEASE
*
New England Aquarium, Central Wharf, Boston, MA 02110-3399, USA.
+
Packleader Detector Dogs, 14401 Crews Road KPN, Gig Harbor, WA 98329-4192, USA.
†
Natural Resources DNA Profiling and Forensic Centre, Department of Biology, Trent University, 1600 East Bank Drive, Peterborough, Ontario K9J
7B8, Canada.
‡
Center for Conservation Biology, Department of Biology, Box 351800, University of Washington, Seattle, WA 98195-1800, USA.
Despite the wealth of data available from these analyses,
this approach has been restricted by the difficulty of
opportunistically locating scat at sea, limiting the number of
available samples. This led to evaluating the use of domestic
dogs (Canis familiaris) professionally trained to detect
wildlife scat (Wasser et al., 2004) to increase sample
collections from right whales. In terrestrial studies,
detection dogs significantly increased scat collection rates
from kit foxes (Vulpes macrotis mutica; Smith, D.A. et al.,
2003), grizzly (Ursus arctos) and black bears (U.
americanus; Wasser et al., 2004). In those studies, dogs
located scat from targeted wildlife with 100% accuracy
(based on genetic species confirmation), and increased
sampling rates four-fold, compared to experienced human
observers. This paper describes the use of detection dogs to
locate faecal samples from right whales over three years.
Faecal sampling efficiency of surveys with dogs is
compared to opportunistic methods and species identity is
confirmed genetically for a subset of samples.
METHODS
Study area and survey methods
This work was conducted during August and September,
2003-05 in the waters around Lubec, Maine (training) and in
the Bay of Fundy, Canada (surveys), where right whales
congregate seasonally to feed (Murison and Gaskin, 1989).
Faecal sample collection surveys using detection dogs were
conducted aboard a 6.4m boat with a global positioning
system (GPS) chart plotter. The chart plotter was used to
mark the location of tracklines and positions where dogs
detected scent from right whale scat, and helped orient the
boat relative to wind and tide direction to locate samples.
The crew included one dog and three to four people (dog
handler, driver, photographer/data recorder). In addition,
opportunistic faecal sample collections occurred aboard a
9.0m vessel with a crew of six to eight people conducting
standardised right whale photo-identification surveys.
Surveys used two detection dogs alternately in 2003 and
2004, and a single dog in 2005. Given the demands of
working on a boat, dogs that had good physical stability,
persistence in locating samples and a calm disposition were
selected. Scat detection dog training follows techniques
used for narcotic, search and rescue and bomb detection
dogs (Wasser et al., 2004). When the dog detects the
targeted scent there is a characteristic change in behaviour
(recognised by tense body posture and ear position),
motivated by the expectation of a reward. Scent from right
whale scat was added to these dogs’ repertoires through
initial exposure using a scent box (Wasser et al., 2004),
followed first by searches on land, then from the bow of a
boat. Previously collected scat samples from male and
female right whales of varied ages were used for training.
Initial training occurred over a period of nine days, and
‘refresher’ work for both handlers and dogs occurred
annually for one or two days prior to the start of each field
season.
All surveys using dogs were conducted with a Beaufort
sea state 53 and wind speeds 510 knots. Boat transects
were conducted perpendicular to the wind direction at a
speed of five to seven knots, downwind from aggregations
of right whales or areas where right whales had been
previously sighted. The dogs were positioned on the bow for
the duration of the trial. On land, the dog leads the handler
directly to the sample by following the scent cone along an
increasing odour gradient. On the water, since the dog could
not lead the handler, the helmsman steered according to the
direction indicated by the dog (as interpreted by the handler)
until the sample was located (Fig. 1). If the dog lost the scent
during the approach, perpendicular transects were resumed
until the dog’s behaviour indicated that the vessel was back
in the scent cone from the sample (Fig. 2). When faecal
samples were successfully collected, the dog was rewarded
immediately by playing with a tennis ball on a string.
Sample collection
Floating pieces of clumped right whale scat were collected
using a 300mm nylon dipnet (Sea-Gear Corp., Melbourne,
Florida, USA; Rolland et al., 2005). Scat samples were
identified in the field by size, shape, brown-orange colour,
characteristic odour and presence of fine baleen hairs. Salt
water was drained off the faeces, samples were stored in
polypropylene jars and placed on ice until frozen at 220°C
for subsequent analyses. The date, time and position of
collection were recorded for each sample. When defecation
was witnessed, the whale was photographed for subsequent
photo-identification analysis (Kraus et al., 1986).
Comparison of sample collection methods
The sampling efficiency of the detection dog surveys was
calculated by dividing the number of faecal samples
collected per day by the total time that the dog was working.
Hours of dog survey effort were defined as the total time the
dog was working ‘on watch’ during transects. These results
were compared with opportunistic faecal sample collections
made during right whale photo-identification surveys.
Opportunistic collections occurred when whales were
observed defecating at the surface or observers detected scat
by odour. Hours of opportunistic effort were defined as the
time observers were ‘on watch’ between the first and last
whale photographed that day. Samples collected per hour of
survey effort were calculated over three years (2003-05).
Comparisons between opportunistic surveys and detection
dog surveys were only made on days when both vessels
were working to control for variability in weather conditions
and whale density.
Data analysis was performed using SPSS 13.0 (SPSS Inc.,
Chicago, Illinois, USA). The data were not normally
distributed, thus non-parametric tests were used. Differences
Fig. 1. The dog handler signalling to the helmsman the direction to
steer as indicated by the detection dog during a search for a right
whale scat sample.
122 ROLLAND
et al.:
FAECAL SAMPLING USING DETECTION DOGS
were considered significant if p<0.05. The number of
samples collected per day and the sampling efficiency using
detection dogs were compared to results from opportunistic
collection methods using a Mann-Whitney U test.
Differences in sampling efficiency between years were
analysed for each method separately using the Kruskal-
Wallis test. The detection distance for each sample located
by the dogs was estimated by calculating the distance
between the GPS positions of the first observed change of
behaviour (indicating scent acquisition) and the location of
sample collection. These are estimates of distance because
tidal motion may have moved the scat (closer or farther
depending on the stage) relative to the location of the dog’s
first detection.
Genetic analyses
Species identity was determined genetically for 54 samples
collected in 2003 by extraction and amplification of
mitochondrial control region DNA. DNA was extracted in
duplicate from frozen, lyophilised faecal samples using a
modified Qiagen DNeasy extraction protocol (Qiagen,
Valencia, CA). Nucleic Acid Purification Grade Lysis Buffer
(1X, 1.6ml; ABI) was added to ~70-90mg of the each
sample, then samples were vortexed (1min) and incubated
(65°C, 1hr). Following incubation, 25ml Proteinase K
(>600mAU ml
–1
; Qiagen) and 600mL AL buffer (Qiagen)
were added. Tubes were inverted and incubated for an
additional hour. Ethanol was added (100%, 600ml), the tubes
were mixed, and the contents were run through a silica spin
column. Samples were washed and eluted following steps
four through seven of the Qiagen DNeasy protocol,
incubated (65°C, 10min) to evaporate any residual ethanol
and frozen at 220°C.
The mitochondrial control region was amplified using the
polymerase chain reaction (PCR) with the primers UP098
and LP282 (Malik et al., 2000; Rastogi et al., 2004).
Amplification consisted of a 25ml reaction (0.3mg bovine
serum albumin, 1X PCR Buffer, 0.2mM of dNTP mix, 2mM
magnesium chloride, 0.3mM each primer, 0.1U Taq DNA
polymerase and ~1.5ng template DNA) with the following
cycling conditions: 94°C for 5min; 50 cycles of 94°C for
30s, 52°C for 60s, 72°C for 60s; 60°C for 45min. Extraction
and PCR negative controls were included to test for
contamination.
RESULTS
Results from the detection dog and photo-identification
surveys were compared for 19 days (2003-05) on which
both detection dog and opportunistic survey vessels were
working. Detection dog surveys located significantly more
samples (n=97) compared to the opportunistic method
(n=30; Mann-Whitney U test, Z=-3.418, p<0.001).
Detection dogs located many scat samples in areas where
the human crew did not observe whales in close proximity.
The mass of faeces collected varied from approximately 20g
to 0.5kg or more. Mean sampling efficiency of the detection
dog surveys from 2003-05 was 1.1 samples hr
–1
(range: 0.80
to 1.43 samples hr
–1
), significantly greater than 0.25
samples hr
–1
(range: 0.15 to 0.32 samples hr
–1
) for
J. CETACEAN RES. MANAGE. 8(2):121–125, 2006
123
Fig. 2. An example of the search pattern the research vessel followed (?) to locate a right whale scat sample
with a detection dog. As the vessel enters the scent cone coming from the sample (striped area), the dog
detects the odour (0) as indicated by a change in the dog’s ear set and body position, prompting the boat
driver to steer into the wind. The dog loses the odour when the vessel leaves the scent cone (X). The
vessel then resumes a transect perpendicular to the wind until the dog has another detection, turning into
the wind again to find the sample. The human crew smelled the sample just before it was collected (4).
The distance from the first scent detection by the dog to the final position of sample collection was
~0.5km.
opportunistic surveys (Table 1; Mann-Whitney U=5.000,
Z=-5.129, p<0.001). Although the sampling efficiency of
both methods appeared to be higher in 2005 (Table 1), there
were no significant differences between years for either
method, indicating consistency in the survey methodologies.
Estimated detection distances for the dogs ranged 22m to
1.93km (just over one nautical mile). In 2003, the only year
that this was measured, humans detected seven samples (by
smell) at 56-359m, while the dogs detected the same
samples at 150-563m. All faecal samples found by the dogs
and humans in 2003 have been confirmed to be from right
whales by mitochondrial DNA analyses, and the remainder
are currently undergoing analysis.
Statistical comparisons only included a subset of samples
collected on days when both research vessels were working
in the Bay of Fundy. Another 72 faecal samples were
obtained between 2003-05 on other survey days, in other
habitats or by other vessels in the Bay of Fundy (total
samples from 2003-05 = 199). Prior to using detection dogs
(1999-2002), an additional 86 samples were collected
opportunistically, bringing the total samples for all faecal-
based studies to 285. All samples were collected for
reproductive and stress hormone analyses. In many cases
sufficient faecal material was collected to allow for sub-
dividing of samples for multiple assays, so that 128 of these
samples are also being examined for marine biotoxins, and
111 for parasites. Additionally, all samples will eventually
be characterised genetically using mitochondrial and
nuclear markers to confirm the species of origin and
determine individual whale identity.
DISCUSSION
These results demonstrate that scat detection dogs can work
from boats to dramatically increase faecal sampling rates
from free-swimming right whales. Sampling efficiency of
detection dogs was over four times higher than opportunistic
collection methods over a three-year period. In addition,
dogs detected samples from as far as one nautical mile away,
greatly increasing the area that can be sampled. The success
of this method depended upon the involvement of a
professional dog trainer, an experienced handler and dogs
and a boat driver with intimate knowledge of the local tide
and wind patterns. It also involved use of a dedicated vessel
for detection dog surveys, because of methodological
conflicts between visually-based photo-identification
surveys and detection dog survey protocols. Nevertheless,
using dogs to collect large numbers of scat samples from
right whales has significantly increased sample sizes,
enhancing the utility of the diversity of faecal analyses in
quantitatively assessing this population’s status.
These assays and faecal collection methods are
potentially useful in multiple species, and can address a
wide array of questions. Faeces have been collected
opportunistically from bottlenose dolphins (Tursiops
truncatus) for genetic studies (Parsons et al., 2003), sperm
whales (Physeter macrocephalus) for feeding ecology
research (Smith and Whitehead, 2000), blue whales
(Balaenoptera musculus) and humpback whales (Megaptera
novaeangliae) to study marine biotoxin exposure (Lefebvre
et al., 2002) and North Atlantic right whales for
environmental toxicology (Weisbrod et al., 2000). Faecal
analyses provide estimates of exposure to synthetic
chemicals and biotoxins, both issues of concern to cetaceans
worldwide because of increasing human impacts on the
marine environment.
In addition to the assays described here, DNA markers
from prey species in scat are being used in cetaceans to
identify dietary components and diversity to understand
marine food webs with more accuracy than previous work
relying on analysis of hard parts of prey in faeces or stomach
contents (e.g. Jarman et al., 2002). Recent advances in
extraction and amplification of host nuclear and
mitochondrial DNA from scat samples permits PCR-based
studies using genetic markers to determine species, sex and
individual identity (Wasser et al., 2004). Although faecal
DNA tends to be more degraded than that obtained by
biopsy, in this study 100% of the faecal samples analysed
yielded sufficient DNA for species determination.
Many cetaceans are at-risk or poorly studied, and
researchers require physiological and biomedical data to
assess population health and reproductive status. Such
information is not easily obtained using conventional
methods. Enhanced sampling of cetacean scat by using
detection dogs, coupled with endocrine, toxicological and
molecular analyses, opens a new window into the
physiology, health and genetic status of free-swimming
whales that can contribute greatly to their conservation and
management.
ACKNOWLEDGEMENTS
The authors are very grateful to Phillip Clapham for his
early support for this work. We are indebted to the New
England Aquarium Right Whale Research team, Laurie
Murison and many other researchers and assistants who
collected samples in the Bay of Fundy for these studies.
Special thanks to Kerry Lagueux for assistance with
graphics. Fisheries and Oceans Canada granted permits to
approach and photograph right whales in the Bay of Fundy.
Our thanks to the North Atlantic Right Whale Consortium
for access to the North Atlantic Right Whale Catalogue and
database and to Todd O’Hara and Nick Gales for helpful
comments on the manuscript. This work was supported by
contracts (to RMR) from NOAA Fisheries.
REFERENCES
Doucette, G.J., Cembella, A.D., Martin, J.L., Michaud, J., Cole, T.V.N.
and Rolland, R.M. 2006. PSP toxins in North Atlantic right whales
(Eubalaena glacialis) and their zooplankton prey in the Bay of
Fundy, Canada. Mar. Ecol. Prog. Ser. 306:303-13.
Hamilton, P.K. and Martin, S.M. 1999. A Catalog of Identified Right
Whales from the Western North Atlantic: 1935-1997. New England
Aquarium, Boston. 27pp. [382 plates].
124 ROLLAND
et al.:
FAECAL SAMPLING USING DETECTION DOGS
Hamilton, P.K. and Marx, M.K. 2005. Skin lesions on North Atlantic
right whales: categories, prevalence and change in occurrence in the
1990s. Dis. Aquat. Org. 68:71-82.
Hughes-Hanks, J.M., Rickard, L.G., Panuska, C., Saucier, J.R., O’Hara,
T.M., Dehn, L. and Rolland, R.M. 2005. Prevalence of
Cryptosporidium spp. and Giardia spp. in five marine mammal
species. J. Parasitol. 91(5):1225-8.
Hunt, K.E., Rolland, R.M., Kraus, S.D. and Wasser, S.K. 2006. Analysis
of fecal glucocorticoids in the North Atlantic right whale (Eubalaena
glacialis). Gen. Comp. Endocrinol. 148:260-72.
International Whaling Commission. 2001a. Report of the Workshop on
Status and Trends of Western North Atlantic Right Whales. J.
Cetacean Res. Manage. (special issue) 2:61-87.
International Whaling Commission. 2001b. Report of the Workshop on
the Comprehensive Assessment of Right Whales: A worldwide
comparison. J. Cetacean Res. Manage. (special issue) 2:1-60.
Jarman, S.N., Gales, N.J., Tierney, M., Gill, P.C. and Elliot, N.G. 2002.
A DNA-based method for indentification of krill species and its
application to analysing the diet of marine vertebrate predators. Mol.
Ecol. 11:2679-90.
Kraus, S.D., Moore, K.E., Price, C.A., Crone, M.J., Watkins, W.A.,
Winn, H.E. and Prescott, J.H. 1986. The use of photographs to
identify individual North Atlantic right whales (Eubalaena glacialis).
Rep. int. Whal. Commn (special issue) 10:145-51.
Kraus, S.D., Hamilton, P.K., Kenney, R.D., Knowlton, A.R. and Slay,
C.K. 2001. Reproductive parameters of the North Atlantic right
whale. J. Cetacean Res. Manage. (special issue) 2:231-6.
Lefebvre, K.A., Bargu, S., Kieckhefer, T. and Silver, M.W. 2002. From
sanddabs to blue whales: the pervasiveness of domoic acid. Toxicon
40:971-7.
Malik, S., Brown, M.W., Kraus, S.D. and White, B.N. 2000. Analysis of
mitochondrial DNA diversity within and between North and South
Atlantic right whales. Mar. Mammal Sci. 16(3):545-58.
Murison, L.D. and Gaskin, D.E. 1989. The distribution of right whales
and zooplankton in the Bay of Fundy, Canada. Can. J. Zool.
67(6):1,411-20.
Parsons, K.M., Durban, J.W. and Claridge, D.E. 2003. Comparing two
alternative methods for sampling small cetaceans for molecular
analysis. Mar. Mammal Sci. 19(1):224-31.
Pettis, H.M., Rolland, R.M., Hamilton, P.K., Brault, S., Knowlton, A.R.
and Kraus, S.D. 2004. Visual health assessment of North Atlantic
right whales (Eubalaena glacialis) using photographs. Can. J. Zool.
82(1):8-19.
Rastogi, T., Brown, M.W., McLeod, B.A., Frasier, T.R., Grenier, R.,
Cumbaa, S.L., Nadarajah, J. and White, B.N. 2004. Genetic analysis
of 16th century whale bones prompts a revision of the impact of
Basque whaling on right and bowhead whales in the western North
Atlantic. Can. J. Zool. 82:1647-54.
Rolland, R.M., Hunt, K.E., Kraus, S.D. and Wasser, S.K. 2005.
Assessing reproductive status of right whales (Eubalaena glacialis)
using fecal hormone metabolites. Gen. Comp. Endocrinol.
142(3):308-17.
Smith, D.A., Ralls, K., Hurt, A., Adams, B., Parker, M., Davenport, B.,
Smith, M.C. and Maldonado, J.E. 2003. Detection and accuracy rates
of dogs trained to find scats of San Joaquin kit foxes (Vulpes macrotis
mutica). Anim. Conserv. 6:339-46.
Smith, S.C. and Whitehead, H. 2000. The diet of the Galapagos sperm
whale Physeter macrocephalus as indicated by fecal sample analysis.
Mar. Mammal Sci. 16(2):315-25.
Wasser, S.K., Davenport, B., Ramage, E.R., Hunt, K.E., Parker, M.,
Clarke, C. and Stenhouse, G. 2004. Scat detection dogs in wildlife
research and management: application to grizzly and black bears in
the Yellowhead Ecosystem, Alberta, Canada. Can. J. Zool. 82:475-
92.
Weisbrod, A.V., Shea, D., Moore, M.J. and Stegeman, J.J. 2000.
Organochlorine exposure and bioaccumulation in the endangered
northwest Atlantic right whale (Eubalaena glacialis) population.
Environ. Toxicol. Chem. 19(3):654-66.
Date received: November 2005
Date accepted: April 2006
J. CETACEAN RES. MANAGE. 8(2):121–125, 2006
125