Caja PDF

Comparta fácilmente sus documentos PDF con sus contactos, la web y las redes sociales.

Compartir un archivo PDF Gestor de archivos Caja de instrumento Buscar PDF Ayuda Contáctenos



De Angelo et al 2010 JWM Track identification.pdf


Vista previa del archivo PDF de-angelo-et-al-2010-jwm-track-identification.pdf

Página 1 2 3 4 5 6 7 8 9 10 11 12 13

Vista previa de texto


STUDY AREA
Between 2004 and 2008, we obtained track records from
jaguars, pumas, and maned wolves kept in captivity. Our
task was carried out in collaboration with 18 zoos from 6
countries: Zoo Bata´n (Bata´n, Argentina), Zoologico Bosque
Guaranı´ (Foz do Iguac¸u, Brazil), Refugio Biologico Bela
Vista (Foz do Iguac¸u, Brazil), Zoolo´gico Itaipu´ Paraguay
(Hernandarias, Paraguay), Parque Ecolo´gico Urbano de Rı´o
Cuarto (Rı´o Cuarto, Argentina), Zoolo´gico de la Ciudad de
Buenos Aires (Buenos Aires, Argentina), Temaiken (Buenos Aires, Argentina), Zoolo´gico de Roque Sa´enz Pen˜a
(Roque Sa´enz Pen˜a, Argentina), Zoolo´gico Tatu´ Carreta
(La Cumbre, Argentina), Jardı´n Zoolo´gico de la Ciudad de
Co´rdoba (Co´rdoba, Argentina), Zoolo´gico Santa Fe de
Medellı´n (Medellı´n, Colombia), Parque Zoolo´gico Caricuao (Caracas, Venezuela), Parque Zoolo´gico Las Delicias
(Maracay, Venezuela), Lincoln Park Zoo (Chicago, IL,
USA), Sedgwick County Zoo (Wichita, KS, USA),
Caldwell Zoo (Tyler, TX, USA), Philadelphia Zoo
(Philadelphia, PA, USA), and Little Rock Zoo (Little
Rock, AR, USA).

METHODS
Sample Collection and Processing
We collected tracks from 28 jaguars, 29 pumas, and 8 maned
wolves (mainly ad of both sexes for all the species, although
we also included 2 tracks of a juv jaguar and 3 tracks of 2 juv
pumas of around 1.5 yr old). We used the same methodology
to obtain large dog tracks from urban areas (35 individuals,
most of them stray dogs). Authors and zoo personnel
recorded tracks by following general recommendations (see
Wemmer et al. 1996, De Angelo et al. 2008) instead of
standardized methodology, because our main objective was to
evaluate identification methods for tracks collected in the
field by different people, under diverse circumstances, and
using different protocols. For the same reason, we used tracks
1142

for as many individuals and zoos as possible, looking for all
kind of tracks (front, rear, right, and left feet), diverse
substrates, and different animal size and behavior (to favor
including track variability caused by animal pace and wt). We
collected all tracks from free-walking animals (i.e., the paw
was not pressed into the substrate by handlers). We obtained
all dog tracks and 78% of tracks from zoos from unprepared
substrates, whereas remaining zoo tracks were from sleekly
prepared surfaces on which animals walked freely to yield
better prints (mainly in zoos lacking an adequate surface for
printing tracks). Substrate conditions varied among zoos and
among the sites where we collected dog tracks, including
mud, dust, and sand (wet or dry), for all groups considered.
We tried to collect as many tracks as possible from each
individual. However, because we obtained few tracks in most
of the zoos, we tried to select 1 track/foot from each
individual, in order to have all individuals represented by the
same number of tracks. We used 3 methods for recording
tracks: plaster molds, digital photos with a metric rule as size
reference, and track contour in acetate paper. Multiplicity of
zoos, individuals, collectors, and methods generated a variety
of tracks of different qualities, from which we only discarded
for further analysis those tracks that had conspicuous
deformations precluding an accurate measurement.
To give digital format to collected data, we photographed
the plaster molds with digital cameras and scanned track
tracings, including a metric rule. Then, the first author
digitalized all tracks using the spline tool in AutoCADE
2004 (AutoDesk, Inc., Fresno, CA). For digitalization and
measurement, we scaled tracks according to the sizereference metric rule. Subsequently, we rotated and placed
tracks over a base line defined by a tangent line between the
outer heel lobes (Smallwood and Fitzhugh 1989; Fig. 1).
We analyzed 167 tracks, including plaster molds (n 5 98),
photographs (n 5 57), and track tracings (n 5 12) from the
3 groups of interest. We randomly separated 35 tracks out of
the total (10 jaguars, 10 pumas, and 15 big canids) to be
used only for independent validation for discriminant
models and identification keys (independent cases). We
used all other 132 tracks in traditional method validation
and for multivariate model and identification key development (46 tracks from jaguars, 46 from pumas, 33 from dogs,
and 7 from maned wolves).
M

(Smallwood and Fitzhugh 1989, Childs 1998, Shaw et al.
2007). For this reason, many authors have described
qualitative traits to differentiate puma from dog tracks
(Smallwood and Fitzhugh 1989, Shaw et al. 2007; Appendix
A), and to distinguish jaguar from puma tracks (Schaller and
Crawshaw 1980, Aranda 1994, Childs 1998; Appendix A).
Nevertheless, due to the subjectivity and exceptions found in
application of qualitative traits, some authors proposed
quantitative identification measures (Belden 1978, Smallwood and Fitzhugh 1989, Aranda 1994, Childs 1998, Shaw
et al. 2007; Appendix B). However, these quantitative
criteria were created and evaluated based on either few
known individuals or unknown field tracks classified by
qualitative traits.
In view of these shortcomings, we used a diverse set of
tracks from definitely known origin with the aim of 1)
evaluating performance of qualitative and quantitative
methodologies described to differentiate tracks of big canids,
pumas, and jaguars; and 2) developing accurate and robust
quantitative multivariate criteria to classify tracks of these
groups and species.

Evaluation of Traditional Identification Methods
To evaluate accuracy of traditional qualitative track
identification we made a classification exercise with 67
participants from Argentina, Brazil, and Paraguay in a
jaguar conservation meeting (Third Workshop of Collaborative Effort for Jaguar Monitoring in Atlantic Forest, 5–6
Oct 2005, Puerto Iguazu´, Misiones, Argentina). Although
participants considered themselves capable of identifying
tracks of these carnivores (jaguars, pumas, and big canids),
we started the exercise by showing examples of tracks from
each group and by reviewing traditional qualitative differentiation characters (Appendix A). From our track database,
we randomly selected 10 photographs of tracks from each
group and then asked volunteers to classify tracks into these
The Journal of Wildlife Management N 74(5)