|UFDC Home||myUFDC Home | Help|
This item has the following downloads:
POPULATION ECOLOGY OF THE VICUrjA (Vlcugna vlcugna) AT THE SALINAS
Y AGUADA BLANCA NATIONAL RESERVE, AREQUIPA, PERU: BASELINE
DATA FOR SUSTAINABLE MANAGEMENT
JENNIFER E. DAVIES
A THESIS PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REOUIREMENTS FOR THE DEGREE OF
MASTER OF ARTS
UNIVERSITY OF FLORIDA
Jennifer E. Davies
I would like to thank Dr. Lauren Chapman for all the knowledge and guidance she
brought in the completion of this project. Her dedication to science, kindness, and
integrity are truly inspiring. She definitely sets the standard in achievement both
personally and professionally. I would also like to thank Dr. Charles Wood for all his
help and support during my time in Florida, and for agreeing to serve on my committee.
I would also like to thank Dr. Emilio Bruna for his support, encouragement, and
knowledge, and for graciously agreeing to serve on my committee.
In addition, I would like to thank all those at CONATURA and Dr. Catherine
Sahley for providing assistance at the field site in Arequipa, use of their database, and for
various crucial aspects of the project. The field research may not have been possible
without the invaluable friendship and support of Vajk Lucas and Mariela Caceras land
Maya) in Arequipa. They shared their home to an unexpectedly long-term visitor, who
would come home covered in dust and smelling of alpacas. I am honored to have been a
part of their lives. Comments and knowledge offered by Colin Chapman and Ron Same
were also invaluable and added greatly to the thesis. I also greatly appreciate the support
offered by those in the Tropical Conservation and Development program throughout my
years of study, namely Dr. Marianne Schmink and Hannah Covert. Technical support
land overall support) was provided by Joe Savastano. Financial support for this research
was provided by the Department of Latin American Studies, the Tropical Conservation
and Development program at the University of Florida, the Wildlife Conservation
Society, and CONATURA. Finally, I would like to thank my family and friends for their
encouragement and patience during the completion of this project.
TABLE OF CONTENTS
ACKNOWLEDGMENTS ........................................ iv
LIST OF TABLES ................... ................... ................... .........
LIST OF FIGURES ........................................ ix
1 INTRODUCTION ........._____ ........._____ ........._____ ..........
General Ecology ........................................
2 STUDY SITE DESCRIPTION.............................
3 METHODS .........___ .........___ .........___ ..........
Data Analyses ................... ................... ................... ..........
Group Characteristics ................... ................... ................... ..........
Distribution and Density.................................
Density and distribution relative to the laguna. ................... ................... ......13
Group size distribution relative to the laguna ................... ................... ........13
Responses to Changes in Water Distribution ................... ................... ................14
Spatial Patterns and Inter-Group Movement ............... ................... .................15
4 RESULTS .........___ .........___ .........___ ..........
General Ecology ................... ................... ................... ..........
Group Size and Population Composition ................... ................... ................... ...16
Group Size and Composition by Group Type ............. ............. ...............17
Spatial Relationships and Intergroup Movement ................... ................... .................18
Density and Distribution............................
Density and distribution relative to the laguna. ................... ................... ......18
Group size distribution relative to the laguna ................... ................... ........19
Group composition relative to distance from laguna ................... ..........
Responses to Changes in Water Distribution ................... ................... ................24
Average distance from laguna as a function of changes in water
distribution ................... ................... ................... ...........
Density and distribution as a function of changes in water distribution ......29
Spatial Patterns and Inter-Group Movement ................... ................... .................29
General Ecology ........................................
Group Size and Type Composition ........................................
Mating Strategies of the Vicui~a ................... ................... ................... .......
Group Composition Within Family Groups ................... ................... .................. 39
Group Size and Density.................................
Spatial Relationships and Intergroup Movement ........................................
Distribution Patterns ................... ................... ................... ..........
Seasonal Changes in Distribution and Density ........................................
Spatial Patterns and Inter-Group Movement ................... ................... .................49
Mean age of identified males found at laguna in 2002 ................. ...............50
Site fidelity and dispersal ........................................
Distance moved from family groups .............. .............. ............. ...52
Description of shifts in group composition ................... ............... ............52
6 IMPLICATIONS FOR CONSERVATION ........................................
7 CONCLUSIONS ........................................
LIST OF REFERENCES .........____ .........____ .........____ .........
BIOGRAPHICAL SKETCH ........................................
LIST OF TABLES
4-1.Composition of identified vicui~a (Vlcugna vlcugna)................................
4-2.Identified male groups ofvicui~a (Vlcugna vlcugna) ........................................
4-3.Life history characteristics of tagged male vicui~a (Vlcugna vlcugna) .......................36
4-4. Examples of movement within female vicui~a among family groups. ................. .....36
5-1.Published denisty estimates of vicui~a (Vlcugna vlcugna) ................. ................. ......43
LIST OF FIGURES
2-1.Transect design for vicui~a (Vlcugna vlcugna) survey. ........................................
4-1.Percentage of group types in the vicui~a (Vlcugna vlcugna) population ................... ..17
4-2.Vicui~a (Vlcugna vlcugna) group density ........................................
4-3.Distribution of vicui~a (Vlcugna vlcugna) group types around the laguna ..................2 1
4-4. Size of vicui~a (Vlcugna vlcugna) groups in relation to the laguna ................... ..........23
4-5.The number of females in family groups of vicui~a (Vlcugna vlcugna) ................... ...25
4-6. The number of young vicui~a (Vlcugna vlcugna) in relation to distance. ................... .26
4-7. The number of males in vicui~a (Vlcugna vlcugna) groups ................... ................... ...27
4-8. Average distance of vicui~a (Vlcugna vlcugna) groups ................... ................... .........28
4-9.Distribution ofvicui~a (Vlcugna vlcugna) groups ........................................
4-1O.Density (groups per sq km) of vicui~a (Vlcugna vlcugna) ................... ................... ...31
4-11.Percentage of individual vicui~a (Vlcugna vlcugna) group types ................... ...........32
4-12.Spatial patterns of followed vicui~a (Vlcugna vlcugna) groups.................................3
6-1.Historic range of the vicui~a (Vlcugna vlcugna) in South America.............................55
Abstract of Thesis Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Master of Arts
POPULATION ECOLOGY OF THE VICUrjA (Vlcugna vlcugna) IN THE SALINAS Y
AGUADA BLANCA NATIONAL RESERVE, AREQUIPA, PERU: BASELINE DATA
FOR SUSTAINABLE MANAGEMENT
Jennifer E. Davies
Chair: Dr. Lauren J. Chapman
Major Department: Latin American Studies
The installation of corrals to facilitate the capture and protection of the vicui~a
(Vlcugna vlcugna) has been encouraged by government agencies in Peru to meet
sustainable development objectives in rural Andean communities. Corrals may have an
impact on vicui~a populations and their habitats by altering the density of the vicui~a.
However, little is known about the home range, movement patterns, or habitat
requirements of free-ranging vicui~a, which precludes an accurate estimate of the size a
corral should be to sustainably manage these wild camelids. The objective of the thesis is
to provide quantative data on aspects of the population ecology of the vicui~a that will
have application to corral management initiatives. Repeat samples of a 38-km transect
and group follows were used to estimate four major ecological traits of vicui~a groups in
the Salinas y Aguada Blanca National Reserve in Arequipa, Peru. These included 1)
group characteristics, 2) movement patterns, 3) responses to declining water availability
in the major laguna for vicui~a in the reserve, and 4) spatial patterns and inter-group
movement. Analyses of these four ecological traits were used to derive patterns of
habitat use, the relationship to water resources, and density variation across space and
time. Family groups made up 67% of all groups within the population, had an average
group size of 5.6 individuals, and had a group composition of 1 male, 3.7 females, and
1.6 young. Overall density of vicui~a in the study area was 3.7 vicui~a km2. Water
distribution and availability had a major impact on vicui~a movement, but no detectable
effect on group and population composition. Group density was highest < 4 km from the
laguna, suggesting that the water source was a key feature in the vicui~a's habitat.
Movement patterns indicated differential habitat use among group types around the
laguna. Large family groups and bachelor herds were more frequently found in
proximity to the laguna, while smaller groups (non-defined groups and solitary males)
were distributed more evenly. Family groups showed a low degree of spatial overlap
around the laguna suggesting that the family groups in the reserve are territorial and
maintain their boundaries while at the laguna.
The vicui~a (Vlcugna vlcugna) is a New World camelid that inhabits the high, dry
puna, or altiplano zones in the Andean mountains of Peru, Chile, Argentina, and Bolivia
between 3,700 4,900 m (Koford, 1957; Franklin, 1983; Vila, 1992). This species is one
of three wild ungulates occurring in these arid habitats. Others include the guanaco
(Lama guanlcoe) and the guemal (H~ppocamelus antlslensls). It is widely reported that
the vicui~a is a dietary specialist and therefore restricted to these elevations (Franklin,
1978; Lucherini, 1996). However, some authors suggest that since the arrival of humans
and their livestock during the Incan period, the vicui~a has been displaced to higher
elevations (Hofmann et al., 1983).
The vicui~a is sedentary and spends most of its time foraging and ruminating on the
hard bunch grasses of the puna (Bosch and Svendsen, 1987; Franklin, 1983; Lucherini,
1996; Menard, 1982; Renaudeau d'Arc et al., 2000; Vila and Cassini, 1993). Habitat use
appears restricted, as vicui~as have been found to avoid open rocky areas and only use
areas that support preferred forage (Franklin, 1983; Renaudeau d'Arc et al., 2000). This
vegetation preferred by vicui~as includes species often associated with the scarce ground
water supply found in the arid puna environment. The vicui~a requires a frequent supply
of water not obtained by either oxidative metabolism or from the food they ingest (Bosch
and Svendsen, 1987; Franklin, 1978, 1983; Menard, 1982; Vila and Roig, 1992; Vila and
Cassini, 1993). Vicui~a have been reported to exhibit daily foraging and drinking patterns
(Franklin, 1983; Renaudeau d'Arc et. al., 2000; Vila and Cassini, 1993) in response to the
drastic fluctuations in temperature, low precipitation, and medium to poor forage quality.
In their study of seasonal activity patterns Vila and Cassini (1993) found a daily peak in
the use of water sources at noon. This pattern in activity was more defined during the dry
season (Vila and Cassini, 1993), suggesting a strong relationship to the distribution of
water. However, a fine-grained analysis on water requirements and effects of water
distribution on vicui~a spatial patterns has not been undertaken.
The social organization of the vicui~a is based on family groups and male groups
(Franklin, 1983; Vila and Cassini, 1993). Several studies propose that male vicui~as in
family groups are territorial (Bosch and Svendsen, 1987; Franklin 1978, 1983; Koford,
1957; Vila and Roig, 1992), and there is evidence for variation in group type and size
with general habitat quality. However, direct links between group type and water
resources are not understood well.
A more complete picture of vicui~a population ecology is critical given increasing
pressures on their populations, and new intensive game management initiatives. Vicui~a
fiber is well suited for the drastic extremes of the Andean climate and is considered to be
the finest in the world. The fiber is 13-14 microns in diameter, which is finer than
cashmere (15-19 microns in diameter). This makes vicui~a a target species for rural
development strategies, because the fiber is viewed as such a lucrative, renewable natural
resource. In fact, vicui~a fiber has been hailed as the "gold of the Andes," a viable way to
stimulate the economies of impoverished indigenous communities in the area (Cattan and
Glade, 1989; Rabinovich et al., 1985; Sahley et al., 2001).
The main threat to vicui~a populations is the poaching for hides and their fiber
(Lichtenstein et al., 2002). This contributed to their rapid decline to an estimated 10,000
individual vicuna in the 1960's (CONACS, 1997). After being listed under Appendix I
of the CITES convention since the 1970's, the vicui~a was reclassified to the Appendix II
species list in 1995 (Wheeler and Hoces, 1997). Recent increases in the number of
vicui~a are assumed to reflect protectionist practices centered on creation of national
parks and sustainable use of wild vicui~a populations to harvest fiber (Cajal, 1991;
CONACS, 1997). Three parks were established in Peru to facilitate protection of wild
camelids: 1) Salinas y Aguada Blanca National Reserve (approximately 366,000 ha),
located just north of Arequipa, 2) Huascaran National park in Ancash (28,000 ha) and 3)
Pampas Galeras National Reserve (75,000 ha). "A vicui~a sheared is a vicui~a saved"
became the slogan of stakeholders to encourage sustainable use of the vicui~a.
Stakeholders claim that conservation objectives can be met with the protection of vicui~a
by park guards from poachers, and the sale of sheared vicui~a fiber as an economic
alternative through community initiatives. These measures have indeed been successful
in bolstering vicui~a populations in a limited number of communities such as Pampa
Galeras, but do not reflect an increase in distribution along the vicui~a's entire range in
In addition to the development of protected areas, new management schemes that
intensively manage dense vicui~a herds in structured corrals have been initiated. The
corrals facilitate capture and protection of vicui~a; and their utilization has been both
encouraged by the government and adopted by numerous rural communities. These
corrals are typically 500-1,000 ha in size and may contain 250-1,000 vicui~as
(Lichtenstein et al., 2002). The installation of corrals may have a detrimental impact on
the vicui~a populations and the xeric habitat they inhabit by anthropogenically altering
their natural density.
Corrals that limit movement may have an impact on the ungulate population by:
minimizing genetic dispersal, increasing the rate and incidence of disease, promoting the
breakdown of family units, decreasing breeding or infant survival, and/or altering plant
communities. However, we understand very little about the home range size, movement
patterns, or territory requirements of the vicui~a. This precludes an accurate estimate of
the size a corral should be to sustainably manage these wild camelids. Such data on
habitat requirements are crucial to the establishment of effective management programs
(Hixon, 1980; Renaudeau d'Arc et al., 2000).
The goal of this study was to quantify key features of the population ecology of the
vicui~a and to consider implications of these characteristics for corral initiatives. Field
work was conducted in the Central Andes within the Salinas y Aguada Blanca National
Reserve (Arequipa, Peru) from April 25'" to July 12" of 2002. I assessed group
composition, density, and movement patterns of the vicui~a, as well as the aggregative
relationship to the distribution of water. My specific objectives were as follows. First, I
quantified group composition characteristics to provide baseline data for comparison with
~ee-ranging and corralled vicuna. These included: representation of age/sex classes
within groups, sex ratio, and group size. Second, I used a repeat transect design to
quantify general movement patterns and spatial distribution in a 20 km2 area that
included representative habitat types and a major water source (a laguna). Data on spatial
relationships were combined with group composition data to analyze movement patterns
for group types relative to the laguna. Third, I explored the response to changes in water
distribution by evaluating relationships between spatial distribution patterns and changes
in the size of the major water source tinder of dryness). Finally, I examined movement
patterns of individually identified vicuna groups and quantified aspects of group stability.
This permitted an evaluation of site fidelity among tagged males around the laguna,
distance traveled over time, and group stability.
STUDY SITE DESCRIPTION
The study was conducted between April 25 and July 12 of 2002 within the Salinas
y Aguada Blanca National Reserve in the department of Arequipa, Peru (15"45'05" and
16"19'15 south and 70"51'20" and 71"34'0" west). The reserve was created in 1979 and
contains 366,936 ha of dry puna characterized by high mountain plains, volcanoes,
hillsides, and cliffs, with elevation between 3,400 to over 6,000 m. The daily
temperature in the reserve averages between 2 to 8" C (absolute minima reaching 18"
C), and dramatic diel temperature fluctuations are characteristic (Caviedes and Knapp,
1995; Pulgar, 1996).
Six Holdridge life zones are represented in the reserve. These include: 1)
subtropical desert mountains, 2) subtropical sub alpine mattoral desert, 3) subtropical
sub alpine humid paramo, 4) subtropical alpine humid tundra, 5) subtropical alpine very
humid tundra, and 6) subtropical nival. Also characteristic of this region is the presence
of snow at the highest elevations (typically the volcanoes) throughout the year. This is an
important source of water for both the vegetation and wildlife. Total annual precipitation
in this area fluctuates between 200 mm and 1,000 mm indicating a high degree of
interannual variability in water availability. The April to July period of this study
covered the end of the wet season, an extremely dry period, and an unexpected wet
period associated with a July snowfall.
Plants found in the study site display adaptations to dry puna, such as slow growth
rates, small plant size, a bunched distribution, and a shallow root system. The common
plants in the reserve include bunched grasses such as the common "ichu" grass (Stlpa
Ichu), several kinds of cactus such as "ulluyma" (OpuntlaJloccosa), the "huajoro"
(Opuntla lagopusl, and the "pajuro" (Opuntlna ignescens). This region also supports
several species of scrub brush. These include the arbustos enanos such as the "shauli",
and the arbustos de Culli, (Colli, quei~ua, and quinal) (Polylepls Racemosa) a tree-like
plant that is heavily exploited for wood.
The puma (Puma concolorl, vicui~a (Vlcugna vlcugna), and guanaco (Lama
guanlcoe) are the only large non-domesticated mammals in Salinas y Aguada Blanca
National Reserve. However, no signs of puma were found in the focal study area. The
only mammalian predator assumed to be in the reserve is the Andean fox jDuslcyon
culpaeusl. In addition to the large wild mammals found in this region of the reserve,
domestic animals inhabit regions near as well as in the reserve. Alpaca, sheep, goats, and
horses are all competitors of the wild species.
The domestic livestock are kept spatially separated from the vicui~a on the
mojedales (wetlands). In other areas of the vicui~a's native range, these wetland areas are
considered important habitat for wildlife, including the vicui~a(Franklin, 1983;
Renaudeau d'Arc et. al., 2000; Rundel and Palma, 2000). However, in this reserve, only
the domestic animals have access to these water and vegetative resources. The major
water source for the vicui~a in the reserve was the laguna. No other water source was
found in the reserve that was used by vicui~a (Davies, pers. obs.).
Due to its location within the reserve, land use by the community of Tambo
Cai~ahuas also has a significant impact on wildlife populations. An important feature in
this landscape is the road that runs through both the community and the reserve (Figure
2-1) Tlus is a malor thoroughfare between the cities of Areqinpa (the second largest city
mn the country) and Juhlaca, a city with sigmuficant commerce There are also several
tounst attractions and active nunes m tlus area that attract many large trucks and
additional traffic As it crosses through the resenre adjacent to the malor water source,
tlus road potentially contnbutes pomnt-source polhition and road-loll mortahties that could
have long lastmng, detnmental impacts on the wlldhfe populations
iM i---- Sumbay
S / -- Transect
Chachanl Volcano -- Paved Highway
a 0 aa ilometers
Figure2-1 Transect design for vacufia (Vlcugna vicugna) sunrey at the Salmas y Aguada
Blanca National Reserve, Peru, durmg Apnl-July 2002 Patterns of land use
are mdicated as well as the mamn roads that pass through the reserve
Current sustainable management strategies m Tambo Cafialulas mvolve the hive
capture, shear and release (called the Chaccu) of fr-ee-rangmng vcufias approximately
once every 2 yr CONATURA, an Areqinpa-based non-governmental organization
assists tlus communuty with logistical support, traimung, and organization Dunng the
time when these vicui~as are captured and sheared, biologists from CONATURA tag
individuals with cow tags and take measurements on age, sex, condition, and length and
diameter of the fiber. A total of 71 vicui~as were tagged between 1997 and 2000. In
addition, this NGO conducts scientific field studies and has established a monitoring
program to help provide data for management decisions.
Repeat samples of a 38-km transect and group follows were used to estimate four
major ecologicaltraits of vicui~a groups. These included 1) group and population
composition, 2) density and distribution, 3) the response to changes in water distribution,
and 4) spatial patterns and intergroup movement. Analyses of these four ecological traits
were used to quantify spatial and temporal patterns of habitat use, dispersal, relationship
to water resources, and density variation.In addition, a 3-yr CONATURA database was
used to reference individual histories on the vicui~as observed in the field survey. The
database comes from a long-term monitoring project and is comprsied of 15 surveys
between 1999 and 2001. Data collected by CONATURA include location, group size
and composition, and identification of some tagged individuals.
One large 38-km transect was established that covered a representative area of the
reserve including the major water source (the laguna), a large pampa or intermontane
plain, and regions near the highway. This transect was surveyed 13 times over the 4-
month study period. For each group or individual vicui~a sighted, the following
information was recorded: group type (i.e., bachelor herds, family groups, non-defined
groups and solitary males), group composition (# males, # females, # young), the identity
and gender of marked individuals, group activity, time of day, and GPS location recorded
in UTM coordinates with a Garmin 12XL GPS unit (KS, USA). Analyses of movement
patterns, distribution, density, and dispersal were based on obtaining exact positions of
vicui~a family groups, bachelor herds, and solitary individuals. This was done using a
variation of the distance method (Buckland et al., 1993). A Garmin 12x1 GPS (Global
Positioning Unit) in UTM was used from the transect to obtain coordinates of the vicui~a
groups. The center of the group was taken as the location of the group sighted, and the
bearing and distance were also recorded from this point. These data were then
transformed using the equation (d)*Cos(B)= a x and (d)*Sin (8) = ay. a x was then used
with the equation: ax = xl + xz~ and ay = yl + yz, and the new UTM coordinate was
found for the exact position. Specific information on marked individuals and behavioral
information were obtained with a Bushnell Spacemaster spotting scope. To determine
family group composition and to identify marked individuals, I also recorded tag
numbers of marked individuals (sexes can be differentiated because males have tags on
left, while females have it on the right). Differentiation of yearlings and adult vicui~a is
difficult, and this information was not collected. From the transect study, distinct group
types, [family groups, bachelor herds, solitary males, non-defined groups (groups of
unknown composition)] were identified and followed to assess movement patterns among
different social associations.
The study was conducted in two periods. The first sampling period ran from April
15 May 23. During this period the laguna was full, and the data collection focused on
the group census. The second period ran from June 24 July 12, during which time the
laguna and the surrounding plains became increasingly dry until a snowfall on July 10.
The size of the laguna was mapped immediately after the rainy season when the laguna
was full (April 25), and after a long dry period (June 28). The laguna was approximately
50 times larger in April than in June. In the first sampling period, transects were used to
identify groups and document their composition. In the second period, I used the same
transect design, but also followed known family groups to provide data on movement and
Group sizes observed during the wet and dry sampling periods were compared with
a t-test and then combined to produce one transect data set. From the combined transect
data, group size was averaged to obtain mean group size for all group types. To
determine the sex ratio within the vicui~a population at the Salinas y Aguada Blanca
National Reserve, averages of the number of males and females were calculated from the
combined transect data. A one-way analysis of variance (ANOVA) was used to test for a
difference in mean group size among bachelor herds, family groups, and non-defined
groups. The Scheffe apostenon test was used to detect differences in mean group size
between any two group categories. Group size values were log transformed to normalize
the data and stabilize the variance. Regressions were used to test for relationships among
group composition characteristics such as group size, number offemales, and number of
Distribution and Density
Vicui~a density was calculated by counting the number of vicui~a within a sighting
distance of 500 m from the transect, taking into account sampling effort, and then
dividing by the effective area sampled. Effective strip width (500 m) was defined by
measuring the perpendicular distance from the transect to each vicui~a group observation.
A sighting distance was determined from a frequency histogram of the distance a vicui~a
group was observed (Chapman et al., 2000). The frequency of vicui~a sightings were
consistently higher at distances less then 500 m. All vicui~a sightings greater then 500 m
were taken out of the overall density estimate. This sighting distance of 500 m
corresponds with figures used in previous studies that estimates vicui~a density (C.
Sahley, unpublished data). The total numbers of vicui~as in groups selected using this
criterion were averaged to include in the density estimate. To determine the area
effectively sampled, the sighting distance was multiplied by the transect length. Portions
where the transect overlapped were taken out of the measurement. The density of
vicui~as was then calculated as the number of individual vicui~as sighted divided by the
effective area sampled. Group density was simply calculated with the aid of the map
generated with Are View, and calculated as the number of groups within certain distances
~om the laguna, corrected for sampling effort.
Density and distribution relative to the laguna
Are View was used to create a series of maps describing the geographic locations
of vicui~a groups in relation to the laguna in the transect survey. A series of 20 concentric
rings (1 km apart) around the laguna's center were superimposed on the map of the
transect data. These maps were used to calculate the distribution of group size, group
composition, and group type in relation to distance from the laguna. Group density was
calculated as the number of groups within a certain distance from the laguna corrected for
sampling effort. Linear regression was used to detect the relationship between density
(dependent variable) and distance from the laguna (independent variable).
Group size distribution relative to the laguna
Group size was extracted from the transect data and mapped with the aid of Are
View to find patterns of distribution around the laguna. Vicui~a groups were arbitrarily
categorized into three size classes, small groups (1-5), medium groups (6-15), and large
groups (16-40). A chi-square test of independence was used to detect differences
between group sizes close (<4 km) and far (>4 km) from the laguna.
The number of males, females and the number of young in groups were quantified
in relation to their distance from the laguna. Regression analysis was used to detect
relationships between the number of males, females, and young (dependent variable) to
distance from the center of the laguna (independent variable). Are View was also used to
map this data for a visual display of distribution in relation to the laguna.
Responses to Changes in Water Distribution
The study was conducted in two periods (April 15-May 23 and June 24-July 12),
therefore, a t-test was used to detect differences in group size between the wet and dry
sampling periods. The distances of vicui~a groups from the laguna were categorized into
monthly averages to reflect changes in distribution in relation to changing water
distribution. An Are View map of the trasect data was superimposed on a map of the
study area. The distances vicui~a groups were observed from the laguna was recorded for
each month, correcting for differences in sampling effort.
Temporal changes in density and distribution were also calculated by quantifying
variation between wet and dry periods. A series of Are View-generated maps of the
transect data with 1 km concentric circles around the center of the laguna was used to
quantify variation in density and distribution of vicui~a groups relative to the center of the
laguna in the two major sampling periods. A chi-square test for independence was used
to detect differences in the distribution of vicui~a groups and group type around the
laguna between the wet and dry periods.
Spatial Patterns and Inter-Group Movement
To estimate spatial characteristics among family groups, locations of identified
dominant males were taken from the transect data and the group follows as these groups
moved to the laguna to drink. These observations were then mapped with the aid of Are
View to show group movement patterns over time. In addition, the life histories of
tagged males that I observed were extracted from the CONATURA database to estimated
site fidelity and to detect patterns in movement and group characteristics. These
individually identifiable groups were followed based on opportunistic observations to
obtain several geographic positions for 2-3 hr during the morning (sunrise to noon) and in
the afternoon (noon to nightfall). These groups were found and followed within an
approximate 5-km radius from the laguna. Finally, tagged females observed in
indentified family groups were extracted from the CONATURA database to explore the
validity of group stability as assumed in the literature.
Group Size and Population Composition
Family group size (t=-1.295, P=0.197, mean wet=6.07, n=136, SE+/-0.24; mean
dry=6.6, n=79, SE+/-0.31) and the number of females (t=0.108, P=0.914, mean wet=3.7,
n=136, SE+/-0.16; mean dry=3.6, n=79), did not differ between the two major sampling
periods. However, there were more young seen in the first period (mean=0.98, n=191)
than the second period (mean=1.22, n=149, t-test, t=-2. 845, P=0.005). Since young only
comprised 22% of the population, and other characteristics did not differ between
periods, I combined sampling periods to provide an overall evaluation of group
Family groups comprised 67% of the total groups sighted, and bachelor herds and
solitary males represented 12% (6% each) of the sightings. Non-defined groups
(indistinguishable groups ofvicui~a) comprised 21% of the observed group types (Figure
The sex ratio for adult vicui~as was 1.5 females for every 1 male. For family
groups, this ratio increased to 3.7 females for every 1 male. This ratio is based on adults
in identifiable group types, without consideration of age class.
fanily groups bachelor solitary non defined
herds r~Elles groups
Figure 4-1. Percentage of group types in the vicui~a (Vlcugna vlcugna) population of the
Salinas y Aguada Blanca National Reserve, Peru. Family groups were the
groups most frequently found in the survey and were comprised of a single
male and a variable number of females and young. Non-defined groups were
comprised of unidentified individuals with no young. Bachelor herds were
groups made up of several identified males and no females or young. Solitary
males comprised of a single male. Total group sightings = 317.
Group Size and Composition by Group Type
Family group size averaged 5.6 (antilog mean, n=215) and ranged from 2 to 13.
Bachelor herd size was larger than that of family and non-defined groups (ANOVA F=
102.98, p<0.001, Scheffe p<0.001), averaging 15.9 (n=18) and ranging from 8 to 40.
Non-defined groups were classified as indistinguishable groupings ofvicui~a whose
group size averaged 2.9 (n=65) and ranged from 2 to 13.
There were an average of 3.7 females and 1.6 young per family group. Family
group composition thus averaged i. 3.7. 1.6 (male: female: young). Thirty-two percent
of all family groups had four or more females. For observed males, 39.2% were in
family groups; 57.4 % were in bachelor herds, and 3.5 % were solitary.
Both the number of young and the number of females were correlated with group
size(young: r = 0.84, p<0.001, n=215; females: r = 0.93, p<0.001, n=215). The number
of young was also positively correlated with the number of females (r = 0. 60, p<0.001,
n=215). However, this relationship was not as strong as the relationship between the
number of young and group size, suggesting a greater amount of variation in the number
of young per family group. In addition, for males there was only one male in every
family group despite the variation in family group size.
Spatial Relationships and Intergroup Movement
Density and Distribution
Density and distribution relative to the laguna
At the Salinas y Aguada Blanca National Reserve vicui~a density averaged 3.7
vicui~asl km2 over the period of the investigation. Vicui~a density increased closer to the
laguna (r = -0.50, p<0.009; Figure 4-2). Forty-eight percent of all vicui~a groups in a 20
km radius were found within 2 km from the center of the laguna.
For data combined across the 4-month study period, all group types were present at
the laguna. However, habitat use of group types differed between the area close to the
laguna (within a 4-km radius) and areas more distant (4 to 20 km). The proportion of
group types was not independent of distance (12=15.16, p<0.005). Family groups
dominated in the first 4 km from the laguna, comprising 71% of all group types. Non-
defined groups made up 20%, followed by bachelor herds (6%) and solitary males (3%,
Figure 4-3). Four to 20 km from the laguna, family groups made up 57% of all group
types (a 14% decline), while the percentage of non-defined groups increased to 28%.
There were no bachelor herds more then 4 km from the laguna, and solitary males
comprised 15% of the groups 4 to 20 km from the laguna.
Non-defined groups and solitary males were more evenly distributed throughout
the study area than bachelor herds and family groups (X2=9.07, p<0.005). Seventy
percent of all family groups and 90% of all bachelor herds were found within 4-km of the
laguna. Only 56% of all non-defined groups and 27% of solitary males were found in
this distance class. In comparison, 4 to 20 km from the laguna, only 10% of the bachelor
herds were observed, and only 30% of the family groups. While 44% of all non-defined
groups and 73% of all solitary males were observed in farther regions from the laguna.
When translated to group densities, patterns were similar.
Group size distribution relative to the laguna
Vicui~a groups were arbitrarily divided into three size classes: small groups (1-5
individuals), medium groups (6-15 individuals), and large groups (16-40 individuals).
Small groups accounted for 54% of the whole vicui~a study population, while medium
and large groups comprised 44% and 2% of the groups, respectively.
To evaluate group size relative to the laguna, I quantified the frequency of group
size types as a function of distance from the laguna using l-km increments (Figure 4-4).
There was a tendency for larger groups to be more frequent closer to the laguna than in
the drier regions (XZ=51, O l
with 16 or more individuals (large groups which are exclusively bachelor herds) were
found within 4 km from the center of the laguna (n=5). For medium-sized groups (6-15
individuals, mostly family groups), 69% were found within 4 km of the laguna (n=94).
And, 59 % of small groups were found within 4 km from the center of the laguna (n=l 1 5)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Distance from laguna (km)
Figure 4-2. Vicui~a (Vlcugna vlcugna) group density as a function of increasing distance
~om a laguna in the Salinas y Aguada Blanca National Reserve, Peru. Forty-
eight percent of all vicui~a groups were found 2 km or less from the center of
~~ ~~~~ -~
Figure 4-3 Dlstnbutlon ofvlcui~a (Vlcugna vlcugna) group types around the laguna at
the Salmas y Aguada Blanca National Reserve, Peru Data were collected
over 13 samplmg dates between Apnl 25 and July 12, 2002 by observmg
groups from a 38-km transect
To evaluate group size relative to the laguna, I quantified the frequency of group size
types as a function of distance from the laguna using l-km increments (Figure 4-4).
There was a tendency for larger groups to be more frequent closer to the laguna than in
the drier regions (XZ=51, O l
with 16 or more individuals (large groups which are exclusively bachelor herds) were
found within 4 km from the center of the laguna (n=5). For medium-sized groups (6-15
individuals, mostly family groups), 69% were found within 4 km of the laguna (n=94).
And, 59 % of small groups were found within 4 km from the center of the laguna
Family group size as a function of distance from the laguna was also quantified.
Small family groups (1-5 individuals) made up 32% of all family groups, medium family
groups (6-10 individuals) made up 47% of all family groups, and large family groups
(11-15 individuals) represented 21% of all family groups. Larger groups were more
frequently observed closer to the laguna (< 4 km vs. >4 km; XZ = 16.37, p<0.005). All
large family groups were found closer than 4 km from the center of the laguna. For
medium sized groups, 65% were found within 4 km from the laguna, while 58% of small
family groups were found in this distance class.
Group composition relative to distance from laguna
The number of females and the number of young in groups were quantified in
relation to their distance from the laguna. The number of females decreased with
Figure 4-4 Size ofvlcui~a (Vlcugna vlcugna) groups m relation to the laguna Vlcui~a
groups were categorized into three size classes small(l-5), medium (6-15),
and large (16-40) Tnere was a tendency for large groups to be found closer to
the laguna than m the drier regions of the study site Group size data were
collected through the repeat census (number of group slghtmgs=317) of a 38-
km transect between Apnl 25 and July 12, 2002
from the laguna (r = -0.64, p<0.001), as did the number of young (r = -0.70, p<0.001,
Only 19% of all family groups had more then the average number of young. Large
grouping of males (bachelor herds) were almost exclusively found near the laguna
(Figure 4-7), and there was a strong relationship between the size of bachelor herds and
water distribution (r = -0.60, p<0.001).
Responses to Changes in Water Distribution
Average distance from laguna as a function of changes in water distribution
The distances from the laguna that vicui~a groups were observed, were categorized
into monthly averages to reflect changes in distribution in relation to changing water
distribution. In April and May, the average distance of vicui~a from the laguna was 4072
m, (SD+/-1158.19) and 4127 m, (SD+/-4544.65) respectively. However, average
distance declined to 2568 m, (SD~-549.88) in June, which coincided with much lower
water levels in the laguna. In July, the average distance increased to 3996 m, (SD+/-
2119.72) (Figure 6a), possibly due to the unexpected snowfall in July, that increased the
extent of water around the plains and mountains.
This analysis included the farthest regions sampled, which may contain vicui~a
groups using water sources other then the laguna. To control for this potential bias, I
repeated the analysis removing groups more than 12 km from the laguna. The pattern
was similar, although averages were slightly lower (Figure 4-8).
Final data set for a
Distance to Center
I c. 1 O 2K
I No Data
,, ",, ~
Figure 4-5. The number of females in family groups ofvicui~a (V~cugna mcugna) as a
function of distance from the laguna in the Salinas y Aguada Blanca National
Resenre, Peru. Family groups were divided into two categories: those with
less then the average number of females (0-3) and those with more then the
average (4-10). Data points represent obsenrations from repeat sampling of a
38-km transect on 13 occasions between April 25 and July 12, 2002.
Distance to Laguna
I No Data
Figure 4-6 The number of young mcui~a (Vlcugna vlcugna) m relation to distance from
the laguna at Salmas y Aguada Blanca National Reserve, Peru Family groups
that had less than average number of young (0-2) and those that had more then
average (3-10) were compared Data pomts represent observations from
repeat samplmg of a 38-km transect on 13 occasions between April 25 and
July 12, 2002
Number of Males
Distance to Laguna
~-~r O 2K
I No Data
Figure 4-7 The number of males in vlcui~a (Vlcugna vlcugna) groups m relation to
distance from the laguna at Salmas y Aguada Blanca National Reserve, Peru
Because family groups consist of only one male, most groups (67%) had only
one male Vlcui~a groups with more then one male were bachelor herds Data
points represent observations from repeat samplmg of a 38-km transect on 13
occasions between April 25 and July 12, 2002
t~a april M sy Juur July
Q) (b) ,,,,
O I I N
E 3soo -I 1 00
April M ay June July
Figure 4-8. (a) Average distance of vicui~a (Vlcugna vlcugna) groups from the laguna at Salinas y Aguada Blanca National Reserve,
Peru, and (b) average distance of groups to the laguna not including distances greater than 10 km. Average distances were
based on repeat sampling of a 38-km transect on 13 occasions between April and July, 2002.
Density and distribution as a function of changes in water distribution
A higher density of vicui~a groups was observed with delcining distances from the
laguna (Figure 4-9). When the laguna was full, 65% (n=82) of all vicui~a groups were
found within the first 4 km from the center of the laguna. In contrast, when the laguna
was dry, 83% (n= 102) of all vicui~a groups were found within this 4-km radius (Figure
Family groups and bachelor herds seem to have been most affected by decreasing
water availability. The distribution within group types was quantified and divided into
those <4 km and >4 km from the laguna when full and when almost dry (Figure 4-11).
Seventy-five percent of all bachelor herds were found less then 4 km to the laguna when
full, and 100% when the laguna was almost dry. Sixty percent of all family groups were
present <4 km when the laguna was full and 75% were present when almost dry (Figure
4-11). These differences in group distribution were marginally significant (family
groups: XZ= 2.85, 0.1
For non-defined groups and solitary males, the proportion of groups <4 km and >4
km from the laguna was independent of season suggesting a more even distribution that
did not change with season (non-defined: XZ = 0.3, 0.9
when the laguna was full, and 56% and 33%, respectively were found in this distance
class when the laguna was almost dry (Figure 4-11).
Spatial Patterns and Inter-Group Movement
To examine spatial distribution patterns within the population, identified family
groups were followed (Table 4-1,4-2). There was a low degree of spatial overlap among
(a') Wet season*
(b) Dry season
Figure 4-9 (a') Dlstributon of vaicuia (Vicugna vicugna) groups wilth respect to the
laguna in the Sahnas y Aguada Blanca Nallonal Resenre, Peru (a) after the
wet season (Apn1 25 to May 23, 2002) when the laguna was full and (b)
dunng a dner penod (June 25 to July 12, 2002) when the laguna was reduced
to 2% of the wet season area
~h 3 1 ~ Almost dry
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Area of ring
Figure 4-10. Density (groups per sq km) of vicui~a (Vlcugna vlcugna) groups around the
laguna at the Salinas y Aguada Blanca National Reserve when the laguna was
full and when almost dry.
Figure 4-11. Percentage of individual vicui~a (Vlcugna vlcugna) group types found <4
km and 4-20 km from the laguna when it was full (April 25 to May 23, 2002)
and when almost dry (June 25 to July 12, 2002) at the Salinas y Aguada
Blanca National Reserve, Peru.
1 2 3 4 5 6
m >4 km <4 km >4
Laguna almost dry
Table 4-1. Composition of identified vicui~a (Vlcugna vlcugna) family groups in the
Salinas y Aguada Blanca National Reserve, Peru (April to July 2002).
Original identification and tagging were done by CONATURA.
Group composition Tagged individuals in family group
1 male, 7 females, 6 young Blue #140 Orange #18
1 male, 6 females, 4 young Blue/Al #142 Orange/Al #20
n=ll Purple #103
1 male, 5 females, 2 young Orange #6 Orange #9
1 male, 3 female, 2 young Blue #142 Blue #144
n=6 Green #153
1 male, 1 female, Oyoung Orange #4
Table 4-2. Identified male groups of vicui~a (Vlcugna vlcugna) in the Salinas y Aguada
Blanca National Reserve, Peru (April to July 2002). Original identification
and tagging were done by CONATURA.
Group type Group size Identified tagged individuals
Solitary male n= 1 Blue #127
Solitary male n= 1 Blue #146
Bachelor herd variable White #52
that were followed (Figure 4-12) and a high degree of site fidelity on behalf of the
identified male over the duration of the study. However, there was distributional overlap
with other non-identified groups.
The mean age of males from each group type was extracted from the CONATURA
database on the identified males observed during the 2002 field survey. Solitary males
averaged 11.5 yr; while young males (<3 yr) were found in bachelor herds, and males in
family groups averaged 9.5 yr. The latter males made up the effective breeding
population, as access to females is limited to those in a male's defended family group
(Koford, 1957; Vila and Cassini, 1994). The histories of the tagged males followed in
this study were also extracted from the CONATURA database to establish trends in age,
group type, and group size as well as site fidelity to the laguna (Table 4-3). Identified
vicui~as were repeatedly found at the laguna for up to 5 yr, indicating a high degree of site
fidelity to the laguna.
Within the social context of establishing corrals, data on the distances that vicui~as
move is critical. Distances moved by the followed groups were compared between the
morning and evening. In the morning, vicui~as moved an average distance of 713 m hr~'
(SE+/-162. 70), while they only moved 508 m hi' (SE+/- 45.67) in the afternoon.
Tagged females observed in identified family groups were extracted from the
CONATURA database to explore the validity of group stability as assumed in the
literature. Our study indicates that females vary in strategies. Some exhibited a high
degree of mobility among family groups, and others remained with the same male over a
long period. For example, tagged female orange #17 was observed to change family
groups four times within 2 yr (Table 4-4). She was observed with male blue/al #142,
switched to male blue #146, and then returned to male blue/al #142 In the same month,
she switched agam to blue #143 and to a family group with no
Rgure 4-12 Spatial patterns of followed vlcui~a (Vlcugna vlcugna) groups around the
laguna at Salmas y Aguada Blanca National Reserve, Peru A low degree of
spatial overlap was observed between family groups in regions <4 km from
Male # 146
Male # 2
Male # 143
Male # 6
Male # 142
Table4-3. Life history characteristics of tagged male vicui~a (Vlcugna vlcugna) in the
Salinas y Aguada Blanca National Reserve, Peru in 2002. Characteristics
extracted from the CONATURA database.
Individual Year Longevity at # years Estimated Group Group type
tagged laguna observed age size
Blue 127 1997 5 2 12 1 Solitary
Blue 146 1997 5 4 11 1 Soitary
Blue 140 2000 2 2 13 14 Family group
Blue 142 2000 2 2 10 11 Family group
Blue 143 2000 2 2 9 8 Family group
Orange 4 1999 3 2 9 2 Family group
Orange 6 1999 3 4 8 6 Family group
White 52 2000 2 2 2 3 to 28 Bachelor herd
White 54 2000 2 2 2 3 to 28 Bachelor herd
White 55 2000 2 2 2 3 to 28 Bachelor herd
White 58 2000 2 2 2 3 to 28 Bachelor herd
White 59 2000 2 2 2 3 to 28 Bachelor herd
White 60 2000 2 2 2 3 to 28 Bachelor herd
White 61 2000 2 2 2 3 to 28 Bachelor herd
White 64 2000 2 2 2 3 to 28 Bachelor herd
Table4-4. Examples of movement within female vicui~a (Vlcugna vlcugna) among
family groups in Salinas y Aguada Blanca National Reserve, Peru from 1999
to 2002. Female Orange #9 remained with the same group over a 4-yr period.
Female Orange #17 switched among four groups. Dates tagged males and
females observed from 1999-2001 extracted from the CONATURA database.
Female Date Male Total group size
Orange #9 21-Dec-99 Orange #6 2
20-Apr-00 Orange #6 3
22-Jun-00 Orange #6 7
13-Dec-00 Orange #6 6
23-May-02 Orange #6 6
26-Jun-02 Orange #6 8
Orange #17 24-Nov-00 Blue/A1#142 2
13-Dec-00 Blue #146 2
30-Mar-O 1 Blue #146 2
1 i-Apr-02 Blue/A1#142 11
15-Apr-02 Blue #143 7
25-Apr-02 Blue #143 8
24-May-02 No Marks 4
other individuals marked the next month. In contrast, tagged female orange #9 was
consistently observed with the same male (orange #6) over 3 yr.
The vicui~a of Salinas y Aguada Blanca were dominated by family groups that
move daily and seasonally, but show minimal spatial overlap. Water seems to be a
fundamental environmental character influencing movement patterns and seasonal
changes in local density. These characteristics are very important when considering the
potential of current management initiatives. They also provide basic descriptors of the
ecology ofvicui~a and are useful in establishing interdemic patterns ofvariation.
The main objective of the study was to offer baseline ecological data on population
parameters of free-ranging vicui~a, and compare this data with other studies on vicui~a
and ungulates. The population characteristics presented in this study included: 1) group
size and group type composition within the population, 2) density and distribution, 3)
movement patterns in response to water distribution, and 4) spatial patterns and
Group Size and Type Composition
I used four categories to describe basic social units in the vicui~a of Salinas y
Aguada Blanca. These included: family groups, bachelor herds, solitary males, and non-
defined groups. These categories were created based on the different roles males play
within the population. Males in family groups are potentially the only males that are
breeding; thus family groups make up the effective breeding population. In Salinas y
Aguada Blanca, 67% of all vicui~a groups were in family groups, while only 12% were
male groups, indicating that approximately two-thirds of the males in this population
maintain a group of females and are breeding.
These trends in group type composition support an earlier study on the
socioecology of vicui~as. In his long-term study of vicui~as in the Pampa Galeras reserve,
Franklin (1983) described vicui~a population structure in relation to behavior and spatial
distribution. He recognized three different vicui~a group categories: family groups, male
groups, and solo males. In this study, Franklin found that 75% of all males were in
territorial groups while only 24% were in other male groups. Franklin also suggested that
male vicui~a in family groups are territorial and maintain boundary-oriented territories.
The limited overlap in range use documented in my study also suggest that over half of
the males in this vicui~a population may be territorial. These results on group type
composition in vicui~a populations coincide with the general population characteristics of
ungulates: within the social distribution of ungulates, generally 2/3 of the adult males
within a population are territorial (Owen-Smith, 1977).
Mating Strategies of the Vic~a
The life history strategies of ungulates evolved within the context of density
dependence and carrying capacity. As a result, strategies common to most ungulates
include the asymmetry of reproductive success between the sexes, a dimorphism in body
size, and a skewed sex ratio towards females (McCullough, 1999). A skewed sex ratio of
1.5 females: 1 male was found in the vicui~a population at the Salinas y Aguada Blanca
National Reserve. Similarly, the vicui~a in Pampa Galeras exhibited a similarly skewed
sex ratio of 1.2 females to 1 male (Franklin, 1983). However, in both these studies, it is
uncertain how this sex ratio changes with age. In many mammals, the sex ratio will be
skewed towards males at birth, then shift towards females in older age groups (Smith,
Earlier studies report resource-defense polygyny as the mating system of the vicui~a
(Bosch and Svendsen, 1987; Franklin, 1983; Vila, 1992). The structure of the family
group includes one adult dominant male, a number of females, and that year's young.
Our study detected no relationship between the number of males and group size,
providing evidence that the population in Salinas y Aguada Blanca also exist in
Group Composition Within Family Groups
Data collected during the field study suggest that vicui~a populations adopt a rigid
family group structure independent of environmental gradients and varying management
strategies. The mean group composition at the Salinas y Aguada Blanca National
Reserve (1 male, 3.7 females, and 1.6 young) falls within the range reported in earlier
studies. In Abrapampa Argentina, the mean group size was 1 male, 3-4 females, and 2-3
young (Vila, 1995), while a study done by the same author in 1992 in Catamarca,
Argentina reported a mean group composition to be 1 male, 3.6 females, and 1.9 young
Both of these sites have dry and harsh climates, similar to the Salinas y Aguada Blanca
National Reserve. In the Lauca National Park in Las Cuevas, Chile, the mean group
composition was 1 male, 3.1 females, and 1.6 young. In Pampa Galeras, a region with a
high density of vicui~a (67. 9 vicui~alkm2) and with more sources of water then my field
site, the ratio was 1 male, 3 females, and 2 young(Franklin, 1983). The mean group size
among vicui~as in corralled habitat in the humid puna where water is not such a limiting
resource was also 1 male, 3.4 females, and 1.7 young.
Group Size and Density
The average group size of family groups (antilog mean=5.6) in the Salinas y
Aguada Blanca National Reserve is similar to the averages reported from other studies.
In Pampa Galeras, the average group size was 6, (Franklin, 1983), and Vila and Roig
(1992) in the Laguna Blanca National Reserve, Argentina, found the average group size
to be 6 to7. Group size of bachelor herds ranged from 8 to 40 individuals in our study.
The dynamic nature of bachelor herd group size seems to be characteristic of vicui~a as
Franklin (1983) reported a range of 2 to 155 individuals and Vila and Roig, (1992)
reported a range of 2 to 22 individual bachelors.
There was a strong positive relationship between the number of females and the
number of young and total group size. However, the relationship between the number of
young in a group and the number of females is not as strong. This suggests that not every
female is giving birth every year. These females may be newly-dispersed young females,
or older, non-reproducing females. It could potentially be disadvantageous for the herd
as a whole to support too many young simultaneously due to the increasing energy
requirements of lactating females and the need to be alert more frequently (Vila and
Cassini, 1994). Other alternatives to explain this relationship include the mortality of
young and the inability of the female to conceive that year. No published data exist on
fertility among free-ranging vicui~a, however, my data suggest that population growth
rate and harvest models should not only incorporate a time lag, but also not assume one
young per female a year as a parameter (Cattan and Glade, 1989).
In general, there is a positive relationship between group size and an open habitat
structure (i.e., no cover) among ungulates. Large group size among ungulates is believed
to be a strategy to avoid predators and a way to optimize foraging behavior(Kie, 1999).
Increasing the number of females in a family group implies a requirement for increased
vigilance on behalf of the dominant male. However, in an analysis of vicui~a time
budgets, Bosch and Svendsen (1987) reported no difference in the amount of time a male
defends females between large and small family groups. In fact, the more females a male
has in his harem, the more individuals there are to watch for predators/aggressors while
the group feeds; thus increasing the amount of time each individual can spend foraging
instead of being alert. Therefore, under the spatio-optic conditions high Andean
mountain plains afford, the dominant male can increase the size of his family group by
adding females with little to no increase in energetic costs in terms of vigilance. As the
male vicui~a increases the number of individuals in his herd, the collective herd increases
their overall grazing efficiency. In addition, the number of females in a group is a good
predictor of the mating success among male vicui~a(Vila, 1995, Vila and Cassini, 1994).
Thus a large group size may be optimal under these arid and open conditions.
Not all vicui~a, however, occur in large groups. Over half (54%) of the vicui~a
population in the Salinas y Aguada Blanca National Reserve occurred in groups with 1 to
5 individuals. Franklin (1983) suggested that a section of the vicui~a population that
occurred more frequently in marginal habitats were temporary family groups with an
average of 4 individuals (1 male: 3 females: 1 young). These marginal grazing territories
were classified as feeding territories of hard bunch grasses with little to no water sources.
This variation in group size in the vicui~as of Salinas y Aguada Blanca National Reserve
may reflect spatial variation in habitat quality and the dependency on water sources for
large groups. Domestic livestock are given priority for the use of wetlands, leaving few
permanent water sources, such as the laguna, for the vicui~a population. This may
explain the tendency of large groups, especially family groups to be observed more
frequently than small groups closer to the laguna, suggesting differential habitat use.
Density estimates reported for vicui~a populations are highly variable among sites
ranging from 2.3 vicui~alkm2 at San Guillermo in Argentina, to 67.9 vicui~alkm2 at Pampa
Galeras, in Peru (Table 4). In our field study at Salinas y Aguada Blanca, the vicui~a
population occurred at a density of 3.7 individuals/ km2, falling within the range reported
for free-ranging vicui~a. The source of variation among vicui~a population estimates may
reflect density increases in corralled populations, temporal variation due to poaching and
other anthropogenic factors, natural decline, or differences in sampling techniques among
Other managed and non-managed ungulates also exist at very different densities.
Thompson's gazelle (Gazella thomsonl), another species adapted to harsh environments
exists locally at the Ngorongoro crater in Tanzania at 70 individuals/ km2, but more
generally at 14 individuals/ km2(McCullough, 1999). Ungulates inhabit a tremendous
diversity of ecosystems, and therefore face different kinds of environmental constraints.
However, they all demonstrate territorial behavior and exist in polygynous herds for at
least part of the year. Although no direct comparisons can be made to this study, these
data in conjunction with published estimates for polygynous ungulates suggest that there
is a high degree of both interspecific and intraspecific variation in the density of these
In light of current management initiatives in Peru, it is also important to consider
the landscape and long-term population effects of maintaining vicui~a at artificially high
densities. Ecosystem management initiatives have only recently explored some of the
positive and negative feedback systems between herbivorous ungulates and vegetation
(Weisberg et al., 2002). Organisms at very high densities have to deal with limited
resources and decreases in overall population health. It is notable that densities in
corralled vicui~a populations are approximately
Table 5-1. Published denisty estimates ofvicui~a (Vlcugna vlcugna) along its range in
Author Place Year Density
Korford Peruvlan Andes 1957 25 vlcuna km Corraled vlcuna
Brack Pampas Galeras 1980 67.9 vicui~a km~2
Cajal San Guillermo 1991 2.3 vlcuna km Free-ranging vlcuna
Villa Catamarca 1992 5.5 vicui~a km~2
Davies Arequipa 2002 3.7 vicui~a km~2
10 to 27 times greater then estimates from free-ranging populations (Table 5-1). The
long-term effects of such intensive use of puna habitat by vicui~a remains largely
McCullough (1999) states that vicui~a are long-lived, large bodied animals with
relatively low reproductive rates and a high level of female investment in raising young.
He also proposes that most ungulate populations, due to density-dependent responses,
have a tendency to exist at densities close to carrying capacity. Populations close to
carrying capacity are prone to explosions in growth rate and subsequent over-
consumption of resources that may induce a possible population crash (Caughley and
Sinclair, 1994; Smith, 1996).
Spatial Relationships and Intergroup Movement
Understanding mechanisms underlying spatial and temporal patterns of distribution
is important in making both effective conservation and management decisions. Although
the Andean puna landscape appears to be barren and homogenous, different types of
habitat can be identified, and differential use in this habitat has been observed among
vicui~a (Renaudeau d'Arc et al., 2000). The optimality of habitat use is based on the need
to balance the intake of high quality food, growth, reproduction, and the need to avoid
predation, disease, etc.
In the vicui~a population of Salinas y Aguada Blanca, the laguna is an integral
feature in the landscape and seems to have a major influence on movement and spatial
patterns. Areas closer to the laguna may be of higher habitat quality based on water
availability for both vicui~a and vegetation, and on the increased density of vicui~a near
this water source.
Large groups were more frequently located near the laguna. Both the number of
females and young increased as distance to the laguna decreased. However, at Salinas y
Aguada Blanca, large bachelor herds (mean size >19) were also found closer to the
laguna, further supporting the important role of the laguna for this population. Other
studies have reported bachelor herds to move greater distances, as they forage in marginal
lands and avoid lands that territorial males occupy (Franklin, 1983).
This relationship to water in vicui~a re-asserts what is known about the relationship
between ungulates and xeric habitats. In dry environments, water is a major factor
limiting movement patterns and distribution. In our study we assume that habitat quality
increases as distance to laguna decreases. Working from this assumption, it could be
argued that medium and large groupings of vicui~a (n > 6) tend to be found more
frequently in areas close to the laguna, and possibly maintain a low degree of spatial
overlap around the laguna because of the high habitat quality. Although there is more
competition and possibly opportunity for predation, conflict, and invasion, there are also
more vigilant individuals in habitat of higher quality. In addition, less time and energy is
spent traveling to a water source if a territory is maintained nearby. Family groups were
the dominant group type in regions near the laguna; and all large family groups (n > 6)
were found less than 3 km from the center of the laguna.
The analysis of the socioecology of the vicui~a in Pampa Galeras conducted by
Franklin (1983) provides a useful comparison. The habitats are relatively comparable as
both have similar annual precipitation (Pampas Galeras: 281 to 742 mm; Salinas y
Aguada Blanca: 200 to 1,000 mm) and vegetation structure. In his study, Franklin found
that the permanent territorial family group consisted of an average of 6 individuals with a
mean group composition of 1 male, 3 females, and 2 young. These were groups that had
well-established high-quality territories. High quality territories were those that
contained water sources and vegetation associated with puna swampland such as quinsa,
chiula, and mojadales. Smaller family groups with 3-4 individuals were often found in
territories with no source of free water. This poor quality habitat was characterized by
the overbearing presence of roquei~o, piedras de loma and the ichu grass, vegetation that
neither requires nor contains much water.
Within the context of corral installation, the mechanisms behind differential habitat
use among vicui~a groups are important to explore. Increasing density either through the
use of management and husbandry techniques or by eliminating dispersal, despite
population growth, could put greater pressure on these vicui~a groups and the puna
ecosystem. Large groups could potentially be forced to reduce their feeding territories
due to over crowding, especially directly around the laguna. This could in turn, reduce
mean group size. Reducing mean group size may result in increased pressure on the
remaining individuals in the family group to spend more time being vigilant and less time
feeding. This process of overall group size reduction may already be happening in the
Salinas y Aguada Blanca National Reserve as 54% of all vicui~a groups have only 1-5
individuals. These animals are already relegated to habitat patches that appear to be of
poor quality and little to no water. The wetland areas (Bofedales, Mojadales) are
dedicated to supporting the hundreds of alpaca and sheep from the community. Not only
are the vicui~a prohibited access to these water and vegetative sources by the presence of
livestock, but are also deterred by the paved road separating the regions where the vicui~a
are and the wetland areas.
Seasonal Changes in Distribution and Density
The vicui~a must drink water regularly, even daily, to sustain their biological
processes. However, these ungulates have evolved within this context of severe water
limitation, and have adopted certain behaviors to cope with water restriction. Many of
these adaptations may be behaviors that are expressed in movement patterns such as
spatial distribution and group composition in relationship to water. For example, a
consistent pattern of movement from the edges of the mountains in the morning to water
sources in the afternoon has been observed in various habitats across the vicui~a's range
(Vila, 1992; Franklin, 1983; Vila and Cassini, 1993). To further explore this relationship
to water, the size of the laguna was used as a dryness index. Although these two
sampling periods may not represent two different seasons at their extremes (extreme wet
vs. extreme dry), there was a definate decrease in the water available. The perimeter of
the laguna was measured in April and decreased to 2% of its size when it was measured
again at the end of June.
In the Salinas y Aguada Blanca National Reserve, we detected a marked decrease
in the average distance of vicui~a from the center of the laguna between the end of the
rainy season (May) and the onset of the dry season (June). The vicui~a were
approximately 1.6 km closer to the laguna when it is almost dry. However, the transect
included territories that were > 10 km from the laguna. Due to potential habitat
~agmentation, the vicui~a groups on these territories could potentially be using a different
water source. After these observations (distance > 10 km) were taken out of the analysis,
vicui~a groups were found to be approximately 1.8 km closer to the laguna when it was
dry then when it was full. The distance of vicui~a groups from the laguna is almost 2 km
less at the onset of the dry season then at the end of the wet season. This marked
decrease coincided temporally with a decrease in precipitation in the whole region, as
expressed by a decrease in water distribution and water levels at the laguna. The
decrease in the average distance of groups from the laguna was reflected in an increase in
vicui~a density near the laguna in the dry season. A complementary relationship between
vicui~as and water distribution in the wet and dry season was also observed in a similar
habitat, a high, dry Andean grassland of Argentina by Vila and Cassini, (1993). Their
probability of finding individual vicui~as near water remained constant throughout the dry
season, but decreased in the afternoon during the wet season.
In our study, the average distance of vicui~a groups from the center of the laguna
decreased by 729.53 m in July. This was most likely due to the snow fall that occurred
throughout the month of July. Although the snow would melt by the afternoon, it served
to distribute water along the entire pampa and surrounding foothills. Vicui~a cannot
depend solely on the water from snowfall; however, the increased availability of water
gives both the vegetation a boost, and helps vicui~a with their intake of water as it collects
on the plants they graze on. The results of this study suggest that movement behaviors in
response to changing water distribution is a key element for survival in vicui~a.
Larger family groups and bachelor herds occupy the habitat immediately around
the laguna more frequently then smaller groups. I also detected differential habitat use
among vicui~a group types between the wet and dry seasons. How different group types
of vicui~a use the patches of habitat directly around the laguna offers some insight to
adaptive behaviors and population dynamics in free-ranging vicui~a. For example, in
general, over half of the family groups were found <2 km from the center of the laguna,
and 100% of the bachelor herds were found in this distance class as well. However, non-
defined groups (antilog mean -2.9), were observed to be more evenly distributed along
the study area. This suggests that these groups could be occupying dry, marginal habitat.
This decrease in representation of non-defined groups (both during the dry and wet
season) in habitat directly around the laguna further supports Franklin's notion that these
are small temporary family groups occupying marginal areas.
When the laguna was full, family groups were observed to dominate within the first
l-km from the center of the laguna. This was followed by the bachelor herds, the solitary
males, and lastly, the non-defined groups. When the laguna was almost dry, there was an
increase in the number of bachelor herds within this first l-km radius. Although it has
been proposed that bachelor herds move greater distances, there was a definite trend in
movement towards the laguna. Most bachelor herds in this study were considered large
(17.5 individuals) and were observed a distance of 3 km or less from the center of the
laguna. This suggests that group size may limit how far vicui~a groups can travel and
forage from the laguna.
Franklin (1983) also reported seasonal changes in vicui~a density. However, he
attributed these fluctuations to expulsion of young during certain seasons. In Pampa
Galeras, the young from family groups are expelled between October and March,
decreasing family group density. From April to May, there is an observed increase in
density, corresponding to the birthing season. From June to August in Pampa Galeras,
there is another unexplained decrease in density. Although our field study only covers a
fraction of the time data was collected on vicui~a density in Pampa Galeras, it could offer
a potential explanation to the observed decrease in density as vicui~a groups are observed
to limit their movements around water holes.
Spatial Patterns and Inter-Group Movement
Identified family groups were followed to gain insight into movement patterns at
the group and individual level. Earlier studies e.g., (Franklin, 1983) estimated home
range and relative territory size, however, these studies did not follow individually
marked individuals over long periods of time. Our study showed that vicui~a groups
exhibit a low degree of spatial overlap around the laguna.
The first formal study on vicui~as conducted by Koford (1957) in the Peruvian
Andes suggested that male vicui~a were highly territorial and would frequently engage in
tights and chases with intruding males. Territoriality is a common characteristic among
ungulates and leads ultimately to differential mating and reproductive success among
members within a population (Owen-Smith, 1977). In Pampa Galeras, Franklin (1983)
also proposed territoriality as a characteristic of the vicui~a; he observed males from
family herds frequently participating in antagonistic interactions with males from other
groups. He classified two different habitat types as vicui~a territories. One was a
sleeping territory on flattened ridges, and the other was a feeding territory on the lower
slopes and flatlands with higher vegetative biomass. He found this territoriality to be
based on its regular defense and on the maintenance of boundaries on behalf of the
dominant male in the family group. In addition, Franklin found that resident neighbors
respected these boundaries, while outside vicui~a groups did not. The nature of this
territoriality, found both from Franklin's study and from the data collected at the Salinas
y Aguada Blanca National Reserve, appears to be "boundary- oriented" (Owen-Smith,
1977). In our study, the low degree of spatial overlap between family groups, and to a
lesser degree, between solitary males and bachelor herds, suggests that vicui~a groups are
maintaining their daytime feeding territories at the laguna.
Mean age of identified males found at laguna in 2002
There are no published data describing age structure within vicui~a populations.
The age a male establishes residency on a territory and the length of time that male
retains a territory remains largely unknown (Bosch and Svendsen, 1987). At the time of
tagging during the chaccu, CONATURA estimates an individual's age by the wear on
their teeth. Therefore, I was able to compile preliminary data on relationships among age
classes, population structure, and movement. Our results indicate a specific age-class
structure within the vicui~a population, with the bachelor herds being the youngest, males
in family groups in a middle range, and solitary males in the oldest age class. This
tinding is a departure from the observations at Pampa Galeras. Franklin (1983) suggested
that solitary males are sexually mature attempting to establish territories. However, age
estimates found in this population suggest that solitary males may be in a stage of
senescence, not in dispersal and establishment. In addition, solitary males were also
found at farther distances from the laguna and in more marginal habitat patches than
other group types when water was scarce.
Site fidelity and dispersal
Not all tagged individuals were observed at the laguna during the course of the
study in 2002. This may reflect observation bias, death or, dispersal. We recorded only
one case of dispersal during our study. A solitary male, (Blue 127; age=12) was
observed approximately 17 km from the laguna where it was originally tagged. Although
one example of dispersal is not enough to make any conclusions about movements in
~ee-ranging vicui~a, it does suggest that the older, solitary males may have larger home
ranges or dispersal distances than other males found in family groups or bachelor herds.
The clustering behavior of young males around the laguna appears to be counter-
intuitive. Earlier studies (Koford, 1957; Franklin, 1983) suggest that these young males
disperse long distances to establish their own territories and family groups, thereby
encouraging genetic shuffling. The distance traveled by bachelor groups and their spatial
distribution suggest that they are potentially optimizing their opportunities to obtain
females at higher densities found around the laguna, while simultaneously being limited
in movement due to the scarcity of water.
Our data suggest that median-aged males maintain territories for several years.
Although not every tagged male was observed during the study, there is evidence that
these males have been coming to the laguna consistently for several years and exhibit a
high degree of site tenacity.
Distance moved from family groups
How much space a vicui~a group needs to fulfill its daily requirements is a central
question within the context of the installation of corrals. The size of corrals will vary
dramatically with habitat quality, density, water resources, and food supply. To begin to
answer this question, vicui~a family groups were followed, and their movements were
documented. The results suggest that vicui~a do not move very far, and that the corrals
would not limit their movement greatly. However, observations were taken only in
regions at or adjacent to the laguna, and did not represent movement across the vicui~as
entire home range.
In addition, differential movement patterns were observed as vicui~a groups moved
greater distances in the morning then in the afternoon. Vicui~a movement was high in the
morning while the groups were heading toward the water source as compared to the
afternoon when they move back to their sleeping areas. These results correspond well
with the study of Vila and Roig (1992) that focused on daily movements and behavior of
family groups. Their study found that vicui~as moved greater distances in the morning
towards the water source and grazed in the afternoon.
Description of shifts in group composition
In this study males tended to maintain spatially discrete ranges around the laguna,
that remained relatively constant for the 4 months of observation. I hypothesize that
males establish territories, and females will occasionally switch family groups. In this
way, there is the potential for genetic shuffling between resident vicui~as in this
population, while maintaining a stable spatial distribution over time. Other studies have
assumed that groups are basically stable, comprising of the same individuals over time
(Franklin, 1983; Vila, 1995). Females in my study were found to adopt different mating
strategies. Some were sighted over several years with the same identified male, and other
females switch between different males at different rates.
Female vicui~as exhibit iteroparous reproduction, giving live birth to one young at a
time, repeatedly during her lifespan. Young ruminants depend completely on their
mother's milk for the required amino acids, vitamins and the inoculation of rumen
bacteria to develop ruminant fermentation (van Soest, 1994). Thus, the female vicui~a
has different energy requirements and constraints in maximizing reproductive success
then males (Kie, 1999; Vila, 1995). The number of family groups a female belongs to in
her lifespan may not affect her fitness. What matters more for her reproductive success is
the habitat quality of the dominant male in her family group, i.e., the quality, quantity,
and heterogeneity of the vegetation for embryo development, lactation, and the continual
growth of her young (Robinson et al., 1999).
IMPLICATIONS FOR CONSERVATION
An increase in the numbers of vicui~a has been reported over the last 30 years in
Peru. There were reportedly only 10,000 vicui~as left in 1964 (CONACS, 1997); as a
result of intensive management practices, the vicui~a population in Peru reportedly
increased to 150,000 individuals by 2000 (Lichtenstein et al., 2002). Despite the increase
in numbers, the distribution of the vicui~a has been dramatically reduced mainly to the
three national reserves in Pampa Galeras, Huaraz, and Salinas y Aguada Blanca. Vicui~a
no longer exist along the entire historic range in Peru (Figure 6-1). It is unknown if the
increase in overall vicui~a numbers is due to strict protectionist measures, or to local
dynamics in a few communities that have dedicated themselves to promoting the growth
rate among vicui~a to facilitate "rational use" of vicui~a fiber as a sustainable management
strategy. Thus, increases in numbers may potentially only reflect a geographically
limited increase in population density.
This centralized increase in vicui~a numbers is exacerbated by new management
initiatives that involve the installation of corrals. This may potentially change the
distribution of vicui~as and alter important ecological characteristics of their population
ecology. In addition, it invariably segregates potentially inter-breeding vicui~a
populations, encouraging homozygosity. It is important to understand the mechanisms of
this population growth among vicui~a, because changes in social behavior associated with
changes in density may change patterns of habitat use, and may negatively alter the puna
Study site II
Figure 6-1. Historic range of the vicui~a (Vlcugna vlcugna) in South American Andes.
Figure adapted from The Camelid (W. Ross Cockrill, 1979).
Information on behavior patterns of free-ranging vicui~a is scarce. Much of the data
on their habitat requirements are from studies that focused solely on biomass, husbandry,
and management (Wilson, 1994). Much of the current data on vicui~a ecology comes
from highly managed vicui~as in corrals and does not directly answer any of the questions
surrounding the issue of limiting movement in regard to group composition, movement
patterns, spacing patterns, and territoriality.
Like other density-dependent ungulates, the vicui~as may exhibit population
oscillations. Therefore great precaution must be taken in determining corral placement,
size, and the number of vicui~as to maintain in those corrals. If the population falls below
a viable level, there could be detrimental consequences for the vicui~a population as a
result of decreased genetic heterogeneity. For example, populations with low genetic
diversity are at risk of lowered immunity to the spread of disease throughout the entire
population, and more vulnerable to stochastic abiotic factors (Caro, 1998). On the other
hand, if the population grows rapidly, the puna ecosystem could be at risk of over-
grazing, homogenous vegetative composition, soil erosion, and desertification.
Currently, the focus in conservation and sustainable development programs seems to be
in increasing the population of vicui~a. Much effort has gone into determining how many
vicui~a can subsist given forage production, given that their use of fibrous feeds is
extremely efficient because of their highly specialized mode of fore-gut digestion (Cajal,
1991; Cueto et al., 1985). However, there are no studies that estimate how long the puna
ecosystem can maintain a high density of vicui~a before nutrient levels in the soil decline,
vegetative structure is degraded, or hydrological regimes altered.
The ability of vicui~a to have access to water and forage of sufficient quality and
quantity is essential in maintaining a viable vicui~a population. The data suggest that
vicui~a in the Salinas y Aguada Blanca National Reserve respond behaviorally to harsh
conditions by moving in response to resource availibility. Limiting vicui~a movement
through the implementation of corrals may have negative impacts for both the health of
the vicui~as and their habitats. The nutritional quality and quantity of biomass in puna
ecosystems has been reported to decline from the end of the wet season through the dry
winter (Pfister et al., 1989). This impacts vicui~a and their movement significantly as the
typical functional response in ruminants is to increase the time spent foraging to
compensate for the lowered nutritional value.
My results show decreased movement centered around the laguna during the dry
season. This limitation in movement around the laguna may be a result of an increased
water requirement. Or, there may be more vegetation of higher quality around the water
source available, decreasing the energy needed to satisfy water and nutritional
requirements. There are other constraints operating in conjunction with the decrease of
water availability and changes in nutritional quantity and quality. As temperatures
decrease during the dry winter, there is a higher cost of thermoregulation (Vila and
Cassini, 1993). Movement in search of forage among vicui~a also results in heat
production and lowered energy retention, especially when both the quantity and quality
offorage is reduced(Murray, 1991).
Limiting movement through the installation of corrals during the dry season may
not have a significant impact on the social structure of the vicui~a, or on natural
movement patterns. However, overgrazing in these areas near the laguna could have
significant long-term consequences on the quality of forage for subsequent years. In
addition, the dry season is also marked by frequent snowfall, as observed in our study
from July 10 to July 12. At this time, vicui~a groups were found father from the water
source, and most identified family groups were not found at the laguna at all. The
increased distribution of water possibly allowed for greater movement, subsequently
releasing local vegetation around the laguna from foraging pressure. The installation of
corrals would obviously limit a functional response among vicui~a to availability to
Another potential consequence of corral installation in the Salinas y Aguada Blanca
National Reserve is related to land use and the geographic position of certain key
landscape features. The area where vicui~a are found is bisected by the paved road
running from Arequipa to the Colca and Juliaca-Cuzco and presents a potential source of
vicui~a mortality. There were several instances during the study where vicui~a groups
were observed crossing this highway in search of water in the reserve.
Placing a corral in this reserve would exclude other vicui~a groups traveling in
search of water and forcing them to find other sources, either in regions near domestic
livestock, or in farther regions. This could potentially decrease the amount of vicui~as
available for shearing during the roundup.
The primary aim of the rational use of the vicui~a as a sustainable management
program is two-fold. First, it claims to protect the vicui~a, while secondarily providing a
legal, alternate source of income for impoverished Andean communities. A major threat
to the vicui~a is anthropogenic in nature as Andean community members claim that there
is competition for feed resources between naturally occurring vicui~a and their domestic
livestock (Urquieta et al., 1994). The implementation of corrals has been hailed as a
viable compromise that will alleviate this competition, provide protection for wild
vicui~as and facilitate the capture process for shearing. However, these corrals may serve
to alter the puna ecosystem and the social and foraging behavior of the vicui~a that may
lead to negative consequences.
Baseline data on the population characteristics of free-ranging vicui~a populations
are critical to the evaluation of effective conservation and management practices and to
our understanding of the importance of local habitat features to population density and
distribution patterns. Group characteristics of the vicui~a of the Salinas y Aguada Blanca
National Reserve are comparable to studies of other free-ranging vicui~a, with densities
notably lower than corralled populations. The changes in distribution and density over
the 4-month study suggest that the location of a permanent water source may be an
integral factor contributing to patterns of distribution and movement over the seasonal
cycle. The major conclusions drawn from this study are summarized below.
1. In the Salinas y Aguada Blanca National Reserve, the mean family group
composition is 1 male, 3.7 females, and 1.6 young. These figures fall within the
range of other studies that include both corralled and free-ranging populations,
suggesting that the vicui~a have a rigid family group structure within different
populations that occur across ecological gradients and management programs.
2. In my study average density was 3.7 vicui~aslkm2. This falls within the range of
studies of free-ranging vicui~a and is much lower than corral estimates.
3. Water distribution and availability seem to have a major effect on vicui~a
movement. Groups were found to be 2 km closer to the laguna in drier conditions.
This may have management implications, as water is a major limiting factor in the
puna. Although domestic livestock and vicui~a herds have not been found to
compete for forage, they could be competing for water resources.
4. Overall, large groups and large family groups are found closer to the laguna. This
indicates that large groups are limited in their movements by water availability, and
that large groups tend to establish territories in higher quality habitats near the
5. The data on the Salinas y Aguada Blanca vicui~a in combination with other studies
on ungulates, suggest that large family groups in open habitats is an optimal
strategy for feeding and reproducing, as smaller vicui~a groups are more frequently
in marginal habitats.
6. There may be a rigid population structure and family group structure within vicui~a
populations found in different habitats. Several group characteristics such as group
type distribution in the population, group size, and group composition remained
stable despite changing water distribution.
7. There was a low degree of spatial overlap around laguna. This suggests that the
dominant males in family groups are maintaining boundaries in relation to other
family groups when traveling to the laguna. This also provides further evidence for
territoriality among males in family groups.
8. Individual vicui~as and groups were found to exhibit differential movement patterns
throughout the day. Vicui~as move greater distances in the morning, presumably to
reach the laguna by midday. The rate of movement decreases in the afternoon.
These differential movement patterns are important when estimating how far a
vicui~a will move, and subsequently how large their territory requirements are in
the context of corral initiatives.
9. Our study provides evidence that some females will stay with the same male in a
family group for an extended period oftime, while others will switch annually
among family groups.
10. With the use of age estimations provided by the CONATURA database, an age
class structure was found in the vicui~a in the Salinas y Aguada Blanca National
Reserve. Bachelor herds were found to be young males that are non-territorial
dispersers. Other studies report that solitary males are in process of establishing
their territories, however, this study indicates that solitary males are older and in a
stage of senescence. Although there was overlap, the mean age of solitary males
was higher then those in family groups
LIST OF REFERENCES
Bosch, P. C. and G. Svendsen. 1987. Behavior of male and female vicuna as it related to
reproductive effort. Journal ofhlammalogy 62. 425-429.
Brack, A. 1980. Situacihn actual de la poblacihn de vicui~as en Pampa Galeras y zonas
aledai~as y recommendations para su manejo. Proyecto Especial Utilizacihn
Racional de la Vicui~a.
Ministerio de Agricultura y Alimentacihn, Lima.
Buckland, S.T, D. R. Anderson, K. P. Burnham, and J. L. Laake. 1993. Distance
Sampling. Chapman and Hall London.
Cajal, J. 1991. An integrated approach to the management of wild camelids in
Argentina. In: Latin American Mammology. History, Biodiversity and
Conservation. M. A. Mares and D. J. Schmidly (eds.). University of Oklahoma
Care, T. led). 1998. Behavioral Ecology and Conservation Biology. O~ord University
Press, New York
Cattan, P and A. Glade
1989. Management of the Vlcugna vlcugna in Chile: Use of a
matrix model to assess harvest rates. Biological Conservation 49 131-140.
Caughley, G. and A. R. E. Sinclair. 1994. Wildlife Ecology and Management Blackwell
Caviedes, C. and G. Knapp. 1995. South America. Prentice Hall, Englewood Cliffs,
Chapman, C. A., S. R. Balcomb, T. R. Gilespie, J. P. Skorupa, and T. T. Struhsaker.
2000. Long-term effects of logging on African primate communities: A 28-year
comparison from Kibale National Park, Uganda. Conservation Biology 14. 207-
Cockril, W. R. 1979. The Camelid. An All Purpose Animal. Volume I. Proceedings
of the Khartoum Workshop on Camels. Scandinavian Institute of African Studies,
Consejo Nacional de Camelidos Sudamericanos (CONACS) website maintained by
Universidad Agraria de la Molina. 1997. Antonio Tovar
Convention for the Conservation and Management of the Vicuna. Website maintained by
Center for International Earth Science Information Network. http://sedac.ciesin. org.
Cueto, L. J, C. F. Pence, E. Cardich, and M. Rios. 1985. Management ofvicui~a: Its
contribution to rural development in the High Andes of Peru. FAO Conservation
Guide No. 11
Franklin, W. L. 1978. SocioecoloW of the vicuna. Dissertation, Utah State University.
Franklin, W. L. 1983. Contrasting socioecologies of South America's wild camelids:
The vicui~a and the guanaco. In: Advances in the Study of Mammalian Behavior.
J.F. Eisenberg and Devra G. Kleiman (eds). Special publication No. 7. The
American Society of Mammologists, Stillwater, OK.
Hixon, M. A. 1980. Food production and competitor density as the determinants of
feeding territory size. American Naturalist 115. 510-530.
Hoffman, R. K., K. Otte, C. F. Pence, and M. A. Rios. 1983. El Manejo de la Vicui~a
Silvestre. Tome I. Eschborn.
Kie, J. G. 1999. Optimal foraging behavior and risk of predation: effects on behavior
and social structure in ungulates. Journal of Mammology 80. 11 14-1129.
Koford, C. B. 1957. The vicui~a and the puna. Ecological Monographs 27. 153-219.
Lichtenstein, G., F. Oribe, M. Grieg-Gran, and S. Mazzucchelli. 2002. Manejo
comunitario de vicui~as en Peni. Estudio de case del manejo comunitario de vida
silvestre PIE Series No. 2
Lucherini, M. 1996. Group size, spatial segregation and activity of wild sympatric
vicui~as Vlcugna vlcugna and guanacos Lama guanlcoe. Small Ruminant Research
McCullough, D. R. 1999. Density dependence and life-history strategies ofungulates.
Journal of Mammology 80. 1130-1146.
Menard, N. 1982. Some aspects of the socioecology of the vicui~a (Lama vlcugna)
Molina. Terre et la vie-revue d'ecologie appliquee 36.15-35.
Murray, M. G. 1991. Maximizing energy retention in grazing ruminants. Journal of
Animal Ecology 60. 1029-1045.
Owen-Smith, N. 1977. On territoriality in ungulates and an evolutionary model. The
Quarterly Review ofBiology 52. 1-38.
Pfister, J. A., F. San Martin, L. Resales, D. V. Sissson, E. Flores, and F. C. Bryant. 1989.
Grazing behaviour of ilamas, alpacas and sheep in the Andes of Peru. Applied
AnimalBehavioral Sciences 23. 237-246.
PulgarVidal, J. 1996. GeografiadelPeru. Las ochoregionesnaturales. D~cima
edicihn. Peisa. Lima, Peni.
Rabinovich, J. E., M. Hernandez, and J. L. Cajal. 1985. A simulation model for the
management of vicui~a populations. EcologicalModeling 30. 275-295.
Renaudeau d'Arc, N., M. H. Cassini, and B. L. Vila, 2000. Habitat use by vicui~as
Vlcugna vlcugna in the Laguna Blanca Reserve (Catamarca, Argentina). Journal of
Arid Environments 46. 107-115.
Robinson, J. J, K.D. Sinclair, R. D. Randel, and A. R. Sykes. 1999. Nutritional
management of the female ruminant: mechanistic approaches and predictive
models. In: Nutritional Ecology of Herbivores. Proceedings of the Vth
International Symposium on the Nutrition of Herbivores. H-J. G. Jung and G. C.
Fahey, Jr. (eds). American Society of Animal Science, Savoy, Illinois.
Rundel, P.W. and B. Palma. 2000. Preserving the unique puna ecosystems of the
Andean Alitplano: A descriptive account of Lauca National Park, Chile. Mountain
Research and Development 20. 262-271.
Sahley, C., J. T.Vargas, and J. Sanchez Valdivia. 2001. Community ownership and live
shearing of vicui~as in Peru: Evaluating management strategies and their
sustainability. Asociacion para la investigacion y conservacion de la Naturaleza,
Smith, R. L. 1996. Ecology and Field Biology. (5" ed.). Harper Collins, New York
Urquieta, B, R. Cepeda, J.E. Caceras, L.A. Raggi, and J.R. Rojas. 1994. Seasonal
variation in some reproductive parameters of male vicuna in the High Andes of
northern Chile. Journal ofArid Environments 26. 79-87.
van Soest, P. J. 1994. Nutritional Ecology of the Ruminant. (2nd ed.). Comstock
Publishing Associates, a division of Comell Universtiy Press, Ithaca and London.
Vila, B. L. 1992. Mother-offspring relationship in the vicuna, Vlcugna vlcugna
(Mammalia: Camelidae). Ethology 92. 293-300.
Vila, B. L. 1995. Spacing patterns within groups in vicui~as, in relation to sex and
behavior. Studies on Neotropical Fauna and Environment 30.45-51.
Vila, B. L and M. H. Cassini. 1993. Summer and autumn activity patterns in the vicui~a.
Studies on Neotropical Fauna and Environment 28. 251-258.
Vila, B. L and M. Cassini. 1994. Time allocation during the reproductive season in
vicui~as. EtholoW 97. 226-235.
Vila, B.L. and V.G. Roig. 1992. Diumal movements, family groups and alertness of
vicui~a (Vlcugna vlcugna), during the late dry season in Laguna Blanca Reserve
(Catamarca, Argentina). Small Ruminant Research 7. 289-297.
Weisberg, P. J., N. T. Hobbs, J. E. Ellis, and M. B. Coughenour. 2002. An ecosystem
approach to population management of ungulates. Journal of Environmental
Management 65. 181-197.
Wheeler, J. and D. Hoces. 1997. Community participation, sustainable use and vicuna
conservation in Peru. Journal of Mountain Research and Development 17. 283-
Wilson, T. R. 1994. Towards committed and collaborative camelid research. Journal of
Arid Environments 26. 95-103.
Jennifer Elena Davies was born on May 4, 1975, in San Diego, California. She is
the only child of Thomas M. Davies Jr. and Eloisa Davies Monzhn. She earned her
degree of Bachelor of Arts in Nature and Culture, with a minor in Spanish from the
University of California, Davis, in 1999. There she translated her respect and love for the
natural world to academic study and research in human-animal interactions. To continue
her commitment to interdisciplinary research in the natural sciences she moved to
Gainesville to enroll in graduate studies in the Center for Latin American Studies/
Tropical Conservation and Development in the fall of 2001.