0

SPATIAL ASSOCIATIONS OF DAMA GAZELLES IN A TEXAS HILL

COUNTRY RANCH

Susan Cooper1. and Elizabeth Cary Mungall2.

1. Associate Professor, Wildlife Ecology, Texas A&M AgriLife Research,

1619 Garner Field Road, Uvalde, TX 78801. s-cooper@tamu.edu

2.. Science Officer, Second Ark Foundation, 105 Henderson Branch Rd.

W., Ingram, TX 78025, USA. elizabeth@mungall.us

December 8, 2016

Photograph by Christian Mungall

1

Contents

Introduction to Dama Gazelle, Conservation Needs and Status in Texas ............................... 4

Goals of this Study ................................................................................................................... 5

Animals Used in this Study ...................................................................................................... 5

Fitting GPS Collars on the Gazelles .......................................................................................... 7

Data Collection Frequency from GPS Collars........................................................................... 8

Animal Association Analysis .................................................................................................... 9

Home Range Size ..................................................................................................................... 9

Pasture Use by Dama Gazelles ................................................................................................ 9

STUDY AREA DESCRIPTION ........................................................................................................ 10

Location ................................................................................................................................. 10

Landscape .............................................................................................................................. 10

Vegetation ............................................................................................................................. 12

Management ......................................................................................................................... 13

Initial Dispersal of Dama Gazelles into the Research Pasture. .............................................. 13

Establishing the Research Herd. ............................................................................................ 15

Annual Distribution of Dama Gazelles within the Research Pasture. ................................... 17

Group Associations and Distribution of Dama Gazelles with the Original Young Adult Male

Present, December 19, 2014 - June 22, 2015. ....................................................................... 19

Group Associations and Distributions of Dama Gazelles in the Absence of Adult Males, June

22, 2015 - August 5, 2015. ..................................................................................................... 21

Group Associations and Distributions of Dama Gazelles with New Adult Males, August 5,

2015 – December 19, 2015. .................................................................................................. 23

Association and Distribution of the Two Adult Males, after the Females have shed their GPS

Collars, December 19, 2015 to June 30, 2016. ...................................................................... 27

2

Habitat Selection by Dama Gazelles in the Research Pasture ............................................... 28

DISCUSSION ............................................................................................................................... 30

Home Range Size and Interaction between Male Dama Gazelles ........................................ 30

Association within the Female Group and Interaction with the Males ................................ 32

Effect of Habitat Selection on the Distribution of Dama Gazelles ........................................ 33

Recommendations for the Conservation and Management of Multi-Male Groups of Dama

Gazelles in Rangeland Pastures. ............................................................................................ 34

SUMMARY ................................................................................................................................. 35

ACKNOWLEDGEMENTS ............................................................................................................. 35

LITERATURE CITED ..................................................................................................................... 36

RECOMMENDATIONS FOR FUTURE STUDY ............................................................................... 38

Development of MOVEPROG ................................................................................................ 38

Use of SOCPROG in combination with MOVEPROG: ............................................................. 39

APPENDIX 1. Further Analysis of GPS Data from this Study ..................................................... 41

Introduction ........................................................................................................................... 41

Overview of software development aimed at analysing animal movement ........................ 42

Development of the animal movement analysis package MOVEPROG ................................ 42

List of Figures

Figure 1. Dama gazelles in the holding pen wearing new, colored, GPS collars. ........................... 7

Figure 2. Location of the study site in Uvalde County, Texas ....................................................... 10

Figure 3. Habitat map of the 202 ha. research pasture inhabited by the dama gazelles. ........... 11

Figure 4. Vegetation on hillsides and valleys within the research pasture. ................................. 12

Figure 5. Group of eight dama gazelles in the research pasture on Morani River Ranch. ........... 14

Figure 6. Rumen contents of a dama gazelle. ............................................................................... 16

3

Figure 7. Distribution of all locations of dama gazelles throughout the pasture over 1 year. ..... 17

Figure 8. Annual 50% KHR core areas of all dama gazelles. ......................................................... 18

Figure 9. Photograph of the bowl-shaped valley in the southwest of the pasture favored by

dama gazelles. ................................................................................................................... 19

Figure 10. Incorporation of female dama gazelle BF into the original group............................... 20

Figure 11. Core areas of female and immature dama gazelles with the original adult male

present. ............................................................................................................................. 21

Figure 12. Core areas of female dama gazelles when no adult males were present. .................. 23

Figure 13. Core areas of female dama gazelles after the introduction of three new adult males.

....................................................................................................................................................... 25

Figure 14. Core areas of the four male dama gazelles from August 5, 2015 to December 19,

2015. ................................................................................................................................. 26

Figure 15. Annual core areas of the two adult male dama gazelles. ............................................ 28

Figure 16. Google map of the research pasture showing elevation and habitat features. .......... 29

Figure 17. Dama gazelle collaring team. ....................................................................................... 37

List of Tables

Table 1. Dama gazelles used in this study. ..................................................................................... 6

Table 2. Association matrix for dama gazelles during the first eight days in the research pasture.

........................................................................................................................................... 15

Table 3. Association matrix between dama gazelles with the original young adult male present.

December 19, 2014 - June 22, 2015. ................................................................................ 20

Table 4. Association matrix between dama gazelles in the absence of adult males. June 22, 2015

- August 5, 2015. ............................................................................................................... 22

Table 5. Association matrix between dama gazelles with adult males present. August 5, 2015 -

December 19, 2015. .......................................................................................................... 24

4

INTRODUCTION

Introduction to Dama Gazelle, Conservation Needs and Status in Texas

Dama gazelles (Nanger dama) are critically endangered in their native range in the Sahelo-Saharan Zone

in Africa. Fewer than 300 are known to exist in the wild (RZSS & IUCN Antelope Specialist Group 2014)

but there are about 500 dama gazelles in zoological parks and most importantly about 1,000 live on

exotic wildlife ranches, mainly in Texas. These dama gazelles living in semi free-ranging conditions in

extensive rangeland pastures provide unique opportunities for conservationists to obtain much needed

information on the behavior and ecology of the species. This is especially useful because such

information is difficult to obtain from wild populations due to rarity of the species and difficulty of

access to the remote areas in which wild dama gazelles still live.

Insight into the social interactions and spatial distribution of dama gazelles within pastures is needed for

successful species management and restoration efforts. Adult male dama gazelles are aggressive

towards each other and this causes problems in the breeding and management of this species. In

confined areas, males will fight to inflict damage and even kill rivals when females are present. (Barbosa

and Espeso 2005, Mungall 2014) Standard management practice is to maintain only one adult male per

breeding herd, however under this system reproductive output is less than optimal. Keeping males in

adjacent fenced pastures appears to improve reproduction (E. C. Mungall personal communication) but

is not an ideal option. The ability to establish groups of dama gazelles including multiple males is a

sought after conservation goal both for management of the species on exotic wildlife ranches and for re-

establishment of dama gazelles in protected reserves within their native African range.

This project aims to provide an understanding of the spatial requirements of dama gazelles for

establishment of successful multi-male breeding populations with minimum social friction. Phase 1 of

the project conducted at Stevens Forest Ranch in Terrell County, West Texas tracked the spatial

distribution, habitat selection and interactions of five male dama gazelles in an extensive (8,998 ha)

semi-desert pasture (Mungall and Cooper, unpublished data). Given ample space, these male dama

gazelles occupied overlapping home ranges, (95% Kernel Home Ranges), averaging about 1,800 ha.

(4,400 ac.). Within these large home ranges, adult males maintained non-overlapping, core use areas

(50% KHR) of 440 ha. (1,088 ac.). Consequently, adult males had minimal contact with each other. In

contrast, subadult males were more social and shared both home range and core areas. The only

aggression recorded between males was when a maturing young male sought to establish his adult

5

breeding ranges, which is a natural and inevitable phenomenon. While the West Texas study provided

useful information, most exotic wildlife ranches and even African restoration sites cannot provide such

extensive areas. Phase 2 of the project, and the focus of this report, is to determine the spatial

distribution, habitat selection and interactions of both male and female dama gazelles when kept in a

sizeable but less extensive rangeland pasture more typical of areas available for conservation projects

involving dama gazelles. Animal home range size is influenced by resource abundance and distribution

on the landscape. Therefore, it is predicted that in a more mesic environment with greater vegetative

production, the size of a home range required by dama gazelles will be smaller and the animals can be

successfully kept at greater density than in the arid landscape of West Texas. The opportunity to track

the associations and interactions of female dama gazelles will also provide useful insights into

behavioral interactions and spatial requirements of breeding groups that will be important for

restoration of the species.

Goals of this Study

The goals of this study are:

1. To identify home range size and interaction between male dama gazelles sharing a moderate

sized (202 ha., 499 ac.) pasture.

2. To identify association within female groups and the interactions between male and female

dama gazelles.

3. To describe habitat selection and effect of landscape structure on the distribution of dama

gazelles in a Texas Hill Country pasture.

4. To provide recommendations for the conservation and management of multi-male groups of

dama gazelles in rangeland pastures.

METHODS

Animals Used in this Study

The dama gazelles used in this study belonged to Kevin and Cole Reid of Morani River Ranch and were of

the eastern dama gazelle Nanger dama ruficollus (also called addra). The main group consisted of an

established group of 8 females, 1 immature male and 1 young adult male that had been kept in an

enclosure adjacent to the research pasture. For this study an unrelated adult male from a neighboring

6

pasture and a newly purchased adult male, adult female and immature male were added to the group. A

further 3 adult males were added to the pasture in August, 2015, part way through the study (Table 1).

Table 1. Dama gazelles used in this study.

Animal ID Collar color Gender Age GPS collar data Start Date End Date*

Original Group

GF Green Female Adult 12/19/2014 12/19/2015

LF Lime Female Young Adult 12/19/2014 12/19/2015

OF Orange Female Adult 12/19/2014 12/19/2015

SF Silver Female Young Adult 12/19/2014 01/02/2015

VF Violet Female Adult 12/19/2014 12/19/2015

YF Yellow Female Adult 12/19/2014 11/4/2015

PF Pink Female Adult 12/19/2014 12/19/2015

TF Teal Female Young Adult 12/19/2014 01/02/2015

Rm Red Male Immature 12/19/2014 12/19/2015

GM Green Male Young Adult 12/19/2014 6/22/2015

Introduced animals

BF Bubblegum pink Female Adult 12/19/2014 12/4/2015

Bm Blue Male Immature 12/19/2014 01/06/2015

YM Yellow Male Adult 12/19/2014 01/11/2015

FM First black Male Adult 12/19/2014 12/27/2014

TM Teal Male Adult 08/05/2015 10/26/2015

BM Black Male Adult 08/05/2015 06/30/2016

NM Navy Male Adult 08/05/2015 06/30/2016

* End dates are when the GPS collar data indicated cession of movement i.e. collar shed or animal died.

7

Fitting GPS Collars on the Gazelles

A week prior to initiation of the study Mr. Cole Reid and his staff transported the dama gazelles from

their home enclosure to a pen at the ungulate handling facility on Morani River Ranch. On December

15th, 2014, experienced animal handlers sent individual dama gazelles through a system of darkened

runways into a drop-floor squeeze chute. Once in the chute each gazelle was blind-folded. Dr. Elizabeth

Cary Mungall of Second Ark Foundation took morphometric measurements of horn dimensions and neck

size; Dr. Cecil Armin of South West Texas Veterinary Medical Clinic took a 5cc blood sample for DNA

analysis; then Dr. Susan Cooper and her team from Texas A&M AgriLife Research fitted each animal with

a GPS collar (Model: Lotek GPS3300S, Lotek Wireless Inc. Ontario, Canada). The animals were then

transported to a 0.3 ha. (0.74 ac) holding pen adjacent to the research pasture. On December 18th 2014

an additional unrelated adult male, adult female and juvenile male were acquired. These animals were

measured and fitted with GPS collars while in the ungulate handling facility, then they were added to

the group already in the holding pen. The gates of the holding pen were opened on December 19th 2014

allowing the gazelles to transition into the 202 ha. (499 ac.) research pasture, yet still have access to

water and pelleted feed in the holding pen, a location familiar to them. In June 2015, an additional 3

adult male dama gazelles were acquired. These animals were kept in a small adjacent pasture for a

month until they were acclimated to their new rangeland environment. Early in the morning of July 14th

2015, Cole Reid and his ranch manager John Fredericks sedated these gazelles for transportation to the

holding pen. At this time, Dr. Mungall took morphometric measurements and Dr. Cooper fitted each

with a GPS collar. These male gazelles were then transported to the holding pen where they stayed until

the gate was opened on August 5th 2015 allowing them to transition quietly into the research pasture.

Figure 1. Dama gazelles in the holding pen wearing new, colored, GPS collars.

8

Data Collection Frequency from GPS Collars

The GPS collars were individually colored for identification of individual animals during subsequent

monthly welfare checks and to aid associated behavioral observations. Animals are identified by a

unique 2-letter code designating the color of their collar and their gender e.g. BM Black male, LF Lime

female. Immature animals are designated by a lower case for the gender, e.g. Rm. The initial set of

collars were programmed to take 1 locational GPS fix every 30 minutes for the first 8 days after the

gazelles were released from the holding pen. This was to monitor how the animals dispersed and

explored their new environment. Thereafter, the collars collected 1 location every 3 hours for 1 year

providing up to 2,920 locations per animal. This frequency of information permits estimation of

habitat use and interactions between animals without the mathematical problems of autocorrelation of

locations inherent in more frequent regimes of data acquisition (Perotto-Baldivieso et al. 2012). To

obtain a full year of data the locations acquired at 30 min intervals were trimmed to the appropriate 3

hourly records and included in the annual data set. The collars of the 3 males introduced in August were

programmed to take 1 location every 3 hours with no initial 30 minute data schedule. Animal locations

obtained from the GPS collars were accurate to within 2m (6ft). The collars included automatic timed

drop-off units that released the collars after 1 year thus allowing retrieval of the collars and GPS data

without recapture of the animals.

Due to changes in numbers of dama gazelles in the pasture during the study this report is divided into 5

sections to provide insights into the changing social dynamics of the group.

1. Introduction to the Research Pasture 12/19/2014 – 12/28/2014.

2. Original young adult male present 12/19/2014 – 06/22/2015.

3. No adult males present 06/22/2015 – 08/05/2015.

4. Three new adult males present 08/05/2015 – 12/19/2015.

5. New males after the females and Rm have shed their GPS collars 12/19/2015 – 06/30/2016.

At the end of the study the GPS collars were downloaded and imported into ArcGIS 10 (ESRI Redlands

CA) and all GPS points were reviewed to identify and remove any inaccurate locations. Such errors are

usually due to temporary poor configuration of the GPS satellites at that time. Videos were created of

the sequential locations of the gazelles during the first 8 days of their introduction to the pasture, and of

the early distribution of the newly introduced adult males, to provide a descriptive overview of animal

movements and associations.

9

Animal Association Analysis

Data from the collars were transformed into UTM coordinates and Euclidian (straight line) distances

between all pairs of gazelles. This was calculated for each GPS fix time using the formula [Sqrt (E1-

E2)2+(N1-N2)2 where N = northings, E = eastings]. Dama gazelles are browsers using scattered bushes as

their food source and thus typically they do not maintain the compact herd formation seen in grazing

animals. Based on the median values of distances between members of the original group of dama

gazelles we selected an inter-animal distance of 50 m to develop association matrices to identify social

groups (Ginsberg and Young 1992; Whitehead 2008). When the frequency of animals being within 50 m

of each other was equal or greater than 0.5, the animals were considered to be associated in a social

group, whereas when the frequency of interactions within 50 m was less than 0.25, the animals were

considered to be separated from each other.

Home Range Size

Distribution of the gazelles within the pasture was depicted by the fixed volume Kernel Home Range

(KHR) calculated in ArcView 3.2. This technique is a standard measure of animal distribution based on

the mathematical probability of an animal using an area (Seaman and Powell, 1996). Due to pasture

shape and the distribution of animals throughout the pasture, the 95% KHR did not provide meaningful

results. However, the 50% KHR, which distinguishes the core use areas in which 50% or more of all

animal locations occurred, was of use in identifying areas most often used by the gazelles. Shared use of

the pasture by the animals was examined by the distribution and degree of overlap of annual 50% KHR

core use areas. Due to differences in group composition, core use areas were calculated on an annual

basis as well as separately for the time periods with 1 male present, no males present and 3 new males

present.

Pasture Use by Dama Gazelles

To determine whether the distribution of gazelles in the pasture was influenced by landscape features,

such as soil-based ecological sites, hill slope (particularly steep inclines of >11 – 27 degrees or 20 – 50%),

natural drainages and areas of dense vegetation, or man-made features such as roads, boundary fences,

feeders and water troughs, the frequency of occurrence of gazelles within mapped areas, or within 10 m

of man-made features, was compared with the distribution of 3,000 random points generated in

ArcView 10. Chi- squared goodness of fit test (χ2) was employed to identify whether the distribution of

gazelles relative to these features was different from random distribution, differences were considered

significant at P ≤ 0.05.

10

STUDY AREA DESCRIPTION

Location

The study was conducted in a 202 ha. (499 ac.) rangeland pasture on the Morani River Ranch, in Uvalde

County Texas, located on the southern border of the Edwards Plateau (also known as the Texas Hill

Country) in South Central Texas (Fig. 2). This ecoregion is the hub of the exotic wildlife activity in Texas

and is home to most of the roughly 1,000 dama gazelles present in Texas (E. C. Mungall pers. comm.).

The climate of this region is classified as humid subtropical, and is characterized by hot, humid, summers

and mild to cool winters. The warmest month is July with an average temperature of 28.6°C and the

coolest month is January with an average temperature 10°C. Precipitation is highly variable but averages

654 mm/year at the closest weather station located at Montell (Weather Underground. 2016).

Figure 2. Location of the study site in Uvalde County, Texas

Landscape

The landscape was characterized by rocky, limestone hills. Within the research pasture, the elevational

range was 332 to 396 m (1,090 – 1,300 ft.) with the steepest slopes on the northern sides of the hills.

The dominant ecological site (176 ha., 435 ac.) in the pasture fell within the Ector (ER) soil series (USDA-

NRCS 2014) with very shallow, rocky, calcareous, clay loam soil over fractured limestone bedrock. The

Ector soil series extends through much of the Texas Hill Country and was prevalent at the Stevens Forest

Ranch in Terrell County, Texas, used in Phase I of this study. Within the Safari pasture, 3 small areas of

slightly different composition existed (Fig. 3). The 10 ha. (25 ac.) area around the ephemeral creek in the

narrow eastern arm of the pasture was characterized by Dev (DE) soils, consisting of limestone cobbles

mixed with very gravelly clay loam soil. The west side of the pasture was almost square in shape. In the

southwestern corner a shallow valley was characterized by 11 ha. (27 ac.) of slightly deeper Eckrant-

Kavett complex (EK) soils. Limestone rockland (LS) or exposed limestone occurred in a 4 ha. (10 ac.) area

11

in the southeast of the pasture and in small areas on the hill tops. Estimated potential agricultural

productivity of the dominant Ector soils is low at 1,500 kg/ha. Productivity in the valleys is reported at

2,000-2,500 kg/ha. due to slightly deeper soils and greater water availability, whereas productivity on

the exposed limestone was estimated at only 1,100 hg/ha (USDA-NRCS 2014). Actual amount of woody

browse produced on these soils has not been measured but appeared to exceed potential herbaceous

production.

Figure 3. Habitat map of the 202 ha. research pasture inhabited by the dama gazelles.

Legend: The background is an aerial photograph. Dark areas indicate shrub cover. The pasture boundary is in red, roads and tracks are white and drainages are depicted in blue. The predominant ecological (range) site in the pasture is Ector (ER), with areas of Dev (DE) soils in the west shaded in green, Eckrant-Kavett complex (EK) in the east shaded in light green, and limestone rockland (LS) shaded in brown. Steep slopes are identified by yellow cross hatching. Locations of supplemental feed and water are depicted by orange and blue triangles respectively.

12

Vegetation

The vegetation of this rocky pasture was dominated by shrubs. Shrub cover was estimated from aerial

photographs to be approximately 50%. On these photographs, dark pixels correspond to woody

vegetation and light pixels indicate the more reflective bare ground. Blackbrush acacia (Acacia rigidula /

Vachellia rigidula) was the dominant woody plant species and was the most important forage plant for

the gazelles. This semi-evergreen, thorny shrub is abundant on rocky ridges in southwest Texas and

northern Mexico. Pricklypear cacti including Opuntia engelmanni var. lindheimeri and O. polycantha

were also an important component of the vegetation and provided energy rich, edible fruits for the

animals. Other common plants on the hillsides were the shrub, coyotillo (Karwinskia humboldtiana), the

leaves and berries of which are highly toxic, and the succulent leatherstem (Jatropha dioica). Neither of

these plants provided food for the gazelles (Fig. 4). In the valleys and along drainages, a more diverse

mix of shrubs existed including Texas persimmon (Diospyros texana), Ashe juniper, (Juniperus ashei),

spiny hackberry (Celtis pallida) and catclaw mimosa (Mimosa borealis). A few live oak trees (Quercus

virginiana) grew along the dry creek on the northeast side of the pasture at the lowest elevation. These

plant species all provide some food to browsing ungulates, although they are not classified as preferred

browse species (Wright et al. 2002). Herbaceous cover was sparse and species composition of annual

plants was dependent on rainfall.

Figure 4. Vegetation on hillsides and valleys within the research pasture.

Legend: Hillsides are predominantly vegetated with blackbrush (in flower) and pricklypear cactus, lower elevations have a more diverse mix of shrubs with live oak trees.

13

Management

The diet of the gazelles was derived predominantly from browsing native shrubs. Five feeders were

situated in the pasture. These feeders were filled twice a week with high protein pelleted feed but

access was dominated by a herd of longhorn cattle and a large herd of 35-40 scimitar-horned oryx (Oryx

dammah). Thus, once the gates to the holding pen were closed there was little chance for smaller

animals like the dama gazelles to obtain any appreciable amount of the supplemental feed. Hence the

gazelles subsisted almost entirely on native vegetation. Water was available to the animals at 5 concrete

tanks, one near each feeder. During rainy periods water drained off the hillsides and collected into rocky

pools in the creek bed on the northern edge of the pasture.

RESULTS

Initial Dispersal of Dama Gazelles into the Research Pasture.

The gates to the holding pen were opened at 2 pm on December 19th 2014, permitting the gazelles to

transition gently into the 202 ha. ($99 ac.) research pasture. Weather was mild and dry, with

temperatures ranging from 12 to 3 oC (53 to 38 oF) and winds of 6 mph. Visualization of the sequential

GPS points showed that at 2:30 pm all the gazelles, except introduced male YM, ran out along the main

dirt road for a distance of 1,246 ± 47 m (0.77 miles) taking the “Y” turn southwest into the main part of

the pasture. For the following two days, the group split into various non-consistent subgroups and

explored the pasture, but they did not circle the fence lines as some animals do. At night, the gazelles of

the original group came together in an area not far from the road in the southwest part of the pasture.

This bowl-shaped valley would become a favored location of the gazelles throughout the coming year

(Fig. 5). The only exception was the female PF. On the outward run, this animal followed the group for

687 m (0.43 miles) and then stopped in a bushy area half way up the hill. Despite the other gazelles

passing nearby, she stayed in this area for two days before rejoining the group. Female TF separated

from the group on December 21st. During the night she returned to the hillside near the holding pen

where she remained for the rest of the introductory period.

On the night of December 23rd there was a change in the weather. A cold front arrived and

temperatures fell to minus 3oC (27 oF) with winds of 7-10 mph gusting to 26 mph (Weather Underground

2016). In response, the gazelles in the original group all moved to the shelter of the valley constituting

the northeast arm of the pasture, and at night they were recorded in the holding pen where pelleted

feed was available. The cold weather continued for 3 days during which these gazelles remained in the

northeast arm of the pasture, often crowded along the fence closest to their former enclosure. Gates to

14

the holding pen were left open to ensure their access to supplemental feed, which the animals appeared

to use mainly at night.

Introduced animals not from the original group did not fully integrate with the original group, nor did

they create a new group. The adult male YM stayed near the front gate and was observed persistently

fighting through the fence with another adult male dama gazelle in the adjacent pasture. In total, he

spent 38% of his time within 10 m of the fence. At night, he often returned to the holding pen,

presumably to eat the pelleted feed. When female TF returned to near the holding pen, YM spent the

nights with her. Then when the original group returned to this area during cold weather he associated

with this larger group. The imported immature male Bm initially ran out of the holding pen with the

group but returned to the holding pen on the first night. He remained in or near the holding pen for the

entire time. He associated first with YM, then rejoined the group when they returned to the area near

the holding pen. The imported female BF and imported male FM initially moved out with the original

group. They spent the first two nights near but not in the original group. When the cold weather arrived

these two gazelles did not follow the group back to the more sheltered area in the northeast of the

pasture. Instead the female BF moved south to the valley below the zebra feeder where she remained

alone for the next three months. The male FM remained at the base of the hill near Eland feeder where

the group had initially congregated.

Figure 5. Group of eight dama gazelles in the research pasture on Morani River Ranch.

15

Associations of dama gazelles during the first eight days in the research pasture.

Association scores identify social relationships between animals. During the first eight days after release

into the research pasture, the original group of dama gazelles tended to stay together, as defined by an

association score greater than 0.5 (Table 2, pink shading). Median distance between the young adult

male GM, immature male Rm and 6 of the 8 females GF, LF, OF, VF, YF, and SF was 35 ± 18 m. The lower

association scores of females PF and TF reflect their separation from the group for two and seven days

respectively. The introduced gazelles, Bm, and YM had some contact with the original group of gazelles,

whereas BF and FM did not associate closely with each other or any other individuals.

Table 2. Association matrix for dama gazelles during the first eight days in the research pasture.

Original group Introduced animals

GM Rm GF LF OF Sf VF YF PF TF BF Bm YM FM

GM 0.73 0.53 0.75 0.65 0.56 0.67 0.63 0.27 0.09 0.05 0.31 0.21 0.02

Rm 0.73 0.57 0.87 0.62 0.61 0.68 0.68 0.21 0.07 0.05 0.36 0.28 0.01

GF 0.57 0.69 0.55 0.73 0.38 0.72 0.64 0.29 0.07 0.05 0.21 0.30 0.00

LF 0.87 0.94 0.55 0.58 0.68 0.56 0.63 0.17 0.10 0.05 0.33 0.28 0.01

OF 0.62 0.70 0.73 0.58 0.44 0.58 0.65 0.29 0.10 0.06 0.07 0.07 0.00

Sf 0.61 0.71 0.38 0.68 0.44 0.45 0.52 0.16 0.12 0.05 0.28 0.18 0.01

VF 0.68 0.80 0.72 0.56 0.58 0.45 0.68 0.28 0.07 0.04 0.33 0.24 0.01

YF 0.68 0.76 0.64 0.63 0.65 0.52 0.68 0.28 0.09 0.10 0.30 0.29 0.00

PF 0.21 0.27 0.29 0.17 0.29 0.16 0.28 0.28 0.03 0.05 0.07 0.10 0.00

TF 0.07 0.21 0.07 0.10 0.10 0.12 0.07 0.09 0.03 0.00 0.10 0.14 0.01

BF 0.05 0.07 0.05 0.05 0.06 0.05 0.04 0.10 0.05 0.00 0.00 0.00 0.00

Bm 0.36 0.41 0.21 0.33 0.07 0.28 0.33 0.30 0.07 0.10 0.00 0.28 0.00

YM 0.28 0.28 0.30 0.28 0.07 0.18 0.24 0.29 0.10 0.14 0.00 0.28 0.00

FM 0.01 0.01 0.00 0.01 0.00 0.01 0.01 0.00 0.00 0.01 0.00 0.00 0.00

Legend: The association matrix is color coded for ease of reference. Pink: Animals associated in a social group (1-0.50), Lilac: Animals having limited association with other animals (0.49-0.26). Blue: animals not associated in a social group (≤0.25).

Establishing the Research Herd.

The original main aim of the project to assess interaction between adult male dama gazelles, was

hampered by environmental problems in establishing a suitable group of dama gazelles in the research

pasture. The newly purchased male FM died within 8 days of release into the pasture. From his

progressively restricted movements revealed by the GPS collar, it is suspected that this naïve animal had

16

eaten toxic coyotillo berries. Theses berries are usually ignored by wild animals but when eaten by

livestock cause rapid and incurable, progressive paralysis and death (Marsh 1928, Lyons and Machen

2000). On the night of December 29th freezing temperatures returned followed by four days of cold with

freezing rain. This freak weather event claimed the lives of the immature male Bm, despite emergency

intervention by the ranch. Female TF, who had earlier separated from the group and the undersized

female SF, also died due to the cold. Throughout the region, many domestic livestock, as well as

numerous other exotic ungulates, perished due to this ice storm and abnormally cold, wet conditions.

Thus, these unavoidable losses were no fault of the ranch management. On January 14th, adult male YM,

was found newly dead. His GPS collar revealed that he had shown minimal movement since the early

morning of January 11th. No injury or visible anomalies were detected. Coyotillo poisoning and cold

stress are both a possibility but cause of death was unconfirmed. This left the young adult male GM as

the only adult male in the pasture.

In June, GM displayed progressive loss of condition, hiding behavior and separation from the herd.

Despite ranch staff administering treatment for internal parasites, which were causing problems in local

livestock operations following a period of heavy rainfall, this male died on June 22nd. This left no adult

male dama gazelles in the pasture until the introduction of 3 new males in August. Of these new males,

TM was found newly dead on October 26th. No injuries or abnormalities were detected and the GPS

collar indicated that he had been with other gazelles and moving normally prior to his death.

Examination of rumen contents ruled out ingestion of toxic plants but did highlight that the animal had

been eating a very limited diet consisting predominantly of blackbrush acacia leaves and pricklypear

fruits, which would provide adequate but not ideal maintenance nutrition (Fig. 6). The high energy

provided by cactus fruits should offset any

deficiency in energy intake associated with a diet

rich in blackbrush foliage (Hanselka and Paschal

1991, Ramirez et al. 1993).

Figure 6. Rumen contents of a dama gazelle.

Legend: Rumen contents indicate that this dama gazelle had a restricted diet dominated by blackbrush acacia and unripe pricklypear cactus fruits.

17

Annual Distribution of Dama Gazelles within the Research Pasture.

Each of the dama gazelles moved throughout the entire pasture (Fig. 7). Thus, the traditional 95% KHR

estimate of home range, often used to delineate animal home range, was not applicable to this study

because home ranges enveloped the entire pasture.

Figure 7. Distribution of all locations of dama gazelles throughout the pasture over 1 year.

Legend: GPS locations of all dama gazelles taken every 3 hours for up to one year for each animal.

Even 50% KHR’s used to depict the core areas favored by the gazelles tended to overlap the pasture

boundaries, however core areas are of use in identifying the areas that are most intensively used by the

gazelles and in defining shared use of space by individuals (Fig. 8).

18

Figure 8. Annual 50% KHR core areas of all dama gazelles.

Legend: Annual 50% KHR core areas of each animal are outlined in the color of their GPS collar. Black, navy blue,

teal blue, and olive green depict adult males, Red is the immature male. The remainder are the females, with light

pink denoting the introduced female BF. Individuals may have more than one core area.

The main area incorporated in the annual core areas of the gazelles encompassed the southwestern part

of the pasture where the land slopes relatively gently in a sheltered, bowl formation (Fig. 9). From this

area, several gazelles followed the valley northwards to include the northwest corner of the pasture in

their core areas. The other annual core areas, used mainly by the female subgroup of LF, VF and GF and

male NM, were in the northeast arm of the pasture, particularly the northeast area near the ranch

entrance.

The average size of the gazelles’ annual core areas within the boundaries of the pasture was 56.85 ±

12.81 ha. with the largest but non-contiguous core areas belonging to females LF, VF and GF and male

NM. Divided by gender, annual core area size was 55.47 ± 9.83 ha. for males including Rm and 57.83 ±

19

14.63 ha. for females. Thus, gender did not influence the distribution of the gazelles or size of their core

use areas.

Figure 9. Photograph of the bowl-shaped valley in the southwest of the pasture favored by dama gazelles.

Group Associations and Distribution of Dama Gazelles with the Original Young Adult Male

Present, December 19, 2014 - June 22, 2015.

Association

During the first six months of the project the group of dama gazelles in the research pasture consisted of

the young adult male GM, immature male Rm, six females GF, LF, OF, PF, VF, YF and the introduced

female BF. The original group of gazelles remained as a cohesive group with median distances between

individuals of 37 ± 13 m. Average group association (Table 3.) was 0.58 ± 0.04 (i.e. for 58% of recorded

locations the animals were within 50 m of each other). Strongest associations occurred between Rm and

LF and between GF and VF. The introduced female BF remained apart from the herd for the first three

months but gradually integrated and finally joined the group on the 96th day and remained as a

consistent group member thereafter (Fig. 10).

20

Table 3. Association matrix between dama gazelles with the original young adult male present. December 19, 2014 - June 22,

2015.

GM Rm GF LF OF PF VF YF BF

GM 0.64 0.46 0.63 0.58 0.42 0.58 0.66 0.15

RJm 0.64 0.54 0.83 0.56 0.53 0.64 0.64 0.25

GF 0.46 0.54 0.52 0.52 0.52 0.75 0.50 0.25

LF 0.63 0.83 0.52 0.53 0.52 0.66 0.60 0.21

OF 0.58 0.56 0.52 0.53 0.46 0.55 0.66 0.15

PF 0.42 0.53 0.52 0.52 0.46 0.62 0.47 0.27

VF 0.58 0.64 0.75 0.66 0.55 0.62 0.60 0.25

YF 0.66 0.64 0.50 0.60 0.66 0.47 0.60 0.16

BF 0.15 0.25 0.25 0.21 0.15 0.27 0.25 0.16 Figure 10. Incorporation of female dama gazelle BF into the original group.

Distribution

When the young adult male GM was present, the dama gazelles of the original group established two

main shared core areas within the pasture (Fig. 11.). All the gazelles used the main area of

approximately 50 ha. (124 ac.) in the southwest and an area of 19 ha. (47 ac.) in the northeast. The male

and two females, OF and YF, also spent enough time in the northwest corner to create a small 7 ha. (17

ac) core area there. The core area of female PF was extended north due to her frequent use of the

dense brush patch adjacent to the road. Introduced female BF initially resided in a valley by the Zebra

feeder in the southeast of the pasture and gradually integrated into the large southwest core area of the

original group of gazelles. BF did not spend enough time in the northeast part of the pasture for this

area to be included in her core area. The distribution of the male GM and the females was very similar

with 88.62 ± 15.38% overlap of core areas (overlap ranged from 98.14% for Rm to 53.68% for BF). Core

area overlap between females of the original group was 86.47 ± 8.83%.

0

400

800

1200

1600

0 30 60 90 120 150 180 210 240 270 300 330 360Min

imu

m d

ista

nce

to

an

oth

er d

ama

gaze

lle (

m)

Days in pasture

21

Figure 11. Core areas of female and immature dama gazelles with the original adult male present.

Group Associations and Distributions of Dama Gazelles in the Absence of Adult Males, June

22, 2015 - August 5, 2015.

Association

For six weeks during the summer there was no adult male dama gazelle in the pasture. Cohesion of the

group of dama gazelles declined (Table 4). Mean association score fell to 0.42 ± 0.23. Three females, LF,

VF and to a lesser extent GF, tended to separate from the main group and spent more time in the

northeast arm of the pasture. Association within this subgroup was less consistent than within the main

group. Female LF was often alone or with VF. VF also reduced association with other members of the

original group but retained close association with GF. Meanwhile GF was associated with both the main

group and with VF in the subgroup. Introduced female BF was now fully integrated into the original

group of dama gazelles. The immature male Rm remained in the main group which now consisted of, PF,

YF and BF, his association score with these 4 females was 0.66 ± 0.08.

22

Table 4. Association matrix between dama gazelles in the absence of adult males. June 22, 2015 - August 5, 2015.

Rm GF LF OF PF VF YF BF

Rm 0.50 0.10 0.69 0.70 0.38 0.77 0.72

GF 0.50 0.22 0.51 0.51 0.64 0.35 0.35

LF 0.10 0.22 0.08 0.04 0.36 0.02 0.09

OF 0.69 0.51 0.08 0.63 0.39 0.73 0.54

PF 0.70 0.51 0.04 0.63 0.29 0.62 0.59

VF 0.38 0.64 0.36 0.39 0.29 0.19 0.24

YF 0.77 0.35 0.02 0.73 0.62 0.19 0.58

BF 0.72 0.35 0.09 0.54 0.59 0.24 0.58

Distribution

The distribution of the group of dama gazelles at this time highlights the developing rift in the female

group once there was no longer an adult male present (Fig. 12). Overlap of individual core areas within

the group declined to 58.06 ± 32.12% (range 100 - 2.38%). All the gazelles used the main core area in the

southwest of the pasture. However, the subgroup of the three females LF, VF and GF, tended to use the

northern side of the pasture more and they expanded the core area in the northeast arm of the pasture.

Core area overlap of the five animals restricted to the southwest of the pasture remained high at 84.19

± 15.19%. Core use overlap between animals in the subgroup was lower at 62.84 ± 18.52%. This lower

degree of spatial overlap within the subgroup was due to the more extensive movements of LF. Overlap

of core areas of GF and VF was 83.77 % but their overlap of with the core areas used by LF was only

52.37%.

23

Figure 12. Core areas of female dama gazelles when no adult males were present.

Group Associations and Distributions of Dama Gazelles with New Adult Males, August 5,

2015 – December 19, 2015.

Association of females

Three new adult males were introduced to the pasture in early August. Once the new adult males were

introduced to the pasture, immature male Rm, now a subadult, no longer had close association with any

of the females. His association score with the females decreased to 0.07 ±. 0.05. The split in the female

group continued to strengthen (Table 5), as the two females LF and VF spent more time on the north

and east sides of the pasture. Association between these two animals increased from 0.36 to 0.75 and

they now had little interaction with the larger group of females. GF mainly stayed with the group of 4

females, hence the frequency of her association with VF declined. Association score within the main

group of females also declined to 0.45 ± 0.15, slightly below the threshold for being considered a

cohesive group. Strongest associations were between members of the original group, OF, PF and YF.

24

Table 5. Association matrix between dama gazelles with adult males present. August 5, 2015 - December 19, 2015.

Males Females

NM BM TM Rm GF LF OF PF VF YF BF

NM 0.08 0.30 0.10 0.21 0.29 0.08 0.07 0.33 0.09 0.08

BM 0.08 0.14 0.07 0.26 0.02 0.39 0.38 0.03 0.34 0.37

TM 0.30 0.14 0.28 0.13 0.14 0.09 0.13 0.15 0.12 0.11

Rm 0.10 0.07 0.28 0.06 0.01 0.11 0.07 0.01 0.15 0.08

GF 0.21 0.26 0.13 0.06 0.21 0.42 0.42 0.26 0.43 0.40

LF 0.29 0.02 0.14 0.01 0.21 0.02 0.04 0.75 0.06 0.10

OF 0.08 0.39 0.09 0.11 0.42 0.02 0.54 0.02 0.50 0.41

PF 0.07 0.38 0.13 0.07 0.42 0.04 0.54 0.06 0.47 0.41

VF 0.33 0.03 0.15 0.01 0.26 0.75 0.02 0.06 0.04 0.04

YF 0.09 0.34 0.12 0.15 0.43 0.06 0.50 0.47 0.04 0.48

BF 0.08 0.37 0.11 0.08 0.40 0.10 0.41 0.41 0.04 0.48

Association of males

The three new males were introduced as a group but did not stay together, even though they had

stayed close together (by a feeder) when in the adjacent pasture while acclimatizing before

introduction. Median distance between these males was 501 m. The male with the widest horns, BM,

who looked best developed and so may have been older, tended to be associated with the main group

of 4-5 females, median distance between BM and these females was 172m (564 ft.) and mean

association score was 0.35 ± 0.05. NM was most closely associated with the two females LF and VF in

the north and east part of the pasture with a similar association score of 0.31 ± 0.03 and median

distance of 316 m (1,037 ft.). These association scores were well below the threshold of 0.5 which was

the threshold indicative of consistent member of a group. The third adult male TM, also had little close

contact with any females, with an association score of 0.12 ±0.02 and a median distance of 555 m (1,820

ft. / 0.34 miles) from females. TM had most contact with immature male Rm and male NM, although the

association score of 0.29 ± 0.01 was not high enough to constitute formation of a bachelor group.

Distribution

In the four months after the three new adult males had been introduced to the pasture, the distribution

of most female gazelles remained relatively unchanged although spatial separation of the main and

subgroups became more defined. This was mainly due to the core areas of VF expanding in the

25

northwest and northeast to match those of LF (Fig. 13). The core areas of these two females now

overlapped by 85.52%. Together these two females had minimal overlap of 3.52 ± 3.82% with the main

group of four females in the southwest of the pasture. Female GF continued to associate with both the

main group and the subgroup so had 2 core areas. Her core areas overlapped 52.39 ± 27.42% with the

main group of 4 females and 44.86 ± 7.19%. with the subgroup. The 4 females YF, BF, OF, PF maintained

the same core use area in the southwest of the pasture, although the size increased from 26 ha. (64 ac.)

to 41 ha. (101 ac.). Overlap of their individual core areas was 80.40 ± 9.70%

Figure 13. Core areas of female dama gazelles after the introduction of three new adult males.

26

Figure 14. Core areas of the four male dama gazelles from August 5, 2015 to December 19, 2015.

Initially the new males moved over the entire pasture but soon began to developed distinct core areas

(Fig. 14). BM established his 36 ha. (89 ac.) core area in the southwest of the pasture and NM utilized 65

ha. (160ac.) in the northern regions of the pasture, distributed over two core areas. These two males

had the most separate of core areas with only 16.18 ± 6.64% of overlap. The third adult male, TM,

utilized 48 ha. (118 ac.) on the west side of the pasture, his core area overlapped extensively with those

of the other males by 50.69 ± 11.05%. The immature male Rm, occupied a large 58 ha. (143 ac.) core

area in the south and center of the pasture, which overlapped with the core areas used by the adult

males by 44.76 ± 17.11%.

The core area of adult male BM overlapped 88.37 ± 7.46% with the core areas of the four females BF,

OF, PF and YF and overlapped the core areas of GF by 46.13%. He did not spend much time in the north

27

and east parts of the pasture so had little contact with the subgroup LF and VF. Overlap of their core

areas was minimal at 2.42 ± 1.55%.

Since, NM mainly used the northwest and north central parts of the pasture, his core areas overlapped

with those of LF and VF by 55.89 ± 3.74% and GF by 32.13 %. Overlap by NM with the core areas of the

other 4 females was 16.25± 5.61%.

Although the third adult male TM and young male Rm had little close association with females their

home ranges overlapped extensively with those of the females. The core area of TM overlapped 59.03 ±

6.99% with the main group of 4 females, 46.15% with GF and 34.71 ± 6.88% with the subgroup of LF and

VF. The core area of Rm had 73.63 ± 7.18% overlap with the main group of 4 females, 42.46% overlap

with GF and only 8.26 ± 6.93% with LF and VF.

Association and Distribution of the Two Adult Males, after the Females have shed their GPS

Collars, December 19, 2015 to June 30, 2016.

Association

The two adult males, BM and NM, retained their GPS collars for slightly more than six months after the

females and Rm had shed their collars. During this time these males remained separate with an

association score of only 0.03. Frequency of close contact, within 10 m, between the two males was

1.35% or 21 recorded instances. The median distance between these two adult males was 884 ± 475 m.

Distribution

The two adult males established separate core areas with minimal overlap of only 1.08 ha. (2.52 ±

1.36%). Although the males moved around the whole pasture BM had a compact 50% KHR core area of

31 ha. (77 ac.) in the southwest part of the pasture while NM had more diffuse split core areas totaling

80 ha. (197 ac.) along the northern side. Thus, the males effectively divided the pasture into two

separate sections (Fig. 15). Visual observations indicated that BM continuing to associate with the larger

group of four females in the southwest while NM remained in the north and expanded his core area into

the northeast to overlap the area used by the female subgroup of LF, VF and occasionally GF.

28

Figure 15. Annual core areas of the two adult male dama gazelles.

Habitat Selection by Dama Gazelles in the Research Pasture

The distribution of gazelles within the pasture is likely not controlled entirely by social interactions.

Habitat preferences and the distribution of critical resources also shape animal distributions. The most

obvious and consistent habitat selection displayed by the dama gazelles was avoidance of steep, rocky,

slopes (χ2 = 69.00, df 10, P <0.001) with inclines of 11 – 27 degrees (20 - 50%). Part of the division of

core areas may well be determined by the steeper slopes bisecting the pasture between the northwest

and east (Fig. 16).

29

Figure 16. Google map of the research pasture showing elevation and habitat features.

Legend: Elevational map of the research pasture with contour lines in yellow. Supplemental feed and water sites are shown by red and blue triangles respectively. Roads are in white and drainages are marked in blue.

Soil based ecological site characteristics had relatively little effect on the overall distribution of dama

gazelles in the pasture (χ2 = 1.42, df 3, P >0.05). Some, but not all, individual animals favored the low-

lying drainage areas with deeper DE and EK soils (χ2 = 27.49, df 10, P <0.01). Animals selecting the DE

area in the east side of the pasture were females LF, VF, and GF. Preference for the EK in the

southwestern part was due to extensive use of this area by the male BM. The areas of deeper soil

tended to be of gentle slope and more heavily vegetated. Thus, they may provide more shelter from

inclement temperatures. However, the gazelles did not show selection for or against more densely

vegetated areas plotted from aerial photography of the ranch (χ2 = 14.55, df 10, P >0.05).

In this rocky pasture, the gazelles were often seen using the smooth caliche dirt road to move across the

pasture. As a group, it appeared that gazelles favored these roads as travel routes (χ2 = 40.79, df 10, P

≤0.001), but further examination indicated that much of this statistic was due to the distribution of the 3

females using the narrow northeast arm of the pasture. In this area, the road parallels the creek bed

along the valley floor and is bounded by the fence to the north and by steep hillside to the south. Thus

30

the distribution of these 3 females near the creek bed (χ2 = 31.12, df 2, P ≤0.001) cannot be separated

from their proximity to the road (χ2 = 28.00, df 2, P ≤0.001) or vice versa. Animals using the larger

western part of the pasture did use the roads but only in proportion to their extent within the habitat

(χ2 = 12.79 df 8, P >0.05). Also, the gazelles frequenting the western part of the pasture showed no

overall attraction the to the creek and drainage areas (χ2 = 3.27 df 8, P >0.05).

The ranch provided supplemental food and water at 5 locations within the pasture. There was no

evidence that the gazelles made much use of these locations. (Feeders χ2 = 12.44, df 10, P >0.05: Water

χ2 = 0.07 df 10, P >0.05). The only animal showing any attraction to feeders was the adult male BM. It

should be noted that because the herds of larger animals crowded out the smaller animals as soon as

feed was placed in the feeders, there was usually little feed available for the dama gazelles.

DISCUSSION

Despite initial difficulties establishing a multi-male group of dama gazelles on the study ranch, due to

the effects of unusual adverse weather conditions, the study achieved its goals which were:

1. To identify home range size and interaction between male dama gazelles sharing a 202 ha. pasture.

2. To identify association within the female herd and the interactions between male and female dama

gazelles.

3. To describe habitat selection and effect of landscape structure on the distribution of dama gazelles in

a Texas Hill Country pasture.

4. Provide recommendations for the conservation and management of multi- male groups of dama

gazelles in rangeland pastures.

Home Range Size and Interaction between Male Dama Gazelles

In Phase 1 of this project, we showed that in extensive semi-desert pastures contact between adult male

dama gazelles was minimal due to their maintaining extensive home ranges each incorporating a non-

overlapping 50% KHR core area of about 440 ha. (1,088 ac.). In the current study the entire pasture size

of 202 ha. (499 ac.) is more typical of the space available for maintaining dama gazelle herds on

rangeland (Mungall 2004) but is less than the size of the core area of a male dama gazelle in West Texas;

thus, greater contact between males, and greater potential for conflict, may be expected.

31

In this study, both male and female dama gazelles moved throughout the entire pasture. Calculated 50%

KHR core areas tended to overlap the pasture boundaries to varying extents but have value in describing

shared areas and spatial separation of individuals. Within the boundaries of the pasture, size of the core

area size of both male and female dama gazelles was 57 ± 13 ha. (140 ±32 ac)

Initially, when there was only one young adult male dama gazelle (GM) in the pasture, he was closely

associated with females of the original group that had been kept together before transfer to the

research pasture. These dama gazelles moved throughout the entire pasture but mostly favored the

shallow, west facing, bowl-shaped valley of 47.5 ha. (117 ac.) in the southwest of the pasture, and to a

lesser extent, the northwest corner of the pasture. They also had a second core area in northeast corner

of the pasture near to their former enclosure.

Later, when two different adult males were present in the pasture, these animals effectively partitioned

the pasture between themselves by each developing core areas with minimal spatial overlap. The male

with slightly larger horns (BM) maintained a single core area based on the favored valley which had

been the southwest core area of the initial male and was favored by the larger group of females. The

other male (NM) mainly used the north side of the pasture where he developed three core areas. These

included the northwest and northeast core areas of the initial male, plus a large area of 42 ha. (104 ac.)

in the central region extending to, but not overlapping with, the core area of BM. Although both males

still ranged over the whole pasture, these separate core areas served to limit the frequency with which

they came into contact with each other. Once the core areas were established, only 3% of paired GPS

locations placed these males within 50 m. of each other. While this frequency of close contact is higher

than contact rates between males in West Texas, it is still strongly indicative of adult males maintaining

social separation.

The third introduced adult male (TM), was only present in the pasture for three months. TM behaved

more like a nonbreeding male in that his core area on the west side of the pasture overlapped

extensively with the core areas of the other males, yet he had relatively little contact with them. The

immature male Rm that had shared distribution with the initial adult male had matured to a subadult by

the time the new males were introduced. Once the new adult males were introduced to the pasture this

young male no longer associated with the females. Despite extensively spatial overlap of core areas, the

rate of contact of Rm with the dominant two adult males was less than 1%, which is indicative of his

avoiding contact with these males.

32

Thus, even in a pasture no larger than a single core area used by a dama gazelle in arid West Texas,

there was adequate space for two adult males to partition the pasture into separate core areas and

hence have very little contact with each other. While there was not enough space for the younger male

to have a core area separate from the adult males, there was adequate space for him to avoid the adults

temporally. No fighting injuries were reported for the male dama gazelles present during the study.

Whether there is space for three or more adult males to continue to coexist peacefully remains to be

seen.

It should be noted that the three new adult males had been kept together as a bachelor herd for over a

month prior to their release into the research pasture, so they may already have established their social

order in the absence of females. The high aggression of the initial male YM towards another male in the

neighboring pasture suggests that not all males may be suitable for inclusion in multi-male herds.

Association within the Female Group and Interaction with the Males

Dama gazelles are browsers feeding on scattered bushes. Thus, they do not display the tight herd

structure seen in grazing ungulates. In this Texas Hill Country pasture with approximately 50% shrub

cover, the median (most common) distance between members of the original group of dama gazelles

was 37 m (120 ft.).

While the young adult male was present, the group remained together with animals staying within 50 m

of each other for 58% of the time. However, the females did not always move as a cohesive group and

hence the size and shape of core areas of individual females varied slightly. The adult female introduced

to the pasture at the beginning of the study (BF) gradually integrated into the group over a period of

three months, showing that unrelated females can be integrated into established groups. Association

within the female group was greatest when the young male was present. Social disruption by this male

may have been minimal because he was already established within the group when the study began.

The fact that no offspring were produced as a result of his presence suggests that he may not yet have

been mature enough to be accepted by the females as a breeding male (E. C. Mungall pers. comm.).

Never the less, his presence appears to have kept the group together. Once this male was no longer in

the pasture, a split started to develop within the group of females. Three females gradually separated

out to form a subgroup inhabiting the north of the pasture. Composition of this subgroup was not

consistent, since one female regularly moved between the subgroup and the main group.

33

When new adult males were introduced to the pasture, the larger horned male (BM) established his

core area in the southwest overlapping the main group of 4-5 females and the other (NM) was found

mainly in the northern area, initially in the northwest and north central areas but then expanding his

range into the northeast area frequently used by the subgroup of 2-3 females. Although the males used

the same core areas as groups of females, the association scores indicate that the relationship between

the new males and the respective female groups was nowhere near as close as that of the original male.

During the four months that both females and new males all had working GPS collars, the association

scores of the females generally did not increase, in fact, some decreased, this suggests that the males

were not keeping the female groups together. Video of animal movements indicated that males tended

to follow females rather than direct their movements. However, despite the low association scores

between males and females some breeding did occur, and in the following months 3-4 offspring were

produced (C. Reid pers. comm.).

Effect of Habitat Selection on the Distribution of Dama Gazelles

Physical attributes of landform, habitat and man-made structures may also influence the distribution of

animals. Within the pasture, the ecological (rangeland) site type, and thus vegetation, was rather

uniform. The location of core areas used by the female gazelles and successive males indicate that the

area most preferred by the animals was the shallow bowl-shaped valley in the southwest, offering

gentle slopes of varying aspect with low lying areas to provide water and shelter. Yet, the smaller

northern core areas were quite dissimilar. The favored area in the northwest consisted of a west facing

hillside. The area in the northeast incorporated a cobbled dry creek-bed with taller tree cover. However,

within this area the gazelles were often seen on the hillside to the south of the ranch entrance gate.

The prior study in West Texas indicated that dama gazelles avoided the most rugged areas of the

pasture with steep slopes. This same selection was evident for gazelles in the smaller pasture, even

though slopes were less extreme. The gazelles clearly selected against hillsides with slopes of 20-50%.

Avoidance of the north facing ridge of steeper slope that bisected the pasture most likely helped define

the separation of the core areas used by the two male dama gazelles sharing the pasture.

In thorny shrubland habitat, it is not uncommon for animals to use ranch roads for ease of travel

(Cooper et al. 2008). Both the original group of dama gazelles, and later the new males, followed the

roads when first released into the research pasture. The dama gazelles were often seen on the road but

in fact use of roads was in proportion to their surface area within the pasture. When in the narrow

eastern arm of the pasture, the gazelles showed selection for the creek and road but this was likely due

34

to confinement by the fence to the north and the steep hillside to the south. Within the remainder of

the pasture, no such habitat selection was evident. Proximity to the main road was associated with the

gazelles traversing the gentler slopes of the saddle between the hills when crossing the pasture from the

north side to the favored bowl area in the south.

When the gazelles were first released, they returned at night to the holding pen where pelleted

supplemental food and water were available. Yet the GPS collars indicate that after the holding pen was

closed the gazelles spent little time near the supplemental feed and water locations in the pasture. As a

desert adapted species, the gazelles probably obtained much of their water from the vegetation and

from rainwater pools on the rocks. Access to the feeders was often blocked by the aggressive presence

of the herd of longhorn cattle and a large herd of scimitar-horned oryx with which they shared the

pasture. The gazelles were able to maintain themselves on native vegetation, although the limited

diversity of plant species revealed from the rumen sample from male TM suggests that nutritional

deficiencies may occur.

Recommendations for the Conservation and Management of Multi-Male Groups of Dama

Gazelles in Rangeland Pastures.

Phase 2 of this project has shown that dama gazelles can be kept in multi-male herds within a pasture as

small as 202 ha (499 ac.). Sharing of smaller pastures by adult males is likely aided by the greater

productivity of land in the Hill Country as opposed to the semi-desert areas of the previous phase of the

study. Greater vegetative productivity provides ample food resources and better visual screening cover

for competing males. At present, there are only two fully adult male and one newly adult (Rm) males in

the pasture. To learn more, further monitoring would be required as males born in the research pasture

mature and join the breeding population, or as further new males are introduced.

35

SUMMARY Adult male dama gazelles shared use of a moderate sized (202ha., 499 ac.) pasture without

conflict by concentrating their activities in spatially separate areas.

When there is not enough room for males to have separate, non-overlapping core areas they

avoid each other by temporal separation.

Keeping multi-male groups of dama gazelles in moderate sized pastures is feasible if there are

adequate resources and visual cover for the males to achieve spatial and temporal separation

from each other.

Female dama gazelles associated in loose groups, into which new females could be incorporated

or subgroups could break off.

While the presence of an established male may have aided cohesion of female groups, the

introduction of new males did not influence the distribution of female groups.

Core home range size, distribution and habitat selection did not differ between male and female

dama gazelles.

This study only included a maximum of three adult males and one maturing male. While no

conflict was seen between males, further observations will be needed to determine the total

number of male dama gazelles that can be kept simultaneously in a pasture of this size.

ACKNOWLEDGEMENTS

This GPS-radio collar investigation of dama gazelles has been a project of the Second Ark Foundation in

collaboration with the Exotic Wildlife Association. The support which allowed this project to go forward

came from the Bisbee’s Fish & Wildlife Conservation Fund (Wayne Bisbee and Brian White), with added

support from Fossil Rim Wildlife Center (Pat R. Condy). Kevin and Cole Reid generously offered their

Morani River Ranch in the Texas Hill Country as a study site.

Susan Cooper and her team of Shane Sieckenius and Andrea Silva from Texas A&M AgriLife Research in

Uvalde, Texas, did an expert job fitting the collars and then were instrumental in the tracking and data

analysis. Working with Cole Reid for the collaring sessions and later were Principal Investigator Elizabeth

Cary Mungall and photographer-analyst Christian Mungall for the Second Ark Foundation, the AgriLife

team, ranch personnel John Fredericks, Johnny Flores, Kenneth Haynie, Logan Bostic, and Tim Crumley.

Special videographer for the initial collaring session was Anthony DellaFlora assisted by John Britt.

Veterinarian was Cecil Armin DVM assisted by Amy Sieckenius.

36

LITERATURE CITED

Barbosa, A., and G. Espresso. 2005. International Studbook Gazelle dama mhorr. Consejo

Superior de Investigaciones Cienrificas Press, Madrid, Spain.

Cooper, S. M., H. L. Perotto-Baldivieso, M. K. Owens, M. G. Meek, and M. Figueroa-Pagán 2008.

Interaction between white-tailed deer and cattle in a semi-arid grazing -system.

Agriculture, Ecosystems and Environment 127:85-92.

Ginsberg, J. R. and T. P. Young. 1992. Measuring associations between individuals or groups in

behavioural studies. Animal Behaviour 44:377-370.

Hanselka C. W. and J. C. Paschal. 1991. Prickly pear cactus: a Texas rangeland enigma.

Rangelands 13:109-111.

Lyons, R. K.; and Machen, R. V. 2000. What we know about coyotillo. http : / /hdl .handle .net

/1969 .1 /86894.

Marsh, C. D., 1928 Coyotillo (Karwinskia humboldtiana) as a poisonous plant. USDA Technical

Bulletin No 29. USDA Washington DC.

Mungall, E. C. 2004. Dama gazelle phone census. Exotic Wildlife, January−February. Pp. 11–12.

Mungall, E. C. 2014. SAF project update: help reduce fights. Exotic Wildlife, Summer. P. 13.

Perotto-Baldivieso, H. L., S. M. Cooper, A. F. Cibils, M. Figueroa-Pagán, K. Udaeta, and C. Rubio.

2012. Detecting autocorrelation problems from GPS collar data in livestock studies.

Applied Animal Behaviour Science 136:117–125.

Ramirez, R.G., E. Rios, and J. Garza. 1993. Nutritional profile and intake of forage grazed by

Spanish goats in a semi-arid land. Journal of Applied Animal Research 3:113–122.

RZSS & IUCN Antelope Specialist Group. 2014. Conservation review of the dama gazelle (Nanger

dama). Royal Zoological Society of Scotland, Edinburgh, UK.

Seaman, D. E., and R. A. Powell. 1996. An evaluation of the accuracy of kernel density

estimators for home range analysis. Ecology. 77:2075–2085.

USDA Natural Resource Conservation Service Soils. 2014.

http://websoilsurvey.nrcs.usda.gov/app/

Weather Underground. 2016. http://www.wunderground.com/history.

Whitehead, H. 2008. Analyzing Animal Societies: Quantitative Methods for Vertebrate Analysis.

University of Chicago Press, Chicago. IL.

37

Wright, B., R. Lyons, S. Cooper, and J. Cathey. 2002. White-tailed deer browse preferences for

South Texas and the Edwards Plateau. Texas Cooperative Extension Publication B-6130.

Figure 17. Dama gazelle collaring team.

The initial dama gazelle collaring team at Morani River Ranch, left to right: Principal Investigator

Elizabeth Cary Mungall, Operations Manager Cole L. Reid, videographer Anthony DellaFlora, ranch hands

Johnny Flores, Logan Bostic, and Tim Crumley, Texas A&M AgriLife collaborator Susan M. Cooper,

AgriLife associates Shane Sieckenius and Andrea Silva, project photographer Christian Mungall, not in

picture photographer John Britt.

38

RECOMMENDATIONS FOR FUTURE STUDY

Use of GPS collars on animals provides a wealth of data. This report provides an initial

examination of the relationships within and between male and female dama gazelles but much

more information can be gathered from the data, given time and more sophisticated analytical

tools. For example, the brief but powerful interaction of males when the three adult males

were added to the study pasture forced a significant change in social dynamics of the dama

gazelles, this was identified the tracking capability of Dr. Christian Mungalls’ program

MOVEPROG (See Appendix) but was not picked up by the initial analysis. In the future,

MOVEPROG can be used to characterize the interaction of males with females, answering the

question of whether males stay constantly with females or just interact briefly, possibly

following or chasing specific females at particular times. Likewise, female separation for

parturition is not easily identified in traditional methods but in future studies such events may

be identified in MOVEPROG (extended as required). Such questions are important for managers

working with these animals. In addition, development of MOVEPROG programs for the study of

dama gazelles can easily be adapted to describe the social behaviour of a variety of ungulate

species. Plans are underway to extend MOVEPROG to look not only at individuals but also to

identify separations between classes of animals.

Development of MOVEPROG

MOVEPROG has a geospatial component (latitude, longitude, and elevation). Thus, it can be

used to identify locations were animals pass through rapidly with long, direct tracks or linger

with short, many-angled tracks. Thus, development of the program have implications for

greater understanding of resource selection and factors influencing fine scale distribution of

animals, such as:

1. Habitat selection: Short, twisting movements keep animals in good food patches, for example

in patches of oaks when the acorns are dropping versus long movements used to cross

unproductive areas. Such movements can be demonstrated by fractal geometry but are far

easier to identify and understand in MOVEPROG.

39

2. Predator avoidance: As a prey species, a gazelle would be expected to move rapidly through

dense brush by the creeks where there is cover for predators but to relax and move slowly on

the open hillside with good visibility. Lack of affinity of gazelles for densely vegetated patches

suggests this is occurring but MOVEPROG could identify the mechanism by which gazelles avoid

dangerous areas. Similar behavior would be appropriate for young males moving through

territory held by more a dominant male.

3. Thermoregulation: Long (rapid) movements take animals through uncomfortable areas and

short movements keep them in comfortable areas. As we saw in this project when the gazelles

moved onto the tops of the sunny eastern slopes at sunrise, presumably to warm up.

Such movements are important to the ecology and management of dama gazelles but are hard

to describe by conventional methods. Development of MOVEPROG will be a big step forward in

such analyses.

Use of SOCPROG in combination with MOVEPROG:

The software program SOCPROG (written by Hal Whitehead of Dalhousie University, Halifax,

Nova Scotia) is a series of programs, written in MATLAB, to analyze data on social and

population structure of individually tracked animals, such as the dama gazelles in this study.

The program can be used to assess group membership, interaction rates, attraction-avoidance,

dominance hierarchies, and more. SOCPROG, however, as currently implemented, is not set up

to display and analyze track data obtained from GPS position data taken at constant time

increments.

Social interactions, or placement of animals within the group, is not always constant but may

change during different phases of activity. MOVEPROG can identify times of differing activity

based on rates of movement (travel vs. sedentary), and then SOCPROG can be used to identify

animal interactions during these times. For example, when a group of animals is moving rapidly,

or as if with purpose, they may display a social hierarchy with established leaders and followers,

but once they reach their destination there may be a different social organization, e.g. A

dominant female may lead a group to a feeding area but once there she may take a place in the

center of the group for maximum safety. Alternatively, a male may lead his group to water then

40

take a place on the periphery of the group as if to stand guard. While the long-range

movements of animals in this study are limited by fencing, for future studies of free-ranging

animals a combined analysis using MOVEPROG and SOCPROG in combination could be quite

illuminating.

In any case, it should be recognized that this Second Ark Foundation and Exotic Wildlife

Association dama gazelle study made possible by support from Bisbee’s Fish &Wildlife

Conservation Fund, supplemented by Fossil Rim Wildlife Center, has presented an unusual

opportunity because all of the individuals of the study species wore GPS collars, not just a

sample. Given the cost of such collars and the difficulty of catching the animals, this is a unique

and unusual situation. Thus, it presents an unusual opportunity to explore ways to investigate

social relationships over time, because the movements and interactions of all individuals can be

accounted for simultaneously. This has implications for the advancement of everyone’s GPS

collar research for many species going into the future. The Second Ark Foundation and the

Exotic Wildlife Association should be justly proud of their role in advancing this increasingly

popular technology.

41

APPENDIX 1. Further Analysis of GPS Data from this Study

Introduction

In this study, we conducted the initial analysis of the data using the software program ArcView by ESRI

in combination with Excel spreadsheets. Given time, much more information can be extracted from the

geospatial data set than is presented in this report. The data set, with its approximately 26,000 sets of

location data generated by the GPS collars, is unusually complete: it is rare that all the adult animals in a

study have collars. The rarity and size of the data set is such that standard tools for extracting all

possible information are not generally available, and special approaches are needed to do this.

Two approaches seem appropriate to extract all useful information from the dama gazelle data set that

are of relevance to ranchers. These are: 1. to write specific code to analyse groups and how they move,

and 2. to take advantage of the few existing social analysis packages that are available.

1. To write software that will efficiently permit an overview of the data so that a) a summary of

routine behaviors can be obtained, and b) periods worthy of detailed analysis can be identified.

Amongst items to be determined first are the routine situations - how the animals move in

groups or as individuals –then during key happenings that might be called “incidents”.

2. To keep in mind the possible use of a well-known animal social analysis package called SOCPROG

(written by professor Hal Whitehead of Dalhousie University) that can be used to permit an

analysis of the dama gazelle social dynamics during the periods selected using the first software

package. Of particular interest, is to determine how groups of females form and split off, how

groups of females move with an ‘attendant’ male, how (if they do) groups of females move

through the core areas of males, and for how long these movement patterns persist – to name

but a few.

It should be noted that any future detailed analysis will draw heavily on the contents of the present

report.

42

Overview of software development aimed at analysing animal movement

A considerable amount of progress has been made since MATLAB was obtained in May, 2016 – mostly in

the area of writing software and testing the results for relevance. The work will be described in the rest

of this appendix.

The work to date is of an exploratory nature, and has been undertaken by Dr. Christian Mungall without

benefit of funding. Consequently, Christian (an offshore engineer by training) has elected to write his

software package (called MOVEPROG) in MATLAB – a software development package familiar to

mathematicians, engineers, physicists, and the like, by MathWorks. (MATLAB stands for Mathematical

Laboratory). MATLAB consists of some 70,000 functions which can be combined to provide numerical,

graphical, and cluster analysis capabilities with the goal being to provide additional, fine scale insights

into social structure of dama gazelles on rangeland pastures.

Development of the animal movement analysis package MOVEPROG

Thanks to a grant facilitated by Dr. Pat R. Condy from Fossil Rim Wildlife Center, the Second Ark

Foundation was able to supply the funds for use of MATLAB with selected add-ons during this project.

The license for MATLAB began in May 2016 and will run for one year. Of the many supplementary

materials available for MATLAB, a spreadsheet link, a statistics and learning toolbox, and a mapping

toolbox were eventually chosen. With MATLAB able to handle sorting through the thousands of data

points generated by the GPS collars, Christian could identify time periods likely to show events

appropriate for closer examination. MATLAB could then interface smoothly with programs like Excel to

investigate and graph special cases. The researchers are very grateful to all those people involved with

making MATLAB available for this work.

MOVEPROG consists of a suite of functions written with the express purpose of facilitating the

preliminary analysis of GPS collar data (latitude and longitude data, elevations and temperatures), with

emphasis on looking at track data. Christian has used his GPS track analysis software MOVEPROG to

describe periods of regular and irregular activity.

MOVEPROG starts with what might be termed a ‘data conditioning session’ in which missing data are

replaced by MATLB-compatible variables. (Erroneous positions had previously been identified and

manually set to a missing data value for the most part, in Excel.)

Following this, several display options are provided as in the following illustration.

43

Some examples follow of the identification of the formation of one or

more female groups and the locations of “singletons” (animals – male or

female) that remain by themselves away from groups. Use was made of display type #20.

As is the case for much of the previous work in this report, the development of MOVEPROG commenced

with the implementation of a tool in Excel that lets one identify brief out-of-the-ordinary activities that

get lost in more basic summarization of the data. An example from Excel is shown below when, in the

period around August 2015, three adult males were introduced into a pasture that contained seven

females and one subadult male called Rm. Closely after the introduction, the maturing young male Rm

was evicted from the female group by the incoming adult males (see three following figures created in

Excel). The figures show first a period of quiet, then identify the chasing of the resident young male (Rm)

by newly introduced adult males and the resultant distancing from the female group.

0.0

200.0

400.0

600.0

800.0

1000.0

1200.0

1400.0

2/10/20152/25/20153/12/20153/27/20154/11/20154/26/20155/11/2015

Rm distance to males

44

The above figure demonstrates (using Excel) a peak in distance between maturing young male Rm and

the new males that occurred on August 5th, the very day that the males were introduced to the pasture.

Distances between Rm and female members of his natal group increase at this time. Rm returns to the

group for a brief period before finally breaking his formerly close ties with the females. TO BE NOTED IS

THAT THE ABOVE TYPE OF DISPLAY IS TIME-CONSUMING TO SET UP IN EXCEL. IF THIS IS PROGRAMMED

IN MATLAB, THE USER CAN BE GIVEN THE CHOICE OF WHICH ANIMAL TO CHOOSE AS THE SUBJECT, AND

WHICH ANIMALS TO CHOOSE AS THE TARGETS.

Given the evidence of this ‘incident’, it is natural to want to look at the composition of the female group

and the location of the young male before the incident, and the composition of the female group(s) and

the location of the three adult males and the young male after the incident. Use is made of the locations

0.0

200.0

400.0

600.0

800.0

1000.0

1200.0

1400.0

6/10/2015 6/25/2015 7/10/2015 7/25/2015 8/9/2015 8/24/2015 9/8/2015

Rm distance to males

0.0

200.0

400.0

600.0

800.0

1000.0

1200.0

1400.0

6/10/2015 6/25/2015 7/10/2015 7/25/2015 8/9/2015 8/24/2015 9/8/2015

Rm distance to females

45

of the animals at midnight to provide an approximation of the groups. (This can be checked by looking at

successive plots of the animal locations over a number of consecutive time steps.)

CASE 1: FEMALES, YOUNG MALE

Full and Midnight scatter plots for period 04/12/2015 to 04/18/2015 when females, adult male GM and

young male Rm present (in ‘reasonable’ harmony). First: a plot of groups of animals over 7 consecutive

midnights:

This is followed by a plot of one particular midnight: Note two groups of females (one with the young

male), a group of two females, and a singleton female.

46

Next follows a close-up of the female group with the young male.

This was the (relaxed) status quo of the situation before the introduction of three adult males.

47

CASE 2: FEMALES, THREE ADULT MALES, YOUNG MALE

Full and Midnight scatter plots for period 08/23/2015 to 08/29/2015 when females and Rm and three

new adult males present. Rm has been partially ‘driven out.’ First a plot of the locations of all animals at

three hour intervals over a period of one week. Note how the young male is by himself for the majority

of the time. (The term “Gscatter” in the title refers to a particular type of MATLAB plot.)

Next follows a plot showing the

midnight positions of all the

animals for seven consecutive

nights. The formation of two

groups of females, each with one

or more adult males, is clear.

Then follows a plot for one

particular midnight, along with

two close-ups displaying the two

different female groups.

48

49

The above show, without the need for explanation, the way in which groups can be visually portrayed.

The reader may wonder to what extent midnight positions – with their presumed low activity – are

representative of groups during daylight hours. This is graphically addressed in the next three plots.

POSITIONS FOR 03:00, 06:00 AND 09:00 HOURS ON 08/27/2015

First a plot for 08/27/2015 03:00. The dots represent the positions at the indicated times; the ‘tails’

represent the movement over the preceding three hours (and are thus an indication of the movement

activity of the animals prior to the indicated time. This is followed by a plot for 08/27/2015 06:01.

50

Next follows a plot for 08/27/2015 09:01 along with two close-ups.

The plots provide a good simulation of what an observer might view from an elevated position. However

– thanks to precise GPS position information – accurate identification, locations, and estimates of

velocity and acceleration can by determined – and much else.

51

In summary, it is hoped that the above examples show what can be achieved by custom-written

software like Christian Mungall’s program MOVEPROG. Of significance is that the analyst – given basic

programming skills – can produce accompanying analysis software that can be used to quantify aspects

such as the degree of ‘groupiness,’ (to coin an expression) and explore the effectiveness of the analysis

methodology. Then, when satisfied, that analysis software can be applied to months of data, plotted,

and used to explore group patterns with a view to selection periods for detailed analysis – using

MATLAB, ArcView, R, or other tools of choice.

Some examples of dama gazelle analysis from the SOCPROG package

The following examples of SOCPROG output were derived by Christian while attending a SOCPROG

workshop at Dalhousie University, Halifax, in August 2016. The output is of the first month of dama

gazelle GPS collar data. Figures include data input screens and results screens.

Typical input data screen shot:

52

File: damaoneyearinputSOC.xls

Sampling period: Day*8

Restrictions: Record<2431

minimum distance of 100 and minimum time of 0.1

Association index = Simple ratio

BF 1.00

GF 0.06 1.00

GM 0.09 0.88 1.00

LF 0.13 1.00 0.84 1.00

OF 0.09 1.00 0.88 1.00 1.00

PF 0.06 0.84 0.75 0.84 0.84 1.00

RJ 0.06 1.00 0.88 1.00 0.97 0.81 1.00

VF 0.09 1.00 0.88 1.00 1.00 0.88 1.00 1.00

YF 0.13 1.00 0.88 1.00 1.00 0.84 0.97 1.00 1.00

YM 0.00 0.56 0.41 0.56 0.53 0.53 0.47 0.56 0.56 1.00

BF GF GM LF OF PF RJ VF YF YM is of first month

Screen shot at start of Association Analysis:

53

Input screen for creating a network diagram:

Resulting Network diagram: Principal coordinates, circular

54

Cluster analysis results: cophenetic correlation coefficient = 0.99566

This concludes examples of what can be achieved using the SOCPROG package.