POTENTIALITIES FOR USING CERTAIN MODERN TECHNOLOGIES FOR THE TRACKING AND MONITORING OF FREE-ROAMING HORSES

Abstract

With reference to the implementation of the areas-`Traditional practices for seasonal grazing of animals` and `Conservation of endangered local breeds`, the interest in free grazing of various farm animals in Bulgaria has been significant in recent years. Horses are particularly suitable for this type of breeding. Pastures are often located in remote areas with limited access, which makes it difficult to visit and inspect the herds and facilities in the pastures on a daily basis. In order to find modern technological solutions to solve these problems and reduce costs and efforts of farmers, we tested several modern devices that are traditionally used in other areas, and their application in animal husbandry in Bulgaria is an innovative approach. These are GPS (Global Positioning System) for tracking animals, photo traps, as well as drones. As a result, we found that they have a successful application in monitoring horses which are raised free grazing, save costs and time, do not cause stress and side effects. These devices can be used to control access to pastures and limit theft, harassment and other encroachments on herds.

Full text

SCIENTIFIC PAPERS
SERIES D. ANIMAL SCIENCE
Volume LXVI, No. 2
BucharesT
2023
Faculty of Animal Productions
Engineering and Management
University of Agronomic Sciences
and Veterinary Medicine of Bucharest
SCIENTIFIC PAPERS. SERIES D. ANIMAL SCIENCE. Vol. LXVI, No. 2, 2023

370
POTENTIALITIES FOR USING CERTAIN MODERN TECHNOLOGIES
FOR THE TRACKING AND MONITORING OF FREE-ROAMING HORSES
Maya POPOVA1, Pavlin ZHELEV2, Gradimir GRADEV3
1Agricultural University - Plovdiv, 12 Mendeleev Blvd, Plovdiv, Bulgaria
2Eco karta LTD, Pastrogor, Bulgaria
3Green Balkans, 9 Stara Planina Str., Stara Zagora, Bulgaria
Corresponding author email: m_popova1@abv.bg
Abstract
With reference to the implementation of the areas - `Traditional practices for seasonal grazing of animals` and
`Conservation of endangered local breeds`, the interest in free grazing of various farm animals in Bulgaria has been
significant in recent years. Horses are particularly suitable for this type of breeding. Pastures are often located in remote
areas with limited access, which makes it difficult to visit and inspect the herds and facilities in the pastures on a daily
basis. In order to find modern technological solutions to solve these problems and reduce costs and efforts of farmers,
we tested several modern devices that are traditionally used in other areas, and their application in animal husbandry in
Bulgaria is an innovative approach. These are GPS (Global Positioning System) for tracking animals, photo traps, as
well as drones. As a result, we found that they have a successful application in monitoring horses which are raised free
grazing, save costs and time, do not cause stress and side effects. These devices can be used to control access to pastures
and limit theft, harassment and other encroachments on herds.
Key words: behavior, conservation, GPS, sensors, wild horses.
INTRODUCTION
Karakachan horses are an old autochthonous
breed, preserved in relatively the same form in
which it existed for centuries (Petrov, 1940;
Sabeva, 2009; Popova & Etarska, 2020). In this
regard, research on Karakachan horses will give
us the opportunity to collect data on a breed that
has preserved genes from ancient times.
Today in Bulgaria, this breed is grown predo-
minantely, freely all year round in the typical
habitats - pastures in high parts of mountains. In
most cases, they inhabit remote pastures,
isolated from settlements and buildings. These
horses have a hard time accepting rearing in
stables and other buildings. This makes it
difficult to carry out direct studies on the
Karakachan horses, and many of the characte-
ristic features of this resistant breed remain
unknown compared to other modern barn-raised
breeds allowing detailed studies of the latter.
With the use of modern technologies, we can
afford to study horses in remote locations and in
their typical habitats. The collection of data on
behavioral and defensive responses, social beha-
vior of the horses, such as group dynamics (group
sizes and membership), locomotion dynamics,
grazing intensity, in addition to allowing for a
better knowledge of breed characteristics, will be
of benefit also for purely practical applications
(White & Garrot, 1990; Anderson & Lindzey,
2003; Cagnacci et al., 2010).
At the same time, the development and
application of modern technologies directly in
the field will promote among the owners the
possibilities of modern animal husbandry and its
benefits, as well as the creation and application
of information technologies through the creation
of databases (Osechas, 2008; Hampson et al.,
2010; Bachmann et al., 2014; Burla et al., 2014;
Collins et al., 2014; Mann et al., 2014; Radoi et
al., 2015; Burov, 2018).
The purpose of the research is to study basic
ethological signs and the choice of habitat over
a large area by the Karakachan horses from the
national gene pool, through the use of GPS
trackers and other technologies for monitoring
and following horses in ecologically and
biocompatible breeding.
MATERIALS AND METHODS
The study was conducted in the areas of the
village of Prisadets (on the Bulgarian/Turkish
Scientic Papers. Series D. Animal Science. Vol. LXVI, No. 2, 2023
ISSN 2285-5750; ISSN CD-ROM 2285-5769; ISSN Online 2393-2260; ISSN-L 2285-5750

371
border - within Bulgaria) and the village of
Levka, falling within SPA Sakar (BG0002021),
which overlaps with SCI (BG0000212), part of
the ecological network NATURA 2000
(MOEW, 2013). In biogeographical terms, the
area falls into the Southern biogeographical
region and, more specifically, according to the
biotic basis, it refers to the "Dolnomarishko -
Dolnotundzhansky" subregion (Gruev &
Kuzmanov, 1999), as Mediterranean influence
penetrates the sub-region along the Maritsa and
Tundzha rivers. This defines the climate as
milder and allows the horses to be kept outdoors
all year round.
The object of observation were two herds of
horses from the Karakachan breed, which are
free-roaming horses all year round. The first
herd consists of 29 mares, 1 stallion and 10
foals. The second herd consisted of 1 stallion
and 12 mares.
The observations were carried out within 12
months. Twelve (12) and twenty-four (24) hour
field observations were conducted.
Ethological studies were carried out - by the
method of visual observation as well as tracking
the choice of habitat with GPS-trackers, etc.
Five GPS trackers were purchased, 2 of which
were placed on the stallions of the 2 herds and
the rest were placed on mares from the same
herds to attempt to track the sexual behavior of
the free range horses.
The frequency of reporting Data from the GPS
trackers was set to once every 15 minutes for a
period of 12 months.
ArcMap ver.10.0 was used to process the data
from the GPS trackers from the *.csv file sent by
the transmitters and convert them into a *.shp
file. Google Earth Pro ver. 7.3.6.9285. was used
to visualize the positions of the horses on a
satellite image of the Earth's surface. We used
Microsoft Excel for the primary processing of
the information.
RESULTS AND DISCUSSIONS
Animal tracking and activity monitoring using
wireless sensors provides high-resolution data to
retrieve their location and study their behavior.
These quantitative measures are useful for better
formulating ideas and informing practices in
animal ecology, such as resource use, home
range, animal dispersal, and population
dynamics, which have so far relied on visual
observation (Cagnacci et al., 2010).
Dorio stallion herd. The stallion Dorio was
marked on 18.09.2021 (Figures 1 and 2).
Figure 1. Placing a GPS-tracker on a Dorio stallion
Figure 2. Dorio stallion with tracker attached
Data processed until 31.08.2022 total positions
collected from the trackers are from 402 to
32,224, but if we subtract the positions of the
mares, from the tracker of the stallion Dorio
there are a total of 13,288, of which daily
positions (from 06:00 to 21:00) – 1,271, and the
night positions (from 21:00 to 06:00) are 966
(Figure 3). The total area of the inhabited
territory is 13,659 decares, with a perimeter of
17,590 km.
Interestingly, the Dorio stallion has few
positions down in the low by the river, which is
due to the episodic range capture. Signals (270
positions) from there were received mostly at
dusk and at night, which is very little against the
background of all 13,288, and it cannot be said
with certainty that they mainly spent the night
there and drank water at that time. It is also
noticeable that there are not many positions
between 11 a.m. and 1 p.m., which may mean
that this is when the horses go down to the river
to drink water.
There are no clearly specialized places to spend
the night, apart from entering the village of

372
Prisidets, when they are most likely being
chased by wolves. There is no mobile operator
coverage in the area with fresh running water
and rock salt applied. For this reason, it is not
possible to model the behavior for the water
drinking residence time and the diurnal range.
More field observations should be made.
Most positions are on the crests of the ridges
(when looking in Google Earth and verifying the
positions on the ground) or on the northern
slopes. This suggests that in low areas the device
loses range.
The old border electrical signaling system serves
as a fence (in the West) of the animals' territory,
although 4 permanently open portals fall into
this area. There is a small clustering of positions
(8-10 positions) around the portals apparently on
the rare occasion they have passed through
them. One portal is near the village and they use
it for overnight stays in the village. The other
(northernmost) was used once on 17-18.10.21
when staying outside their normal area.
To the Еast, the newly built border fence and the
Tundzha River limit the horses.
In the North, the old border electrical signaling
system rests on the Tundzha river and also limits
the territory to some extent, despite the open
portals, and in the south the terrain is more
rugged, but they still have the opportunity to
move. On the maps (Figures 3 and 4) it is clearly
visible how the old border electrical signaling
system limited the territory and the horses went
out only through the open portal of the village of
Prisadets and through one at the northernmost
positions. It can be seen that the remaining
positions have been cut off along the border
electrical signaling system.
It is noted that there are positions in the old
almond plantations in the territory. This means
that here, in addition to the mainly open
meadows, the old almond orchards are also
sporadically used (Figure 5). It can be seen on
the map that there is also a reservoir here, which
is visited by the herd mainly in the afternoon
hours (most often between 14:00-16:30), which
is also confirmed by the field observations.
Trackers were placed also on mares from the
same herd to try to track sexual behavior. The
GPS trackers used have a special feature that
tracks more atypical temperature deviations and
movements to determine if the mare has been
grazed or will be foaled. The results of the
observation did not show such activity on the
GPS, although the mares foaled and on field
observations we observed graze and jumping.
Figure 3. Map of the positions Dorio's herd
Figure 4. Map of all tracking positions of mare and
stallion from the stallion Dorio's herd
Figure 5. Map of the most frequently used positions of
the stallion Dorio's herd

373
This may be due to the loss of range in the area,
as well as the fact that the GPS's used did not
provide accurate temperature information.
Because of the location of the temperature
sensors, averaged values were sent between the
horses' body temperature and that of the
environment.
Data from the mares' trackers was processed
from 22.03.2022 - to 27.04.2022 - entirely in the
spring period, when the foals and coverings
from the stallion were expected. Total positions
collected by the trackers are 2,884, of which day
positions (from 06:00 to 21:00) – 1474, and
night positions (from 21:00 to 06:00) are 1,410
(Figure 6). Apart from 6 evenings, there are no
other clearly defined day and night positions in
the village.
Figure 6. Map of all tracking positions of the mare from
the herd of the stallion Dorio in the period from
22.03.2022. until 27.04.2022
When looking at all stallion and mares positions
on one map it looks like they are moving
together in the same territories, therefore the
horses are moving in a herd and not separated.
When analyzing the data more thoroughly, when
looking at a specific day where there are
positions from both GPS trackers at equal
intervals, it is observed that the two animals are
moving at a certain distance from each other.
This distance in the herd can be interpreted as a
sign that the animals are calm, not attacked by
predators to cluster close together, but graze
scattered over the terrain. But it can also be due
to the fact that when the GPS trackers are
located close to each other, they give some
deviation in the coordinates.
Unfortunately, the terrain has deep ravines and
there is no range for the transmitters there. Their
main watering place and salt blocks are in the
deepest part and we cannot ascertain exactly
how long they spend there.
From the maps it can be assumed that horses
prefer open areas, avoiding scrubby areas and
forests. All known positions are only on the
higher parts of the terrain 150-215 a.s.l. and
there is not enough information on the positions
in the lowlands, where there is water and salt -
60 m above sea level, because of the technology
limtations.
The herd of stallion Oliver. The pastures
occupied by Oliver's herd are fenced with an
electric herder, and the pastures are alternated on
a rotational basis.
The pastures are alternated using an electric
herder, with a minimum of 70-100 acres per
horse per year. The annual care (review of the
state of health, prevention, individual marking,
etc.) and the relocation (Figure 7) between the
individual pastures is carried out with the help
of volunteers.
Figure 7. Map of the positions of stallion Oliver's herd
until 08.12.2021
Pasture 1 (Figure 8) is used during the autumn
(October and November 2021), winter
(December 2021, January and February 2022)
and spring (from March to May 2022) seasons,
and the date of moving to this pasture is 28.10.
2021; date of moving from the pasture –
25.05.2022; collected data – 12,244 positions.
There are clearly marked overnight positions on
the highest part of the hill in an open meadow.
The pasture is 331 acres with a perimeter of 2.34
km, with an approximate altitude of 222 to
267m. The pasture is a hill with a sloping top
and is a mixture of oak coppice forest and forest
glades. Both day and night positions are in the
meadows and there are almost none in the forest
area.
There is one watering place in the pasture - a
probe with a pump that collects water in cisterns.

374
There is a certain dependence on the watering
time, as the horses visit it either around 11-12
hours before lunch or around 16-17 hours in the
afternoon, but only once a day.
Figure 8. Pasture 1 ‒ Autumn, Winter and Spring –
stallion Oliver's herd
Pasture 2 (Figure 9) was used for a short time as
the date of moving to this pasture is 25.05.2022,
and the date of moving from the pasture is
09.06.2022; data were collected from 1,775
positions. For daytime positions we take the
hours from 5:40 to 21:00 and there are 1,062 in
number, and evening positions -703 in number.
The pasture is 357 acres with a circumference of
5.5 km, with an approximate altitude of 147 to
225 m. The pasture is a mixture of overgrown
gully and meadows with all the horse positions
being in the meadows. Both the day and night
positions are there and there are hardly any in
the thicket.
There are no clearly defined roosting positions
in this pasture. There is one place for the horses
to drink water (a micro-dam), but there is no
clearly defined time to visit for drinking water.
Figure 9. Pasture 2 ‒ summer season ‒ of stallion
Oliver's herd
Pasture 3 (Figure 10) - summer season as the
date of moving to this pasture is 9.06.2022, and
the date of moving from the pasture is
16.07.2022; a total of 3,193 items were
collected.
For daytime positions, we take the time range
from 5:40 AM to 9:00 PM, with daytime
positions being 2,059 in number and evening
positions being 1,134 in number.
The pasture is 223 decares with a perimeter of
3.1 km, with an approximate altitude of 183 to
216 m, with a flat section of about 200 m above
sea level prevailing. The pasture is a mixture of
forest and woodland meadows, with all horse
positions in the meadows. Both the day and
night positions are in the open meadows and
there are almost none in the wooded area.
In this pasture, there are no clearly defined
overnight positions and there is one watering
place (micro-dam), but there is also no clearly
defined time for visiting a watering hole.
Figure 10. Pasture 3 ‒ summer season – of stallion
Oliver's herd
The entry of free-ranging horses into settlements
and crops is a major problem. Thanks to the
outlined virtual fences, the reason for entering
the nearby settlement was established, namely
an attack by wolves. As a result, night lighting
was spent where the horses would gather when
attacked by wolves.
As a result of the study, we can summarize the
benefits of using GPS trackers, which are:
possibility to outline virtual fences (Figure 11);
signals are sent when leaving the fences and
immediate measures can be taken; the detection
of the horses becomes much faster.
As a result of the conducted research, we found
that GPS-trackers facilitated the work related to
the selection of suitable pastures according to
the number and age of the stallion, rotation of
the pastures used, limited the number of victims
to predators, limited the entry of free herds into
the settlements, by building specially designated

375
places lit up at night, for the horses to visit in
order to escape from predators at night, etc. All
this is a prerequisite for reducing the costs of
raising herds, for their quick localization if
necessary, and timely reaction to incidents.
Also, these technologies can provide data on
crossing virtual fences, separating certain
individuals from the general group, etc.
Figure 11. Oliver's virtual herd fence
CONCLUSIONS
The used GPS-tracker model is suitable for
tracking the location of free-range horses, which
helps to easily find them in hard-to-reach areas
and large territories. With enough observations
of the herd, certain behavioral responses such as
places to hide from predators, water sources,
diurnal and nocturnal positions can be learned
and tracked. In the event of an attack, the GPS-
tracker immediately signals unusual movement
or leaving the virtual borders of the herd. The
herd's entry into foreign territory can easily be
seen and prevented.
Because of the location of the temperature
sensors, averaged values were sent between the
horses' body temperature and that of the
environment. Reproductive indicators as well as
clinical indicators (temperature, pulse, etc.)
cannot be tracked.
Due to the lack of coverage near the Turkish
border, there is still no established watering
period and no clearly defined roosting places in
the herd of stallion Doryo.
From the maps it can be assumed that horses
prefer open areas, avoiding schrubby areas and
forests. It follows that forest and shrub areas
should not be counted as part of pasture acres.
Despite some difficulties, it is possible to carry
out quality monitoring of animals using GPS-
trackers. Owners will be able to calmly monitor
on their mobile devices where and how much
their horses move in the field, in the mountains,
they will have the opportunity to identify
atypical behavior.
ACKNOWLEDGEMENTS
The study was conducted under research project
№ 06-21 on topic: "Potentialities for studying
the behaviour of free-roaming horses using
modern technologies for monitoring and
tracking". The project is funded by the Centre
for Research, Technology Transfer and
Protection of Intellectual Property at the
Agricultural University - Plovdiv in partnership
with "LIFE for Lesser Kestrel" LIFE19
NAT/BG/001017 implemented with the support
of Program LIFE of the European Union.
We would like to acknowledge the translation
assistance from Ellie Holden.
REFERENCES
Anderson, C. R., & Lindzey, F. G. (2003). Estimating
cougar predation rates from GPS location clusters.
Journal of Wildlife Management, 67, 307–316.
Bachmann, M., Wensch-Dorendorf, M., Hoffmann, G.,
Steinhöfel, I., Bothendorf, S., & Kemper, N. (2014).
Pedometers as supervision tools for mares in the
prepartal period. Applied Animal Behaviour Science,
151, 51-60.
Burla, J.B., Ostertag, A., Westerath, H.S., & Hillmann, E.
(2014). Gait determination and activity measurement
in horses using an accelerometer. Computers and
Electronics in Agriculture, 102, 127-133.
Burov, I. (2018). Application of information technologies
in selective horse breeding and training of farmers, E-
magazine "Education and Development", 1, 48-56,
(online).
Cagnacci, F., Boitani, L., Powell, R. A., & Boyce, M. S.
(2010). Animal ecology meets gps-based
radiotelemetry: a perfect storm of oppurtunities and
challenges. Phil. Trans. R. Soc. B, 365(1550), 2157–
2162.
Collins, G. H., Petersen, S. L., Carr, C. A., & Pielstick, L.
(2014). Testing VHF/GPS Collar Design and Safety in

376
the Study of Free-Roaming Horses. PLoS ONE 9(9),
e103189. doi:10.1371/journal.pone.0103189
Gruev, B., & Kuzmanov, B. (1999). General
biogeography. Plovdiv, BG: University Press of
Plovdiv Publishing House, 305-306.
Hampson, B. A., Morton, J. M., Mills, P. C., Trotter, M.
G., Lambb, D. W., & Pollitta, C. C. (2010).
Monitoring distances travelled by horses using GPS.
Australian Veterinary Journal, 88 (5), 176-181.
Mann, J., Radoi, I.E., & Arvind, D.K. (2014). Prospeckz-
5-A wireless sensor platform for continuous tracking
and monitoring horses in the wild.
17th Euromicro
Conference on Digital System Design, Verona, Italy,
700-703, doi: 10.1109/DSD.2014.102.
Microsoft Corporation (2018). Microsoft Excel 2010.
Retrieved from https://office.microsoft.com/excel
MOEW (2013). Information system for protected areas
from the Natura 2000 ecological network. MOEW,
http://natura2000.moew.government.bg
Osechas, O., Thiele, J., Link, J. A. B., & Wehrle, K.
(2008). Ratpack: Wearable sensor networks for animal
observation. Proceedings of the IEEE 30th Annual
International Conference on Engineering in Medicine
and Biology, EMBS, 538-541.
Petrov, A. (1940). The Karakachan horse. Sofia, BG:
Yearbook of the Faculty of Agronomy and Forestry,
15 pages. (Bg)
Popova, M., & Etarska, H. (2020). Program for breeding
and conservation of the Karakachan breed of horses
2020 – 2030, Karlovo, 26 pages (Bg)
Radoi, I. E., Mann, J., & Arvind, D. K. (2015). Tracking
and monitoring horses in the wild using wireless
sensor networks. IEEE 11th International Conference
on Wireless and Mobile Computing, Networking and
Communications (WiMob), Abu Dhabi, United Arab
Emirates, 732-739, doi:
10.1109/WiMOB.2015.7348035.
Sabeva, I. (2009). Breeding program for the conservation
of the autochthonous primitive breed of the
Karakachan horse 2009–2019. Karlovo, 26 pages
(Bg)
White, G. C., & Garrot, R. A. (1990). Analysis of Wildlife
Radio-Tracking Data. San Diego, USA: Elsevier
Publishing House, p. 383.