The Effect of Extra Food Stimulation on Asian Elephants(Elephas maximus) Kept at Kolmarden Zoo

Bachelor’s ThesisJohanna Sjoberg

LiTH-IFM- Ex–2516–SE

Supervisor: Mats Amundin, Kolmarden Zoo

Examiner: Anders Hargeby, Linkopings universitet

Linkopings universitetDepartment of Physics, Chemistry and Biology

SE-581 83 Linkoping, Sweden

Contents

1 Introduction 21.1 Background . . . . . . . . . . . 21.2 Aim of the study . . . . . . . . 3

2 Matherials and methods 32.1 Subjects and husbandry . . . . 32.2 Experimental treatments . . . . 42.3 Statistics . . . . . . . . . . . . 5

3 Results 5

4 Discussion 74.1 Bua . . . . . . . . . . . . . . . 74.2 Saonoi . . . . . . . . . . . . . . 84.3 Saba . . . . . . . . . . . . . . . 84.4 Conclusions . . . . . . . . . . . 8

5 Acknowledgements 8

1

Stereotypic behaviors in all animals are more often than not associated with poor welfare.Limited access to perform species specific behaviors is often a reason for the developmentof stereotypies. Elephants with their great intelligence and need of social contact, coupledwith a destructive a behavior are especially difficult to house in captivity. To decrease theoccurrence of stereotypic behaviors in elephants, environmental enrichment in form of foodenrichment is a good option, since elephants have a great need of foraging. The aim of thisstudy was to investigate if an extra supply of food enrichment could decrease the presence ofstereotypic behaviors at night in three Asian elephants at Kolmarden zoo. Already existinghay nets attached to wires in the roof were used and connected to a timer. The hay netswere lowered to vision trunk reach between 6:00 am and 6:30 am during five days and thefrequencies of selected behaviors were compared with the frequencies of the same behaviorsduring five preceding baseline nights. The animals were filmed using mounted cameras withIR lights. There was a significant decrease of stereotypic behavior for one of the elephants,but all three showed an increase in foraging whereof the increases were significant for two ofthem. One of the elephants showed no stereotypic behaviors at all during the study. To keepin mind is that the elephants have different backgrounds and experiences which might haveinfluenced the results.

Keywords: Asian elephant, Stereotypic behaviors, Food enrichment, Foraging, Social interac-tions, Night behaviors, Videotaping

1 Introduction

1.1 Background

It is always difficult to meet the behavioralneeds of wild animals which are held in cap-tivity. Limited areas or shortage in stimuliprevents animals from performing their nat-ural behaviors. Due to this, captive animalsoften develop so called stereotypic behaviors[9]. Stereotypic behaviors appear as repeatedmovements with no function, but are often per-formed to compensate for the lost ability toperform a natural behavior [5]. Stereotypic be-haviors can be seen in many different kinds ofanimals and are often considered an indicationof an inadequate welfare. Stereotypic behav-iors are not always an indication of an animalsuffering from physical or psychological stress.It can sometimes be an expression of elation[10]. Stereotypic behaviors in captive animalsas a result of limited abilities to perform theirnatural behaviors or as a result of poor welfare

cannot just disappear or be cured by changedconditions. Once a stereotypy has been per-formed and learned by the animal, it will per-sist even under improved conditions, albeit inlower extension [10]. To lower the occurrenceof stereotypic behavior and promote the ani-mal’s natural behavior and by this increase itswell-being, different types of environmental en-richments can be used [12]. Elephants are thelargest land living animal on earth [10] withan average lifespan of more than 60 years [11].With their size, their complex behaviors andgreat need for social interactions coupled witha developed tendency to destroy their environ-ment it is difficult to maintain a good welfarefor the animals [10]. Elephants have a greatneed of foraging and in the wild they spendup to 16 hours a day, searching and processingfood. Their main food consists of roots, bark,fruit, leafs and different kinds of herbs [8]. Thefeeding behavior is peaking in the morning, af-ternoon and around midnight [4]. Because ele-

2

phants have a great need of foraging, food en-richment is a good type of environmental en-richment for them. Earlier studies have shownthat hiding and dispreading food increases thetime spent on foraging in elephants held in cap-tivity [12]. Elephants in different parts of theworld are sometimes held restrained by chains.This greatly reduces their chances of being ableto perform foraging and locomotion behaviors[4]. The reduction in the ability for the animalto move and interact with conspecifics lreducesthe animal’s psychological health and may mo-tivate it to behave abnormally. Although mostelephants held in captivity even without chainsare restricted in their expression of their nat-ural behaviors, a reduction in stereotypic be-haviors have been shown in elephants left un-chained in enclosures compared to chained ele-phants [3]. Stereotypic behaviors in elephantscan be expressed as repeated swaying of thewhole body from one side to the other, weavingwith the head, swinging the trunk, lifting thefeet and bobbing up and down with the head[2]. They can often be seen in some particu-lar situations as just before feeding and whilewaiting for performance in elephants [3].

1.2 Aim of the study

The aim of this study was to identify thestereotypic behaviors in Asian elephants heldin Kolmarden zoo, occurring during the nightin connection to foraging. Another aim wasto find out if time control of existing foodenrichment could lower the presence andfrequencies of stereotypic behaviors and thusincrease the well-being of the animals.

Hypothesis: The presence and frequency ofstereotypic behaviors during the night andearly morning in three Asian elephants kept inKolmarden zoo will be reduced with increasedaccess to food enrichment

2 Matherials and methods

2.1 Subjects and husbandry

The animals used in this study were threefemale elephants located at Kolmarden Zoo,Saba 43 years old, Saonoi 15 years old andBua 14 years old. Saba was wild caught andarrived at Kolmarden in 2008. Before shecame to Kolmarden, she was kept in a zooin France. There she was kept alone andwas often chained, making it hard for herto associate with other elephants. Bua andSaonoi were born in a elephant labor campin Thailand and came to Kolmarden in 2004.Normally the oldest elephant of the herd takesthe role as a leader, but because Saba was socautious and anti-social, Saonoi has taken therole as a flock leader.

The first thing that happens in the morningis that the elephant keepers arrive at about07:30 am and the animals are given breakfastconsisting of pellets (Saba 4 kg, Saonoi 4 kgand Bua 2,5 kg) with minerals and seed straw(5 kg each). After that each of the elephantsgets a shower and the elephant keepers lookover the animals. The rest of the day has dif-ferent approaches. When weather permits, theanimals are held outside as much as possibleand then have the opportunity to reside in abigger enclosure. The elephants are trained atleast two times a day by the zoo keepers andthe training consists of for example pedicureor behaviors that will promote the handling ofthe elephants.

About 16:00 pm the elephants are givenseed straw in nets suspended in a wire fromthe roof which can be hauled out of trunkreach or lowered so the elephants can eat.Saba gets ca 20 kg, Saonoi ca 15 kg and Buaca 15 kg of straw. In addition the elephantshave close to free access to tree branches.During both day and night the elephants have

3

free access to seed straw.

During the days and depending on weather,the elephants are kept outside in a 3000 m2

public display or in a 600 m2 off exhibitenclosure contiguous to the indoor enclosures.The animals always have access to the indoorquarters during the night and during thewinter season they are mostly housed insideall the time. Saba is usually kept alone in oneof the enclosures since she does not appreciatethe company of the other elephants. Theelephants are switched between the indoorenclosures every night. One of these enclosuresis 250 m2 with a roof height of 6 m. Theenclosure is separated into one bigger and onesmaller room, all with concrete floor and withconcrete walls on three sides and a wire withelectricity towards the visitors side. In thisenclosure the elephants have access to watervia a concrete bowl. The other enclosure is170 m2 , with the floor covered with a bed ofsand and a roof height of 6 m.

Both Saba and Bua shows stereotypic behaviorcontinuously, mainly so called head bobbing,were the animal swings its head up and down.Saonoi rarely shows stereotypies, but when shedoes, she weaves her head back and forth.

2.2 Experimental treatments

Data were collected in the spring of 2011,during ten nights between the 5th and the25th of May. The first five nights consti-tuted the baseline of the study, were the ele-phants where observed under normal condi-tions with their usual routines, without accessto the extra food enrichment. The animalswere taped by video cameras five hours pernight between 16:30-17:30, 22:00-23:00, 03:00-04:00, 06:00-07:00 and 07:00-08:00. The videocamera used in the room with sand on the floorwas an Ikegami ICD-47E (Ikegami Electronics

Figure 1: Elephant foraging by reaching forone of the hay nets with the trunk.

GMBH, Ikegami Strasse, DE-414 60 Neuss 1T)supplemented with an IR lamp, Videor Tech-nical MFL-I/LED 60 (MonCam AB, 141 05Kungens Kurva). The cameras used in the en-closure with concrete floor were LG Mercury/2VR-DN3V10IRH (Svensk Sakerhetsvideo AB,Gruvgatan 35 B, 421 30 Vastra Frolunda) andwere supplemented with two extra IR lamps,Eneo IR-LED 150-A-50 (MonCam AB, 141 05Kungens Kurva). The remaining five nightsconstituted the experimental treatment. Dur-ing these nights the animals were given extraaccess to food enrichment via net bags filledwith seed straw (see Figure 1), one in each en-closure. Between 06:00 am and 06:30 am eachmorning, the nets were lowered to within trunkrange via a timer (multifunction time relay)CMFR-66 (www.conrad.se), connected to thewires.

The films were later analyzed to identify theselected behaviors. The observation methodused was instantaneous sampling with 2.5minute intervals.

4

Table 1: The ethogram used in the study. The column to the left shows the functional termand the column to the right shows the descriptive term

Funtional term Descriptive term

Stereotypic behaviour Repetitive movement, weaving or head bobbing for example, can occurat the same time as other behaviours

Foraging Manipulating food or water with the trunk/mouth, feeding/drinking forexample, can occur at the same time as other behaviours

Social behaviour Touching other individual with any body part

Locomotion Locomotion of at least body length

Standing Standing up

Lie down Laying down

Other Other behaviours not included in this ethogram, dust bathing for exam-ple

Not visible Individual is not visible

No data Individual is not in the exhibit

To identify the selected behaviors withoutany confusion between the three observers anethogram was made (see Table 1).

2.3 Statistics

The calculations of the recorded data weremade in Microsoft Excel 2007. The frequenciesof every behavior performed by each of theelephants were measured separately and themean value of each behavior was calculated.The mean values of each behavior during thebaseline were then compared with the meanvalues of each behavior during the experimentusing students T-test. The significant levelused in the test was 0.05.

The frequency results were transferred intopercent and showed in bar graphs (see figure2-4). The variance for every behavior wascalculated and is inserted as error bars in thegraphs.

To measure the percentage of observationsthe elephants spent on foraging during their

time awake, the data points where the animalswere laying down were excluded. The amountof data points where each individual spentforaging was divided with the total amount ofdata points. The results were transferred intoa table (see Table 2).

3 Results

The analysis of the baseline showed a frequentoccurrence of stereotypic behaviors in Buaand Saba. The stereotypic behaviors that Buashowed took place almost exclusively in themorning, a couple of hours before the elephantkeepers arrived to the building to provide theelephants with food, while Saba performedher stereotypic behaviors distributed over thewhole night. Saonoi showed no stereotypicbehaviors at any time during the study.

When given extra access to food enrichmentthrough the hay nets all of the elephantsshowed an increase of foraging behavior,

5

Figure 2: The average in percent, with variance for all the behavoirs registrated for Bua duringthe observations. * marks a significant difference (p < 0.05) between baseline and enrichment.

Figure 3: The average in percent with variance for all the behavoirs registrated for Saonoi duringthe observation. * marks a significant difference (p < 0.05) between baseline and enrichment.

Figure 4: The average in percent with variance for all the behavoirs registrated for Saba duringthe observation. * marks a significant difference (p < 0.05) between baseline and enrichment.

6

Table 2: Percentage of observations spent on foraging by each of the individuals. Numbers arethe fraction of data points registered as foraging of the total amount of data points.

Individual Before After

Saba 45 % 80 %

Saonoi 59 % 65 %

Bua 45 % 64 %

whereof the increases for Bua and Sabawhere significant (see figures 2 (P < 0.015)and 4 (P < 0.0017)).There was a small difference between the base-line nights and the nights with extra food en-richment in time spent lying down with anincrease in all of the elephants, but none ofthese were significant. Saba showed a signifi-cant reduction in stereotypic behaviors duringthe nights when given extra access to food en-richment compared to the baseline nights. Buashowed no difference in frequency of stereo-typic behaviors during this study.

When excluding the data points where theelephants were observed lying down, there wasan increase in percentage spent foraging for allof the elephants. Two of these results were sig-nificant (see Table 2).

4 Discussion

The aim of the study was to test if extra ac-cess to food enrichment during the night andearly morning could decrease the occurrenceof stereotypic behaviors in the three elephantskept at Kolmarden zoo. During the baselinetwo of the elephants showed stereotypic behav-iors whereas one (Saonoi) did not show anystereotypic behaviors during the whole study.All of the elephants showed an increase in for-aging and this led to a percentage which moreresembles the percentage of time spent foragingby elephants living in the wild [8]. This indi-cates that the extra food enrichment is help-

ing the elephants in regard to meet their nat-ural need to forage. Some of the nights werenot used because the animals were allowed tostay outside and consequently impossible toobserve. A possibly disturbing factor to takeinto account is that one night during the study,a lamp was left turned on in the hayloft nearthe sand box. This night the elephants actedworried and they spent less time lying down.Of course it cannot be known if the elephantswould have spent more time of that night lyingdown if the light had been turned of, but it can-not be ruled out that the light might have beena disturbing factor on the elephant’s sleep.

4.1 Bua

Bua showed stereotypic behaviors in form ofhead bobbing both during the baseline and theexperiment with no significant change. Thereason for this might be that Bua’s form ofstereotypic behavior is not caused by poor wel-fare or limitation of food, but instead is an ex-pression of exitedness or frustration, since wesaw a trend that she only performed stereo-typic behaviors just before given access to food[10]. On the other hand there was a significantincrease in foraging during the experimentalperiod (see Table 2) which should indicate thatthe extra access to food enrichment was bet-ter fulfilling her need for foraging during oneday, which should be about 16 hours/day (ca66%/day), based on results obtained in wildelephants [8].

7

4.2 Saonoi

Saonoi did not show any stereotypic behav-iors during the study. She showed an in-crease in foraging and a decrease in lying down,which might indicate that the extra access tofood makes her occupied a larger part of thenight, than during the baseline where no extrafood was provided. This extra supply of foodmay provide Saonoi with enough occupationto meet her need of foraging. To take into ac-count is that Saonoi might still have performedstereotypic behaviors, albeit the observer wasuable to detect them or when the stereotypicbehavior was misinterpreted as another behav-ior. Saonoi did not show a typical example ofher stereotypic behavior before or during thestudy and this might therefore be difficult todistinguish.

4.3 Saba

Saba showed steretypic behaviors both duringthe baseline and the experiment period andwe were informed by the elephant keepers thatSaba spent much of the days with head bob-bing in a way that made her unable to contact.Saba showed a significant decrease in stereo-typic behavior with the extra access to foodenrichment and a significant increase in forag-ing. Therefore it can be supposed that the ex-tra access to seed straw kept her busy duringthe night and that the time she usually spenton stereotypic behavior during the baseline wasnow put on foraging. Her need of foraging dur-ing one day might then has been more fullfilled.However, in Saba’s case it must be included thefact that she was kept alone much of the timebefore she came to Kolmarden zoo and there-fore limited access to food may not be the onlyreason for her stereotypic behavior. Her lostopportunity to interact socially with other ele-phants in the past may have affected her tohave difficulties, interacting with the other twoelephants in the zoo and therefore not being

able to function normally in the group. Be-cause social interactions with conspecifics isvery important for elephants to learn how tobehave with other individuals [10], this mightbe a key factor for Saba’s low rank and motiva-tion for performing stereotypic behaviors. An-other factor to take into account is that Sabawasheld alone during the nights. This might ef-fect her behavior, since elephants have a greatneed to interact with conspecifics [5]. In Saba’scase this is hard to execute, because she is un-able to rest in company of Saonoi.

4.4 Conclusions

This study has shown that food enrichment canbe a way to increase and also better meet theneeds of time spent on foraging in the Asianelephant. Food enrichment can also be a pos-sible way to decrease the occurrence of stereo-typic behaviors, but this depends on manyother factors, such as the animal’s background,limitations to perform other natural behaviorsand interacting socially. Further experimentshave to be made to establish whether food en-richment is the ultimate solution to decreasestereotypic behaviors in elephants and conse-quently increase their welfare.

5 Acknowledgements

This study was supported by Kolmarden zoo.I would like to thank Per Jensen for helping usget in touch with Kolmarden. I would also liketo thank my supervisor Mats Amundin for tak-ing the time and giving me the possibility todo this exciting research. Some other personsI would like to thank is all the elephant keep-ers at Kolmarden which have helped us withinstalling all of the equipment and helping usrun the test when we had no possibility, forbeing kind to us and letting us participate inthe daily management and above all for beingso god and caring with the elephants.

8

References

[1] Brockett R.C., Stoinski T.S., Black J.,Markowitz T., Maple T. L. 1999. Noc-turnal Behavior in a Group of UnchainedFemale African Elephants. Zoo Biology18:101-109.

[2] Elzanowski A., Sergiel A. 2006. Stereo-typic Behavior of a Female Asiatic Ele-phant (Elephas maximus) in a Zoo. Jour-nal of applied animal welfare science 9(3),223-232.

[3] Friend T. H., Parker M. L. 1999. The ef-fect of penning versus picketing on stereo-typic behavior of circus elephants. Ap-plied Animal Behaviour Science. Pages213-225.

[4] Gruber T. M., Friend T. H., GardnerJ. M., Packard J. M., Beaver B. andBushong D. 2000. Variation in StereotypicBehavior Related to Restraint in CircusElephants. Zoo Biology 19:209-221.

[5] Kurt F., Garai M. 2001. Stereotypies inCaptive Asian Elephants - A Symptomof Social Isolation. Scientific Progress Re-ports. 57-63.

[6] Moss C. J. 1983. Oestrous behaviour andfemale choice in the African elephant. Be-haviour volume 86. Pages 167-196.

[7] Moss C. 1988 Elephant Memories. Pages8-12.

[8] Stoinski T.S., Daniel E., Maple T.L. 2000.A Preliminary Study of the Behavioral Ef-fects of Feeding Enrichment on AfricanElephants. Zoo Biology 19:485-493.

[9] Thomas S., Gloyns R., Angele C., Mar-shall A., Barber N. 2001. The effectivenessof a long-term environmental enrichmentprogramme for elephants at Paignton Zoo

Environmental Park. 3rd Annual Sympo-sium on Zoo Research. 9-16.

[10] Veasey J. 2006. Concepts in the care andwelfare of captive elephants. Int. Zoo Yb.40: 63-79.

[11] Vidya T.N.C, Sukumar R. 2005. Socialand reproductive behaviour in elephants.Current Science 89:1200-1207

[12] Wiedenmayer C. 1997. Food hiding andenrichment in captive Asian elephants.Applied Animal Behaviour Science 56 77-82.

9