HEADACHES AND HEARTACHES: THE ELEPHANT MANAGEMENT DILEMMA

Ian John Whyte (PhD)
Initially published as:
WHYTE, I.J. (2001). Headaches and Heartaches - the elephant management dilemma. In: Environmental Ethics: Introductory readings. Eds. Schmidtz, D. & Willot, E. Pp. 293-305. New York: Oxford University Press.

The version presented here is an update prepared for a reprint of this volume.

The dilemmas of managing elephants in reserves are generally poorly understood. The different attitudes to basic management philosophies have lead to many acrimonious debates and deep rifts have developed between some of the respective proponents. While those from the anti-culling lobby have condemned the killing of elephants, the non-interference policies are not without their own ethical dilemmas. It is not the purpose of this article to try to favour either one of these two elemental points of view, but to try to set the situation in Kruger National Park (South Africa) against the background of other national parks in Africa. The intention is to get to the heart of the elephant dilemma so that readers themselves may understand the issues and have some empathy with elephant managers and the decisions with which they are faced.

There are many viewpoints emanating from a diverse variety of sources on the subject. These come from scientists from academic institutions involved in elephant research, animal rights and animal welfare groups, “absentee or ex-situ conservationists”, ex-patriot conservationists and other interested and affected parties. It is the purpose of this paper to examine the problems specifically from the viewpoint of managers tasked with elephant management. These are the people who ultimately are accountable for the outcomes of any management actions undertaken. The people and groups who work on the fringes seldom fall into this category, and they are not accountable for their sometimes radical proposals and solutions. They can walk away from any resulting adverse ecological outcomes without the serious personal consequences faced by managers. They are therefore likely to view the issues in an entirely different light to those faced with the hard realities.

Before the advent of firearms, the elephants of Africa were probably not greatly influenced by Man, though there is a view held by an increasing number of prominent conservationists that indigenous Africans were well capable of influencing the dynamics of elephant African elephant populations well before the advent of firearms. There is also archaeological evidence of proboscidean “overkills” elsewhere in the world. Be this as it may, the ability to easily kill large elephants increased as firearms began to proliferate, and the populations gradually started to decline. Initially this was due to the demand for ivory in Europe and Asia and there was no control over the off-take. This situation possibly persisted for two hundred years or more, and in spite of the establishment of reserves and national parks throughout Africa, many (if not most) elephants roamed outside of these parks and remained vulnerable. The numbers of elephants on the African continent in earlier times are not known, but by 1979 they had been reduced to an estimated 1.3 million. By this time the range states (countries which have elephant populations) north of Zimbabwe and Namibia were in the grip of a poaching epidemic which reduced the continental population to 609 000 by 1989.

Concern over this dramatic trend resulted in the banning by CITES (Convention on the International Trade in Endangered Species) of international trade in elephant products in 1989. Although one study four years later concluded that in most countries the ban had not halted the illegal off-take, it is widely believed that the ban had achieved much in reducing the poaching. The worst of the poaching was experienced in the countries to the north of the Zambezi and Cunene rivers. South of this, although increased poaching was experienced, this was not at a scale which significantly reduced elephant numbers. In Kruger the first case of elephant poaching was recorded in 1975. Since then the incidence has been sporadic except during the years 1981 - 1985 when a sharp increase in the incidence was experienced. During this time, 193 were known to have been shot. Active anti-poaching measures solved the problem to a large degree and elephant poaching has occurred only sporadically since. Populations in the southern range states have continued to grow and by 2007 were estimated at a definite total of around 298 000, with the possibility that there could be as many as 358 500. The disparity between these two geographical zones has had the result that northern range states generally favour the ban while southern countries do not. Southern wildlife departments maintain that they could benefit enormously from the financial returns that the sale of elephant products could bring, particularly as elephant management in these countries is now impeded by financial constraints. So while the northern African range states welcome the ban as perceived protection for their populations, most southern African range states are seeking ways of limiting their elephant numbers. But why would you want to limit elephant numbers? Why manage elephants at all? Why not just let them be?

The problem
In many African countries today, elephant populations are confined to national parks and reserves. This is also true of most other wildlife species, both plant and animal. In South Africa nature reserves and national parks are conservation "islands" whose boundaries are hard-edged up against the activities of people. In these conservation islands, it is usually the wildlife manager's job to try to protect all of these species as part of biodiversity maintenance programs, but there are some reserves that have been proclaimed specifically for the protection of particularly endangered species and biodiversity may be considered of lesser importance. But there are some countries such as Kenya who can still claim that as much as 70% of their wildlife exists outside of national parks. Wildlife managers in these countries have their own set of problems, usually concerned with interactions between man and wild animals, and the dilemmas of conserving elephants in closed systems have not yet really become part of their conservation priorities. Large parts of east Africa, particularly in Kenya and Tanzania, is Masai country. Masai were semi-nomadic pastoralists who traditionally had little interest in wildlife. They saw wild animals as "God's cattle" and their whole lives were centred around their own herds of cattle and goats. As agriculture was not practised at all, competition with wildlife was confined to incidents where people or their livestock were killed or injured by animals, but by and large they could co-exist in relative harmony. But this situation is changing slowly for two main reasons. Dramatic human population increases and hunger for land no longer affords people a nomadic life style, or their herds free access to rangeland, and many are switching from pastoralism to agriculture. Agriculture is rarely compatible with wild animals as crops planted by people are a highly attractive to the larger herbivores. In the early 1990's the African Elephant Specialist Group considered human/elephant conflict to be the greatest current problem facing elephant conservation. These clashes will gradually force wild animals more and more into sanctuaries set aside for them. Secondly, college education of the Masai youth means that they have tasted the offerings of civilisation and are no longer satisfied with the pastoralist's simple life style. They want jobs, houses, and access to electricity and running water, and to the trappings of the modern world, none of which are compatible with pastoralism.

These two forces will probably ensure that the wildlife in these countries will also eventually occur only in reserves, and the problems of maintaining biodiversity in these “conservation islands” will be brought into sharper focus.

As a general rule of thumb, the smaller the conservation island, the greater the degree of management it will require as some of the essential ecological processes will be missing. An example of this is that many small parks can not accommodate the larger predators, and in the absence of the population checks which these predators impose on prey populations, man has to take over this role. But if the "island" is large enough (and the Kruger National Park is considered to be one of these) the vast majority of its component species and ecosystem processes will require little or no management at all. Throughout time, these species and processes have co-evolved to form a complex matrix of competition, inter-dependence and above all, survival. This is the way it should be and it is accepted that managers should interfere with these processes only when it is considered unavoidable.

In nature, nothing is static. Rainfall and temperature are the engines of ecological processes, and these are never constant. There are years when rainfall is abundant and others of drought. In most of Africa these conditions are cyclic and there are periods of a few years of above average rainfall followed by periods below average. In Kruger, records show that rainfall cycles average about 20 years in duration - ten dry years followed by ten wet ones, though it is usual to experience one or more dry years during the wet part of the cycle and vice-versa, and the length of a wet or dry cycle may also vary considerably.

Different species respond differently to the prevailing conditions, some increasing during the wetter years, but some also favour the dry conditions and their populations flourish during the drier parts of the cycle. In Kruger wildebeest and zebra favour the drier years when the grass has been grazed short and trampled to create more open conditions. These animals rely on their good eyesight to avoid predators, so when the grass becomes long and rank during the wetter years, lions have enough cover to get up close and make their kills more easily.

But most species prefer the wetter times and have developed ways of reducing the threat of predators in long grass. When food is abundant, buffalo congregate in large herds and the adults are aggressive to predators and protective of the younger animals and calves. But when droughts come, each individual buffalo is forced to compete with the others for the meagre grass reserves that are available. Food shortages force them to split off in smaller groups to allow more effective foraging. Being in a weakened physical condition, these small groups are much more vulnerable to lions. A single large herd of buffalo will find itself in only one lion pride's territory at any one time, but when a herd splits, more prides have access to the herd and individuals are in poorer physical condition, rendering them less able to defend themselves against these predators. Furthermore, water is in short supply and buffalo have to drink. Lions wait at the water and in their more vulnerable condition, buffalo can be killed almost at will. Many cases have been recorded of lions making multiple buffalo kills during these times. It is a predator's instinct to kill, and each time the buffalo herd approaches the water, the lions kill another. During the severe droughts of 1992/3, Kruger's buffalo population declined by more than 50% from nearly 30 000 to less than 15 000.

Kudu have other strategies for avoiding lions in long grass. They have excellent eyesight, smell and hearing and live in small groups who forage silently. During droughts they lose condition and are forced to forage in habitats they would normally avoid. As with buffalo, they are forced also to drink at the few remaining waterholes, which is where the lions are waiting. In most cases these population fluctuations are driven by the short-term (± 20 year) climatic cycles.

But elephants do not conform to this pattern. In today's imperfect world, Africa is no longer what it was before the advent of technological man, and the continent-wide ecological processes which used to operate, can now no longer do so. The functioning of most of these processes is lost in the mists of time and can now only be speculated upon. The population cycles of elephants is one of these. Elephants had few natural enemies in those times, and the questions of how their populations were regulated are unanswered? It is sure of course, that no population can continue to grow forever, as eventually they will exceed the resources upon which they are dependant. There must come a time when conditions are less favourable and the population will enter a phase of decline. Did elephant populations build up over centuries in local "events" to the point that food became limiting and then die out to a much lower level? Or did they move away when food became scarce? If so, where did they go? And what would have happened to the elephants already occupying the area to which they moved? Did disease play a role? Was pre-technological Man really capable of limiting elephant numbers? These are all questions which nobody can answer with surety, and maybe nobody ever will.

What is known is that elephants have the intellect and constitution to exploit a wider range of food resources than any other animal (except Man). When grass and browse are not available, they can eat twigs and branches or use their tusks to prise bark off trees, and can even push trees over to reach the leaves in the canopy or to expose the roots. The latter two activities will almost invariably result in the death of the tree. Elephants are selective feeders, and it is the favoured food species that are initially affected. But as these are depleted they are forced to switch their feeding activities to other less palatable plant species. These in turn will also be depleted as the elephant population grows.

Natural limitation of elephant numbers will only begin to occur once even these resources have been depleted and the elephants become nutritionally stressed, but by this time the environment will have been subjected to severe impacts. In this process, the question is what would have happened to the other species - both plant and animal - occupying these habitats? Some may have even been extirpated (local extinction).

In the old Africa, this would not necessarily resulted in the extinction of species as other populations would likely have occurred elsewhere and recolonisation by both plants and animals could have taken place naturally, even though it may have taken hundreds of years. In evolutionary time, time scales like this are irrelevant. Some species that were unable to adapt or recolonise may have been pushed to extinction, but this is nature's way - adapt or die!

But the old Africa has now gone forever. No longer can species range far and wide, and no longer can most terrestrial life forms naturally recolonise areas where they have been extirpated. And so elephants, with their ability to drastically modify habitats, are a threat to many species within these confined reserves. A reserve like Kruger, which is large enough to allow for minimum management of most species, is still considered to be too small to accommodate elephant population fluctuations without environmental damage. And thus, if the objective of a national park is the long-term conservation of all the indigenous biota occurring there, then something will ultimately have to be done to limit elephant numbers.

A theory that has been proposed recently is that if an elephant population is left long enough without artificial limitation of its numbers, it will eventually attain a state where it will stabilise numerically. At this point the population would have achieved what is known as a “stable age structure” and a decline of the population growth rate to zero. Put simply, in a growing population the most numerous age class will be the youngest. i.e. there will be more individuals of less than a year than one year olds, more one year olds than two year olds and so on through all the age classes. For a population’s growth rate to decline to zero (no increase in the number of individuals in the population), the number of deaths (mortality) must equal the number births (natality). For this to happen, either the number of births must decline, or else a significant number of younger animals must die. In this case the age structure would show much reduced proportions of calves and juveniles. Over time, this structure would remain largely unaltered, giving rise to the concept of a “stable age structure”. There are no epidemic diseases known to affect elephants and predation is negligible. Lowered fecundity rates and increased mortality of juveniles therefore implies that the individuals in the population would have to be severely stressed nutritionally. Elephants are extremely successful and efficient foragers capable of exploiting nearly all parts of both grass and woody plants, and for them to reach a state where they are not in condition to breed or to where they may die from food deprivation, the habitat would be in a severely degraded state.

There is as yet no evidence to suggest that an elephant population will attain such a stable age structure, or evidence which may indicate the environmental conditions which would give rise to such an age structure. The question for the Manager is whether he can risk allowing the elephant population to increase until a possible stable age structure and zero growth rate in the population was achieved? What would have been lost from the habitats in his charge in the process?

History of the Kruger National Park elephant population
Elephant populations throughout Africa were decimated by the early hunters whose writings were more concerned with the thrill of the hunt than of natural history. In the Lowveld areas of South Africa where Kruger is situated, all the elephants had been shot out before the park was proclaimed in 1898. Nobody knows how abundant elephants were in this area, or how they utilised their range, but there is evidence from some sources which suggests that elephant numbers were never very high in the Kruger area before the advent of the white man and his guns.

The San (Bushmen), whose characteristic rock paintings are still visible in rock shelters in both the south-western and northern areas of the KNP, were associated with the latter part of the Late Stone Age between 7 000 B.C. and 300 A.D. Elephants must have occurred in the KNP area during this time as one of the paintings in a shelter along the Nwatindlopfu River shows a group of five of these animals. Of the 109 shelters containing rock art so far discovered in the KNP area, this is the only one featuring elephants. One other shelter with elephant paintings has been found in the area, some 30 kms to the west of the KNP.

It would be expected that a large, dramatic and dangerous animal like the elephant would feature prominently in Bushman folk lore. Elephants were a popular theme for paintings elsewhere in southern Africa as they provided a lot of meat and were associated with water and rain which the Bushmen artists were keen to influence. Bushmen were capable of hunting elephants by "ham-stringing" (as was shown in a painting near Molteno in the eastern Cape) or through the use of poisoned arrows – particularly on the younger animals. San art is believed to have been of considerable spiritual significance to the artists, depicting the spiritual experiences of shamans during states of altered consciousness induced by ritual dances. San art was not narrative of their lifestyles nor "menus" representative of their diet, and that the incidence of different species in their art may not be reflective of their relative abundance. But elephants have been shown to be of "special symbolic importance" to San peoples elsewhere in South Africa. Human figures with trunks and even with heads of elephants have been painted. Given then that elephants were of considerable cultural significance to the San people, their relative scarcity in the rock art of the KNP and surrounding area may indicate that these animals were relatively rare during the San era.

There is also a lack of evidence of old elephant utilisation of baobab trees. Elephants strip off the bark of these trees for food and scars persist for hundreds of years. One baobab in Kruger, carries the inscription "BRISCOE 1890" carved in its bark. This carving, now 110 years old, is still as clearly visible as if it was carved only a year or two ago, and will probably persist for another 100 years or more. If elephants were utilising baobabs 100 or even 200 years or more ago, it could be expected that the scars would still be clearly visible. Yet more than 50% of trees in the far north of the park show no sign of utilisation by elephants. Baobabs outside the Park show no signs of old elephant damage, but clearly show the impacts of earlier people who cut "panels" of bark from the trees for domestic use.

A third clue comes from the writings of early travellers to the area who made little mention of these animals. Francois de Cuiper and his party were the first to visit the area in 1752. His mission from Delagoa Bay (now Maputo in Mozambique) was to establish trade with the interior in gold, ivory and copper. They saw few elephants and were informed by the indigenous people that if they wanted ivory and gold, they would have to go far to the north to the area now known as Zimbabwe. Copper could be obtained from the area now known as Phalaborwa. This last information has proved to be correct for gold and copper, so the information on ivory was probably also correct. Louis Trichardt on his trek through Kruger in 1838 also made no mention of elephants in his diaries, though he mentioned an elephant hunt once they arrived in Delagoa Bay. Joâo Albasini was the first white settler in the Kruger area. He arrived in Delagoa Bay in 1831 and established himself as a hunter and trader in Mozambique and the eastern Transvaal. He formed a company in Lorenco Marques whose objective was to hunt elephants and increase the trade in ivory. This lasted only six months and he then established himself as a hunter and trader in the Phabeni area in the southern part of Kruger. After two years there he abandoned the store and moved to Ohrigstad. Though no records exist, the outcome of these ventures hint at an elephant population at too low a density to sustain viable hunting and trade.

Of the later hunters who left any written record, it is perhaps significant that none who hunted elephants (eg. Selous, Finaughty) did so in the KNP area while those that did hunt there (Vaughn Kirby, Glynn) did not hunt elephants.

A final speculative clue may come from the floral and faunal diversity which still exists in the Kruger today. Over two thousand plant species have been recorded, some of which are known to be vulnerable to elephant. If there had been successive episodes of high densities of elephants over time, this complex diversity may have been much reduced.

Why elephant densities may have been low is not known, but one theory speculates that perhaps numbers could have been held in check by early man. If densities were low, and man had the means to kill elephant calves, say using poisoned arrows, they would not have to kill many to impose limits on their population growth. But however many elephants that there may have been, they were shot out by hunters, and by 1903 James Stevenson-Hamilton, the park's first warden, reported that there were no elephants to be found. They responded quickly to the new sanctuary, and by 1905 his staff had found evidence of their occurrence near the confluence of the Letaba and Olifants rivers which is located roughly midway between the Park's northern and southern boundaries. From there they gradually recolonised the whole Park. Diaries of the earlier Rangers recorded the first sightings of elephants in new localities. The recolonisation process to the northern and southern extremities of the Park occurred at about the same rate. Northward to the Luvuvhu River took until 1945 (40 years to cover 290 kms) while the spread southward to the confluence of the Crocodile and Sigaas rivers, was slightly slower, taking until 1958 (53 years to cover 280 kms).

Numbers increased steadily during this time and concern over the evident impact they were have was shown as early as 1942. In that year an early park ranger by the name of Steyn made the following tongue in cheek comment in his annual report: "With regard to the question of the control (culling) of elephants in certain areas where it may become necessary, the following idea has come to mind, i.e. to use a 10 or 12 ton armoured car to remove them from any region. This will naturally only be possible after the present war and I leave the details to the imagination of the reader".

Early estimates of population size were made by Park rangers based on their observations, but without the modern aids like helicopters, these estimates were clearly severe underestimates. The concerns intensified until 1967 when the first comprehensive aerial census was conducted and the population estimate of 6 586 proved to be nearly 3 times larger than the 1964 estimate of 2 374. The decision was then taken to limit the population to around 7 000 and culling was initiated.

This figure concurred with those of biologists working elsewhere in Africa at around that time. The elephant population density in Murchison Falls National Park in Uganda was 5 elephants per square mile and it was felt that this considerably exceeded the carrying capacity. In Tsavo National Park, Kenya, the early recommendation was that one per square mile was the right number (this recommendation was never implemented). The 7 000 for Kruger, which has an area of 20 000 square kilometres (8 000 square miles), was just below this. This number gave an average area of 2.7 square kilometres per elephant. As this policy was implemented to prevent the loss of biodiversity from the Kruger's ecosystems, it was undoubtedly successful as no species are known to have been lost from the Park.

So how was the Kruger elephant population managed to maintain it at the prescribed level of around 7 000? A unique technique was developed for the censusing of Kruger's elephant population, which is the most intensive and accurate census of elephant conducted anywhere in the world, though recent intensive fixed wing census techniques in Tsavo using multiple teams and GPS (Global Positioning System) technology probably give comparable results. The Kruger census is conducted annually during August and September to capitalise on the late dry season conditions. At this time of year the animals tend to congregate in the vicinity of watercourses and waterholes, and visibility is at its best due to the trees having shed their leaves. A helicopter is preferred as it can fly at extremely low speeds and can hover as well. In large, loose aggregations of elephants, the pilot can manoeuvre the aircraft systematically from group to group at low level until all have been exactly counted. In contrast, fixed-wing aircraft are forced to maintain forward speed which necessitates circling of the groups which is confusing to observers. Because of the helicopter’s manoeuvrability, a flight pattern which follows the watercourses is flown. Kruger is particularly suitable for this as its undulating savanna terrain is drained by a well spaced network of watercourses which are clearly visible from the air. The pilot begins by flying along one bank of a major watercourse, keeping close enough to it to allow careful scanning of this denser vegetation but yet far enough from it to allow adequate scanning of the ground as far as the watershed. He then turns up each tributary and sub-tributary, following it up one bank to its source and back down the other bank. In this way each drainage system is systematically covered before moving on to the next. The watercourses give the pilot visual cues as to where to fly and systematic ground coverage thereafter is almost automatic. The census is conducted at an altitude of about 200m above ground depending on the terrain and visibility. The whole census is completed in 18 days and about 130 hours of flying time. Research has shown that census totals have never exceeded 7% of the expected result. Once the census result was known, a committee (known as the Standing Committee for Wildlife Management) made up of the senior nature conservation staff of the Kruger met to decide on an appropriate culling quota to conform with the elephant management policy. This quota was then achieved either through capture and translocation to other conservation areas where they could be accommodated, or through culling.

But whereas the Kruger managers decided to limit elephant population growth, a laissez faire (non-interference) policy was adopted in east Africa. This policy's roots almost certainly originated from compassion for the elephants themselves. To sit quietly in the close proximity of a herd of elephants who are going quietly about their business is an emotional experience that can not easily be described to anyone unfamiliar with these animals. Their sheer size alone induces a feeling of awe, and you will not have to sit for long before their intelligence, playfulness, compassion and tolerance become evident. All of these attributes of elephants combine to instil in those lucky enough to have experienced them, a feeling of empathy which intensifies the longer that exposure to elephants lasts. These emotions are not comfortable bedfellows with the concepts of killing these wonderful animals.

Elephants are also considered to be a "keystone species" - one upon which other species are dependant for their own survival. Elephants open up areas of thick woodland affording habitats to species favouring those of a more open character. They also pass seeds through the gut to be deposited in seedbeds of fertile dung assisting germination. Trees pushed over give refuge and food to many lesser vertebrates and invertebrates. These refuges are essential to their survival and without these, populations of these species would be severely compromised. Elephants also excavate holes in sandy riverbeds to gain access to the water below. These water holes give other species access to water which would otherwise be unavailable allowing them to survive in arid environments out of the rainy season.

But to stand under the canopy of a massive old baobab tree (Adansonia digitata) and to ponder a little on the size and age of such old giants, is an emotional experience of a different kind, but one which in its own way is no less soul stirring than that which may be gained from elephants. It is perhaps also their size which makes the initial impression, but the aura of age is tangible. The age that these trees may attain is not known as the pithy wood shows none of the rings which allow the determination of age in other tree species. The "Briscoe" baobab is not a particularly large tree suggesting that it is not one of the older ones, yet it was more than a hundred years ago since the carving was carried out. The tree can have changed little in the intervening 120 years, so some of the older ones must be many hundreds if not thousands of years old!

Baobabs are also a keystone species and a little examination of some of these trees will reveal their significance in the environment. The convolutions in the trunks of these benevolent old giants form cracks and holes which provide shelter to many small animals and birds and offer ideal sites to rear their young. To some species, the presence of baobabs is critical - in Kruger the only known nesting sites of both the Böhm’s Spinetail (Neafrapus boehmi) and the Mottled Spinetail (Telecanthura ussheri) are in hollow baobabs. Thus without these trees, these birds would simply not occur in the area. Mosque Swallows (Hirundo senegalensis) and Grey-headed Parrots (Poicephalus fuscicollis) also favour these trees for breeding and any decline in the number of baobabs would also have its effects on the populations of these birds also. Barn owls (Tyto alba) would occur at much lower densities if the nesting holes offered by baobabs were not available, and White-headed Vultures (Aegypius occipitalis) favour baobabs for building their nests as well.

So, quite apart from their aesthetic appeal, both elephants and baobabs play important but different ecological roles - the elephant in its capacity to accelerate nutrient cycling and alter their environments, and the baobab because of its importance to many species of animals as a source of food and shelter. But elephants also feed on baobabs, stripping the bark off and even chipping away the pithy wood with their tusks. In extreme cases the trees may become so weakened that they eventually topple and die. Most tree species die when ring-barked, but baobabs have a remarkable capacity to regenerate bark. They are therefore resistant to utilisation by elephants up to a point, but one of the major causes of mortality in baobabs is drought, particularly once they have had excessive amounts of bark removed by elephants which probably accelerates moisture loss. Anyone visiting the northern parts of Kruger will not fail to notice that most baobabs have suffered considerable bark removal. During aerial censuses of the Kruger elephant population, dead baobabs are also recorded and this has showed that over the past few years, more than 200 have died due to the combined effects of drought and elephant utilisation.

But baobabs are not the only plants which are vulnerable to utilisation by elephants. The knobthorn (Acacia nigrescens) is another keystone species as it is favoured by a wide number of raptors and other birds for nesting. Wahlberg's (Aquila wahlbergi) and Tawny (A. rapax) eagles, White-backed vultures (Gyps africanus) and some of the small goshawks (Accipiter spp.) also favour the mature knobthorns, while quelea finches (Quelea quelea) and Wattled starlings (Creatophora cinerea) are communal breeders favouring stands of the stunted forms of this species. They breed in thousands (even millions) in these stands and are a staple food of migrant Steppe and Lesser Spotted eagles. Luckily the knobthorn is a very common species but few mature knobthorns in Kruger are free of some form of elephant utilisation and many have been ring-barked resulting in the death of the tree. Absence of mature knobthorns would have severe consequences for many of these species. Marula trees (Sclerocarya birrea) are often debarked or pushed over and even the branches up to the thickness of a man's forearm are eaten. Numbers of marulas have also declined over the years. Other plant species at risk are fever trees (Acacia xanthophloea), kiaat (Pterocarpus angolensis) and star chestnut (Sterculia rogersii). These are just a few of the species favoured by elephants but where elephant densities increase to the point where food availability becomes limiting on the elephant population, nearly all plant species are at some degree of risk. In some parks elsewhere in Africa, this has meant loss of species from systems with unlimited elephant populations.

In Amboseli National Park (Kenya) for example, an extremely high density of elephants has led to a decline in the woodland in the park resulting in the extirpation (local extinction) of both lesser kudu (Tragelaphus imberbis) and bushbuck (Tragelaphus scriptus). Other species favouring woodland such as gerenuk (Litocranius walleri), giraffe (Giraffa camelopardalis), baboons (Papio species) and monkeys (Cercopithicus species) have also declined inside the Park. In Tsavo National Park (Kenya), dense woodland was changed into open savanna and baobabs are now very rare where they were once common.

The Tsavo story is an interesting one. The huge elephant herds that modified Tsavo's landscapes this way are now severely depleted - by poachers, not managers! Apart from being one of Africa's most famous national parks, it gained initial fame through the book, "The man-eaters of Tsavo" which tells the story of two lions which killed at least 28 labourers and even held up the construction of the railway in 1898. J. H. Patterson, the author of this book was a keen observer and he described the nature of the bush as "interminable nyika... the whole country covered in low stunted trees... the only clearing being the narrow track of the railway". Elephants are hardly mentioned in the book. In 1903, the writer/traveller Mienertzhagen found "very few" elephants in the thick Tsavo bush, but by 1970, this had changed and human developments outside had compressed the population into the park - an area roughly the same size as Kruger. Aerial surveys in that year estimated the population to stand at 45 000. These elephants had a huge impact and by 1974, severe droughts had reduced the population to 36 000. Nine thousand elephants had died and at the same time, 4 000 black rhino also died of starvation. The carcasses of these animals offered an opportunity to local people and corrupt staff to establish an illegal trade in the horns and tusks. Having become established, this trade turned to the poaching of live elephants and rhinos and by 1989, the Tsavo elephant population had been reduced to just 6 000. Thirty thousand elephants had been poached and black rhinos had been all but extirpated. Tsavo today is an open savanna which has favoured the grassland species, and the removal of the trees has had some other interesting consequences. Fountains have appeared where they were not known to occur before - the moisture which once was sucked up by the trees and transpired into the atmosphere, now seeps into the drainage lines. These changes are perceived by some as benefits to the Tsavo ecosystem induced by high densities of elephants, and their role in this transformation has often been lauded. But an unanswered question remains - what would Tsavo look like today had the 30 000 elephants not been removed from the system by the poachers?

The poaching has now been contained and the most recent census of Tsavo's elephants in 2007 yielded a total of close to 9 000.

Elephant management options
Given this background, the obvious basic question about elephant management in any confined reserve is whether or not to limit elephant numbers. If the management authority decides that the maintenance of biodiversity is the objective, what options are available? In order to limit the population growth of any species (and cynically, this applies even to humans), there are only three possible options. These are translocation, contraception and culling or a combination of these. Others such as hunting are just variations of one or more of these options. In western society, non-lethal means of population limitation would clearly be preferable to the killing of elephants, but there is a strong body of opinion in Africa which supports the sustainable use of wild animals (even elephants) to provide benefits to local people and communities. Who is right? Westerners who live far removed from the problems or local people who often suffer terrible depredations from living close to elephants and see the killing of the animals as a desirable solution? For them, if the problem animals are removed, the meat becomes available and revenues can even be generated. In some areas such as in Zimbabwe, problem or excess animals are hunted by professional hunters in schemes known as CAMPFIRE (Communal Areas Management Programme For Indigenous Resources). These hunters pay for the privilege and the money is given back to the people for community development projects.

If non-lethal means are to be employed, there are three options available - range expansion, translocation and contraception. Range expansion would usually be the first choice if it were available, but of course, very little land is now still available to allow for this. Perhaps one of the last large remaining patches of land that could still be zoned for conservation was recently proclaimed on Kruger's eastern boundary in Mozambique. Known now as the Limpopo National Park (LNP), it is half the size of Kruger. Even here there are still many people (an estimated 22 000) still living on the land, and there are many problems associated with this that still need to be overcome. In the modern era people can no longer simply be relocated. There were very few elephants in this area when it was proclaimed and it was initially seen as a considerable opportunity for Kruger to dispose of its excess elephants.

Range expansion
There was a common perception that once the game proof fence between Kruger and the proposed park had been removed, elephants would immediately move off into Mozambique, alleviating the need to limit the population in Kruger. But studies of elephant movement in Kruger have revealed that they show a remarkable degree of fidelity to their home ranges. These home ranges have an average size of about 1 200 square kilometres. Factors which may be expected to stimulate movements, such as rainfall in nearby areas and culling, have little effect on their movements except within their respective home ranges. They may be induced to move to other parts of their home range, but they do not leave these well defined areas. This is certainly because these ranges are the areas where they are most comfortable. The old matriarchs know the area well – they know where water will be found in the dry season, where food availability may be best - information and experience that they have built up over their long life times. They also have the ability to extract sufficient nutrition from the available vegetation even during severe droughts, so they have no real necessity to move.

After the proclamation, 25 elephants (three family units of seven each and four adult bulls) were translocated into the new park. All of these animals returned to Kruger and made their way back to their original home ranges. This illustrated that translocation over a relatively short distance will not be successful unless the elephants are prevented from returning to their original home ranges by very robust, intact and well maintained fences.

However, more recently, three sections of the eastern boundary fence between the Limpopo and Olifants rivers were removed to allow for natural recolonisation of game in this new trans-frontier conservation area. At certain river crossings (Shingwedzi and Nshawu rivers) wash-aways had also occurred creating more gaps in the fence. A problem with simply creating gaps in a fence is that only the animals whose home ranges border on the fence will even be aware that a gap has been created. The vast majority of Kruger’s elephants will not even know that there is a fence there, or that that there is now an opportunity to expand their home ranges to the east. There can therefore be no sudden spill of elephants into the new area. The resident elephants will start the process by exploring the opened up area and will gradually expand this deeper and deeper into Mozambique. This may relieve population pressure in the immediate area of the gaps, but not in the vast remaining area of the Kruger.

In October 2006 an aerial survey of the Shingwedzi River basin area of the LNP was conducted, and a total of 630 elephants were recorded at widespread localities in the survey area. This census result showed that the natural recolonisation process had begun, but censuses within the Kruger showed that no decline had occurred. Population growth within Kruger was negating any emigration to Mozambique.

An aspect of the concept of range expansion is a new idea of corridors connecting conservation areas to enable the establishment of what are called “metapopulations”. A metapopulation is a group of spatially separated populations of the same species which interact through dispersal. In this instance, dispersal is supposed to occur through the established corridors. The promoters of this theory suggest that such an arrangement could allow for stabilisation of population numbers through a model known as a source-sink.

This is a model used to describe how variation in habitat quality may affect population growth. Source patches are high quality habitat in which a population will increase while in the sinks, low quality habitat can not support a population and survival is reduced. The theory holds that excess individuals from the source will move to the sink. The supposition in the case of elephants is that in the sinks mortality rates would be higher than birth rates and higher than in the source population. Those advocating this idea suggest that by emigration from source to sink elephant numbers could be prevented from increasing to levels that are deemed undesirable. They acknowledge that sink population numbers must be limited, and while they advocate natural control, they concede that there may be a need for people to reduce numbers in sinks. This they see as “natural” given that human predation is considered a major historical source of impact on elephant numbers. To manage sinks effectively, local communities could be allowed to hunt elephants in a controlled and authorised manner. They argue that this must be seen in the context of sustainable resource use and must provide tangible benefits (for instance in the form of meat or revenue from sports hunting) to these communities.

While there would be a few benefits emanating from the implementation of this model (benefits to local communities, genetic exchange within the metapopulation, There also seem to be a number of potential problems. These are:
• Mortality rates in sinks must exceed birth rates in the source population and in the overall metapopulation. Normal increase in an elephant population is between 5% and 7%. In a large population this means a considerable mortality must take place (675 – 945 deaths). If the habitat was that poor as to induce mortality of this scale, why would elephants want to go there? Similarly if local communities were allowed to hunt elephants, it would need to be conducted at a similar scale. Elephants are not stupid and with the harassment of hunting, they will be unlikely to remain in the sink and will quickly move back to the source.
• It is not clear how hunting in the sinks differs from culling at source. If the objective is to obviate culling, why is hunting acceptable? Sports hunters may be interested in hunting adult bulls, but what about the family groups? Hunting or culling of individuals from family groups is seen as inhumane due to the special social bonds which exist between family members and the trauma involved in killing individuals. This form of elephant management is banned by legislation in the National Norms and Standards for the Management of Elephants in South Africa. It would therefore not be acceptable for local communities to be allowed to hunt elephants for meat or revenue in these sinks.
• Corridors. The same problem exists for corridors connecting metapopulations as for range expansion - very little land is now still available to allow for this. Can they be established in areas where people are already established? What about human-elephant conflict issues in the corridors? Can elephants be induced to use the corridors if people are present?
• Very few elephants will be in a position to utilise the corridors. As is the case with fence gaps, only the elephants whose home ranges border on the corridor opening will even be aware that a corridor has been created. The vast majority of the rest of the elephant population will be unaffected by its creation. It is unlikely that sufficient elephants will use the corridor to affect the population size within the source.

Translocation
Translocation is the second option and is one that most will find ethically acceptable. Whole families can now be moved together so splitting of families does not occur, and it has the additional advantage of establishing other elephant populations elsewhere. In southern Africa today however, saturation point has almost been reached. Most conservation areas already now have elephants and their managers understand the consequences of too many elephants and do not wish to increase their populations any further. Translocations are very expensive and ultimately the areas acquiring excess elephants from one conservation area will one day be faced with exactly the same dilemmas - the problems are merely transferred. Translocations are therefore still not the ideal solution. They provide temporary respite, but in a population as large as Kruger’s translocations can not be conducted at a rate fast enough to reduce numbers, and there are currently no reserves requiring elephants.


Contraception
Contraception is the third non-lethal option. Two methods have received attention so far, the first being through hormonal control using oestrogens. A project researching this method was terminated in

Very good article - but the last portions is missing.

On paper everything could be made to make sense, but on the ground in real settings, elephants will do unpredictable moves every day just like Global warming.

Remainder of the article, summary and crux

Contraception
Contraception is the third non-lethal option. Two methods have received attention so far, the first being through hormonal control using oestrogens. A project researching this method was terminated in Kruger on humane grounds as there was strong evidence that the hormones were drastically affecting behaviour of the vaccinated cows and attendant bulls. Females were induced into a state of "false oestrus", and bulls were attempting to mate with them while they were not receptive. This lead to harassment of the cows by the bulls to the point that they got separated from their families, and even from their small calves. Three of the calves died during the research period, either through starvation or predation.

The other approach achieves contraception through Porcine Zona Pellucida (pZP) vaccination of adult elephant females. This vaccine is made from the ovaries of pigs obtained from commercial slaughterhouses. The vaccine stimulates the animal's immune system to produce antibodies which bind to the outer membrane (zona pellucida) of the elephant cow's egg cells (oocytes) and prevents penetration of sperm. This method has no hormonal or behavioural consequences. But in an area the size of Kruger with a population of 13 500 elephants, even this method is unlikely to provide a solution for logistical reasons. Computer modelling has shown that to stabilise an elephant population, approximately 70% of all breeding females have to be under treatment at any one time. In Kruger this means about 5 500 females. Currently the technology will give a contraceptive effect for up to about 2.5 years. The expense would almost certainly be prohibitive.

In a smaller reserve however, contraception programs using pZP vaccines have been implemented with great success.

A problem with the use of contraception is that it can not reduce a population over the short term. Contraception prevents additions to the population and thus stabilises the population, but to achieve a decline in numbers, you have to await natural mortality. Elephants are long lived (± 60 years) and many years will have to pass before a significant decline will have occurred.

An unexplored possible negative consequence of limiting an elephant population through contraception would be a decline in the sizes of matriarchal (family) units. Young cows stay with their mothers for as long as the mother remains alive. The normal elephant family is thus a large extended one consisting of the old matriarch, her surviving daughters, and all of their respective offspring. A cow may have as many as eight or ten calves in her life of which half could be expected to be female. A family could thus be comprised of five or six adult females with calves of varying ages in a group size of around 15 or more. This is an important social structure among savanna elephants in which young animals learn essential lessons in life, particularly with regard to birth, death, mating behaviour and rearing of young. If the objective of a contraception program was aimed at stabilizing numbers in the population, each cow would theoretically only be allowed to have two offspring, one of which could be expected to be a male. The outcome of this would be that the average group would thus consist of only the old female, a single daughter and possibly one or two calves. Given the social importance of the family, would this enforced change in elephant society be ethically acceptable?

Culling
Culling is currently the only remaining long-term alternative, and here there are also a few options. Anaesthetic drugs would be ideal but they can not be used for such purposes as the meat may subsequently not be used for human consumption. The leaving of such contaminated carcasses in the field for scavengers is therefore also questionable. The only drug that has been approved which allows for later human consumption of the meat is succinylcholine chloride (scoline) whose component compounds occur naturally in mammalian bodies. Scoline is a neuro/muscular blocking agent which paralyses the animal by preventing the brain’s impulses from reaching the muscles. This was used in Kruger to cull elephants in the past as it had the clear advantage of obviating wounding and provided a far greater safety margin for staff and scientists attending such culls. However, research conducted in Kruger showed that the use of this drug on elephants was inhumane as the heart muscle remained largely unaffected by the scoline. The locomotory muscles were the first to be affected followed by those controlling breathing (the diaphragm). This meant that the animal was fully conscious but paralysed and unable to breathe, and therefore died of suffocation if it could not be brain shot immediately after becoming recumbent. The use of scoline was then discontinued and the only method now considered humane is by a sharpshooter using a rifle and live ammunition from a helicopter at close range. Only whole family units or single bulls would be culled to prevent trauma to surviving family members.

Culling is the only option which can generate revenues adequate for covering the costs involved. Translocations and contraception are expensive as are land acquisitions for range expansion and corridors.

Previous and current policies
The policy of limiting the Kruger elephant population to a level around 7 000 was maintained until 1994 when it was challenged by an animal rights group. A decision was then taken to place a moratorium on culling until the policy had been reviewed. This review compiled a new policy by 1999 which was never implemented due to sensitivities surrounding elephant culling. The policy has now once again been reviewed in 2008, but has also still to be implemented. In the mean time, the population has increased to an estimated 13 500 at the last (2009) census.

The text below has been extracted from the new policy and gives a brief summary. This policy can be viewed in full at:

http://www.sanparks.org/parks/...sues/plans/elephant/

In essence this new policy focuses less on numbers of elephants than on the impacts that they are having. The park has been divided into various areas constituting designated areas of high elephant impact and low elephant impact. The impact zoning approach considered several aspects including various biodiversity values, incidences of and risks from damage-causing animals, tourism expectations, landscape linkages and trans-frontier opportunities. Five key elephant management objectives have been developed for Kruger:

Objective 1 seeks SANParks to manage elephant impact and human interactions through inducing spatial and temporal variation in elephant use of landscapes through “High” and “Low” elephant impact zones. This should be achieved through:
1. Minimizing the number of additional water points and dams;
2. Mimicking the effect of natural water distribution;
3. Expanding land through contracts and agreements; and
4. Removing restrictions such as fences. This objective deals with direct influences that elephants have on the landscape and the associated suite of values.

Objective 2 focuses on reactive responses and associated actions to ensure that management accommodates both the consequences of historic biodiversity, elephant and tourism-related management philosophies and the current expectations as articulated in the broad Kruger park management objectives. This objective thus strives to ensure that the consequences of historic management actions are minimized by proposing short- to medium-term actions, evaluating risks to other objectives, and implementing actions that do not compromise SANParks’ strategic objectives and primary mandate of biodiversity conservation.

Objective 3 focuses on the effects that elephants have on stakeholders through aligning SANParks’ Elephant Management Plan with co-management and contractual agreements and, where appropriate, revisiting and establishing agreements with stakeholders and affected parties. These actions focus on assessing concerns and issues of various stakeholders, acting on these, informing stakeholders and evaluating how SANParks’ actions affect stakeholders.

Objective 4 strives to align SANParks and Trans Frontier Conservation Area (TFCA) Elephant Management Policies through appropriate bilateral approaches.

Objective 5 is directed at expanding understanding through focused research, namely to evaluate, inform and revise elephant management through collaborative research agreements. This provides for the critical evaluation both internally and externally of SANParks’ achievements against the intentions articulated in this Kruger Elephant Management Plan. The actions provide explicitly for the opportunity to generate information as well as to inform, review and accommodate variance in management actions on annual, bi-annual, five-yearly and ten-yearly intervals.

Through the implementation of this policy, much information should be gained on the dynamics of elephant population cycles, and the consequences of large elephant populations on biodiversity and ecosystem functioning.

Summing up - the dilemma
This brings us finally to the headaches and heartaches that are experienced by all managers of elephants. A decision has to be made before elephant damage occurs as to whether the area for which they are responsible will be managed as an elephant reserve or whether the maintenance of biodiversity will be the priority. The two approaches are mutually exclusive. The decision to manage as an elephant sanctuary is a valid one which is usually taken out of respect for the elephants themselves. Elephants are wonderful animals with which most people easily empathise, and there should be areas where they can live out their lives free from the stresses of the various management options. Amboseli is probably a good case in point. The wonderful research that has been conducted there by Cynthia Moss and her colleagues has taught us most of what we know about the complexities of elephant society and behaviour. There is still much to learn about long-term elephant cycles and Amboseli, with its huge data base and intimate knowledge of the elephants and their ecosystem, may be the best place to study these. The Addo Elephant National Park in South Africa is another. This park was created as a sanctuary for the last remaining elephants in the Cape region. Recent expansions to the park have perhaps shifted the focus towards a biodiversity approach, but elephants will probably always remain the conservation priority.

But a decision to allow unlimited growth of an elephant population needs to be taken consciously by the management authority, in the full awareness of the consequences - they must be aware that ultimately, other species will begin to disappear from the system due to over-utilisation or habitat changes induced by elephants. In some cases this will mean extirpation and perhaps even extinction. Ultimately also, when times of drought come (as they inevitably will in Africa), they must be prepared for the die-offs of elephants which will occur. These will be controversial, disturbing and emotional times.

But if the maintenance of biodiversity is the priority, something will have to be done (usually culling) to limit the numbers and densities of elephants before biodiversity is affected. This is also not an easy decision - the killing of elephants is never one that can or should be taken lightly. But here there lies yet another moral dilemma. At what level should the population be held? To maintain a population at any particular level requires the removal from the population of a number of animals equal to the population's annual increase. The average increase in Kruger’s elephant population has been calculated at a rate of 6.2% per year. To maintain a population at a low level would require the annual removal of a relatively few individuals. Maintaining it at a higher level requires the annual removal of proportionately more animals. As we have seen, when such numbers are involved, the only option for the removal of most of them would be through culling. If you agonize over the morality and ethics of culling elephants, then the issue will be greatly compounded by having to cull a much larger number. Is it not better to keep a population low at the level where few animals need to be culled or better still, at a level where most excess animals could be translocated rather than culled?

There is no middle of the road on the issue of elephant management – in large national parks with large elephant populations a choice has to be made for either one of these options: to cull or not to cull? If the choice is for the latter of these options (which is a valid one depending on the conservation priority for the designated area), there is no going back once extensive damage has occurred. When plant and/or animals species have been lost, it may (through the management of the elephant population) be able to maintain the remaining biodiversity, but the restoration of the system to its former richer state of biodiversity and function will scarcely be possible, particularly while elephants are still present in the system. The dilemma is in weighing up the sacrifice of individual elephants against the sacrifice of species. Either way, the decision taken will always trouble the consciences of those involved in the process. You should not have cold feet if you enter the elephant management arena, and be prepared to be castigated for your opinions and decisions whichever side of the debate they may lie.

So, if the decision was yours, what would you do?

(Now retired, Ian Whyte served in Research Departments in Kruger National Park, South Africa for 37 years, completing his service there in 2007 as Program Manager for research into large herbivores).

A very interesting read. It just goes to support the statement again, no matter how large or small an area is, if it has a fence around it, it needs to be managed!

Thank you for the rest of the article. Everyone interested in Africa and its wildlife needs to read it in its entirety.
Who ever said the management of our wildlife was easy! It takes broad shoulders, a thick skin and big nuts to get it right - qualities which it seems are seriously lacking these days!
The greenies will one day have lots to answer for - all the great national parks in southern Africa will one day hold plenty of elephants and very little else.

The original models of parks were to have hunting blocks surrounding them both insulating the surrounding farms and providing more habitat and corridors for migration.

Without them...fences are the only option and truly wild Africa is gone forever. The hunting blocks control the overpopulation and give worth to the land they encompass. This is the "only" model that will work and self-sustain.

Doctari is certainly correct on how tough it is to maintain the concept as intended...I can attest first hand from work done in the LCTF.

BTT, more people need to read this and understand how Africa is changing and will change. It is Elephant specific but applies to a greater or lesser degree to many other species and areas