Elephant Breeding Herd Analyses – Kruger National Park

by Salomon Joubert, September 2015

Introduction and background

Two years ago my wife engaged in a discussion with a group of tourists at the Babalala Picnic spot. All concerned were passionate Kruger Park fans, the ‘other’ party being ex-SANParks Honorary Rangers. Of course, the topics ranged around the excitements of the Park and shared experiences.

Being in the northern areas of the Park and hardly able to avoid plenty of elephants on almost any game drive, the group of ex-Honorary Rangers did not waste time to explain the Park’s approach towards the management of the elephant population. The elephant management philosophy was explained to them by Dr Sam Ferreira during a session between SANParks and the Honorary Rangers. Essentially, water holes were being closed, young and aged elephants were dying because of the longer distances between water and food resources and this had stabilised the elephant population. Come the next drought the deaths would increase and the population would decline. In this way, the elephant population would remain in perfect ecological harmony with the other components of the ecosystem.

On 22 November 2014 Glynis O’Hara, of the Conservation Action Group, published an article liberally quoting Dr Sam Ferreira, a senior research officer based in the Kruger National Park. Dr Ferreira confirmed the story of the ex-Honorary Rangers and, amongst others, was quoted as explaining the rationale of the approach towards the management of elephants as follows:

  • The “natural” way of managing elephants brought the recruitment rate down from 6.5% at the time that culling was terminated (in 1993/4) to 2%. [According to an interview, dated 3 May 2013, with a “Sapa correspondent” (Sapa/str/gm/hdw/dd/jk/ks, 3 May 2013; OrigID: LP 145661) the population growth rate had decreased to 3.5% per annum].
  • Managing elephants was not about manipulating numbers but rather to let natural processes take their course. It was contended that culling did not curtail environmental impacts.
  • The population was currently (Nov 2014) considered to be 16 900, based on a census done in 2012. (According to the Sapa correspondent, quoted above, “Ferreira said the 2012 census counted 13 700 elephants in the park” while it was 8 000 when culling was stopped. Had it continued to increase at 6.5% per annum it would now have totalled roughly 25 000).
  • The new approach was based on the closure of artificial waterholes, roughly 65% since 2003, and the “landscape and vegetation got respite from elephant use”.
  • Natural patterns were now being restored: elephants had to move more extensively to find water while additional land was being made accessible, e.g. into Mocambique and private nature reserves. Due to this the intercalving period had increased from 3 years to 4.2 to 4.5 years, “a classic population response”.
  • In the northern areas of the Park “survival rates have declined, while in the south (where there’s more water) the birth rates have declined.” Though this phenomenon could not be explained the conclusion was drawn that “natural regulation is taking place.”
  • An important consideration is that “now that boreholes are closed … elephants have moved to the rivers, accentuating vegetation impact there, and such ‘lag effects’ have to be studied and monitored.” Thirty-two (32) areas with “lag effects”, referred to as “areas of concern” have been identified.
  • Tourist considerations were important and where excessive impact was taking place elephants could be driven away by simulating human presence.
  • Highest rates of impact did not necessarily coincide with the highest densities but, in many cases, could be ascribed to young bulls in puberty or early adulthood.
  • Of some 150 elephants collared in the Kruger Park all but one have strayed from the Park, albeit only temporarily.
  • Ferreira, as quoted by O’Hara, described the approach adopted in the management of the elephant population as a “giant experiment”.

The rationale for the approach towards the management of elephants in the Kruger Park, as related by the group of tourists and the articles by the Sapa correspondent and O’Hara obviously deviates from the erstwhile approach of managing the population by culling and live sales to maintain it within specified limits (8 000 to 8 500).

Irrespective of what the outcome of this “giant experiment” is yet to prove, an aspect of interest is the statement that young and old were dying, thereby placing a limit on population growth. The pattern of this process was ostensibly also different in different areas of the Park.

We are regular and passionate visitors to the Park, averaging up to 45, or more, nights per year. During our game-viewing trips we frequently encounter elephants and, whether lone bulls or breeding herds, always marvel at these fascinating animals. Due to this we decided that it would make an interesting project to test the statements regarding the age structures of the breeding herds.


Initially the objective was to obtain a total count of the breeding herds and calves under one year of age. However, this was soon changed to include three categories, i.e. juveniles, calves up the age of three years (tusks just about to appear), sub-adults, 4 years to approximately 9/10 years and adults. This approach proved practical and was regarded sufficient to yield satisfactory results.

The method was quite simple. When we encountered a breeding herd and it was possible to obtain a good view, I would scan the herd with a pair of 10 x 42 Zeiss binoculars, calling out ‘adult’, ‘sub-adult’ or ‘juvenile’ and my wife would enter my calls in a note-book. In this way we could determine the numbers of each category.

The kind of problems related to this approach and which had to be taken into account were the following:

  • We fully complied with all the rules and regulations applicable to tourists, with very limited opportunity for manoeuvring for better positions. With patience this handicap was frequently overcome and only herds of which we were confident of an accurate count were taken into consideration. Roughly 80% of the herds encountered could be counted, the others rejected.
  • Particular attention had to be given not to miss very young calves, which were often effectively concealed near the legs of the mother or even under her belly. The dry conditions with the deciduous trees largely leafless and the field layer (grasses) relatively sparse and short were helpful. Nevertheless, some may have been missed which would have resulted in a bias in favour of the adult segment.
  • Particular care had to be taken not to count the same herd twice. This potential problem cropped up due to the fact that we camped in the same area for 7 or 8 days and some of the routes were travelled more than once. This could also apply to waterholes, e.g. at Punda Maria three herds drank at the waterhole in front of the rest camp. One herd had a cow of which the left tusk pointed prominently straight down and only one of the other two herds has been included in the tables. The same approach was taken in other similar situations.
  • A number of herds (or groups) consisting of 4 or 5 members were encountered. Due to the limited manoeuvrability it was not possible to scan the area to ascertain whether these represented groups temporarily isolated from the main herd and were therefore not taken into consideration.
  • It was not always easy to distinguish between older sub-adults and young adults. To remain on the conservative side doubtful cases were allocated to the adult category.

Results and discussion

The results presented in Table 1 (below the text) refer to three districts, i.e. the Southern District, encompassing the area between the Crocodile and Sabie rivers, the Central District, between the Sabie and Olifants rivers and the Northern District, the area north of the Olifants River. Counts in the Southern District were restricted to 8 days, from 11 to 18 March 2015, while the Central and Northern districts were done between 30 July and 30 August 2015.

The major part of this survey was conducted towards the end of a long and dry winter. Due to good rains over the previous years the “carry-over” from this period was obvious throughout the Park. Copious water supplies were available in all the perennial and larger seasonal rivers, grazing was showing signs of a long, dry winter but generally in good supply and, with few exceptions, animals of all species were in good condition. Elephants were, if anything, in prime condition.

The results reveal some interesting trends:

  1. On the face of it, the ratios of adults to immatures and of sub-adults to juveniles are remarkably similar in all three districts.
  2. The spread of immatures in the three districts, i.e. 53.1%, 55.0% and 54.2% for the Southern, Central and Northern districts respectively, does not reflect a population with a recruitment rate of 2% and is probably closer to 6.5% (as at the end of culling in 1994).
  3. Similarly to (b) the ratio of juveniles to sub-adults (17.3 : 35.8; 19.5 : 35.5 and 20.0 : 34.2 for the three districts respectively) also indicates a vigorous and growing population.

These data do not indicate a population that is static (stable). On the contrary, they reflect a population that is vigorous, healthy and growing and at similar rates throughout the Park.

The results of this casual survey therefore clearly contradict the statements of SANParks.

In Table 2 the population totals for a variety of the larger herbivores for 1971 and 1992 are also given. These years represent the end of two protracted periods of severe drought of similar duration and intensity. From these data it is apparent that the unprecedented increase in the number of dams and boreholes provided subsequent to 1971, played no discernable role in creating and sustaining animal populations at higher population levels. The period 1973 to 1980 was also one of the wettest in the recorded history of the Kruger Park, with five consecutive years (1973 – 1978) recording rainfall above the long-term mean.

Table 2: Census figures for elephant indicating no response to 1971 and 1992 droughts and of selected large herbivores illustrating no discernable trends related to the provision of artificial water resources.

The census totals for most of the large herbivore species peaked during the mid-1980’s. This was in spite of a succession of drought years from 1981 to 1983, and most probably due to the favourable conditions provided by the “carry-over” from the wet years. Though several of these populations dropped significantly in the latter part of the drought years the populations, at the end of 1992, stabilised at very similar levels to those of 1971.

It is true, as stated by Dr Ferreira, that the elephant population has not been subjected to a deep drought similar to that of 1991/2 in recent years. However, from the census results of that time, and those from similar conditions in 1971, derived from attempted total counts (Table 2), no adverse effect was recorded for the elephant population that could be ascribed to the drought. The question then arises whether the series of dams and artificial waterholes ever even played a role in the survival of the elephants, either young or old. If so, it could only have been to a very minor extent.

Of concern are the 32 “areas of concern”, where heavy impact has been identified. By what manner and means will these areas be relieved of the impact, they are areas obviously selected by elephants as favoured habitats so the assumption is that they will be deprived of these habitats, if elephants are to be “evacuated” from their choice habitats (which is highly doubtful!) without reducing their numbers where are they supposed to go? Has anything been done in this respect?

Also of concern is the acknowledgement by Dr Ferreira that the cessation of culling is a “giant experiment”. What if this “experiment” does not have the desired outcome?! There is no quick fix, if any at all at this stage.

In October 2010, after Sam Ferreira and I had had some discussions on the elephant situation, I queried the following:

“If population control cannot be exerted I regard the new approach as extremely high risk. This is based on the following:

  • It is quite apparent that the Park does not have the research capacity to properly evaluate and prioritise areas that would qualify for ‘event-driven’ intervention.
  • It is not clear what ‘intervention’ implies. If it does not entail the reduction of the elephant population where are the elephants of the impacted area supposed to go?
  • Justifying the unchecked increase of the elephant population on the basis of an “early signal” (also read “giant experiment”), supposedly signifying the reduced increase and leveling off of the population, is scientifically unfounded.

Any suggestion that the elephant population would eventually level off and continue at a stable level in harmony with other ecosystem components does not take into account elephant population cycles and is therefore equally unfounded on a factual/scientific level.”

Furthermore, “To let the population go on the grounds of unproven and wishful assumptions (however popular this may be in terms of public sentiment) can very well lead to an irreversible situation that offers very little opportunity to manage the population or effect rehabilitation of disturbed areas if and when the “early signal” assumption may, in fact, prove to be false. This would be akin to signing a contract without an escape clause!

If one has to err, err on the conservative side and retain the opportunity of redeeming mistakes, even if they are – as accepted – made with the best possible intentions.”

I am not aware that Dr Ferreira has, at any time, in either a published paper or unpublished internal report, presented the data on which he has drawn his conclusions and on which the current approach towards the management of Kruger’s elephants is based. Such profound statements as quoted above cannot be founded on anything other than intensive and very thorough surveys, an analysis, interpretation and publication of the data. I challenge Dr Ferreira to present his data.

The last elephant census of the Park was conducted in 2012 while a follow-up census was done in August 2015. My bet is that if the private nature reserves and transfrontier area of Mocambique are included (they are all part of the Kruger “open ecosystem”), the total will be in the region of 24 000 – 26 000.

Table 1: Analyses of elephant breeding herds encountered throughout the Kruger National Park during March and August 2015.



NORTHERN DISTRICT (north of Olifants River)

One Comment

  1. Pingback:The Kruger Controversy PART I – The True Green Alliance

Leave a Comment

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.