The first study that I want to discuss here is by Harrington and others (1999). Although the study is 13 years old, it is a great study showing how complicated natural systems can be, and how the effects of our actions have to be monitored to make sure we pick up their unintended consequences. The study deals with roan antelope, which have always been rare in Kruger. The roan population declined very suddenly in the late eighties, and has since struggled to recover. This study looked at the population of roan antelope from the early eighties up to the mid-nineties to see what was hindering the roan from recovery and if anything could be done to help the population grow.
It's important to bear in mind that the Kruger National Park is located on the southern margin of what would have been the original distribution of roan antelope. As such, roan were probably never really abundant in this area, but nonetheless occurred in greater numbers than they do now.
Harrington and others, in their study, looked at population size for roan and other herbivores from aerial surveys, and also looked at the estimated predator numbers from ranger reports. They also considered the composition of roan herds (male to female, adult to young etc). They then compared their numbers to factors like rainfall and condition of vegetation in the same time period to see if they could match certain events to decline in roan populations. They suggested five possibilities that could have led to the decline, and tested each one against their data. The five possibilities considered were:
1. Habitat deterioration through aridity
2. Anthrax (despite immunisation)
3. Stress from vaccination against anthrax
4. Increased populations of zebra and wildebeest in the roan habitat. Zebra and wildebeest populations increased following the creation of dams and waterholes, which may have led to increased competition for food
5. Increased predator populations - with the creation of waterholes, the lion population increased in roan habitats. The lions most-likely followed the zebras and wildebeest. Note that the actual mechanism of the decline is different in 4 and 5 above, but the ultimate cause (more waterholes) is the same.
So, to figure out which of these factors were important in decimating the roan population, Harrington and others first looked at specifics of how the population declined. Between 1977 and 1985 the population was stable at about 300 individuals. In 1986 it grew to 450, after which it declined to 150 between 1988 and 1991. By 1992 it had dropped to 70 and by 1993 it was down to 44 (this is 10% of the high in 1986!). During the decline, both juveniles and healthy adults died, and the percentage of calf survival remained unchanged.
Other important developments noted by the study are the development of dams and waterholes in the habitats favoured by roan:
By 1970, five dams and thirty-two waterholes had been built in the northern part of the park in the roans' range. By 1976 there were 9 dams and 46 windpumps, and by 1994, there were 11 dams and 59 windpumps. By 1995, 1 dam and 12 waterholes were closed down.
With regard to potential competitors, zebras numbered 5000 in the northern roan habitat in 1977. This increased to about 14000 during the period between 1986 and 1991.
As for lions in the same area, ranger sightings of prides increased from 10 a year between 1980 and 1985 to 100 a year between 1986 and 1988. This dropped back down to 20 a year in 1989. Investigation of predator kills also showed that roan accounted for a higher proportion of kills than would be expected by their proportion of the herbivore population.
So what caused the roan decline according to Harrington and his coauthors?
The first hypothesis (aridity) seemed unlikely as a controlling factor, because roan declined more than other herbivore populations. Also, a lack of food, as in a drought, is more likely to kill the very old and very young, while those animals in their prime are more likely to survive. But for the roan, healthy adult individuals were also dying.
The second hypothesis (anthrax) was not the cause - an inspection of carcasses revealed that the animals did not have anthrax.
The third hypothesis (stress from vaccination) was also discounted. Of two populations of roan in northern Kruger (one at Mooiplaas and one at Vlakteplaas), only the Vlakteplaas population was vaccinated. Therefore, if stress was the cause, only the Vlakteplaas population would have declined, and the Mooiplaas population would have been unaffected. This was not the case, thus discounting this hypothesis.
The fourth hypothesis (competition with zebras and wildebeest for food) is intriguing. As with lack of food due to aridity, a lack of food due to competition should affect the very old and young before it affect healthy adults. This was not the case. Additionally, zebra populations increased in roan habitats from the early eighties, but roan only started the decline in the late eighties. This hypothesis doesn't entirely explain the lag. This would therefore have been a minor factor.
The fifth hypothesis (increased predation) can explain why healthy adults were dying. It is also supported by the fact that roan were making up a greater proportion of predator kills than their abundance could explain. Also, it explains why roan declined a few years after zebras increased (because it took time for the lions to follow the zebra populations). As such, Harrington and coauthors considered this a likely causal factor in the decline of the roan. Other things like aridity and increased competition for food may then have exacerbated the problem.
So it seemed the problem had been identified: creation of waterholes allowed zebra and wildebeest to move into areas preferred by roan antelope. The zebra and wildebeest were then followed by lions, which increased predation on roan. The problem was made worse by increased competition for food, and also low rainfall in the late eighties.
After the closure of some waterholes in the mid-nineties, some recovery took place. At Vlakteplaas in 1995, 3 out of 4 of the roan herds bred successfully, with this subpopulation growing from 16 to 29 individuals between 1993 and 1997.
So if the problem has been addressed, why hasn’t the roan population shot up in the last ten years? Well, there’s probably still many other aspects of the problem that we don’t understand, like exactly how roan use the resources in their environment, and what effects the increased grazing has on plant species. Ultimately, we need more information on these things before we can figure out a way to increase the population. Finally, because Kruger is at the edge of the range that roan originally lived in, they were probably never very abundant here, and this also makes their conservation here especially challenging.
Harrington, R., Owen-Smith, N., Viljoen, P.C., Biggs, H.C., Mason, D.R., Funston, P. 1999. Establishing the cause of roan antelope decline in Kruger National Park, South Africa. Biological Conservation Volume 90, Pages 69-78.