Chapter 12
Food resources constitute an ultimate limit on the population size attained, but how this operates is not so simple and direct. For large herbivores, food production depends on rainfall, which varies seasonally and between years. During the wet season, there is more green vegetation than herbivores could possibly eat. Towards the end of the dry season, animals may be starving because the little plant material that remains is mostly dead or dormant, largely indigestible and perhaps toxic (Figure 12.1). They may be forced to crowd near remaining water sources, increasing their exposure to predation. They may have to take greater risks in seeking what food remains. Many large herbivores migrate seasonally, while others remain resident year-round. Demographic segments make distinct contributions to population changes generated by the balance between births and deaths. Over multi-year periods, populations expand and contract in their distribution ranges, governed by various influences. In such spatially and temporally variable contexts, how can any stable abundance level, or ‘carrying capacity’, be maintained? These are the topics to be addressed in this chapter.
Figure 12.1
Trends in large herbivore populations in Kruger NP, from annual aerial counts over 25 years. The five most common species maintained fairly stable populations fluctuating twofold in abundance, while four less common species showed substantial downturns in abundance during the early 1990s. Note the log scale on the y-axis enabling populations to be compared on the same proportional scale.
Seasonal Food Limitations
How herbivore populations respond to changing environmental conditions may not follow the expected patterns. The large ungulate that is most responsive to annual rainfall variation, in Kruger NP1,2 and elsewhere,3 is the African buffalo, despite its large size and digestive efficiency. This is because, as bulk grazers, they are most directly dependent on the amount of grass produced annually. Moreover, being especially water-dependent, they are confined to the vicinity of long-lasting water bodies through the dry season, where grass gets locally depleted. Buffalo herds must then undertake regular journeys between places that retain grass and where water still remains. During the severe drought of 1991/2, when the annual rainfall total was only half of the mean, Kruger Park’s buffalo population fell by 45 percent.2 In the Mara region of northern Serengeti, buffalo numbers dropped by almost 50 percent during the 1984/5 drought, and by over 75 percent during 1993/4 when rainfall deficiencies were worsened by an influx of cattle into the protected area.3,4 In Ngorongoro Crater, drought-related mortality among buffalo in 2000 accounted for 30 percent of the population and was associated with complete loss of offspring in the current year plus a lack of calves in the following year.5 However, buffalo can be quite mobile and, during the exceptionally severe drought that prevailed through 2014–16 in Kruger NP, herds moved to places where local rain showers had generated more grass, alleviating mortality.6
Plains zebras, able to exploit stemmy grass remnants through their non-ruminant digestion, appeared least sensitive to rainfall variation among Kruger Park’s grazers.7,8 Their numbers have also varied little in Serengeti.9 However, their population dynamics were more responsive to annual rainfall variation in private conservancies in central Kenya.10 In Serengeti, the survival of zebra foals actually improved during years with lower rainfall.9
After buffalo, the ungulate next most responsive to annual rainfall variation in Kruger NP is the greater kudu, a browser.1,11 However, the mechanism involved is not simply the lack of food affecting survival through the dry season, but rather the effect of wet season rainfall on calf survival, presumably acting through maternal influences on late foetal growth and milk supply after birth. This is probably due to the influence of rainfall on forb production in the herbaceous layer.
The population dynamics of wildebeest responded only to the dry season component of the annual rainfall, both in Kruger NP7,8 and in Serengeti.12 Dry season rainfall prolongs the period over which green growth persists among short grasses. Higher wet season rainfall producing taller grass may also result in greater cover for stalking lions. In Kruger NP, other grazing ruminants exploiting somewhat taller grass than wildebeest responded positively to both the wet and dry season components of the annual rainfall total.8
The megaherbivores, including giraffe along with elephant and white rhino, responded little to annual rainfall variation in their population trends in Kruger NP7 and elsewhere,13,14 despite their bulk food requirements. Although the elephant population in Tsavo East NP crashed by 20–30 percent during very severe drought conditions,15 this mortality was exacerbated because elephants had been compressed within the park by settlements outside its bounds. Drought-related mortality of this magnitude has not been recorded among elephants elsewhere. In Amboseli NP in Kenya, severe drought conditions caused mortality especially, although not exclusively, among calves.14 Hippos are more prone to drought-related die-offs than other megaherbivores16 because their grazing is restricted to river or lake margins, where forage gets depleted by concentrations of water-dependent ruminants, as well as by what hippos eat. The very largest herbivores have greater metabolic inertia to buffer their population trends against annual variation in food availability, which only temporarily affects fertility and calf survival.
Nevertheless, there are circumstances in which less-large herbivores can incur quite severe population crashes, amounting to 50 percent or more.17 In the Central Kalahari region of Botswana, 90 percent of the wildebeest died when fences blocked access to surface water during the extreme 1982/3 drought, while hartebeest, eland and kudu numbers fell by 50 percent or more.18 In the Klaserie Private Nature Reserve near Kruger NP, >90 percent of wildebeest and warthog, >80 percent of buffalo and zebra, and 70 percent of impala died during this same drought.19 The exacerbating feature there was excessive provision of surface water in the form of the dams constructed on private properties, elevating densities of the water-dependent grazers before the drought. With water no longer limiting spatially, the grass cover became denuded everywhere, removing the resource buffer that might have persisted remote from water. The browsers, specifically kudu and giraffe, incurred less mortality than the grazers because woody plants produced new foliage at the end of the dry season despite the lack of rainfall, alleviating starvation.
In summary, most herbivore populations, apart from the very largest species, fluctuate in abundance in response to annual rainfall variation, indicating the controlling influence of changing food production. However, the range in variation seems typically no more than about twofold, achieving such approximate constancy in the numbers of animals supported from year to year. Greater seasonal food depletion during severe droughts, at least locally near surface water, can generate extreme population crashes, especially where animals cannot move freely to exploit spatial variability in rainfall. Demographic inertia in the maximum population growth rate (see below) limits the extent to which large herbivore populations can grow beyond what the prevailing food resources can support in successive years. However, via their feeding and other impacts these herbivores, especially the largest, can alter the vegetation cover, for better or for worse (see Chapter 13). Predation can potentially restrict how closely herbivore populations approach the food ceiling, at least locally.
What Difference Does Predation Make?
All animals ultimately die, but their deaths may occur somewhat sooner through the agency of a carnivore and reduce their reproductive contribution to maintaining a population. However, with fewer mouths left more food is available to feed the animals that remain and promote their reproductive success. This mechanism compensates to some extent for the losses of animals killed by predators. Hence the additive effect of predation on prey abundance is somewhat less than commonly surmised. The indirect effects of the risk of predation on where herbivores can safely feed and rest can be much greater.20
Moreover, predation usually interacts with resource limitations.21 While food remains plentiful, herbivores can restrict their movements to places where they are most secure from predation. When food runs short, in the dry seasons of dry years, they are forced to take greater risks while seeking food, increasing their exposure to predation. Hence although almost all herbivores die through the agency of a predator, the mortality incurred by the population depends on the prevailing rainfall conditions.7 The main effect of predation is to restrict habitat occupation to more secure places, most of the time. For example, in Kruger NP wildebeest herds remain settled within open glades with short grass providing little cover for predators for most of the year, day and night, until little grass remains, before venturing out into more densely wooded areas.22,23
The reason why wildebeest are orders of magnitude more numerous in Serengeti than in Kruger NP is due not simply to better-quality grazing, but also to the vast extent of open grassland where they are relatively secure from being ambushed by lions. The huge concentrations of migratory ungulates overwhelm the capacity of lions to have much additive impact. About 15 percent of wildebeest deaths and 30 percent of zebra deaths result directly from starvation, and many of the wildebeest eaten by lions are scavenged from hyena kills.24 Nevertheless, the densities of resident ungulates, including topi, impala and warthog, increased in northern Serengeti after lion numbers there were reduced as a consequence of illegal hunting concentrated on buffalo, despite the fact that the species responding positively were not the primary prey species of lions.25
However, there are circumstances in which predation can have a substantial impact on prey abundance, via a nexus of interactions among alternative prey species. In Kruger NP, several of the less-common herbivore species declined to 20 percent or less of their former abundances, shortly following a severe drought (Figure 12.2). Their downward trends were associated with a two- to threefold increase in adult mortality, implicating increased predation specifically by lions.26,27 Lions had increased following a doubling in zebra numbers. Zebras increased because park managers had provided additional waterholes, aimed at alleviating the effects of droughts. Lions turned to hunting the less-common ungulate species, which became more vulnerable to predation than their preferred prey, i.e. wildebeest, zebra and buffalo, as a consequence of the drought for various reasons. Notably, it was the alternative prey species, rather than the primary prey species, that were most susceptible to elevated predation. Furthermore, none of the ungulate species adversely impacted has yet gone locally extinct in the park, although the status of roan antelope is precarious.
Figure 12.2
Food depletion by the late dry season. (A) Grass cover grazed down to bare soil by wildebeest, basaltic region of Kruger NP; (B) zebra grazing remnants of taller grass in same region of Kruger NP; (C) buffalo grazing remnants of grass in Kruger NP; (D) buffalo trekking to water over area barren of grass, Kruger NP; (E) kudu nibbling flowers appearing on trees before leaves in Kruger NP; (F) skinny impala nibbling remaining leaves on a shrub during drought.
This scenario revealed some of the mechanisms dampening fluctuations in prey populations.21 Buffalo, the largest ungulate in the primary prey base supporting lions, gained demographic buffering because prime-aged animals are not easily killed, meaning that lion prides focus on young buffalo in the herds. Medium-sized ungulates gain security by occupying habitats where they are less vulnerable and by showing contrasting responses to rainfall variation. Relative population stability was conferred by the habitat template. When conditions are benign, animals can mostly occupy secure sites. During adverse conditions, they are forced to spend more time outside these refuges seeking whatever food and water remain. Predation and diminishing food availability thus interact to counteract population fluctuations.
Migration: Seasonal Relocation of Populations
All of Africa’s migrants are grazers, because it is the grasses that respond most directly to gradients in rainfall via forage production and quality. Migratory populations attain vastly greater abundance levels than shown by those remaining resident year-round, for various reasons.28,29 Seasonal concentrations swamp the capacity of resident predators, so that many carcasses are left uneaten.30 Predator numbers are restricted by the dearth of prey left after the migrants have departed. Regular seasonal migrations over long distances are, or were, a feature of numerous wildebeest and many zebra populations across Africa.31,32 Mass migrations are still undertaken by white-eared kob (Kobus kob leucotis) and tiang (Damaliscus lunatus tiang) east of the Nile River in southern Sudan.33 Other grazers shift between seasonally separated ranges without moving as far, being confined near surface water in the dry season and dispersing further afield in the wet season.34 Some ungulates move essentially nomadically, i.e. widely but less regularly; for example, eland and various gazelles.35,36 Many ungulate species remain within the same home ranges year-round; for instance, sable and roan antelope as well as various reduncine grazers, plus most browsers. Some populations are partitioned between a portion that migrates and a remnant that stays behind, a phenomenon labelled partial migration.37 Even in Serengeti, wildebeest populations remain resident year-round in the western corridor and in Ngorongoro Crater.
Several benefits accrue to migrants. They (1) gain access to high-quality but ephemeral resources (Figure 12.3), promoting reproductive success,38,39 (2) locally dilute predation,28 and (3) get dispersed more broadly than residents. Lack of year-round surface water can contribute to the abandonment of the wet season range. When migratory movements get blocked, herbivore populations collapse to much lower numbers.40
Figure 12.3
Migratory wildebeest and zebra in Serengeti. (A) Wildebeest dispersed over the eastern short grass plains early in the wet season; (B) zebra spread where the grass is taller further west; (C) zebra flanking a wildebeest concentration on the short grass plain; (D) wildebeest aggregated in the central woodlands during migration; (E) wildebeest spread in northern Serengeti towards the end of the dry season after most grass has been grazed short; (F) aggregation crossing the Mara River back south near the end of the dry season.
Genuine or ‘mass’ migrations represent one extreme of a spectrum characterised by regular seasonal shifts in home range occupation. Importantly, seasonal concentrations of migrants greatly exceed the local densities that could be supported locally year-round. The numerical density of wildebeest in Serengeti gets amplified almost 10-fold to nearly 500 animals per km2 while they concentrate in the 2500 km2 extent of the short grass plains during the time when calves are produced. Similar concentrations develop within the Maasai Mara Reserve in the north where the migrant wildebeest congregate during the dry season, benefiting from the green grass retained there due to locally high rainfall, along with water readily available from the Mara River. In southern Sudan, densities attained by kob and tiang on the Nile River floodplain can exceed 1000 animals per km2 locally during the dry season.29 Historically, huge numbers of ungulates of various species concentrated seasonally in Kenya’s rift valley near Lake Nakuru, especially zebra and Thomson’s gazelle, before hunting eliminated them.41 Migrations were probably more widely prevalent in the past, before modern humans intruded.
Life History: Births, Deaths and Population Growth
Life-history features determine (1) how fast a population can potentially grow, and (2) the recruitment rate needed to counterbalance adult mortality through predation or other causes. Formal population models must accommodate (1) age at first reproduction, (2) number of offspring produced in each litter, (3) inter-birth intervals, (4) mortality incurred at different ages, and (5) the potential lifespan. Population models focus on the female segment, because the number of males makes little difference to the dynamics, unless there are very few or if the birth sex ratio deviates much from 50:50. The potential population growth rate depends on body size, because bigger animals take longer to reach sexual maturity and have longer inter-birth intervals associated with longer gestation periods.42 This is largely a consequence of how the metabolic rate scales with body size, as discussed in connection with food requirements in Chapter 10. Longevity is ultimately curtailed by the wearing down of tooth surfaces and consequent impairment of food processing.43
Ruminants from medium-sized impala (45 kg) to large eland (450 kg) all have gestation periods of 7–9 months, enabling them to synchronise annual births with the wet season.44 Females typically first give birth at three years of age and potentially produce a single offspring annually during a lifespan of ~15 years. The maximum rate of population growth that can be sustained if all females survive with zero mortality from birth to the maximum lifespan is almost 30 percent per year. This rate could be exceeded temporarily if offspring mortality, perhaps in a drought year, leaves a population comprising mostly prime-aged females.
Smaller antelope like springbok and steenbok can attain reproductive maturity by one year of age, have gestation periods of 4–5 months, and may reproduce twice during some years, but seldom live longer than 10 years.45,46 They can potentially attain a population growth rate as high as 40 percent per year. Buffalo (gestation period 11 months) and zebra (gestation period 12 months) cannot maintain annual synchrony of births with the wet season and periodically skip a year.47,48 Furthermore, they both produce their first offspring only at 4–5 years of age, with longevity lengthened to 20–25 years to compensate for the slower recruitment. Their maximum rate of population growth is around 15 percent per year. Giraffe, with a gestation period of 15 months and birth intervals generally around 20 months, no longer maintain birth synchrony with the wet season and live up to 26 years of age. Rhinos have gestation periods of 15–16 months, give birth first around 7 years of age and have a mean birth interval of 2.5 years until they reach a maximum of about 40 years of age. Their maximum population growth rate is lowered to 9 percent per year. African elephants have a gestation period as long as 22 months, first give birth between 10 and 15 years of age and produce offspring with around a 4-year birth interval until 55–60 years of age. Their maximum population growth rate is reduced to ~6 percent per year. Hippos have a gestation period as short as 8 months, but still maintain birth intervals of around 2 years and typically do not reproduce until as late as 10 years of age in the wild. Hence their potential population growth rate would not much exceed that of rhinos. Intriguingly, it seems that each female can potentially produce a maximum of 12 offspring, should she live through her potential reproductive lifespan, whatever the life-history schedule.
The lower growth rates of megaherbivores mean that their populations take longer to recover from setbacks like drought-induced mortality than those of smaller species. However, on the other hand, they are less susceptible to elevated mortality during droughts because of their fibre-tolerant food requirements and greater metabolic inertia. Having mostly escaped predation by large carnivores once they reach maturity, megaherbivores have little capacity to sustain predation imposed by human hunters.
Populations ultimately stop growing once recruitment into the adult segment is barely sufficient to counterbalance mortality incurred during adulthood. The survival of juveniles and old females is most sensitive to limitations in food availability related to rainfall as well as to the population density (Figure 12.4). Various combinations of adult mortality and offspring recruitment can generate zero population growth. If, hypothetically, no adult dies before reaching the maximum longevity, at least 15 percent of the offspring born must attain reproductive maturity in order for the population to persist. If half of the adults die or are killed annually, every single offspring born would need to survive to replace them. These scenarios represent the extreme bounds for population viability. A more realistic demographic combination producing zero population growth for a large ungulate would be 50 percent survival of offspring to maturity coupled with 25 percent annual mortality in the adult segment, including those dying of old age.
Figure 12.4
Annual survival rates of juvenile, yearling, prime and old kudus governed by annual rainfall relative to population biomass density in Kruger NP. Outlying points represent 1981 when a cold spell caused mortality despite high prior rainfall.
(from Owen-Smith (2002) Adaptive Herbivore Ecology)
Observed combinations of adult mortality and juvenile recruitment generating population stasis in Kruger NP are not obviously influenced by body size, except towards the extremes (Figure 12.5). The highest rates of juvenile recruitment were exhibited by the two ungulates that form the major prey species of their main predators, i.e. wildebeest for lions and impala for less-large carnivores. For both populations, females generally gave birth as early as two years of age.49 In Serengeti, wildebeest exhibited a different combination, coupling low adult mortality with low juvenile recruitment.12 This indicates that these wildebeest are food-limited, without much additive contribution from predation. Adult mortality averaged to 9 percent annually, meaning that few died much before 15 years of age. However, juvenile mortality exceeded 70 percent during the first year. For kudus, the annual mortality rate among the adult female segment was almost 20 percent, around twice the annual mortality incurred by the prime-aged segment alone.
Figure 12.5
Combinations of adult mortality and juvenile recruitment generating zero population growth observed in Kruger NP plus one point (green) representing wildebeest in Serengeti.
Ungulates typically use changing daylength as a cue to schedule when mating and hence subsequent births occur. However, this cue does not exist near the equator because daylength remains a constant 12 h there. Consequently, although impala and hartebeest exhibit narrow birth peaks in southern Africa, they reproduce year-round in Serengeti close to the equator.50,51 On the other hand, wildebeest and topi retain narrow birth peaks near the equator despite the lack of the daylength cue. Narrowly restricted births swamp predators so that more offspring survive.52
Intriguingly, in southern Africa different ungulate species time their birth peaks in different stages of the wet season. Tsessebe (Damaliscus lunatus lunatus) and hartebeest have birth peaks during October–November early in the wet season. Impala drop their lambs from late November into December. Wildebeest calve during December. Kudu produce offspring in January–February. Sable antelope give birth as late as February or even March.51 Zebra and buffalo, with inter-birth intervals longer than a year, show rather weak pulses in births late in the wet season. Rhinos exhibit a diffuse birth peak early in the dry season, while elephants have births concentrated early in the wet season. Grazers favouring wetlands, where green grass remains available in the dry season, including waterbuck as well as their kob congeners, have births spread more widely through the year.44
For white rhinos, oestrous cycling is suppressed while food quality is poor, leading to a pulse in matings when nutrition improves after the start of the rains.42 The result is a surge in births 16 months later, early in the dry season. A similar mechanism could be operating for wildebeest and topi in Serengeti. Female ungulates generally do not resume cycling until after they have regained body condition following weaning of the preceding calf.
Are Bigger or Smaller Herbivores More Abundant?
Whether smaller or bigger animals are more abundant depends on how abundance is measured. Smaller animals tend to be more numerous, because individually each requires less food. However, larger animals need less nutrition per unit of their biomass and so can attain greater biomass densities from the same amount of food. From metabolic scaling, numerical population density (N) should decrease with body mass (M) raised to the power three-quarters: N = M–0.75, while biomass density should increase as a function of M0.25 (M1.0/M–0.75). Expressed in words, abundance assessed in terms of biomass density, i.e. weighted by the body mass, should increase with body mass raised to the power one-quarter. However, there is much variation in the maximum population density levels recorded for individual herbivore species, related to their specific food requirements and other factors (Figure 12.6). Simply putting a regression line through the scatter of points can be misleading.
Figure 12.6
Allometric relationships between population density and body mass for various African ungulate populations, distinguishing grazers from browsers. (A) Scatter plot compiled from all available records of local densities within a home range or study area, for various African ungulates; the line represents the allometric constraint generated by a power coefficient of –0.75; (B) numerical densities translated into biomass; constraint lines have slope of 0.25.
The herbivores most closely approaching the metabolic ceiling in local abundance set by food requirements are certain large grazers, notably wildebeest, buffalo and white rhino, along with mixed feeding elephant. Wetland grazers can also attain high densities locally, including kob in Uganda, and lechwe (Kobus leche) in Zambia. Maximum abundance levels attained by browsers are only about 20 percent of those shown by grazers of similar body mass. However, some grazers are no more numerous relative to their size than browsers, notably non-ruminants like zebra and warthog. Contrary to common assumptions, the numerical densities that elephants and white rhinos can attain, amounting to around 5 animals per km2, are not lower than those maintained by many much-smaller ungulates. Below a body mass of around 50 kg, numerical densities decline with decreasing body mass, apart from the outlying point representing the huge densities attained locally by little dikdiks in Kenya.53
The values plotted in Figure 12.6 represent the maximum abundance levels recorded for each species locally within a home range. If abundance levels are assessed over a broader region, like a national park or other large protected area, the relationship between population density and body mass falls away. This is because smaller herbivores are more narrowly restricted in their habitat occupation by their food or security requirements. Among the largest herbivores, regional and local densities converge.
Although the biggest herbivores attain the greatest standing biomass, the turnover of this biomass engendered by mortality is much slower than for smaller herbivores: ~5 percent versus 15–25 percent annually. Hence herbivores weighing between 100 and 500 kg actually produce most potential food for large carnivores, particularly the grazers among them. Moreover, the carcasses of big megaherbivores are utilised less effectively, leaving more flesh behind for scavengers and decomposers.
Distribution Patterns
Populations can grow and shrink additionally via expansions and contractions in their distribution ranges. Dispersal movements played a major role in re-establishing wildlife populations in South Africa’s protected areas after the decimation by hunting plus the rinderpest epizootic. Elephants and various antelope moved into Kruger NP from secluded regions of Mozambique once protection was afforded. Large herbivores and carnivores filtered back into the Umfolozi Game Reserve, where everything except rhinos had been shot in an attempt to eliminate tsetse flies, coming from the nearby Hluhluwe Game Reserve.54 White rhinos reintroduced from there to Kruger NP had spread through most of the park before the recent poaching wave hit. Changes in the geographic distributions of species are engendered by shifts in the home ranges occupied by the individual animals or groups representing these species.
The drastic decline of the sable antelope population in Kruger NP was brought about by the disappearance of herds from previously occupied localities as well as shrinkage in numbers of the remaining herds.55 Roan antelope had been even more restricted in their presence within the park, which was at the southern limit of their historical distribution range. They had occurred in two clusters of distinct herds in the far north plus three isolated herds further south, comprising 250–400 animals in total.55 When their population crashed, one complete cluster plus all of the isolated herds disappeared, leaving a local remnant of just three herds, totalling fewer than 40 animals.26 Both of these antelope species were threatened by the prospect of a downward ratchet toward local extinction.7 While herd sizes remain small, it is more difficult for mothers to defend their calves jointly against attacks by hyenas and wild dogs.55
Overview
Savanna herbivore populations typically vary over perhaps a twofold range in abundance in response to annual variation in the effective carrying capacities, governed by the rainfall dependence of plant growth and hence food resources. Periodic droughts amplify food depletion in the dry season and can generate substantial population crashes, especially if movements are restricted. Direct effects of predation on prey populations seem quite minor, because of compensatory mechanisms. Indirect responses to the risk of predation can be huge, in contrast, by restricting habitat occupation and hence regional population densities. Impacts of resources and predation on herbivore populations are inextricably entangled, because when food runs out animals take greater risks. Megaherbivores, migratory antelope and wetland dwellers most closely approach the population ceilings set by metabolic requirements for food. Migration elevates populations largely by restricting exposure to resident predator populations. Although population biomass densities rise with body size, turnover of this biomass diminishes correspondingly. Accordingly, herbivores in the size range 100–500 kg are most productive in biomass generated annually. Browsers attain local density levels amounting to only 20 percent of those exhibited by grazers of similar size.
South America currently lacks concentrations of large herbivores. Australia has enormous numbers of kangaroos and wallabies, but populations fluctuate hugely due to die-offs during periodic droughts. Predation by dingos seems inconsequential.56 Spectacular migratory concentrations of saiga (Saiga tartarica) and Mongolian gazelles (Procapra gutturosa) develop in the steppe grasslands of central Asia and among caribou/reindeer (Rangifer tarandus) herds in the arctic north. Bison (Bison bison) formerly achieved similarly enormous abundance in North American grasslands. Intrinsically generated population irruptions like those manifested by deer in North America have not been observed in Africa. Seasonal concentrations of grazers near water are a feature especially of African savannas.
SUGGESTED FURTHER READING
Owen-Smith, N. (1988) Megaherbivores. The Influence of Very Large Body Size on Ecology. Cambridge University Press, Cambridge.
Owen-Smith, N. (2010) Dynamics of Large Herbivore Populations in Changing Environments. Wiley-Blackwell, Oxford.
REFERENCES
1.Mills, MGL, et al. (1995) The relationship bewteen rainfall, lion predation and population trends in African herbivores. Wildlife Research 22:75–87.
2.Marshal, JP, et al. (2011) The role of El Niño–Southern Oscillation in the dynamics of a savanna large herbivore population. Oikos 120:1175–1182.
3.Dublin, HT; Ogutu, JO. (2015) Population regulation of African buffalo in the Mara–Serengeti ecosystem. Wildlife Research 42:382–393.
4.Ottichilo, WK, et al. (2000) Population trends of large non‐migratory wild herbivores and livestock in the Masai Mara ecosystem, Kenya, between 1977 and 1997. African Journal of Ecology 38:202–216.
5.Estes, RD, et al. (2006) Downward trends in Ngorongoro Crater ungulate populations 1986–2005: conservation concerns and the need for ecological research. Biological Conservation 131:106–120.
6.Staver, AC, et al. (2019) Grazer movements exacerbate grass declines during drought in an African savanna. Journal of Ecology 107:1482–1491.
7.Ogutu, JO; Owen‐Smith, N. (2003) ENSO, rainfall and temperature influences on extreme population declines among African savanna ungulates. Ecology Letters 6:412–419.
8.Owen-Smith, N; Mills, MGL. (2006) Manifold interactive influences on the population dynamics of a multispecies ungulate assemblage. Ecological Monographs 76:73–92.
9.Grange, S, et al. (2004) What limits the Serengeti zebra population? Oecologia 140:523–532.
10.Georgiadis, N, et al. (2003) The influence of rainfall on zebra population dynamics: implications for management. Journal of Applied Ecology 40:125–136.
11.Owen-Smith, N. (1990) Demography of a large herbivore, the greater kudu Tragelaphus strepsiceros, in relation to rainfall. The Journal of Animal Ecology 59:893–913.
12.Mduma, SAR, et al. (1999) Food regulates the Serengeti wildebeest: a 40‐year record. Journal of Animal Ecology 68:1101–1122.
13.Owen-Smith, N (1981) The white rhinoceros overpopulation problem, and a proposed solution. In Jewell, PA et al. (eds) Problems in Management of Locally Abundant Wild Mammals. Academic Press, New York, NY, pp. 129–150.
14.Moss, CJ. (2001) The demography of an African elephant (Loxodonta africana) population in Amboseli, Kenya. Journal of Zoology 255:145–156.
15.Corfield, TF. (1973) Elephant mortality in Tsavo National Park, Kenya. African Journal of Ecology 11:339–368.
16.Smit, IPJ; Bond, WJ. (2020) Observations on the natural history of a savanna drought. African Journal of Range & Forage Science 37:119–136.
17.Young, TP. (1994) Natural die‐offs of large mammals: implications for conservation. Conservation Biology 8:410–418.
18.Spinage, CA; Matlhare, JM. (1992) Is the Kalahari cornucopia fact or fiction? A predictive model. Journal of Applied Ecology 29:605–610.
19.Walker, BH, et al. (1987) To cull or not to cull: lessons from a southern African drought. Journal of Applied Ecology 24:381–401.
20.Owen-Smith, N. (2019) Ramifying effects of the risk of predation on African multi-predator, multi-prey large-mammal assemblages and the conservation implications. Biological Conservation 232:51–58.
21.Owen-Smith, N. (2015) How diverse large herbivores coexist with multiple large carnivores in African savanna ecosystems: demographic, temporal and spatial influences on prey vulnerability. Oikos 124:1417–1426.
22.Yoganand, K; Owen‐Smith, N. (2014) Restricted habitat use by an African savanna herbivore through the seasonal cycle: key resources concept expanded. Ecography 37:969–982.
23.Owen‐Smith, N; Traill, LW. (2017) Space use patterns of a large mammalian herbivore distinguished by activity state: fear versus food? Journal of Zoology 303:281–290.
24.Scheel, D; Packer, C (1995) Variation in predation by lions: tracking a movable feast. In Sinclair, ARE et al. (eds) Serengeti II: Dynamics, Management, and Conservation of an Ecosystem. University of Chicago Press, Chicago, IL, p. 299.
25.Sinclair, ARE, et al. (2003) Patterns of predation in a diverse predator–prey system. Nature 425:288–290.
26.Harrington, R, et al. (1999) Establishing the causes of the roan antelope decline in the Kruger National Park, South Africa. Biological Conservation 90:69–78.
27.Owen-Smith, N; Mills, MGL. (2008) Shifting prey selection generates contrasting herbivore dynamics within a large‐mammal predator–prey web. Ecology 89:1120–1133.
28.Fryxell, JM, et al. (1988) Why are migratory ungulates so abundant? The American Naturalist 131:781–798.
29.Fryxell, JM; Sinclair, ARE. (1988) Seasonal migration by white‐eared kob in relation to resources. African Journal of Ecology 26:17–31.
30.Cooper, SM, et al. (1999) A seasonal feast: long‐term analysis of feeding behaviour in the spotted hyaena (Crocuta crocuta). African Journal of Ecology 37:149–160.
31.Harris, G, et al. (2009) Global decline in aggregated migrations of large terrestrial mammals. Endangered Species Research 7:55–76.
32.Naidoo, R, et al. (2016) A newly discovered wildlife migration in Namibia and Botswana is the longest in Africa. Oryx 50:138–146.
33.Morjan, MD, et al. (2018) Armed conflict and development in South Sudan threatens some of Africa’s longest and largest ungulate migrations. Biodiversity and Conservation 27:365–380.
34.Naidoo, R, et al. (2012) Home on the range: factors explaining partial migration of African buffalo in a tropical environment. PLoS One 7:e36527.
35.Hillman, J. (1988) Home range and movement of the common eland (Taurotragus oryx Pallas 1766) in Kenya. African Journal of Ecology 26:135–148.
36.Fryxell, JM, et al. (2005) Landscape scale, heterogeneity, and the viability of Serengeti grazers. Ecology Letters 8:328–335.
37.Cagnacci, F, et al. (2016) How many routes lead to migration? Comparison of methods to assess and characterize migratory movements. Journal of Animal Ecology 85:54–68.
38.Kreulen, D. (1975) Wildebeest habitat selection on the Serengeti plains, Tanzania, in relation to calcium and lactation: a preliminary report. African Journal of Ecology 13:297–304.
39.Murray, MG. (1995) Specific nutrient requirements and migration of wildbeest. In Sinclair, ARE, et al. (eds) Serengeti II: Dynamics, Management, and Conservation of an Ecosystem. University of Chicago Press, Chicago, IL, p. 231.
40.Ogutu, J, et al. (2013) Changing wildlife populations in Nairobi National Park and adjoining Athi-Kaputiei Plains: collapse of the migratory wildebeest. The Open Conservation Biology Journal 7:11–26.
41.Ogutu, JO, et al. (2012) Dynamics of ungulates in relation to climatic and land use changes in an insularized African savanna ecosystem. Biodiversity and Conservation 21:1033–1053.
42.Owen-Smith, RN. (1988) Megaherbivores: The Influence of Very Large Body Size on Ecology. Cambridge University Press, Cambridge.
43.Gaillard, J-M, et al. (2015) Does tooth wear influence ageing? A comparative study across large herbivores. Experimental Gerontology 71:48–55.
44.Owen‐Smith, N; Ogutu, JO. (2013) Controls over reproductive phenology among ungulates: allometry and tropical–temperate contrasts. Ecography 36:256–263.
45.Bigalke, RG. (1970) Observations on springbok populations. African Zoology 5:59–70.
46.Mills, MGL. (2017) Kalahari Cheetahs: Adaptations to An Arid Region. Oxford University Press, Oxford.
47.Ryan, SJ, et al. (2007) Ecological cues, gestation length, and birth timing in African buffalo (Syncerus caffer). Behavioral Ecology 18:635–644.
48.Smuts, GL. (1976) Population characteristics of Burchell’s zebra (Equus burchelli antiquorum. H. Smith, 1841) in the Kruger National Park. South African Journal of Wildlife Research 6:99–112.
49.Whyte, IJ; Joubert, CSJ. (1988) Blue wildebeest population trends in the Kruger National Park and the effects of fencing. South African Journal of Wildlife Research 18:78–87.
50.Sinclair, ARE, et al. (2000) What determines phenology and synchrony of ungulate breeding in Serengeti? Ecology 81:2100–2111.
51.Ogutu, JO, et al. (2015) How rainfall variation influences reproductive patterns of African savanna ungulates in an equatorial region where photoperiod variation is absent. PloS One 10:e0133744.
52.Estes, RD. (1976) The significance of breeding synchrony in the wildebeest. African Journal of Ecology 14:135–152.
53.Augustine, DJ; McNaughton, SJ. (2004) Regulation of shrub dynamics by native browsing ungulates on East African rangeland. Journal of Applied Ecology 41:45–58.
54.Owen-Smith, RN. (1983) Dispersal and the dynamics of large herbivores in enclosed areas: implications for management. In Owen-Smith, N (ed.) Management of Large Mammals in African Conservation Areas. Haum Educational Publishers, Pretoria, pp. 127–143.
55.Owen‐Smith, N, et al. (2012) Shrinking sable antelope numbers in Kruger National Park: what is suppressing population recovery? Animal Conservation 15:195–204.
56.Caughley, G, et al. (1980) Does dingo predation control the densities of kangaroos and emus? Wildlife Research 7:1–12.