Lions

Lions (Panthera leo) are social animals that associate to form prides consisting of 2-18 individuals (Funston and Mills 1997; Whitman and Packer 1997). The pride, which is made up of several females together with their associated males, subadults and cubs, may split into several subgroups or semi-permanent subprides scattered throughout the pride area/range. Pride ranges may vary between 20 kmē and 500 kmē in size (Funston and Mills 1997). The area is usually fixed and is determined by the availability of food, water and other prides (Orford 1986; Stander 1991). The area or territory held by a pride may have boundaries overlapping those of neighbouring prides. Within these territories there are smaller, defendable areas known as "home ranges" (Whyte 1985).

Pride size varies from one area/habitat type to another (Van Orsdol et al. 1985; Stander 1997). Van Orsdol et al (1985) compared lion densities in ten different habitat types, focussing on the relationship between range/territory size and prey. Their results showed that range size was inversely correlated with prey abundance during the period of least abundance and that pride size and cub survival was strongly correlated with the lean season food abundance. They also found that the physical features of the habitat influence the level of male-male competition and thus, indirectly the sex ratio within prides. However, this habitat comparison was across several reserves, which leaves room for subjectivity when defining habitat structure. Variables such as climatic differences also confound direct comparison.

Pride size and lion density also varies with the abundance of prey, declining when prey disperses in wet seasons or in periods of drought, largely due to increased cub mortality (Grobler et al. 1984 cit. loc. Orford 1986; Stander 1997). Most lion cub litters are born during the period when the major prey species of the lions produce their young (Smuts et al. 1978). Nutrition has the greatest effect on cub survival after birth therefore it is important that there is a high density of prey available (Van Orsdol et al. 1985). Prey distribution and "catchability" is therefore of importance to the maternal females (Sunquist and Sunquist 1989).

Population demography

A population is a group of individuals of the same species inhabiting a given area. It is a biological unit for which meaningful birth and death rates, sex ratios and an age structure can be determined (Caughley 1977). Population analysis is concerned with the numerical attributes of the population together with the properties of the animals and the properties of the environment that determine these values (Caughley 1977). The trends/fluctuations in wildlife populations are determined by rates of natality and immigration versus mortality and emigration and reflect the interplay of numerous, often variable, environmental factors (Caughley 1977; Mason 1990). The rates of increase and decrease are determined by the population's interaction with the quality of the environment, where quality refers to the weather, suitability of food, suitability of home ranges etc (Caughley 1977).

The distribution of a population through a given area is influenced by the availability of suitable habitats (Pienaar 1974; Stander 1991). Although various habitats may be inhabited by a given species they are never of the same quality. Good quality habitats yield a demographic excess (natality>mortality), while lower quality habitats yield a demographic deficit (mortality>natality) (Dias 1996). However, habitats seldom remain constant for long. Fluctuations occur on a broader scale as a result of climatic fluctuations, year to year variability in weather, plant succession and the activities of man and other animals (Pienaar 1974; Caughley 1977). The distribution of animals changes as a result of the above. This may be caused by the resultant increase in habitat patch size, decrease in patch size, or by the creation or destruction of habitats (Caughley 1977). Animals cope with these changes in a number of different ways. They may disperse, change genetically or phenotypically, or they may be ecologically adaptable to cope with a broad range of habitat types. Dispersal depends on a number of factors and may be a spontaneous (random) movement or driven by a response to unfavourable conditions.

I shall be focussing on three ecological processes that influence lion demography in the Kruger National Park, South Africa, namely habitat structure, habitat type/food availability and rainfall.

Habitat structure refers to the physical appearance of the region, e.g. woodland or grassland. The type of habitat chosen by a species is related to the properties of that area. A suitable habitat should supply adequate shelter to avoid exposure to the elements (Pienaar 1974).

The habitat should also provide protection to young. This is important, especially for lions, where new males taking over a pride will kill young cubs to induce oestrus in the females (Pusey and Packer 1994). In the case of lions, a good habitat provides females with suitable, hidden areas where their young cubs will be safe from new males that may take over the pride and from leopards and hyenas that prey on the cubs.

A good habitat should not only fulfil the above requirements, but should also provide freedom from excessive competition from competitive associated species (Pienaar 1974). This applies to predators as well as prey species. Predators may have overlaps in their diets, competing for the same prey species (Mills and Biggs 1993). Competition for prey was found to be affected not only by diet but also by the number of prey available and the relative number of predators. Mills and Biggs (1993) also found that the five major predators in the Kruger National Park selected different habitats based on habitat structure.

Habitat structure places a constraint on hunting and hunting success. The height and density of the vegetation influence the ease with which lions can stalk their prey without being detected. Short grass provides few hiding places for stalking lions while tall grass and woodlands/thickets supply ample cover (Schaller 1972). Therefore lions should be choosing the areas in which hunting success is high. I shall therefore be determining lion distribution and group composition in four habitat types namely, the Marula plains on basalt in the east, the Combretum woodlands in the north-west, the Acacia thickets combined with the sour bushveld in the south-west and mountainous areas, which will combine the Lebombo mountains and the Malelane mountain bushveld.

The vegetation of a habitat is important to herbivorous prey and therefore, indirectly, to predators. The vegetation is not only a source of cover and protection to herbivores, but also a food supply (Pienaar 1974). Natural populations of large herbivorous savanna mammals tend to be close to the limits set by their food resources (East 1984). Predator populations are limited by prey populations and therefore by the quality of habitat available to their preferred prey.

Prey availability influences predator distribution. Instead of dividing habitat on the basis of vegetation species present in the area, they can be further divided according to prey species present. Lions prey on a number of species including buffalo, wildebeest, zebra, impala, kudu, giraffe and waterbuck (Mills et al. 1995). The presence or absence of their preferred species in a given habitat will therefore influence lion distribution and density.

However, it is not only the presence or absence of prey that is important, but also their abundance. This will be limited by the availability of cover, food and water; e.g. buffalo movement was found by Hunter (1996) to be limited by water availability. They therefore remained within a certain radius (8.7 km) of water pointss (Hunter 1996).

Rainfall is an important factor affecting population trends. A number of studies have shown relationships between herbivore populations and rainfall (e.g. East 1984; Sinclair et al. 1985; Mills et al. 1995). East (1984) found that the total biomass of large savanna mammals is positively related to the mean annual rainfall in 20 wildlife areas of southern and eastern Africa, in which rainfall varied from less than 200 to more than 1100 mm.p.a. The density of wildebeest in the Serengeti was found to increase with the rainfall increase in the dry season (Sinclair et al. 1985).

Mills et al. (1995) showed that herbivore populations react differently to the rainfall cycles experienced in the Kruger National Park and as a result differ are in their vulnerability to predation. They found that wildebeest and zebra are more vulnerable to predation by lions during the wet cycle, while buffalo and waterbuck are more vulnerable in the dry cycle. Mills et al. (1995) discuss the relationship between rainfall, lion predation and their effect on herbivore population trends.

Rainfall also affects habitat structure. Mitchell, Shenton and Uys (1965) recorded a total of 19 prey species taken by lions of which buffalo were the most important. However, there was a strong seasonal trend with more being killed in the dry season than during the rainy season. They explained this by the structure of the habitat. When the grass is long and cover dense, the lions have enough cover to successfully stalk and capture their prey. However, during the dry season there are bush fires that cause a flush of green grass bringing grazing herbivores out on to the plains. Hunting is more difficult as the lions' cover has been destroyed therefore they turn to the buffalo, which remain in the thickets and woodland.

The Kruger National Park (KNP) has been divided into 35 habitat/landscape types by Gertenbach (1983), who defines a landscape as an area with a specific geomorphology, climate, soil and vegetation pattern together with associated fauna. A current map of KNP, however, divides the Park into sixteen habitat types based on the above definition. Although these zones differ in plant species, the physiognomy of some is similar. I will, therefore divide KNP into four habitat types, namely the Marula plains on basalt in the east, the Combretum woodlands in the north-west, the Acacia thickets combined with the sour bushveld in the south-west and mountainous areas, which will combine the Lebombo mountains and the Malelane mountain bushveld.

The Kruger National Park is a summer rainfall region where wet and dry cycles follow each other at approximately 10-year intervals (Gertenbach 1980; Mason 1990). Periods of above and below the long-term average rainfall occur at regular intervals within the wet and dry cycles (Gertenbach 1980). The precipitation in KNP decreases from south to north, except for the area around Punda Maria, which is situated at a higher altitude. There is a minor decrease in rainfall from west to east that corresponds to the decrease in altitude from west to east. This becomes more pronounced towards the escarpment on the western boundary (Gertenbach 1980). Game populations are influenced by the rainfall cycles. All populations, except wildebeest and zebra, tend to decrease during dry cycles (G. Mills, pers. comm.). While population decreases, or in the case of buffalo and waterbuck population increases, have been recorded in the wet cycles (Gertenbach 1980; Mills et al. 1995). Initially lions were blamed for the decrease but it soon became evident that it was the effect of the high rainfall and long grass that played an important role in suppressing these animal populations.

Although studies have been done relating lion biology to prey biomass and rainfall (Van Orsdol et al. 1985; Mills et al. 1995), none have analysed lion demography in terms of the above. Neither have comparisons been made between lion group characteristics within one reserve containing numerous habitat types. My study therefore undertakes to compare lion demography within one large reserve, Kruger National Park, South Africa, that consists of several different habitat types. The KNP provides an excellent opportunity to study a lion population in an area where there is little climatic variation but where there is variation in habitat structure, prey distribution/ abundance and rainfall.

Research Protocol

General Methods

There are approximately 10 000 lion sightings recorded on monthly data sheets (1965-1985) from 22 stations throughout the Kruger National Park. These sightings are distributed throughout all habitats. Volunteer undergraduate students and technicon students under my supervision will enter the ranger lion sighting reports into a database. In addition, I will read through monthly ranger reports to access additional data. The diaries and periods chosen will be decided upon in consultation with my supervisors with well-documented lion sightings given a preference. A preliminary assessment of the data, including the entry of 15 years of data from six stations has indicated these data to be robust.

Lion sightings have not been recorded for the previous 10 years therefore it would be beneficial to obtain more knowledge about lion demography, local movements and on a smaller scale, individual lion pride information. This will be achieved by getting the night drive students at KNP to complete lion sighting forms similar to the pink forms of previous years, but also detailing the exact location of the sighting. The students will mark the lion locations on a map detailing their route for the evening. They will also write down their mileage so that an exact location can be obtained. The number of adult males, females and unknowns, subadult males, females and unknowns and number of cubs will be recorded. The lions' behaviour will be noted as well as the identification of known/identified individuals. If there has been a kill and the prey species, if identifiable, will be recorded. These data will provide a year of current data to add to the historical data. A pilot study of this exercise will be run at Skukuza after which, depending on the success, it will be introduced to more camps.

 

GIS resources

Maps of the Kruger Park detailing landtypes, waterhole and road positions and landscapes will be obtained from KNP. A database will be created listing the area and grid reference of the lion sightings, the number and sex of adult lions seen, the number and sex of subadults seen as well as the number and sex of cubs seen. The lion database that is created from the sightings will then be used together with the maps to determine broad-scale ecological patterns, such as lion group composition or the density of females in the different habitat types.

A grid size of 5 kmē will be used as this complies with the KNP lion sighting data. It is also a convenient working grid size as lion density, prey abundance and habitat structure may vary on a small scale in certain areas. Certain areas may not have been accessible to the rangers and therefore if a larger grid cell size was used, e.g. 25 kmē, extrapolations from the data may be made which are not necessarily true. Using as small a grid cell size as possible/feasible increases the accuracy of/confidence in the results obtained from analysis.

A number of lion variables will be extracted from the sightings data:

1. Total group size: defined as the number of adults, subadults and cubs.
2. Functional group size: defined as the number of adult females (they are generally the hunters of the group).
3. Presence/absence of cubs.
4. Presence/absence of males.
5. Male group size.
6. Sex ratio of adults.

Chapter 1

Aim

To determine the effect of habitat structure on lion spatial ecology.

Methods

The KNP has been divided into 35 landscapes according to geomorphology, climate, soil, vegetation pattern and the associated fauna (Gertenbach 1983). I shall divide the KNP into four habitat types, i.e. the Marula plains on basalt in the east, the Combretum woodlands in the north-west, the Acacia thickets combined with the sour bushveld in the south-west and mountainous areas, which will combine the Lebombo mountains and the Malelane mountain bushveld. Differentiation is not being made on a plant species level as this should be accounted for by the differences in prey densities in areas falling into the same category. Each grid cell (5 kmē) will be allocated a number (e.g. 1/2/3) according to the dominant habitat type. The lion variables will be averaged (number of lions/number of sightings) for each grid cell over three month periods to reduce the effects of pseudoreplication (Hurlbert 1984). The maps of the lion variables will be overlaid on the habitat map to extract group sizes and group compositions (in terms of sex ratios and age structure) in the three habitat types using GIS. A spatial database detailing lion group sizes and compositions in the different habitat types will be produced.

Analysis

Lion group size in the three habitat types will be compared using ANOVA. ANOVA will also be used to compare lion group compositions in the three habitat types. If there is great variation between the results of individual habitat types, a more detailed analysis involving plant species may be necessary. However, such variation may be accounted for by prey density differences (see chapter 2).

Chapter 2

Aim

To determine the effect of prey availability or presence of major lion prey species on lion group dynamics.

Methods

Census data on the major prey species (e.g. buffalo, wildebeest, zebra, waterbuck, warthog, impala and giraffe) of lions will be obtained from KNP. As the herbivore censuses take place once a year during the dry season, maps of the lion variables averaged for each grid cell (5 kmē) during the dry season will be made.

Three analyses will be carried out. Firstly, a map of presence/absence will be created for each prey species. Each map of averaged lion variables will be overlaid on these presence/absence maps. The second analysis will involve overlaying maps of prey numbers and lion numbers and group compositions. The ratio of lions (in terms of total and functional groups) to prey will be extracted to a spatial database. Finally, the presence/absence of cubs will also be extracted to a spatial database detailing the abundance of prey.

Analysis

ANOVA will be used to compare lion group sizes in areas where each prey species is either present or absent. Dry season comparisons of the lion variables can also be done over the years using ANOVA. A correlation analysis will be used to determine the relationship between lion density and prey density. ANOVA will be used to compare the presence/absence of cubs as related to prey abundance.

Chapter 3

Aim

To determine the effect of rainfall on lion group dynamics through its influence on prey species' movement patterns and distribution.

Methods

Rainfall data from the 22 stations in KNP is recorded on a monthly basis. The data will be used to create a map of varying (annual) rainfall regions. As rainfall is largely confined to the summer months (September to April) (Gertenbach 1980), the annual rainfall will be calculated from 1 July to 30 June.

Firstly, the annual averaged lion variables map will be overlaid on this rainfall region map to compare differences between group sizes in the various regions. A second analysis will involve averaging the lion data over ten year periods corresponding with the wet and dry cycles occurring in the Park. Comparisons of group sizes and group compositions between the two cycles will then be made. Analyses comparing group sizes in periods of especially high and low rainfall within the cycles themselves can also be made.

Analysis

The coefficient of variation will be determined for rainfall within each cycle. ANOVA will be used to analyse the differences between group sizes in the different rainfall regions and to analyse the differences between group size and group composition in wet and dry cycles. The differences between group sizes in the high and low rainfall periods will be analysed using ANOVA. The variation in rainfall over the thirty-year study period will also be determined. A categorical analysis, the G-test, will also be carried out to determine the variability in cub presence/absence when rainfall is either high or low.

Output

1. The results of the study will be presented as a Masters thesis in the Department of Biology, UND, at the end of 1999.
2. A lion database for use by KNP wildlife managers.
3. The results will be written for publication in a peer review journal.
4. Presentation at conferences.

Short term goals:

Activity	Date
Initiate Skukuza night drive pilot study	February 1998
Lion sightings from ranger diaries captured	July 1998
"Pink forms" captured onto computer	December 1998
Current lion sightings captured	Feb? 1999
GIS lion layer complete	1999
Thesis	1999

Budget:

Item	Rate	Cost
Travel
Durban to Kruger Park, return *2	1400 km @ R1.5/km	R4200
Total		R4200
Accommodation
Kruger Park	120days @ R25pppd	R3000
Food	120 days @ R25pd	R3000
Total		R6000
Consumables
Photocopies	1000 pages @ R0.20	R200
Computer Disks	20 disks @ R3.50	R700
Stationery	assorted	R150
Total		R1050
Miscellaneous
Assistant in Durban for 6 months	R11/hr * 20 hr * 6 mo	R1320
Total		R1320
Software
Idrisi		R1237.50
Cartalinx		R975
Total		R2212.50
Hardware
GIS computer		R10000
Zip drive		R941
Total		R10941
Grand Total		R25723.50

 

Logistics

This project will be conducted under the supervision of Dr Rob Slotow (principle supervisor; Department of Biology, UND) and Dr Gus Mills (co-supervisor; Predator Ecologist, KNP). Transport (Toyota Hilux LDV) and a notebook computer (Compaq, asset number xxxx) will be provided by the Department of Biology, UND, for the fieldwork. I shall also be requiring accommodation in the KNP research camp.

References

Dias, P.C. 1996 Sources and sinks in population biology. TREE, 11: 326-330.

East, R. 1984 Rainfall, soil nutrient status and biomass of large African savanna mammals. African Journal of Ecology, 22: 245-270.

Funston, P.J. and Mills, M.G.L. 1997 Aspects of sociality in Kruger National Park lions: the role of males. Proceedings of a Symposium on Lions and Leopards as Game Ranch Animals, Onderstepoort, 18-26.

Gertenbach, W.P.D. 1980 Rainfall patterns in the Kruger National Park. Koedoe, 23: 35-43.

Gertenbach, W.P.D. 1983 Landscape patterns of the Kruger National Park. Koedoe, 26: 9-121.

Hunter, C.G. 1996 Land uses on the Botswana/Zimbabwe border and their effects on buffalo. South African Journal of Wildlife Resources, 26: 136-150.

Hurlbert, S.H. 1984 Pseudoreplication and the design of ecological field experiments. Ecol. Monogr., 54: 187-211.

Mason, D.R. 1990 Monitoring of sex and age ratios in ungulate populations of the Kruger National Park by ground survey. Koedoe, 33: 19-28.

Mills, M.G.L. and Biggs, H.C. 1993 Prey apportionment and related ecological relationships between large carnivores in Kruger National Park. Symp. Zool. Soc. Lond. 65: 253-268.

Mills, M.G.L., Biggs, H.C. and Whyte, I.J. 1995 The relationship between rainfall, lion predation and population trends in African herbivores. Wildlife Research, 22: 75-88.

Mitchell, B.L., Shenton, J.B. and Uys, J.C.M. 1965 Predation on large mammals in the Kafue National Park, Zambia. Zoologica Africana, 1: 297-318.

Orford, H.J.L. 1986 Reproductive physiology and hormonal contraception in free-ranging lions (Panthera leo L.) at the Etosha National Park. Masters Thesis, University of Natal, Pietermaritzburg.

Pienaar, U. de V. 1969 Predator-prey relationships amongst the larger mammals of Kruger National Park. Koedoe, 12: 108-176.

Pienaar, U. de V. 1974 Habitat preference in South African antelope species and its significance in natural and artificial distribution patterns. Koedoe, 17: 185-195.

Pusey, A.E. and Packer, C. 1994 Infanticide in lions: consequences and counterstrategies. In: Infanticide and parental care. (Ed S. Parmigiani and F. Van Saal). Harwood Academic publishers.

Schaller, G.B. 1972 The Serengeti Lion: a study of predator-prey relations. University of Chicago Press, Chicago.

Sinclair, A.R.E., Dublin, H. and Borner, M. 1985 Population regulation of Serengeti Wildebeest: a test of the food hypothesis. Oecologia, 65: 266-268.

Smuts, G.L., Hanks, J. and Whyte, I.J. 1978 Reproduction and social organisation of lions from Kruger National Park. Carnivore 1, 17-28.

Stander, P.E. 1991 Demography of lions in the Etosha National Park, Namibia. Madoqua, 18: 1-9.

Stander, P.E. 1997 The ecology of lions and conflict with people in north-eastern Namibia. Proceedings of a Symposium on Lions and Leopards as Game Ranch Animals, Onderstepoort, 10-17.

Sunquist, M.E. and Sunquist, F.C. 1989 Ecological constraints on predation by large felids. In: Carnivore behavior, ecology, and evolution. (Ed. Gittleman, G.L.) Chapman and Hall, London. 283-301.

Van Orsdol, K.G. Hanby, J.P. and Bygott, J.D. 1985 Ecological correlates of lion social organisation (Panthera leo). Journal of Zoology, London, 206: 97-112.

Whitman, K. and Packer, C. 1997 The effect of sport hunting on the social organisation of the African lion (Panthera leo). Proceedings of a Symposium on Lions and Leopards as Game Ranch Animals, Onderstepoort, 177-183.

Whyte, I.J. 1985 The present ecological status of the blue wildebeest in the central district of the Kruger National Park. Masters Thesis, University of Natal, Pietermaritzburg.