The elephant debate deals largely with population size, how elephant numbers change over time, how they may affect vegetation, and how their populations should be managed. Trends in elephant numbers frequently motivate management decisions, and past efforts to alleviate elephant impact aimed at controlling population size. However, methodological and statistical constraints may influence interpretation of trends and lead to incorrect management decisions. Furthermore, inferences about the response of elephant populations to specific management actions are seldom based on scientific evidence. In this thesis I assess the consequences of survey design and monitoring features on the interpretation and statistical reliability of population trends as well as the effect of population management on elephant densities and population growth rates. To do this, I collated information on elephant population estimates and past management actions across Africa. I used information from the northern Botswana elephant population to clarify temporal trends in elephant densities and numbers. Elephant numbers in northern Botswana increased from 1973 to 1993 while densities remained relatively stable. This difference in trends is due to an associated increase in survey area during the same time. In contrast, from 1996 to 2004 surveyed areas remained constant in size and neither elephant numbers, nor densities changed significantly during this time. This apparent stabilisation in numbers may have resulted from density-related elephant dispersal. This case study suggests that in open populations movements may complicate the interpretation of trends, and that differences in the rates of change in numbers and densities may have different management implications. The precision of population estimates, sample size, population size, and the magnitude of the annual rate of population change to be detected, affect power to identify trends. Two-thirds of the 156 time series that I assembled apparently were stable, and only 30 % of these had sufficient statistical power to detect population changes. These apparent stable trends without sufficient statistical power are inconclusive and should not be used to inform management decisions. Past elephant population management practices may have increased densities and growth rates in African elephant populations. Case studies of populations that were exposed to different management actions indicated that fencing of populations and water supplementation may have enhanced growth rates probably by influencing dispersal patterns. Thus, past management practices may have contributed to the ‘elephant problem’ by enhancing local elephant densities and population growth rates. In this thesis, I showed that trends based on elephant numbers may be misleading when the area over which elephants were counted, increased in size. Second, despite much effort and resources devoted to the monitoring of elephant populations for more than 50 years, population estimates and time series including such estimates had low quality, thereby reducing statistical power to detect trends in population change. Third, population growth rates were associated with management, where elephant population densities grew at faster rates when managed. Future conservation efforts should take into account the methodological and statistical constraints that may influence trend analyses of elephant populations and take cognizance of the fact that management decisions need to be evaluated against expected outcomes. Copyright / Dissertation (MSc)--University of Pretoria, 2009. / Zoology and Entomology / unrestricted

No description available.

The ecology of urban birds—populations, foraging ecology, and nest-sites—was studied between 1968 and 1970, mostly in Vancouver, British Columbia. Four census plots in Vancouver, each in a different habitat type, were studied year-round. Additional winter studies were made on two plots each in Sacramento, California and Ottawa, Ontario and on two additional Vancouver plots. Breeding bird densities on the Vancouver plots were lower than those in some other urban areas, but comparable to those in most non-urban habitats. Densities decreased with increasing urbanization. In winter, densities were generally higher than in the breeding season, and were much higher than those in non-urban habitats. Winter densities, unlike breeding densities, increased with greater urbanization. The availability of food is probably a major cause of the high winter densities. Sacramento and Ottawa both had lower winter densities than Vancouver. The species diversity of the urban plots was low at all seasons. This results both from a small number of species and a low equitability (i.e., dominance of the population by a very few species). On the Vancouver plots, species diversity was highest in the breeding season, owing to a higher equitability then. The latter probably results from territorial behaviour, which makes it less likely that one or two species will dominate the community. The urban bird populations were dominated—especially on the most urbanized plots—by House Sparrows, Starlings, and Rock Doves. An attempt is made to explain how the biological features of these three species make them especially well suited to the urban environment. Two types of observations—stopwatch observations and spotchecks —were used, in studies of foraging ecology. This phase of the study was confined to the Vancouver plots. An analysis of the stopwatch observations revealed that nearly every species had a distinctive foraging pattern. The only pair of species which had closely similar patterns were the Starling and the Robin. However, these two had quite different foraging methods and utilized different foods. The spotchecks, being discrete, were amenable to statistical testing. Tests were carried out to see whether the use of different microhabitats corresponded to their availability, and to check for differences between species. In both cases, only tests involving the Crested Mynah failed to show significant differences. This lack of significance is believed to be merely the result of a small sample size. Nest-sites were also studied. Interspecific differences in nest height and placement were demonstrated. A conspicuous feature was the almost total absence of nests near the ground. Cat predation and human disturbance are probably responsible for this. In closing, some general features of urban bird ecology are discussed. The importance of studying the foraging ecology of unrelated species, as well as of related ones, is stressed. While two related species with similar foraging ecology usually occupy different habitats, the same, is probably true of unrelated species to a lesser degree. The foraging patterns of urban birds may be expected to overlap more than those of non-urban birds, and nest-site availability may be particularly crucial to birds in cities. / Science, Faculty of / Zoology, Department of / Graduate

The population dynamics of caribou (Rangifer tarandus) on the Island of Newfoundland were studied 1957 to 1967. Four herds were recognized and censused: the Northern Peninsula — 450 animals in 1958 and 400 in 1966, the Avalon Peninsula — 125 in 1957 and 720 in 1967, the Humber River — 130 in 1956 and 115 in 1964, and the Interior -- 4600 in 1957 and 6200 animals in 1966. The Island population 1900-1910 was estimated at 40,000 animals. After 1915 these herds rapidly declined and reached a low of perhaps only 2000 animals by 1930. The annual rate of increase (r) of all the herds was low. The Humber and Northern Peninsula herds showed no increase while the Interior Herd grew at only 0.044. The Avalon Peninsula Herd showed the greatest rate-of-increase 1961 to 1967, r= 0.120. The potential r of Newfoundland caribou is probably greater than 0.30. A herd of caribou introduced to Brunette Island increased at r=0.352, from 17 to 100 animals in 5 years. In the Interior Herd, birth rates were high and constant and averaged 0.85 calf per doe 2-years-of-age and older and 0.94 calf per doe 3 years and older. Natural mortality rates were low beyond 6-months-of-age. They were 4 per cent for does, yearlings, and calves and 9 per cent for stags 2-years-of-age and older. The kill of stags by hunters was 11 per cent and reduced the proportion of stags in the population. The survival of calves to 6-months-of-age was strongly correlated with growth of populations and appeared the main influence on numbers (correlation coefficient r=0.922, P<0.01). The mortality of calves in the first summer was high; an average of 69 per cent died in the Interior Herd and in the Avalon Peninsula Herd 30 per cent of the calves died. The major cause of mortality of calves was apparently predation by lynx (Lynx canadensis). As early as 2 weeks after calving, 27 per cent of the calves were missing. They were apparently dragged into forest cover by lynx. Of 114 dead or morbid calves located 74 per cent were bitten by lynx, escaped and had developed cervical abscesses from infections of Pasteurella multocida. The two major factors limiting populations of caribou in Newfoundland 1900 to 1967 appeared to be lynx predation of calves and shooting mortality of adults. Poor recruitment and high loss to hunting probably caused the decline of the herds 1915 to 1930. The primary factor limiting numbers in the Interior and Avalon Herds, 1957 to 1967 was lynx predation of calves in their first summer. Illegal hunting was probably important in the Northern Peninsula and Humber River herds 1957 to 1967. / Science, Faculty of / Zoology, Department of / Graduate

The object of this study was to compare the role of changes in reproduction and mortality in regulating population density in the deer mouse, Peromyscus maniculatus, and if possible to discover some of the factors affecting the reproductive rate. Observations were made on natural and experimentally reduced populations on the University Endowment Lands in 1964 and 1965. In both years numbers remained relatively steady during the summer, increased in the fall, when immature animals replaced the adults, and gradually declined over the winter. Animals were about equally abundant at comparable times in both years, fall densities being about 4.72 mice per acre. The stationary state of these populations was associated with a poor reproductive performance on the part of the females, whose breeding success varied between areas, and was greater in 1964 than in 1965. No change in litter size or in prenatal loss was observed during the period of study. The males, in contrast, were sexually active from March to September on all areas in both years. The greatest loss of mice occurred between birth and age at first capture, after which juveniles survived at the same rate as the adults. Survival was poorer during the breeding season than during the winter, and survival of males was poorer than that of females. Populations whose numbers had. been experimentally reduced and whose age structure had been altered, were not significantly different from the natural populations in mean monthly body weights, reproductive performance, or survival. Mean monthly body weights and reproductive performance were lower, and survival of the young from birth to age at first capture was higher in 1965 than in 1964. The proportion of subadults which became fecund was greatest on this area in 1964. The stationary state of these populations was maintained by changes in survival rather than by changes in reproductive rate. The reproductive performance of the females was fairly constant whereas the loss of young from birth to age at first capture varied. Loss of the young is attributed to their death or emigration in response to aggressive interactions within the population. / Science, Faculty of / Zoology, Department of / Graduate

Sadleir (1965) proposes that the survival of juvenile deermice is determined by the aggressiveness of the adult population. During the summer, when adult aggression is high, juvenile survival is poor, but in the fall, when adult aggression is low, juveniles survive well. The purpose of this study is to examine some of the consequences of Sadleir's hypothesis experimentally. Sadleir bases his hypothesis on the observation that the aggressiveness of males changes seasonally. This premise has been reexamined and confirmed. How adult aggression affects juveniles was studied first in the laboratory. Juveniles grow poorly when competing with adults in their home cage. Males appear to be more active aggressors than females, but only aggressive males are capable of inhibiting juvenile growth. Even though juveniles grew slowly when competing with aggressive adults, they seldom died from encounters with adults. In order to avoid the crowded conditions and confinement implicit in the laboratory experiments, the relationship between adult aggressiveness and juvenile growth and survival was reexamined in field experiments. Two partly isolated plots of habitat were used, and on these plots artificial populations of aggressive or docile male deermice were established. Juveniles were then released onto the plots, and their growth and survival followed. In the field, as in the laboratory, juveniles grew poorly when competing with aggressive adults. Since emigration was not restricted in the field, however, juveniles disappeared in significantly greater numbers when the adult population was aggressive than when the adult population was docile. In addition to these experiments, the success of immigrants onto trapped out plots and plots with a resident population was examined. Immigrants were more successful in establishing themselves on trapped out plots. All the data collected support Sadleir's hypothesis, and it seems reasonable to conclude that the correlation he drew between adult aggressiveness and juvenile survival is real. However, the data collected also provide some interesting clues as to the organization of deermouse populations. An organization is proposed in which the social unit is an animal and its immediate neighbours. Within the social unit mutual antagonism is reduced. But the members of the unit maintain a high level of aggressiveness, and are intolerant of any stranger that wanders into their home ranges. The system proposed would prevent immigrants from settling, while conserving energy by reducing antagonism between familiar animals. The system would also effectively regulate population size. / Science, Faculty of / Zoology, Department of / Graduate

La présence des mutations délétères a favorisé l'évolution de mécanismes, au niveau cellulaire et au niveau des organismes (e.g. les régimes de reproduction), permettant de diminuer leurs effets négatifs. Au cours de cette thèse nous avons étudié leur impact sur la taille des populations à travers des modèles tenant compte de l'interaction entre la démographie et la sélection, cette interaction étant souvent mise de coté dans les modèles conventionnels de génétique des populations. Dans un contexte déterministe à un seul locus des mutations somatiques et gamétiques influencent la taille et le fardeau génétique des populations (ces derniers étant dépendants du moment d’expression des mutations dans le cycle de vie). Nos modèles stochastiques avec un grand nombre de locus indiquent que la viabilité des populations dépend des paramètres démographiques et génétiques (taux de mutation, effet délétère des mutations). L'autofécondation est généralement avantageuse, augmentant la taille et la viabilité des populations, mais lorsque les mutations sont de faible effet un régime d'autogamie stricte mène à l'extinction par fonte mutationelle. En permettant l'évolution de l'autofécondation à partir d'une population allogame nous observons des cas de suicide évolutif où les populations évoluent vers l'autogamie stricte et s'éteignent, ce qui pourrait expliquer les taux d’extinctions élevés des espèces auto-fécondantes comparées aux allo-fécondantes. Ces modèles prédisent que la taille des populations pourrait être une conséquence et non une cause de leurs propriétés génétiques, appuyant sur l’importance de prendre en compte leur interaction dans l'étude de l'évolution des populations. / As the ultimate source of genetic variation, mutation has the inconvenience of introducing deleterious mutations. These mutations shape the evolution of species, from genetic mechanisms on the cellular level to reproductive systems, which lessen their effects on fitness. In this thesis we explore how these mutations influence population size by allowing the interaction between population size and selection, which has been little explored in conventional population genetics models. In a deterministic context with a single locus, germ-line and somatic mutations influence population size and the mutation load, both which depend on the timing of the expression of these mutations. Multi-locus individual based models show that population viability depends on the demographic properties and on the rate of introduction and impact of mutations. Though self-fertilisation generally increases population viability, strictly self-fertilising populations go extinct due to mutational meltdown when mutations are of small effect. When selfing is allowed to evolve from an outcrossing reproductive regime, there are cases of evolutionary suicide where strict selfing evolves and leads to extinction. We predict that the genetic properties of populations may not be a consequence but a cause of population size. We have emphasized the importance of taking the demographic consequences of deleterious mutations into account when studying the evolution of populations, as in the case of the evolution of self-fertilisation where the previously undetected evolutionary suicide was observed. This result may explain the observed higher extinction rates in selfing compared to outcrossing species.

La thése poursuit les travaux de la théorie des Jeux évolutionnaires Cette théorie est un cadre de modehsation de la dynamique des populations dans lequel les interactions entre agents sont modélisées par des dilemmes classiques de la théorie des Jeux. Les agents interagissent avec leurs pairs et les meilleurs comportements se diffusent. les moins performants tendent à disparaître. Les modèles spécifiés mettent notamment en évidence des conditions sur les rapports inter-individuels qui permettent de faire émerger des équilibres coopératifs. En supposant que chaque agent a des relations non plus avec tous les agents de la population mais seulement avec un sous-ensemble des agents de la population et toujours avec les mêmes, on augmente considérablement le nombre des dynamiques possibles. Cette démarche fait apparaître un réseau des interactions,.soit un graphe. La contrainte spatiale s'avère une condition favorable au maintien des comportements coopératifs et de la biodiversité des comportements. L'analyse formelle de la convergence n'est généralement plus possible et les modèles sont étudiés par simulation. La these poursuit l'étude de l'impact de la répartition spatiale. Elle introduit un nouveau modèle de répartition spatiale où des communautés d'agents sont en réseau et non plus des agents. Ce modèle permet de mettre en évidence de nouvelles formes d'attracteurs coopératifs et de nouvelles conditions au maintien de la biodiversité. La thèse montre aussi la possibilité de convergence de marchés vers des équilibres non concurrentiels et de maintien de comportements coopératifs, des comportements de cartel / The thesis continues the work of Evolutionary Game Theory. This theory is a framework for modeling the dynamics of populations interactions betwcen agents are modeled by the classic dilemmas from Game Theory. The agents interaet with their peers. Best behaviors spread. less suecessful ones tend to disappear. Evolutionary Game Theory prondes conditions upon which cooperative behaviors can survive and cooperative equilibriums can appear. Assuming that each agent does not interact with all the others but only with a fixed group of neighbors greatly increases the number of possible dynamics. The relations between agents form a social network. Limitation of contacts between agents favors cooperative equilibrium and biodiversity. Nevertheless, formal approach to the models is no longer possible and they are studied through massive computer simulations. The thesis continues the study of the impact of social networks. lt introduces networks of communities, a new model where commnnities are networked rather than agents. This model exhibits new forms of eooperative convergence and new conditions to persistent biodiversity. The thesis also shows the possibility of convergence of markets towards non-competitive equilibriums and the survival of behaviors collectively able to reach cartel equilibrium, a particular form of cooperative equilibrium.

This thesis combined data from a trapping and radio-telemetry study of snowshoe hares at Kluane Lake, Yukon from January 1984 through August 1985 with data collected at the same site from 1977-83 (Boutin et al. 1986; Krebs et al. 1986) to examine possible causes for the 10-year cycle in density of snowshoe hares. In Chapter 2 I used data on causes of mortality, from a radio-telemetry study of a cyclic snowshoe hare population during 1978-84, to consider the importance of predation in causing the hare cycle. I found that predation during winter was the largest source of mortality for snowshoe hares during 1978-84. There was a 1-year lag in the response of predation mortality to changing hare density. There was a 2-year lag in the response to changing density of mortality due to causes other than predation. I incorporated this information on causes of mortality into a simulation model, to see whether observed predation mortality can cause changes in density similar to those of a cyclic population. I fitted the predation mortality data to a function in which total predator response consists of a Type II functional response and a delayed density-dependent numerical response. Using a simulation model that predicted mortality rates with this function, I produced 8-11 year cycles within parameter values measured in this study. In Chapter 3 I compared a non-cyclic snowshoe hare population on Jacquot Island in Kluane Lake, with a cyclic population on the mainland, 40 km to the SE. I use trapping data from both mainland and island sites, for a period that included population increase, peak, and decline (1977-85) to test hypotheses of conditions sufficient to cause a hare population cycle. I also presented results from a radio-telemetry study, conducted on both mainland and island during a population low on the mainland (1984-85). The hypothesis that high rates of recruitment followed by low rates of recruitment, is sufficient to cause a cycle was not supported. Data presented was consistent with hypotheses that any one of the following conditions was sufficient to cause the hare cycle: 1. High rates of survival followed by low rates of survival, particularly of juveniles 2. Delayed density-dependent predation 3. Periodic food shortage. / Science, Faculty of / Zoology, Department of / Graduate

A number of studies on small mammals have shewn that changes in the frequency of alleles at polymorphic loci are correlated with population fluctuations. To determine whether this association between demography and genetics is causal, I altered gene frequencies in two field populations of M. townsendi Using starch gel electrophoresis, I detected a leucine aminopeptidase (LAP) polymorphism in M. townsendi. The fast allele, LAP-F, was present in a control population at a frequency of about .35 from July 1971 to July 1973. By removing homozygous SS voles from one experimental population I maintained an LAP-F frequency of about .75. Removal of FF homozygotes from a second population resulted in an LAP-F frequency of about .25. I monitored demographic variables of the populations while the selection was being applied. The populations went through increasing and peak phases and then declined sharply during the spring of 1973. There were indications that different genotypes had an advantage in survival and reproduction during different phases of population density. The selection that maintained the polymorphism on the control area could be correlated with population density. However, the overall fitness of each experimental population was not affected by its genotypic composition at this locus. / Science, Faculty of / Zoology, Department of / Graduate