Scope 47- Long-term Ecological Research, An International Perspective

10.1 INTRODUCTION

Long-term ecological research in Africa has had a substantial range of foci and approaches. The International Livestock Centre for Africa, headquartered in Addis Abbaba, has emphasized management, veterinary, and other studies related to development. Various United Nations organizations, including UNESCO, FAO, and UNEP, have also supported ecological research in a wide variety of countries and with varied objectives. Our focus in this chapter, for reasons of coherence and conceptual integration, emphasizes research on natural ecosystems, largely national parks and game reserves, highlighting research in Tanzania's Serengeti National Park as a case history approach.

Long-term research in natural African ecosystems has been centered on the continent's abundant large mammal populations, but those populations have also been integrated from the outset into a more comprehensive ecosystem perspective (McNaughton and Georgiadis, 1986). Systematic, quantitative research began in the 1950s and early 1960s, most of it initially undertaken by visiting scholars from the United States, Great Britain, and Europe. That research tended to focus on defining the ranges of migratory herds as an aid to drawing boundaries of the newly developing game reserve and national park systems, evaluating reductions in tree cover associated with concentrations of elephants (Loxodonta africana) in the protected areas, and taking a census of the game populations to ascertain their numbers.

Most long-term ecological research in Africa, in fact, has an explicit management responsibility. It is performed in national parks and other reserves, is financially supported, at least in part, by African governments, and is explicitly charged with advising management personnel and formulating management plans. Among the principal sites of the programs are Kenya's Amboseli National Park, South Africa's Kruger National Park, Uganda's Queen Elizabeth National Park, Ivory Coast's Lamto Research Station, and Tanzania's Serengeti National Park.

Kruger National Park supports one of the most sophisticated long-term ecological research programs, with extensive computerized records of animal populations, vegetation, weather, fire, and other ecological components important in African ecosystems. Regular aerial censuses of animal populations and tree cover are features of the data bases. Among the longest-term ecological experiments known to us is the factorial burning treatments initiated in the Park in 1954 (Gertenbach and Potgieter, 1979). Kruger also is an intensively and extensively managed park, and it is not accidental that intense management is coupled to sophisticated long-term data records. The Park has an entirely fenced perimeter, is divided into firebreaked landscape blocks subject to a regular burning program, and the animal populations are intensively harvested to maintain their bounds within explicit management goals. Annual reports of the Kruger research staff, however, are published in Afrikaans, and much of the information from the program is inaccessible to most potential readers.

10.2 ECOLOGICAL MONITORING

A core component of most African long-term research projects has been ecological monitoring, the process of maintaining regularly updated records of key parameters in a defined geographic area. Management goals have been explicit elements in monitoring programs. The term 'monitoring' has the implicit implied function in research of warning management personnel of both desirable and undesirable trends, as they are defined by the goals of the regional management. The monitoring process involves maintaining and continually updating records of such key components as rainfall, flora, fauna, human activities, and land use. In most African programs, monitoring is designed to establish comprehensive baseline data, and to quantify basic patterns of geographical distributions and abundances of major large mammal species. Observations are designed to encompass the key phenomena believed to reflect trends and parameters of broad environmental and ecosystem conditions. The monitoring records eventually become the bases for determining trends in ecosystem state and dynamics over extensive temporal sequences and spatial scales.

Ecological monitoring relies heavily upon co-operation between monitoring personnel, managers, and project-oriented research scientists, who together evaluate and modify the monitoring program to pin-point important parameters desirable or essential for long-term evaluation. The ecological monitoring strategy is designed to improve its own cost-effectiveness. Measurements implemented on the ground, common in project research and some management activities, provide detailed information but are impractical and expensive to implement over large areas. For example, they are impractical for such key goals as animal counting, and expensive for monitoring the balance of grass, bush, and trees over large areas. Low-level aerial techniques supply less detailed information, but are cheaper and quicker to apply for such goals as animal censuses. Satellite images can provide low-resolution, but spatially extensive, information of certain types on a regular and cost-effective basis for whole regions, or even continents (Tucker et al., 1985). A comprehensive monitoring program combines data from each of these three levels, ground, light aircraft, and satellite.

Ecological monitoring is an essential, integral element of long-term research. It provides the data base necessary for implementing goal-focused project research and for computer modeling. Because the authors of this chapter are directors of the Serengeti Ecological Monitoring Programme and of a project research program of over 15 years' duration in the Serengeti, the chapter emphasizes the research programs in- the Serengeti ecosystem, probably Africa's most intensively studied and comparatively unmanaged ecosystem that preserves reasonable components of the continent's abundant large mammal community (Sinclair and Norton-Griffiths, 1979). The chapter emphasizes three aspects of the Serengeti research:

1. The history of long-term research there, as an instructive model for the hazards of long-term concepts in the absence of reliable funding;
2. The accomplishments, for what they reveal about the implementation of concepts; and
3. Future projections, contrasting the ideal with the financially and logistically feasible in less-developed countries.

10.3 THE SERENGETI

10.3.1 THE ECOSYSTEM

The Serengeti ecosystem (Figure 10.1 ) is defined by the annual movement of the earth's largest population of migratory large ungulates, over one million (Sinclair , 1987) wildebeest (Connochaetes taurinus). This herd moves over approximately 25 000 km² of open and wooded grasslands along a mean annual rainfall gradient from less than 400 mm in the south-east to above 1100 mm in the north-west. Much of the herd's range is within Tanzania, but a portion of the critical dry season range extends across the international boundary into Kenya. The limits of the annual ranges of wildebeest and the other major migratory species, the plains zebra (Equus burchelli), were employed at the outset to define the bounds of the ecosystem (Lamprey, 1979). These bounds encompass the ranges of many resident large herbivores of over 25 species, a diverse mammalian carnivore fauna, and a spectacular avifauna.

The Serengeti National Park lies at latitudes of 1°-3° S within an elevation range of 1135-1800 m. It spans three of East Africa's 10 major geological regions from the fault-scarp of the Rift Valley in the east, across a northern extension of Tanzania's huge central plateau in the middle, to the Lake Victoria basin in the west (Lundgren, 1975). The rainfall gradient is due to the rainshadow created by the Crater Highlands in the south-east, the Lake Victoria Convergence Zone in the west, and the seasonal movements of the Intertropical Convergence Zone (McNaughton, 1983); The large migratory herds of wildebeest and zebra that dominate the fauna spend the peak of the wet season (January to April) in the driest region in the south-east, and the peak of the dry season (July to October) in the wettest region in the north-west.

Figure 10.1 Map of the Serengeti National Park and associated wildlife areas in Tanzania that constitute core elements in the Serengeti ecosystem.

During the initial period of intensive research in the ecosystem, the Park was mapped geologically (Macfarlane, 1969), woodlands were assayed (Herlocker , 1976), soils were surveyed (deWit, 1978; Jager, 1982), animal populations were regularly censused (Sinclair and Norton-Griffiths, 1982), landscape classification systems were applied (Gerresheim, 1974; Epp, 1978), and autecological studies concentrated on all major mammalian herbivore and carnivore species (Sinclair and Norton-Griffiths, 1979). This period of comprehensive research was initiated as the Serengeti Research Project and was quickly converted into an institutional form as the Serengeti Research Programme.

10.3.2 DEVELOPMENT OF A LONG-TERM PROGRAM

Although the Serengeti Research Programme was not conceived initially as an ecosystem study, it did, nevertheless, rapidly evolve into one (Lamprey et al., 1971; Lamprey, 1979), and the pioneering project research by early scientists (Pearsall, 1957; Grzimek and Grzimek, 1960; Talbot and Talbot, 1963) was important in laying a regional foundation for what was to follow. From the outset, research in the Serengeti was greatly influenced by potentially serious management problems associated with dense populations of large grazing and browsing mammals. Elephants, the major non-human agent of habitat change in  Africa (Laws, 1970), were unknown in the ecosystem from its discovery at the turn of the century until 1951, when they invaded the Park due to displacement by growing human populations in the vicinity of Lake Victoria (Lamprey et al., 1967).

Rinderpest, or cattle plague, an exotic disease introduced into Africa prior to the discovery of the Serengeti by Europeans, had a devastating effect on many susceptible ungulate species (Lydekker, 1908), particularly wildebeest and African buffalo (Cyncerus caffer). Conversion of the Serengeti region to a rinderpest-free zone by livestock inoculation in the early 1960s resulted in a resurgence of susceptible species, and the wildebeest, buffalo, and topi (Damaliscus korrigum) populations increased several-fold over succeeding years. Ongoing research sponsored by the European Economic Community has documented continued presence of the virus in Serengeti animals.

Regular, extended censuses of major animal populations has been a fundamental contribution of ecological monitoring (Figure 10.2) that is unlikely to have been achieved in the absence of a formal, supported monitoring program. Justification of the expense of these censuses and integrity of record maintenance is unlikely in more project-oriented research programs. However, such population records can be essential to much project research, and can contribute to scientific progress in their own right through the demographic data that they provide.

Figure 10.2 Thirty years of census records for the Serengeti migratory wildebeest population. Data through 1967 were 'total' counts and afterwards were photographic sample counts with error bars showing standard errors of the estimates. Data are part of the archives of the Serengeti Ecological Monitoring Programme, and were published in Swynnerton (1958), Sinclair (1973), Sinclair and Norton-Griffiths (1982), Sinclair et al. (1985), and Sinclair (1987)

The invasion of the Park by elephants and the resurgence of the ungulate populations created two serious management problems at which research was directed: loss of tree cover by the combined effects of elephants, browsing, fire, and overgrazing (Norton-Griffiths, 1979). The vision and efforts of the trustees of Tanzania National Parks led to establishment of the Serengeti Research Project in 1963 and creation of the Serengeti Research Institute (SRI) in 1966. A comprehensive ecosystem study was envisioned, but that was prevented by limited funds and facilities. Instead, a scheme of research priorities was designed by the SRI and its international Scientific Council. Priorities were based largely on the existing requirements of management staff, and as a matter of policy, a purely academic approach was avoided. An early major priority was the establishment of a monitoring program, and in 1969 the Serengeti Conservation and Research Programme was initiated. This program became the Serengeti Ecological Monitoring Programme (SEMP) in 1970, and was operational for eight years until withdrawal of international funding led to serious curtailment of its activities. 

A broad range of research was carried out at SRI from the 1960s to the late 1970S and SEMP (Norton-Griffiths, 1972) was integral to all of it. A comprehensive rain gauge system was established, among the most dense in Africa, and a complete weather station was constructed at SRI headquarters. The number of ungulates was monitored, and monthly reconnaissance of distributions of animals visible from the air was implemented on a grid system. Studies of major herbivore and carnivore species were commenced, some of which have continued throughout the intervening 25 years.

Over an approximate l5-year period of intense activity, data were collected on rainfall, animal distributions, population sizes and demographics of major large herbivores, ecology and behavior of the major large carnivores, the region's abundant scavenging avifauna, primary productivity, soils, geology, tree cover, fire frequency and extent, human settlement outside the Park and encroachment within the Park, hydrological properties of the major river system, and veterinary aspects of the animals. A technical and support staff was an integral component of the long-term research program. This staff included a botany technician, a photographic technician, a librarian, soils technicians, veterinary and wildlife assistants, vehicle drivers, mechanics, and a general laborer staff.

10.3.3 ORGANIZATION OF LONG-TERM RESEARCH

Research was organized in a typical administrative fashion as the program evolved. The Director of SRI was responsible for principal liaison with the national government, the Director and Trustees of National parks, the Chief Park Warden of the Serengeti National Park, international funding organizations, and the Scientific Council. The Scientific Council was an international body, including several scientists of major stature, created to evaluate ongoing research and to advise on program orientation. The director of SEMP was responsible for executing and maintaining the records of the principal monitoring functions: rainfall and weather records, population estimates of principal herbivore species, reconnaissance flight data on animal distributions, aerial photography for determining trends in tree cover, fire mapping, and base maps of regional geology, soils, and vegetation. The administrator was responsible for day-to-day operation and disposition of responsibilities among the technical and labor staff. An assistant director was responsible for co-ordinating the Institute's overall research objectives and meeting the needs of project research. A clear ecosystem approach was evident in the permissions allowed for this project research, largely funded outside the SRI budget. Some projects dealt with soils, primary production, and vegetation; others with each of the major herbivore species; others with major carnivores, including the longest continuous study, of over 20 years, on lions (Panthera leo); and other important ecosystem components, little studied in most long-term ecological research, such as small mammals, the scavenging avifauna, parasites and pathogens, termites, and dung beetles.

10.3.4 THE  CONSEQUENCES OF HIATUS

Long-term ecological research, by definition, requires reliable, continuous funding. With the exception of Kruger and Amboseli National Parks and Lamto Research Station, continuous funding has been only partially characteristic of research in Africa. Political, economic, and, in some places, military turmoil have hindered the continuity required by long-term research. Most project grants have been subject to significant funding hiatuses throughout the period of active research in the Serengeti.

Because its activities are central to long-term research, the disappearance of SEMP was symptomatic of an interruption of broad, cross-project approaches to Serengeti research. International economic problems, associated with oil prices and exacerbated by Tanzania's war with Uganda, closure of the international border between Tanzania and Kenya, and various political factors made international funding bodies wary , and funding for SEMP outside of Tanzania was no longer available after 1977. Internal funds were used to maintain rainfall records from rain gauges accessible by ground transport, but most of the region became inaccessible in the absence of light aircraft and maintained bush landing strips. The weather station at headquarters fell into disrepair, and crucial reference collections of mammal skins, bird skins, and insects were destroyed by pests. Integrity of the herbarium was maintained by project funds. Research in the Serengeti became largely confined to individual project programs, each with independent sources of funding, and the original research concepts and resultant priorities no longer applied. For eight years, long-term research in the Serengeti continued only in the form of those individual projects.

10.3.5 MONITORING: RE-INITIATION IN THE SERENGETI

During the funding hiatus for SEMP, ecological research in Tanzania was reorganized. SRI became the Serengeti Wildlife Research Centre, a subsidiary of the Serengeti Wildlife Research Institute (SWRI), headquartered in Arusha and no longer a component of Tanzania National Parks. SWRI assumed responsibility for all wildlife-related research throughout Tanzania and no longer focused exclusively or largely on the Serengeti. The need for continuous ecological monitoring of the Serengeti was emphasized at a workshop (IUCN, 1986), and SEMP was re- established in 1986, with a full- and part-time project officer, with funding from the Frankfurt Zoological Society, World Wide Fund for Nature, and the SWRI. In 1987, the Wildlife Division seconded a project officer to SEMP, and Tanzania National parks created the Post of Serengeti National Park Ecologist under a newly formed Ecology Department. This ecologist is attached to SEMP, providing an important direct link to park management.

The principal objectives of the monitoring program have remained much the same over the last 20 years (Norton-Griffiths, 1972):

1. To provide information for park management staff on the status and trends of, animal and plant communities within the ecosystem.
2. To provide information for park staff and land-use planners on land-use patterns and trends around the ecosystem.
3. To provide timely warning of potential management problems.
4. To provide baseline information for the formulation of management plans and policies.
5. To provide a means of evaluating the effectiveness of management policies.

A specific requirement now exists for monitoring throughout Tanzania. As in the Serengeti, monitoring presently takes place at two levels: ground-based sampling by monitoring staff, functioning as part of the management of various conservation areas; and aerial survey. SEMP and the Serengeti Wildlife Research Institute currently meet requests for aerial monitoring from Tanzania National Parks, the Wildlife Division, and the President's Office. Training of Tanzanian personnel is an important component of this country-wide effort.

10.4 LONG-TERM ECOLOGICAL RESEARCH

10.4.1 PROJECT PERSPECTIVES

The longest term African research projects, aside from the Kruger burning plots, have been animal centered and behaviorally oriented. Primate research,  concentrating on chimpanzees (Pan troglodytes ), has been underway in Gombe Stream National Park since 1960 (Goodall, 1987). The lion research project in the Serengeti was begun by George Schaller in 1966 (Schaller, 1972) and has continued to the present, with some discontinuities, through four sets of prime investigators. The Amboseli elephant behavior project is 15 years old and has been largely supervised by a single individual with many assisting participants (Moss, 1988). Principal components of these long-term studies are individual animal recognition, detailed behavioral records, and lineage determinations. Although the focus of these autecological studies has been sociobiological, they nevertheless provide substantial long-term information on basic properties of the systems that the animals occupy. 

The Serengeti Grazing Ecosystem Project has been ongoing for 15 years, and combines field studies (McNaughton, 1976, 1983, 1985) with simulation modeling (Coughenour et al., 1984a,b) and laboratory experiments that are infeasible in the field on parameters necessary for model development and for understanding underlying mechanisms (McNaughton et al., 1983; Coughenour et al., 1985; Ruess and McNaughton, 1987). Although the goals of SEMP, as outlined above, are oriented toward management, the accumulated data bands and previous research at SRI were fundamental to rapid development of the grazing ecosystem project and to its progress. The rainfall and soil survey data provided a basis for rapid initial selection of a study site network, giving a broad span of coverage of the region's principal ecological variables: rainfall, edaphic properties, grazing intensities, and vegetation types. Although most study sites had to be individually assayed for rainfall during the hiatus in SEMP, site placement otherwise has taken advantage of the SEMP rain gauge network. Aerial observations of animal distributions allowed temporary study sites to be placed in a timely fashion to take advantage of specific experimental opportunities (McNaughton, 1985). All in all, experience with the Serengeti Grazing Ecosystem Project reveals a central role for long-term monitoring in the establishment, execution, and progress of long-term project research.

10.4.2 OBJECTIVES AND PRIORITIES

Long-term monitoring either is goal-specific or follows a more general course. In the former case, goals are determined before initiating monitoring activities, and methods established at the outset must be sufficiently accurate and precise to meet the stated objectives. In general, this requires both prior knowledge of the ecosystem and substantially greater personnel and funding than the second approach, in which monitoring activities follow a less-detailed, data-gathering path. The two approaches are usually complementary, with the more general, less specific approach often a necessary precursor to the goal-specific approach. An important aspect of less goal-oriented monitoring, however, is its ability to respond and adjust to explicitly formulated goals that arise out of the monitoring process, and those which management personnel may require.

Long-term ecological research in African ecosystems is, in our view, infused with five, not always explicit, goals. First, it is designed to establish baseline data on prevailing environmental conditions; this focuses primarily upon rainfall records at a representative range of locations. Second, it maintains and updates records on such fundamental underlying properties as geology, geomorphology and landscape regions, soils, and vegetation types. Third, it considers variance about baseline data, either as fluctuations about a mean or as directional trends. Fourth, it studies patterns of state and dynamics that cannot be revealed by short-term research. Finally, whether sociobiological or ecological in primary focus, it is comprehensive and integrative in scope. Animal behavior is placed in a system context, and ecosystem studies include the animals as an integral component. These are neither strictly monitoring nor strictly project or management goals, but a combination of the three.

What is long-term in the African context? For a multivoltine insect, a couple of years of research can encompass many generations and constitute a long-term study. But for ecosystems like those in Africa, whose most conspicuous feature is populations of long-lived mammals, a more extended temporal scale is relevant. For monitoring functions and most project programs, a decade seems a reasonable time span to establish prevailing patterns and general trends. For long-lived animals, behavioral studies with kinship and learning as fundamental elements may require somewhat longer time periods. Similarly, determining changes in the balances among an arborescent overstory, an intermediate bush canopy, and the underlying grass canopy may require periods longer than the decade range.

It is clear that a successful monitoring program cannot work as an independent  unit, but must maintain a constant exchange of ideas and data between the ecologists working on specific projects and the management. Only through a flow of information from all directions can management design realistic policies, the monitoring program achieve its objectives, and the project research be efficiently prosecuted. The SEMP currently combines data collection with an information management system to update baseline records, advise management personnel, and provide ongoing data for project research in the following fields:

1. Climate, principally rainfall;
2. Wildlife numbers, population structure, and distribution;
3. Fire and fire control;
4. Vegetation distribution and changes; and
5. Poaching distribution and intensity.

Beyond aerial measurement of tree cover during the active period of SRI, little vegetation monitoring has been accomplished, in spite of the central role that it could play in assessing long-term ecosystem trends. Two types of permanent vegetation plots were established early in the SRI program, permanent plots in the south-east and the north-east for aerial and terrestrial censuses, and experimental plots in several locations combining fencing and firebreaking. The former have rarely been quantitatively censused and the latter were dismantled or fell into disrepair in the mid 1970s. Quantitative baseline data on plant community composition were not collected on either type of plot when they were first established.  

SEMP is now establishing permanent vegetation plots throughout the Serengeti.  Twenty of a planned forty have been delimited with markers visible from the air. These will be sampled regularly from the ground with standardized vegetation sampling methods. In addition, about 2200 sample aerial photos have been taken on a systematic basis for monitoring tree-bush-grass balances throughout the Serengeti National Park and adjacent areas. 

SEMP, when fully implemented under its current goals, should provide the fundamental data bases required for long-term project research. Implementation of a good vegetation monitoring program on fixed plots at a representative range of locations will strengthen the baseline considerably.

10.4.3 FUNDAMENTALS

Successful long-term ecological research, as practiced in the Serengeti specifically, and throughout Africa generally, has several fundamental properties. Those were largely evident from the outset, but have also gradually evolved in implementation. 

First, there must be a clear goal-oriented vision of what the research is to  accomplish. Although it is arguable whether long-term research in Africa has contributed more to basic science or to providing guidelines for management policies, the latter has been the central explicit objective. The inception of long-term research was coincident with recognition of a number of practical problems associated with political, economic, and demographic changes in Africa. For example, within the game reserves and national parks, the 'elephant problem' was a central concern throughout the continent. Increases in elephant populations due to pressures from human populations outside the parks and protection within them led to a rapid decline in tree communities in most parks (Laws et al., 1975). While in recent years this has become a moot point due to wholesale poaching, nevertheless, advising management remains an explicit charge of most long-term ecological research.

Second, monitoring plays a critical role in long-term research. Monitoring programs commonly concentrate on accumulating baseline data on four factors: (1) weather, particularly rainfall; (2) numerical censuses of the principal animal species, together with some attention to such demographic generalities as cow/calf ratios and death rates; (3) the balance between tree and grass cover; and (4) extent and frequency of fires. In addition, basic geological, soil, and vegetation mapping often forms part of the central data archives of monitoring programs. Experimentation is rarely a component.

Third, project research focused on specific research areas is often an integral element of long-term research. The studies with the greatest continuity and detail have been primarily behavior oriented. But a clear delineation of three types of long-term project research is evident:

1. Projects such as the Serengeti Grazing Ecosystem Project that concentrate on such fundamental ecological processes as energy and nutrient flow, emphasizing primary productivity and its consumption by the herbivore community;
2. Autecological studies of the major herbivore species concentrating on demography, food requirements, and the degree and causes of population regulation; and
3. Predator studies focusing in part on predator-prey relations but with a strong sociobiological component.
     

No one has yet accomplished a functional integration of the trophic web from the soil-related to the top predator and scavenger trophic levels. Rather, the long-term project research as been less comprehensive in scope and, because of the nature of its funding, oriented toward short-term, achievable goals. Experimental manipulations are often a major component of long-term project research.

Certain areas of research were only sporadically implemented, but they have continued over a sufficient period of time to become long-term research. For example, veterinary research was an integral component of the initial SRI program because of the importance of rinderpest as a regulator of the Serengeti's game animals. The research fell into abeyance with the inoculation of cattle and suppression of the disease in the wild animals, but reappearance of the disease (Rossiter et al., 1983) led to its reinstatement, with EEC support.

10.4.4 A CRITICAL APPRAISAL

Our evaluation of long-term research in African ecosystems suggests that such research is uniquely or usefully appropriate when the phenomena being studied take place over long periods of time, like the population demographics of long-lived animals (McNaughton and Georgiadis, 1986), or when the phenomena are functionally and structurally complex, like the relationships among plants, herbivores, and the physical environment (McNaughton et al., 1988). In the first case, a certain amount of time is required merely to document and determine any change of direction, and a variety of specific studies are needed to unravel the causes of the change or of constancy. In the second case, the complexity of the study requires a step-by-step experimental approach, involving a variety of research methods from field experimentation to computer modeling in order to identify the organizing linkages and determine their effects.

Long-term research has been fundamental in the progress of scientific understanding of African ecosystems and in the planning of management policies. The research has contributed to the establishment of principles of stability and resilience of the savanna ecosystems (McNaughton, 1977, 1985; Walker et al., 1981; Ellis and Swift, 1988); to the determination of patterns and causes of change in animal and plant populations (Sinclair and Norton-Griffiths, 1979); to the development of management plans related to park goals, for example, fire plans for arresting the deterioration of woodland cover (Stronach, 1988); to the development of tools, such as simulation modeling, to provide scientific insight (McNaughton et al., 1988) and serve as quantitative, predictive management tools; and to the identification of potentially important limiting factors through long-term information and sophisticated technologies (McNaughton, 1988).

The development and refinement of wildlife census methods was a particularly important contribution of long-term Serengeti research to basic ecological methodology (Jolly, 1969a,b). An ultimate outcome of the research was the production of five handbooks on techniques in wildlife ecology focusing on Africa (Norton-Griffiths, 1978; Bertram, 1979; Grimsdell, 1979; Western and Grimsdell, 1979; Sinclair and Grimsdell, 1982), and fundamental contributions to general approaches to ecological monitoring in long-term research (Clarke, 1986). General descriptions (Lamprey, 1979) and detailed descriptions (Sinclair and Norton- Griffiths, 1979) of the Serengeti research also provide ample evidence of the value of long-term ecological research and its applications to management policies.

Nevertheless, it would be disingenuous to suggest that research has contributed to solutions of some of the most critical practical problems that confront African resource managers. Chief among those problems has been the blitzkrieg of commercial poaching sweeping over Africa. It is correct, we believe, that regular censuses of animal populations have provided early information about the onset of intense poaching almost everywhere that it has occurred; the solutions to the problem, however, lie in economics and law enforcement. In spite of the fact that data-based simulation studies document that the greatest economic return is derived from harvesting mature ivory from dead animals (Pilgram and Western, 1986a,b), the most rapid economic yield to poachers is the complete harvest that is currently underway.

Long-term research can be most useful to management in three ways. First, it can provide records of subtle, chronic changes in ecosystems, such as changes in the balances among woodland, bushland, and grassland. Second, it can provide an early warning of the onset of acute changes, such as those due to poaching or disease outbreak. Third, it can suggest management strategies for dealing with changes. Dealing with the acute changes, such as poaching and disease, is complex and costly, but is well understood in the hypothetical sense.

10.4.5 PROSPECTS

Initial funding of long-term research in Africa came primarily from international funding bodies with a major conservation orientation. Later, project grants obtained funds from national scientific funding bodies such as the US National Science Foundation and the British Royal Society and Research Councils. But international conservation organizations later shifted funding away from research and into direct management activities and educational programs. This led to such consequences as the almost decade-long interruption of SEMP. Project research has continually been disrupted by the vagaries of funding.

The main impediment to long-term ecological research, in fact, has been funding discontinuity. Project proposals, by their nature, move in a generally linear fashion from one specific aspect to another in a general problem. This curtails the opportunity to take a broad perspective. Monitoring programs are so goal oriented and non-experimental in character that they are unable to provide the mechanistic insight into ecosystem function and population regulation that arises out of project research. Simulation modeling, an integral element of mechanistic understanding of large-scale ecological problems, has rarely been a component of research anywhere in Africa.

Mechanistic, quantitative simulation modeling, we believe, should be a core component of long-term ecological research, but only the Grassland Research and Serengeti Systems Model (GRASS) fulfils this requirement in African ecosystems (Coughenour, 1984; Coughenour et al., 1984a,b; McNaughton et al., 1986). Simulation modeling is important in long-term ecological research because it provides an organizing framework for research objectives, can point to areas of desirable information currently unknown, allows model manipulation to substitute for experiments that are impractical, impossible, or dangerous, and provides the only basis for extrapolations beyond the lifetimes of individual projects or scientists. Properly constructed and parameterized, simulation modeling could provide the basis for formulating long-term policies necessary in management. In addition, good models could allow the assessment of different management options and provide a basis for estimating the consequences of such trend deviations as drought, runs of wet years, severe poaching, and epidemic disease in some animal populations. There is a clear need for more thoroughly developed simulation modeling of ecosystems in Africa and in the tropics as a whole.

The United States' Long- Term Ecological Research Program (Callahan, 1984; Chapter 2 in this volume) could serve as a model for funding long-term research in Africa and throughout the world. The program funds multi-investigator teams for five-year, renewable intervals, concentrating on site-specific ecological phenomena. International funding with a similar scope, concentrating, most logically, on World Heritage sites representing a characteristic range of ecosystems, could provide the funding continuity necessary for long-term, integrated research. Combining the functions of monitoring with the experimental approaches characteristic of project research, including a strong modeling component, would do much to place long-term ecological research in the global scientific mainstream.

10.5 ACKNOWLEDGMENTS

We are grateful to the Trustees of Tanzania National Parks and the National Scientific Research Council of Tanzania for permission to live and do research in the Serengeti National Park. We express our appreciation to Mr David Babu, Director of Tanzania National Parks, Professor Karim Hirji, Director of the Serengeti Wildlife Research Institute, and Mr B. Maragesi, Principal Park Warden of the Serengeti National Park. Our research is supported by US National Science Foundation grants BSR 8505862 and NSF BSR-8817934 to SJM, and funds from the Frankfurt Zoological Society and World Wide Fund for Nature to KLC. Appreciation is expressed to Professor A.R.E Sinclair for organizing much of the wide census work conducted during 1986.

10.6 REFERENCES

Bertram, C.R. (1979). Studying Predators. Afr. Wildl. Leadership Fdn., Nairobi, Kenya. 

Callahan, J.T. (1984). Long-term ecological research. BioScience, 34, 363-367.

Clarke, R. (1986). The Handbook of Ecological Monitoring. Oxford University Press, Oxford.

Coughenour, M.B. (1984). A mechanistic simulation analysis of water use, leaf angles, and grazing in East African graminoids. Ecol. Model., 26, 203-220.

Coughenour, M.B., McNaughton, S.J. and Wallace, L.L. (1984a). Modelling primary production of perennial graminoids - uniting physiological processes and morphometric traits. Ecol. Model., 23, 101-134.

Coughenour, M.B., McNaughton, S.J. and Wallace. L.L. (1984b). Simulation study of East African perennial graminoid responses to defoliation. Ecol. Model., 26, 177-201.

Coughenour, M.B., McNaughton, S.J. and Wallace, L.L. (1985). Responses of an African graminoid (Themeda triandra Forsk.) to frequent defoliation, nitrogen, and water: a limit of adaptation to herbivory. Oecologia (Berlin), 68, 105-110

deWit, H.A. (1978). Soils and grassland types of the Serengeti Plains (Tanzania). PhD Dissertation. University of Wangeningen, Wageningen, The Netherlands.

Ellis, J.E. and Swift, D.M. (1988). Stability of African pastoral ecosystems: alternate-paradigms and implications for development. J. Range Manage., 41, 450-459.

Epp, H. ( 1978). A natural resource survey of Serengeti National Park, Tanzania. Ser. Res. Inst. Publ., 237, Arusha, Tanzania.

Gerresheim, K. (1974). The Serengeti Landscape Classification--Map and Manuscript. Afr. Wildl. Leadership Fdn., Nairobi, Kenya.

Gertenbach, W.P.D. and Potgieter, A.L.F. (1979). Veldbrandnavorsing in die struikmopanievel van die Nasionale Krugerwilduin. Koedoe, 22, 1-28.

Goodall, J. (1987). The Chimpanzees of Gombe. Harvard University Press, Cambridge, MA. 

Grimsdell, J.J.R. (1978). Ecological Monitoring. Afr. Wildl. Leadership Fdn., Nairobi,  Kenya.

Grzimek, M. and Grzimek, B. (1960). A study of the game of the Serengeti Plains. Z. Saugertierk, 25, 1-61.

Herlocker, D.J. (1974). Woody Vegetation of the Serengeti National Park. Texas A&M University, College Station, TX.

IUCN. (1986). Toward a Regional Conservation Strategy for the Serengeti: Report of a Workshop held at the Serengeti Wildlife Research Centre. IUCN, Nairobi, Kenya.

Jager, T. (1982). Soils of the Serengeti Woodlands, Tanzania. Agr. Res. Rept., 912. Cen. Agr. Publ. Doc., Wangeningen, Netherlands.

Jolly, G.M. (1969a). Sampling methods for aerial censuses of wildlife population. E. Afr. Agric. For. J., 34, 46-49.

Jolly, G.M. (1969b). The treatment of errors in aerial counts of wildlife populations. E. Afr. Agric. For. J., 34, 50-55.

Lamprey, H.F. (1979). Structure and functioning of the semi-arid grazing land ecosystem of the Serengeti region (Tanzania). In Tropical Grazing Land Ecosystems. A State of Knowledge Report. UNESCO, Paris, 526-601.

Lamprey, H.F., Glover, P.E., Turner, M.I.M. and Bell. R.H.V. (1967). Invasion of the Serengeti National Park by elephants. E. Afr. Wildl. J., 5, 151-161.

Lamprey, H.F., Kruuk, J. and Norton-Griffiths, M., (1971 ). Research in the Serengeti. Nature, 230, 497-499.

Laws, R.M. (1970). Elephants as agents of habitat change in East Africa. Oikos, 21, 1-15. 

Laws, R.M., Parker, I.S.C. and Johnstone, R.C.B. (1975). Elephants and Their Habitats. Clarendon, Oxford.

Lundgren, B. (1975). Land Use in Kenya and Tanzania. Roy Col. For., Int. Rur. Dev. Div., Stockholm.

Lydekker, R. (1908). The Game Animals of Africa. Rowland, Ward, London.

Macfarland, A. ( 1969). Preliminary report on the geology of the Central Serengeti, northwest Tanzania. l3th Ann. Rep. Afr. Geol., University of Leeds, 13, 14-15.

McNaughton, S.J. (1976). Serengeti migratory wildebeest: facilitation of energy flow by grazing. Science, 191, 92-94.

McNaughton, S.J. (1977). Diversity and stability of ecological communities: a comment on the role of empiricism in ecology. Am. Nat., 111, 515-525.

McNaughton, S.J. (1983). Serengeti grassland ecology: the role of composite environmental factors and contingency in community organization. Ecol. Monogr., 53, 291-320.

McNaughton, S.J. (1985). Ecology of a grazing ecosystem: the Serengeti. Ecol. Monogr., 55, 259-294.

McNaughton, S.J. (1988). Mineral nutrition and spatial concentrations of African ungulates. Nature, 334, 343-345.

McNaughton, S.J. and Georgiadis, N.J. (1986). Ecology of African grazing and browsing mammals. Ann. Rev. Ecol. Syst., 17, 39-65.

McNaughton, S.J. and Ruess, R.W. and Coughenour, M.B. (1986). Ecological consequences of nuclear war. Nature, 321, 483-487.

McNaughton, S.J., Ruess, R.W. and Seagle, S.W. (1988). Large mammals and process dynamics in African ecosystems. BioScience, 38, 794-800.

McNaughton, S.J., Wallace, L.L. and Coughenour. M.B. (1983). Plant adaptation in an ecosystem context: effects of defoliation, nitrogen, and water on growth of an African C4 sedge. Ecology, 64, 307-318.

Moss, C. (1988). Elephant Memories. Thirteen Years in the Life of an Elephant Family. W. Morrow, New York.

Norton-Griffiths, M. (1972). Serengeti Ecological Monitoring Program. Afr. Wildl. Leadership Fdn., Washington, DC.

Norton-Griffiths, M. (1978). Counting Animals. Afr. Wildl. Leadership Fdn., Nairobi, Kenya.

Norton-Griffiths, M. (1979). The influence of grazing, browsing, and fire on the vegetation dynamics of the Serengeti. In Sinclair, A.R.E. and Norton-Griffiths, M. (Eds) Serengeti. Dynamics of an Ecosystem. University of Chicago Press, Chicago.

Pearsall, W.H. (1957). Report on an ecological survey of Serengeti National Park, Tanganyika. Oryx, 4, 71-136.

Pilgrim, T. and Western, D. ( 1986a). Inferring hunting patterns of African elephants from tusks in the international ivory trade. J. Appl. Ecol., 23, 503-514

Pilgrim, T. and Western, D. (1986b). Managing African elephants for ivory production through ivory trade regulations. J. Appl. Ecol., 23, 515-529.

Rossiter, P.B., Hessett, D.M., Wafult, J.S., Harstad, L., Chemwa, S., Taylor, W.P., Rowe, L., Nyange, J.C., Otaru, M., Mumbala, M. and Scott, G.R. (1983). Re-emergence of rinderpest as a threat in East Africa since 1979. Vet. Rec., 113, 459-461.

Ruess, R.W. and McNaughton, S.J. (1987). Grazing and the dynamics of nutrient and energy regulated microbial biomasses in the Serengeti grasslands. Oikos, 49, 101-110.

Schaller, G.B. (1972). The Serengeti Lion. University of Chicago Press, Chicago.

Sinclair, A.R.E. (1973). Population increases in buffalo and wildebeest in the Serengeti. E. Afr. Wildl. J., 11, 93-107.

Sinclair, A.R.E. (1987). Long-term monitoring in the Serengeti-Mara: trends in wildebeest and gazelle populations. Report No.5, Serengeti Ecological Monitoring Programme, Arusha, Tanzania.

Sinclair, A.R.E., Dublin, H. and Borner, M. ( 1985). Population regulation: a test of the food hypothesis. Oecologia (Berlin), 53, 266-268.

Sinclair, A.R.E. and Grimsdell, J.J.R. (1982). Population Dynamics of Large Mammals .Afr. Wildl. Leadership Fdn., Nairobi, Kenya.

Sinclair, A.R.E. and Norton-Griffiths, M. (1982). Does competition or facilitation regulate migrate ungulate populations in the Serengeti? A test of hypotheses. Oecologia (Berlin), 53, 364-369.

Sinclair, A.R.E. and Norton-Griffiths, M. (Eds) ( 1979). Serengeti. Dynamics of an Ecosystem . University of Chicago Press, Chicago.

Stronach, N.R.H. (1988). Fire Management Plan for the Serengeti National Park. A Report to Tanzania National Parks. Serengeti Wildl. Res. Inst., Arusha, Tanzania.

Swynnerton, G.H. (1958). Fauna of the Serengeti National Park. Mammallia, 22,435-450. 

Talbot, L.M. and Talbot, M.H. (1963). The Wildebeest in Western Masailand. Wildl. Monogr.,  Washington, DC.

Tucker, C.J., Townshend, J.R.G. and Goh, T.E. (1985). African landcover classification using satellite data. Science, 227, 369-375.

Walker, B.H., Ludwig, D., Holling, C.S. and Peterman, R.M. ( 1981 ). Stability of semi-arid grazing systems. J. Ecol., 69, 473-498.

Western, D. and Grimsdell, J.J.R. (1979). Measuring the Distribution of Animals in Relation to their Environment. Afr. Wildl. Leadership Fdn., Nairobi, Kenya.