Most natural ecosystems constantly undergo changes caused by fluctuations either in the population interactions or in the environmental factors. Ecosystems also vary both in time and space.
Adjoining ecosystems also interact among themselves and involve a continual interchange of materials and energy. A further complicating factor is the artificial manipulation, destruction, disruption or selection of the entire ecosystem or of some of its component parts through human activity.
Being so highly complicated, it is naturally extremely difficult to study the ecosystems by conventional methods of descriptive ecology. To overcome this problem, procedures of systems analysis have been applied to the study of the complex functional aspects of ecosystems, and this branch of modern ecology has come to be styled as systems ecology.
Van Dyne (1966) defined systems ecology as “the study of the development, dynamics and disruption of ecosystems.” The systems analysis approach enables ecologists to analyze and understand the highly complex interactions in a rather simple way, and the main importance of this new branch of ecology lies in the fact that it can help us to formally simplify the complex ecosystems by the use of modern mathematical and computer techniques.
All this is based on the underlying assumption that the state of any given ecosystem at any particular time can be expressed quantitatively, and that changes in the system can be interpreted and described in terms of mathematical expressions.
The systems analysis approach basically involves systematic data collection (based on appropriate statistical sampling procedures, etc.) and a simplification of the three-dimensional, complex interacting factors in the form of appropriate mathematical models.
In one such approach, one mainly deals with the quantities of energy and materials in the form of hypothetical compartments’ of the ecosystem. This is called the compartmental system approach. An alternative approach is the experimental components approach in which instead of concentrating on identification of quantitative descriptive aspects of ecosystems, the emphasis is rather placed on interactions and the careful detailed analysis of different ecological phenomena such as competition, predation, symbiosis, etc.
The chief advantage of mathematical models, as applied to ecology, lies in the fact that by their use one can often predict the behaviour of ecosystems and of the dynamic changes undergone by them through a certain time period. An examination of the mathematical system can often provide useful information about corresponding properties of the real-world ecological situation.
The models can be understood and analyzed in terms of such parameters as feedback and control, stability, and the effects of one part of the system on another. The systems approach enables one to incorporate, within a single interrelated model, information from diverse kinds of studies carried out in the field, laboratory, greenhouse and phytotrons or environmental chambers, etc.
Some of the tools that are being increasingly used by systems ecologists consist of digital and analogue computers, electronic calculators, gas chromatographs, infrared gas analyzers, spectrophotometers, bomb calorimetric, telemetric equipment and mathematical simulation techniques, etc.
Van Dyne (1966) rightly emphasized the tremendous importance of computers in the study of systems ecology. While using computers, it is necessary to state precisely what we already know, what we do not know, and what we would like to know.
Also it will be necessary to assemble, analyze, identify, reduce and store our ecological data and knowledge in a form suitable for retrieval by machines. However, one limitation of these models must be borne in mind: all mathematical models are simplified abstractions of an otherwise complex situation and, lience, at least to some extent, are unrealistic.
Systems ecology is based on a strong interdisciplinary and multidisciplinary approach to problems of environmental biology. It requires an effective and integrated collaboration and cooperation among plant and animal ecologists, microbiologists, physiologists, statisticians and mathematicians, computer technicians and programmers, physicists, chemists, meteorologists, agriculturists, electronics engineers, geologists, geographers and general environmental and population biologists. Its objective is to meaningfully study the total ecosystem rather than a fragmentary or disjointed piecemeal consideration of isolated sectors of an ecosystem.
Systems ecology takes into account the whole complex of physicochemical and biological components within a specific environment and their interactions, and treats all these as a single functional unit, the ecosystem. The behaviour of this system through time may be regarded as the outcome of a large number of physicochemical and biological phenomena that are operating within the ecosystem.
The conceptualization of complex systems into various levels of organization—from molecules, organelles and cells to populations, communities and ecosystems—is an inherent feature of biological science. Each of these hierarchic levels exhibits a distinctive behaviour resulting from the integration of sublevel processes and environmental influences, and for effecting such integration, mathematical modelling is being increasingly used.
Such modelling in conjunction with systems analysis and computer programming has been particularly successfully applied for the analysis and understanding of the following areas of animal science: thermoregulation, blood circulation, morphogenesis, neurological functioning, artificial intelligence, and digestion in ruminants (see Atkins, 1969; Baldwin et al., 1977). In plant sciences, ecologists, agronomists and others have applied integrative systems approach for analyzing the behaviour of vegetation and for the study of tundra (Brown, 1979) and grassland (Innis, 1978) biomes.
Many ecologists concede that environmental systems analysis can be a useful approach to the analysis of ecosystem function. But the question commonly posed is whether the present state of the art permits the technique to be appropriately applied at this time.
The answer to this question is, yes. Simulation methods have already been applied, with considerable success, to many general ecological problems such as the evaluation of pest control methods, complex predator-prey relations, and the food-web dynamics of insects (see Reichle, 1971).
Systems analysis and modelling techniques have been applied in the study of radioisotope transfers to man through terrestrial ecosystems (Booth and Kaye, 1971) and a pasture-cow-milk-man food-web (Booth et al., 1971). While the sufficiency and adequacy of the currently available mathematical and computer methods and techniques has been fairly well established as a result of these studies, nevertheless, it has also been found that certain gaps and lacunae exist in our knowledge of environmental parameters. One of the serious gaps is that rate coefficients for many important ecological phenomena are not yet accurately known.
Some of the problems arising from the damage to our environment caused by industrial and urban development, mining, and changes in agricultural and forestry operations, etc., can be tackled effectively by a judicious recourse to systems analysis.
This is because most environmental problems are complex, span long time scales, and require multidisciplinary approach. Traditional methods to solve environmental problems frequently fail, and complex mathematical analogues of physical processes cannot be applied to ecological systems and in some cases have hindered rather than helped solutions (Jeffers, 1978).
In contrast, systems analysis is essentially a broad framework of thought designed to aid us in making best-possible decisions to manage our environmental problems or ecosystems, summarizes some constituent phases of applied systems analysis. Of all the different phases, the definition and bounding of the problem is perhaps the most important of all (Jeffers, 1978).
Systems analysis approach has also been applied in pest management (Ruesink, 1976). Most models emphasize the description of pest population dynamics, with sub models to simulate the infection and spread of the insects. The amount and sites of disease damage or insect feeding assume an important significance, and combined models of this kind may prove to be fairly reliable predictors of pest development or disease injury to crops and for evolving appropriate strategy to control them. Combined models may also facilitate examination of energy and nutrient transfers in host-parasite systems (Loomis et al., 1979).
A somewhat less moderate application of mathematical equations and models in the analysis of biological population has been advocated by Williamson (1972). While recognizing that the study of natural populations is undoubtedly quantitative and requires a mathematical approach, Williamson nevertheless pleads for the use of somewhat simpler rather than advanced mathematics which is generally not properly understood by an average biologist.
Although the use of advanced analytical techniques is certainly desirable, it is not always essential or a necessary requirement for the systems approach. The concept of integration is even more important than the use of computers and advanced techniques.
And, even more important than these considerations is the basic requirement of a proper and adequate data collection, for there is no use in applying the complicated systems analysis on poor, inadequate or unreliable data. Modelling cannot be a cure for poor data.
In fact, some ecologists have strong reservations about the utility of the modelling approach (see Passioura, 1973). Modern ecology is sometimes criticized as “lacking in depth and unifying concepts and subject to excessive jargon” (Lbomis et al., 1979, p. 362), and one explanation for this state of affairs may be that grand models are sometime thoughtlessly built on poor and unreliable data.
This is not to undermine the importance of the hierarchic simulation models or of the systems methods, which have come to stay, however, the modelling and systems approach must be applied only to those cases where the information and data base is really sound and reliable.
A systems analysis approach is perhaps the main foundation upon which the Man and the Biosphere Programme (MAB) was based (see UNESCO, 1972). The general objectives of this programme are: (a) to develop the scientific basis for the rational use and conservation of the resources of the biosphere, and (b) the improvement of the global relationship between man and the environment.
Its scientific approach includes the analysis of ecological systems, the reciprocal studies of man-environment impacts, and the integration of information over various levels, and the inclusion of mathematical modelling techniques to allow quantitative predictions.
According to UNESCO (1972), the programme is intended to:
“(1) identify and assess changes;
(2) examine the structure, functioning and dynamics of the ecosystems;
(3) study the interrelations between ecosystems and socioeconomic processes;
(4) develop means for measuring environmental changes;
(5) increase global coherence of environmental research;
(6) promote simulation and modelling as tools for environmental management, and
(7) promote environmental education.”
All these objectives are intended to be interpreted in the context of man, the biosphere, and their mutual and reciprocal relations.
Lieth (1972) made an intensive and extensive use and application of systems ecology and mathematical modelling in estimating the primary productivity of different kinds of biomes throughout the world. He has compiled a valuable and informative which compares the production averages of various biome types of the world.
This table is also probably the first table of its kind which gives a breakdown and summary of annual energy fixation for the total vegetation cover of the entire world, the estimates being based on the year 1950 when man-made alterations in natural vegetation, etc., were not as severe as in more recent years. According to Lieth (1974), the net primary production figure for the entire biosphere is about 55?109 tonnes dry matter per year, with about 100 x 106 tonnes being due to the terrestrial vegetation.
At the Brookhaven Symposium on the Primary Productivity in the sea, held in 1980, one dominant idea that emerged was that most oceanic productivity measurements made till then were underestimates of the actual values. It was brought out that even in oligotrophic seas, the growth rate of the phytoplankton closely approaches the maximum rates observed in culture.
If the above view is correct, then how can the high growth rates be actually realized in the sea? It is believed that these high growth rates are maintained by the rapid recycling of limiting nutrients (see Falkowski, 1980). In fact, nutrient recycling is not only important in oceans but also in fresh waters. In fresh waters, for instance, phosphorus recycling can be a very critical factor involved in phytoplankton productivity.
Another idea suggested at the above Symposium was that there may be size-class differences in the utilization of nitrogen in the oceans; thus, nanno- plankton is the dominant users of ammonia. The general concept that emerged from this hypothesis was that the organisms most likely to be recycled may possibly be those that are most likely to be the users of recycled products.
Goodall (1974) has advocated the hierarchical approach to systems analysis and model building. According to him, the systems analysis can be facilitated by separating sub-systems from the major system; when a system includes many interlinked components, including flows of nutrients, energy and information, etc., then any sub-set of these components can be separated as a sub-system. The sub-system can in turn be segregated into still smaller categories.
A hierarchical organization thus seems involved in the concept of a system (including ecosystem). Any complex ecosystem can be dissected into simpler and simpler sub-systems amenable to better characterization than the ecosystem as a whole. The components and processes within an ecosystem can be classified in different (independent but mutually compatible) ways and even the classification used may differ from process to process without in any way affecting the validity of the research results or conclusions.
Although many valuable contributions have resulted from the use of systems analysis in ecology, some workers (see Jeffers, 1974) have advocated constraint and caution in indiscriminately and uncritically applying mathematical and computer methods and techniques in the study of dynamic biological and ecological phenomena.
According to Jeffers, systems analysis means orderly and logical organization of data into quantitative (algebraic and arithmetic) models, followed by the rigorous testing and application of the models to similar sets of unknown situations for predictive purposes.
Needless to emphasize that the model building must be based on accurate and reliable data. It is difficult for verbal expressions to adequately describe the complexity inherent in most ecosystems. Hence an increasing recourse to precise and critical systems analysis and reliable models seems inevitable in the future.
Systems analysis is a good means whereby a synthesis between the collection of reliable data and their interpretation through modelling may be achieved satisfactorily.
Jeffers (1974) has proposed that these objectives could be realized in five successive phases, viz., definition of objectives and preliminary syntheses, experimentation, management, evaluation, and final synthesis. The basis of systems and modelling approaches in ecology is the unifying concept of integration, and this is why the philosophy of systems analysis has been wholeheartedly accepted by the MAB Programme.
Another new concept, the concept of frequency response analysis, has been applied to the study of magnesium cycling in a tropical forest ecosystem (see Patten, 1972).
This concept is based on the premise that an exchange of signals such as sunlight, rainfall, etc., among components is a fundamental property of systems. A known incoming signal leads to output responses which define the natural frequency and damping characteristics of the particular system.
These in turn, can lead to a definition of the properties of the systems especially its resistance or resilience to perturbations. Frequency response analysis can, therefore, serve as a valuable tool for the study of ecosystem stability, resistance or resilience.
Systems analysis techniques have also been applied for the study of food- webs and species niches (see Patten, 1972). A food-web includes the pathways of matter and energy within a community and systems analysis helps determine certain mathematical properties of a food-web e.g., the maximum number of alternative food chains. A niche is defined as a volume in a multidimensional space of environmental variables. Such a definition enables one to infer the important attributes of the niche, such as dimensions, overlap, and relation to species density.