Methoden der Populationsökologie

Abstract

Population ecology describes and analyses the distribution and abundance of individuals in space and time, including age structure, sex distribution, and genotype frequencies. Different methods and expected technical and theoretical developments are described using examples from plankton ecology; their assumptions, difficulties and implications in regard to the theory of scientific knowledge are discussed. Three main fields of population ecology are methodologically well defined: 1. Field description and analysis: There are new trends in automatization of sampling up to continuously recording (e.g. by Coulter-counter), and even automatically species determination might be possible (by holography). Representative samples of hetero-geneously distributed populations can be taken by integrating along a transect (e.g. by Clarke-Bumpus-sampler). The real distribution of the single specimens can be recorded by the echo sounder (Fig. 2). There are now also methods which allow the measurement of population parameters like natality and mortality (Figs. 4 and 5) as well as physiological characters like filtration and respiration rates under field conditions. Field analysis can be performed by changing single abiotic and biotic factors and watching the effects (Fig. 6) or by statistical analysis of field data (e.g. multiple and partial correlation and regression analysis). The difficulty of interpreting correlations is discussed by the example of two different association coefficients in a well known situation of complex interspecific relationships in three rotifers (Fig. 7). 2. Laboratory investigations: The culture of organisms under defined conditions allows the variation of single factors and the study of their effects on life table data (Fig. 8) and population dynamics (Fig. 9). The use and significance of different types of culture methods (e.g. chemostate, recycling cultures, cultures with periodically renewed medium) are demonstrated (Figs. 10–12). The causal relationships between ecological factors and population dynamics can be analysed stepwise. Methods are now available for experimental examination of filtration, assimilation and respiration rates of single specimens of planktonic organisms (e.g. with the radio carbon method or with Cartesian divers). 3. Models: In some cases (e.g. in rotifers) we are able to construct the population dynamics based on physiological and demographic data measured in isolated individuals, and compare these with empirical population curves (Figs. 13–15). The simulations are based on deterministic models (e.g. Lotka-Volterra-equations with simple time lag) using numerical computer techniques. To get more realistic models in future, we will have to use stochastic models to a larger extent and we will have to take into account the spatial heterogeneity of the population (Figs. 16–18), because this widely disregarded aspect is of outstanding importance for the stability and evolution of populations. For the formal description of spatial and temporal heterogeneity two possible ways are proposed: 1. The use of mathematics for describing gradients in a five-dimensional system of co-ordinates (3 dimensions for space, 1 for time, and 1 for population density). 2. A digital information theory has to be developed, in which each specimen is regarded as a unit in space and time, interacting with all other specimens of the population. The connections between the individuals are composed of a variety of relationships (e.g. positive or negative attractions). The intensity of each relation is, among other dependences, a function of the distance between the individuals (Figs. 19 and 20). Such a model could describe individual distance patterns in territorial as well as in heterogeneously distributed populations. Age differences, bisexuality, and genetic polymorphism can be included in this model as well as ethologic characters and social behavior (like social stress), and their influences on mortality, natality, and migration. Such time-space-population models might turn out to be more realistic and precise than most contemporary models. The basic requirements for the development of such population model systems of high complexity are fulfilled by the existence of macro-computers.