Charles J. Puccia

Models for New and Resurgent Diseases

many variables at the same time and these may themselves be very different kinds and even belong to different biological and social disciplines. Models are constructs which we make to study directly instead of the objects of real interest. Making a model requires making theoretical and practical choices in which tradeoffs are necessary among goals such as achieving generality, realism, precision, manageability, cost, and intelligibility. Models can be divided into two categories: tactical models and strategic models. In well-established disciplines tactical models (sensu May, 1983. Stabzlity and ComplRlcity of Mo&l Emsystm, Princeton University Press) are widespread. These rely on specific information, detailed knowledge, accurate measurements and aim to give precise results. For example, a tactical model of striped bass population size would give a lot of detail to the precise growth and reproductive parameters of striped bass in order to help the fisheries managers determine the level of fishing activity consistent with avoiding overfishing. New disciplines can benefit by employing strategic models, even though they are also associated with established disciplines that have accumulated knowledge over long periods. Strategic models are characterized by their wide applicability, describe qualitative behaviors, and yield broad generalizations; they depend less on precise assumptions and data. They are often of a qualitative nature and reach qualitative conclusions. For example, even the sim le verbal model that pro ses that the prevalence of a disease is the result

Models and Field Data of New England Tide Pools1

Abstract Qualitative (signed digraph) or loop models were constructed for rocky shore tide pools. A discussion of the use of loop models for field research is presented. A comparison of pairwise correlation coefficients of our data with predicted correlations of the models is made. These suggest environmental changes in parameters effecting growth rates of populations is primarily through the herbivorous snail. Preliminary experimental manipulation of snails also supports this conclusion.

Qualitative Models for the East Coast Benthos1

Abstract Soft-bottom estuarine environments for the East Coast are analyzed using qualitative models (loop models; signed digraphs). The benthic organisms are divided into functional groups that interact trophically or through sediment mediation. The functional groups are mobile burrowers, filter-feeders, tube-builders, and predators. Alternative models are presented based on accepted hypotheses and some speculation. The physical substrate enters into the models through the biology of the organisms. The tube-builder/mobile burrower subsytem stability becomes an object of interest in these models. The scale of the system being modelled can alter self-damping of the organisms and change results. Finally, the trophic-amensalism hypothesis may predict the correct results but not because filter-feeders are aided directly through sediment stabilization of tube-builders but by limiting mobile burrower activity.

Statistical Patterns of Non-Equilibrium Systems1

Abstract Most applied models assume the system under study is in equilibrium or moving-equilibrium. Recent theoretical work has considered chaotic behavior or sustained-bounded-motion (SBM) of systems. We show how to determine the necessary conditions for SBM. We establish, in some cases, the patterns of covariation or the bounds on variance. It is possible to use the data to identify sources of external perturbation to a SBM system. Alternative measures of stability, resiliency and rates of change of systems are provided. Theoretical models help explain variable time-series data collected for environmental assessment or monitoring.