Mechanistic Approaches to Community Ecology: A New Reductionism

Mechanistic approaches to community ecology are those whichemploy individual— ecological concepts—those ofbehavioral ecology, physiological ecology, and ecomorphology—as theoretical bases for understanding community patterns. Suchapproaches, which began explicitly about a decade ago, are justnow coming into prominence. They stand in contrast to more traditionalapproaches, such as MacArthur and Levins (1967),which interpretcommunity ecology almost strictly in terms of "megaparameters.". Mechanistic approaches can be divided into those which use populationdynamics as a major component of the theory and those whichdo not; examples of the two are about equally common. The firstapproach sacrifices a highly detailed representation of individual—ecological processes; the second sacrifices an explicit representationof the abundance and persistence of populations. Three subdisciplines of ecology—individual, populationand community ecology—form a "perfect" hierarchy in Beckner's(1974) sense. Two other subdisciplines—ecosystem ecologyand evolutionary ecology—lie somewhat laterally to thishierarchy. The modelling of community phenomena using sets ofpopulation-dynamical equations is argued as an attempt at explanationvia the reduction of community to population ecology. Much ofthe debate involving Florida State ecologists is over whetheror not such a relationship is additive (or conjunctive), a verystrong form of reduction. I argue that reduction of communityto individual ecology is plausible via a reduction of populationecology to individual ecology. Approaches that derive the population-dynamicalequations used in population and community ecology from individual-ecologicalconsiderations, and which provide a decomposition of megaparametersinto behavioral and physiological parameters, are cited as illustratinghow the reduction might be done. I argue that "sufficient parameters"generally will not enhance theoretical understanding in communityecology. A major advantage of the mechanistic approach is that variationin population and community patterns can be understood as variationin individual-ecological conditions. In addition to enrichingthe theory, this allows the best functional form to be chosenfor modeling higher-level phenomena, where "best" is definedas biologically most appropriate rather than mathematicallymost convenient. Disadvantages of the mechanistic approach arethat it may portend an overly complex, massive and special theory,and that it naturally tends to avoid many-species phenomenasuch as indirect effects. The paper ends with a scenario fora mechanistic-ecological utopia.     

Time Budgets and Territory Size: Some Simultaneous Optimization Models for Energy Maximizers

A set of optimization models in two variables of choice, territorysize and time spent patrolling for intruders, is presented forenergy maximizers. Models vary in the curvilinearity of therelationship between territory circumference and both intrusionrate and cost of expelling a single intruder. Models are analyzedboth with and without constraints; constraints are on processingrate and on the time spent patrolling, feeding and activelydefending. The models all include the concept of "intruder equilibrium,"an equilibrial density of intruders in a territory resultingfrom a balance between intrusion rate and expulsion by the defender.This equilibrial density can be considered a measure of territorialexclusiveness. The two-variable models predict effects on territory size andpatrol time of variation in food density, intrusion rate, costsof expelling a single intruder in energy and time, food-consumptionrate of an intruder, area of detection while patrolling, totaltime available for territorial and feeding activities, timeto eat a unit of food energy, energy cost of patrol per time,and processing-rate capacity. With increasing intruder rate,optimal territory size usually decreases, whereas optimal patroltime behaves much more irregularly. With increasing food density,optimal patrol time usually decreases, whereas optimal territorysize behaves irregularly. In particular, when intrusion rateand expulsion costs accelerate sufficiently with increasingterritory size and no constraints exist, the higher the fooddensity the smaller the optimal territory size. When food densityis large enough for a constraint to be effective, the oppositerelation can hold and will always hold for a processing constraint. When a particular parameter changes, optimal territory sizeand optimal patrol time may covary or one may increase whilethe other decreases, depending on the parameter and model. A new set of one-variable models is suggested by the two-variablemodels; models optimizing patrol time while holding territorysize constant could correspond to a tightly packed system ofterritories initially determined by settlement patterns. A unifiedonevariable analysis suggests that how food density affectsterritory size when patrol time is constant depends upon whethera constraint is operating: Provided that invasion rate doesnot vary with density of intruders on the territory, time minimizersand constrained energy maximizers decrease territory size withincreasing food density; unconstrained energy maximizers dothe opposite. The addition of a second optimization variable to a one-variablemodel can change qualitative predictions about variation inparticular parameters (e.g., food density) and can increasethe number of parameters predicted to affect optimal territorysize and patrol time.