US Fish & Wildlife Service


Date of this Version



Pages 113-131 in R. G. Stahl, Jr., R. A. Bachman, A. L. Barton, J. R. Clark, P. L. deFur, S. J. Ellis, C. A. Pittinger, M. W. Slimak, and R. S. Wentsel, eds. Risk management: ecological risk-based decision-making. SETAC Press, Pensacola, FL.



The harvest of renewable natural resources is predicated on the theory of density-dependent population growth (Hilborn et al. 1995). This theory predicts a negative relationship between the intrinsic rate of population growth and population density (i.e., number of individuals per unit of limiting resource) due to intraspecific competition for resources. In a relatively stable environment, unharvested populations tend to settle around an equilibrium where births balance deaths. Populations respond to harvest losses by increasing reproductive output or through decreased natural mortality because more resources are available per individual. Population size eventually settles around a new equilibrium and the harvest, if not too heavy, can be sustained without destroying the breeding stock. Resource managers typically attempt to maximize the sustainable harvest by driving population density to a level that maximizes the intrinsic rate of population growth (Beddington and May 1977).

Although the theoretical basis for harvesting renewable resources is fairly straightforward, the practice of harvest management has had its share of difficulties. History is replete with cases where uncontrolled variation in harvests or the environment, naive assumptions about system response, and management policies with short time horizons have led to resource collapse (Ludwig et al. 1993). To be successful, sustainable harvesting depends on an ability to effectively regulate the size of the harvest, on a sound understanding of the biological system and its density-dependent responses, and on management objectives that are congruent with the renewal capacity of the resource. Even with a firm commitment to long-term resource conservation, harvest managers always will be burdened by complex, dynamic systems that are only partially observable, and by management controls that are indirect and limited. It is for these reasons that a coherent framework for managing ecological risk is necessary.

Harvest management decisions involve three fundamental components: (1) unambiguous objectives; (2) a set of alternative harvest actions, including any constraints on those actions; and (3) the predicted consequences of those actions in terms that are relevant to the stated management objectives. The consequences of harvest actions cannot be predicted with certainty, and the associated risk is what makes management decisions difficult. I define risk as the probability of a management outcome, where the probability can be assessed reliably from past experience with the resource or with a similar biological system. Thus, risk differs from true uncertainty, in which past experience provides no guide for the future (Costanza and Cornwell 1992). In keeping with the definitions in this book, ecological risk assessment involves associating empirical probabilities of possible system responses with alternative management actions. Ecological risk management then is the process of using management objectives to value those (probabilistic) responses so that a preferred management action can be identified.