Question 2. Where are we positioned now?--Present status of recruitment modeling.

Question 2 Response--Answers concerning the present status of recruitment (secondary production) modeling were rather diffuse. However, the general theme in the answers was that no present model is satisfactory for making realistic secondary production estimates. One respondent summarized the current state of secondary production modeling as a few models with no unifying theme. However, others indicated that current conceptual models of the processes controlling secondary production appear to be correct, but translation of these poorly constrained concepts into mathematical models often gives unsatisfactory results. It was felt that this arises because the resources (prey fields) of the zooplankton are correctly simulated but ad hoc approaches are used to represent processes at the upper end of the food web, such as mortality and predation.

It was also felt that current models of secondary production are complex and that this complexity obscures the understanding that could be gained from the models. Several respondents recommended that secondary production models be reduced to basic elements to give simple models with few parameters. An issue that arose was how simple do the models need to be to address questions about the basic processes that underlie secondary production.

Several comments were made about processes not presently (or inadequately) included in existing secondary production models. These include: descriptions of the predator and prey fields, realistic predation mortality, the effects of prey spectrum on prey-specific removal, and treatment of spatial heterogeneity in the predator and prey fields. One respondent pointed out that existing secondary production models have not undergone rigorous testing to determine how well they work and evaluate their more general use. For this to happen, modeling approaches need to be available to the community so that observations can be more readily input to models. Also, several respondents indicated the need for models that can include fast and slow processes (e.g., multiple scales). There is also a need to study the structure of secondary production models to understand under what conditions the models are stable and to understand the effect of forcing functions.