Piscator, an individual based fish model

Time schedule: 1998-2000
Researcher(s): E.H.(Egbert) van Nes
Other people involved: M.(Marten) Scheffer and E.H.R.R.(Eddy) Lammens (RIZA)
Conducted jointly with: RIZA
Funded by: RIZA

The composition of fish communities is determined by recruitment, growth and mortality of individuals. The variation in each of these processes and the number of species involved creates an enormous amount of possible fish communities. As we have only a limited knowledge of the causes of variation in these processes, models are still poor tools for prediction. However, models are much more useful as instruments for understanding these processes and for determining how much variation each of these processes cause.

Piscator is an individual based model. A crucial difference with the traditional lumped population models is that the populations are really composed of individuals with their own specific parameters regarding feeding (in particular, size dependent predation) and growing, whereas the model result variables (individual numbers, size distributions, diets) are similar to the really observed values. This implies that we have better information to feed the model but also much better possibilities to check the model.

Piscator is a multi-species model with at default eight interacting fish species, fishery (fykes, seine and gill nets), piscivorous birds and a simple representation of the fish food (zooplankton, benthos), but you can add as many user-defined species as necessary. Especially, the feeding is modelled in a detailed way. Being a complex model, we need to be able to control the complexity of Piscator. Therefore we have:

incorporated the flexibility to zoom in or out on different processes by removing species without changing the computer code.
implemented an option to shut down feedback mechanisms ('Pandora's box').
simplified or ignored insufficiently known processes (e.g., first life stage, number of eggs, migration patterns, etc.).

Piscator has been found useful to quantify and explain qualitative hypotheses of fish biologists about fish dynamics in several Dutch lakes (Frisian lakes (Van Nes et al. in press. and Lammens (in press) ), IJsselmeer (Lammens, 1999a ), Lake Veluwe (Lammens et al., in press; Lammens et al., in prep. ) and Lake Volkerak (Lammens et al., in press ). These studies have helped, for instance to estimate the effect of cormorants on the fishery in IJsselmeer (Lammens, 1999a) and to substantiate the idea that natural growth of the bream population (without immigration) could explain the growth of the bream population in the recently created Lake Volkerak.