The model is forced by tidal height at the sea boundary, light intensity, air temperature, wind speed, cloud cover and boundary conditions for some of the state variables. The time step used is 3 seconds and the variables simu- lated are: water temperature, current speed and direc- tion, dissolved substances, suspended matter and phyto- plankton concentrations and bivalve species biomass. Figure | - Ria Formosa (South of Portugal). Part of the model grid is shown at the upper left corner (resolution of 100m) Different classes simulate different variables and proc- esses, with proper parameter and process equations. Classes can be selected or deselected from shell dialogs determining its inclusion or exclusion in each simula- tion run of the model. Figure 2 — Agent-based Simulation System Architecture Details concerning the calibration procedure followed by the CA are described in (Pereira et al. 2004). Its main algorithm is depicted in Figure 3. The region area is the union of one or more simple regions, each one of these defined by a point or a basic polygon (rectangle, square, circle or arc): The <action_1p> field identifies the <1p> of the action message. Figure 5 — Farmer Agent seeding and inspecting regions
