Simulation of Vegetable Populations Dynamics Based on Cellular Automata

Abstract

Modeling the dynamics of vegetable populations is an extremely challenging problem. The evolution of a vegetable population, that is, of all the weeds, plants and trees that grow in a given area, is mainly influenced by the resources available on the territory (i.e. sunlight, water, substances present in the soil), and how the single individuals compete for them. Traditional models for this case study are continuous and based on differential equations. However, most of the data needed to provide reliable parameters for these models are usually scarce and difficult to obtain. The model we present is instead based on two-dimensional Cellular Automata, whose cells, arranged on a square grid, represent portions of a given area. Some resources are present on the area, divided among the cells. A cell can host a tree, represented in the model by a set of parameters defining its species, its size (that is, the size of its parts such as limbs, trunk, and roots), the amount of each resource it needs to survive, to grow, and/or reproduce itself (that is, produce fruits). The model has been applied to the simulation of populations consisting of robiniae (black locust), oak, and pine trees on the foothills of the italian alps, with encouraging results reproducing real conditions.