In recent decades, thanks to the increase in the computational capabilities of our computers, the fields of ecology and forestry have seen the appearance of new tools within them. Among these are spatially explicit models, in which processes and represented entities are placed in space. These models, now widely used and developed in number, then make it possible to simulate different scenarios to better understand or better predict the evolution of a studied system.
In order to simplify the use of these models and the calculations they generate, the spatial dimension of the model is generally “discretized”. By this act, the space is then divided into more or less small cells depending on the resolution used. The geometric shape most used by modelers to cut out space is that of the square, which makes it possible to create regular grids, in addition to other advantages. This choice of geometry to discretize space can however be questioned, while it has been shown that it could have strong impacts on the results of a model.
We then decided to focus on the impact of this choice on the functioning of spatially explicit models in forest ecology. For this, we will reproduce the operation of several existing models well known in the field, and which represent an important process for forest ecosystems (e.g. natural disturbances). We will then run these models on different grids (e.g. square, hexagonal, irregular, etc.) to observe the effects of these on the results obtained. Results obtained from preliminary work suggest that such effects may indeed be important, but not for all types of processes.