Aquatic plant growth is a complex phenomenon, which involves physical, chemical, ecological and biological processes. Modelling such complex phenomena is rather difficult and hardly accurate because of the limited availability of adequate spatial and temporal measurements for water quality and ecological data. It is shown here that time series of high resolution photographs are one option for monitoring and simulating surface spatial pattern development. Phenomena on a fine-scale tend to have high spatial neighbourhood interactions and local environmental conditions tend to have strong effects on species abundance, e.g., in the case of macrophytes growth. Cellular Automata (CA) models are capable of dealing with spatial variations and local interactions. They have become increasingly popular in ecosystem dynamics modelling since simple local rules can lead to globally complex patterns. In this study a cellular automata (CA) based model is developed by combining time series of high resolution photographs with meteorological data and biological knowledge in order to simulate macrophytes growth. The applicability of the model under different cell sizes and neighbourhood schemes is analyzed. A case study is carried out to test the CA model on a small pond which shows quite reasonable two dimensionally spatial pattern dynamics. This study indicates that Cellular Automata are quite capable of using local interactions to capture the biological characteristics of aquatic plant growth systems.