This study presents a proposal for a hybrid model (based on agents and cellular automata) which aims to analyse the long-term effects of sea level rise and real estate market dynamics on the urban development of coastal cities in Rio Grande do Sul, a state located in south Brazil. The model's operation is based on the Complexity Theory of Cities and seeks to simulate the study area's urban growth by replicating the process of spatial allocation of residential and commercial activities and the variations in built form and territorial attributes -such as land value and attractiveness -that result from this process. To accomplish this, specific modules were developed to determine the land value of each part of the territory through the behaviour of individual agents and to compute land attractiveness metrics through the representation of the cellular automaton as a graph. The proposed model is tested in an experiment for Tramandaí and Imbé -two coastal municipalities in Rio Grande do Sul -considering different scenarios of building restrictions and sea level rise. The results show: i) the model's ability to contribute to the development of planning policies for the study area; ii) indications of its consistency in relation to theoretical statements; and iii) its limitations in reproducing the dynamics that generate diffuse urban growth patterns.