Understanding the mechanisms and patterns of earthquake occurrence is of crucial importance for assessing and mitigating the seismic risk. In this work we make an initial analysis of the viability of Evolutionary Computation as a means of generating models for the occurrence of earthquakes. We use the Japan Meteorological Agency (JMA) earthquake catalog in our analysis. Our proposal is made in the context of the "Collaboratory for the Study of Earthquake Predictability" (CSEP), an international effort to standardize the study and testing of earthquake forecasting models. As a starting point, we propose a methodology for encoding earthquake risk models in a Genetic Algorithm as a real valued genome, where each allele corresponds to a bin in the forecast model. A fitness function based on the log-likelihood of the evolved model against the earthquake occurrence data is used. Because of the high dimensionality of the chromosome, special attention has to be paid to crossover, mutation and validation operators. A careful discussion of these factors is conducted, and the results of our experiments are interpreted using a simple geophysical model.