The influence of external colored noise on the carbon monoxide oxidation on Iridium(111) surfaces is examined. The noise is introduced in the reaction by randomly varying the composition of the gas flow that keeps the reaction going on. Colored noise is studied using two models: a simple discrete time Markov chain, and the Ornstein-Uhlenbeck process. We compute the probability distribution and transition times, for medium and large correlation time of the noise. These results extend previous analyses that have been limited to small correlation times and the presence of a slow manifold, both assumptions that are not supported by experiments. As we will see, the correlation and intensity of the noise leads to qualitative changes in the stochastic behavior of the system.