October 2003 solar proton events (SPE) is rather well covered by the observations; therefore its studies represent a good way for model validation and intercomparison. Here we apply chemistry–climate model (CCM) SOCOL with complete (SOCOL i ) and parameterized ion chemistry to evaluate the accuracy of a commonly used ion chemistry parameterization scheme. We performed ensemble experiments with and without SPE to characterize the effect of the October 2003 SPE on the NO x , HO x , ClO x and O 3 in the middle atmosphere. Preliminary comparison of the simulated effects against MIPAS observations revealed rather good general agreement for most of the species. Comparison of the results obtained with complete and parameterized ion chemistry representation showed that the model with parameterized ion chemistry underestimates the effect of SPE on chemical composition of the middle atmosphere by up to 40% for NO x and N 2 O, up to 70% for HO x and ClO x and up to 600% for HNO 3 . The parameterization is more accurate for ozone, however the model with parameterized ion chemistry underestimates ozone depletion by up to 15% during the SPE in the mesosphere and by 10% 2 weeks later in the stratosphere, which can be important for the long-term effects of SPE on the ozone layer.