Abnormal decrease in the ozonesonde sensor signal occurred during air-pollution study campaigns in November 2011 and March 2012 in Mexico City. Sharp drops around 5 km a.s.l. and above were observed in November 2011, and a broad deficit in the convective boundary layer in March 2012. Various circumstantial evidence indicates that the decrease was due to interference of SO<sub>2</sub> gas to Electrochemical Concentration Cell (ECC) ozone sensors. The sharp drops in November 2011 are considered to be caused by the SO<sub>2</sub> plume from the Popocatépetl volcano to the south-east of Mexico City. Response experiments of the ECC sensor to representative atmospheric trace gases showed that only SO<sub>2</sub> could generate the observed abrupt drops. The vertical structure of the plume reproduced by a Lagrangian particle diffusion simulation also supported this assumption. The near-ground deficit in March 2012 is considered to be generated by the SO<sub>2</sub> plume from the Tula industrial complex to the north-west of Mexico City. Sporadic large SO<sub>2</sub> emission is known to occur from this region, and before and at the ozonesonde launching time, large intermittent peaks of SO<sub>2</sub> concentration were recorded at the ground-level monitoring stations. The difference between the O<sub>3</sub> concentration obtained by ozonesonde and that by UV-based O<sub>3</sub> monitor was consistent with the SO<sub>2</sub> concentration measured by a UV-based monitor on the ground. The plume vertical profiles estimated by the Lagrangian particle diffusion simulation agreed fairly well with the observed profile. Statistical analysis of the wind field in Mexico City revealed that the Popocatépetl effect is most likely to occur from June to October, and the Tula effect all the year.