Modelling simulations are used to identify the most efficient emission reduction strategies to reduce PM_2.5 concentration levels in northern Italy. Results show contrasting chemical regimes and important non-linearities during wintertime, with the striking result that PM_2.5 levels may increase when NO_x reductions are applied in NO_x-rich areas – a process that may have contributed to the absence of significant PM_2.5 decrease during the COVID-19 lockdowns in many European cities.