The aim of this paper is to address computational aspects of the targeted observations problem for atmospheric chemical transport models. The fundamental question being asked is where to place the observations such that, after data assimilation, the uncertainty in the resulting state is minimized. Our approach is based on reducing the system along the subspace defined by the dominant singular vectors, and computing the locations of maximal influence on the verification area. Numerical results presented for a simulation of atmospheric pollution in East Asia in March 2001 show that the optimal location of observations depends on the pattern of the flow but is different for different chemical species. Targeted observations have been previously considered in the context of numerical weather prediction. This work is, to the best of our knowledge, the first effort to study targeted observations in the context of chemical transport modeling. The distinguishing feature of these models is the presence of stiff chemical interactions.