In the measurement of surface seismic data, an accurate attenuation-estimation method is necessary to compensate for the energy loss and phase distortion of seismic waves, and also beneficial for further quantitative amplitude analyses and reservoir parameters prediction. We design a more accurate Q tomography method to estimate Q values. This method inverts for the Q values in the overburden and target layer simultaneously without assumptions on the overburden, which allows us to separate the overburden effect during the inversion. As Q tomography methods estimate Q value by assuming the source wavelet is known, the results of Q value are often affected by the errors of the calculated source wavelet. Thus, from the same source, we take the spectral ratio of two reflections from the top and bottom of a target layer as an alternative for tomography input. For the log spectral ratio (LSR) method and attenuated traveltime tomography (t* tomography), picking two reflections at different traveltimes may be affected in different ways by lateral changes in the overburden (we call it the overburden effect). Thus, the accuracy of the conventional methods may be affected by differences between the ray paths in the overburden of the two picked reflections. In our method, we address this problem using an inversion method, which allows us to separate attenuation effects from the overburden by using traveltime difference in the tomography grid cells. Q values of the target layer and that of the overburden can be inverted simultaneously. We test the method in synthetic data and prove its feasibility and effectiveness by applying it to field data.