To date, TM AFM (tapping mode or intermittent contact mode atomic force microscopy) is the most frequently applied direct imaging technique to visualize surface nanobubbles at the solid-aqueous interface. On one hand, AFM is the only profilometric technique that provides estimates of the bubbles' nanoscopic dimensions. On the other hand, the nanoscopic contact angles of surface nanobubbles estimated from their apparent dimensions that are deduced from AFM "height" images of nanobubbles differ markedly from the macrocopic water contact angles on the identical substrates. Here we show in detail how the apparent bubble height and width of surface nanobubbles on highly oriented pyrolytic graphite (HOPG) depend on the free amplitude of the cantilever oscillations and the amplitude setpoint ratio. (The role of these two AFM imaging parameters and their interdependence has not been studied so far for nanobubbles in a systematic way.) In all experiments, even with optimal scanning parameters, nanobubbles at the HOPG-water interface appeared to be smaller in the AFM images than their true size, which was estimated using a method presented herein. It was also observed that the severity of the underestimate increased with increasing bubble height and radius of curvature. The nanoscopic contact angle of >130° for nanobubbles on HOPG extrapolated to zero interaction force was only slightly overestimated and hence significantly higher than the macroscopic contact angle of water on HOPG (63 ± 2°). Thus, the widely reported contact angle discrepancy cannot be solely attributed to inappropriate AFM imaging conditions.