Wavefront measurements from wind tunnel or flight testing of an optical system are affected by jitter sources due to the measurement platform, system vibrations, or aero-mechanical buffeting. Depending on the nature of the testing, the wavefront jitter will be a composite of several effects, one of which is the aero-optical jitter; i.e., the wavefront tilt due to random air density fluctuations. To isolate the aero-optical jitter component from recent testing, we have developed an estimation technique that uses only higher-order wavefront measurements to determine the jitter. By analogy with work done previously with free-stream turbulence, we have developed a minimum mean-square error estimator using higher-order wavefront modes to compute the current-frame tilt components through a linear operation. The estimator is determined from computational fluid dynamics evaluation of aero-optical disturbances, but does not depend on the strength of such disturbances. Applying this technique to turret flight test data, we found aero-optical jitter to be 7.7±0.8 μrad and to scale with (ρ/ρ SL )M 2 (∼1 μrad in the actual test cases examined). The half-power point of the aero-optical jitter variance was found to be ∼2u ∞ /D t and to roll off in temporal frequency with a power law between f −11/3 and f −10/3 .