With development of new airborne optical systems, there is a significant amount of effort being put into maximizing the farfield performance of these optical systems. To meet this goal, the root mean square of the optical path difference, OPDrms , over the aperture, a measure of the nearfield optical distortions, is being reduced as much as possible. Part of this development process involves testing models of these systems with various configurations in wind tunnels. In these tests, the optical disturbances due to the testing environment are becoming a large percentage of the measured optical disturbances. It is now a necessity to be able to identify, measure, and remove the various noise sources that appear in aero-optical measurements that take place in wind tunnels.One of the primary sources of signal noise in these measurements, is acoustics. Both the wind-tunnel fan and the flow through out the tunnel generate large amounts of acoustic noise that travel throughout the tunnel as acoustic duct modes. The fan generates strong narrow-band signals at the blade-passing frequency and its harmonics. There are also significant broad-band acoustic duct mode signals along with vibration related signals and strong mostly-steady optical lensing signals that add to the overall measurement.A significant portion of this dissertation is dedicated to analyzing wind-tunnel optical wavefront data in multidimensional spectral space. The signal identification and filtering mostly take place in the multidimensional spectrum, where the optical wavefront is not only split into its temporal frequency components but also its spatial frequency components. A combination of three filters is able to remove most of the noise signals while retaining most of the aero-optical signal. A velocity filter, which retains a narrow band of signal that is traveling within a small velocity range, is able to remove most of the broad-band signal noise, especially at higher temporal-frequencies. A backward filter, which retains only the portion of a signal that is traveling in the direction of the flow, removes signal that the velocity filter is not able to, at lower temporal-frequencies. Finally a baseline filter identifies the baseline spectrum and removes narrow-band peaks.