Turbulence in the marine atmospheric surface layer can induce substantial perturbations in propagating infrared (IR) or optical signals. These perturbations result in scintillation, which is a fluctuation in received signal intensity. Scintillation can be quantified by the single parameter C_n², the refractive index structure parameter. It is necessary to measure C_n² from routine measurements in order to evaluate, predict, and compare EO system performance in operational environments and optimize EO systems for expected C_n² values from climatological measurements and models.

The Rough Evaporation Duct (RED) field test, conducted at Oahu, Hawaii, during August-September 2001, presented an excellent opportunity to quantify scintillation and to validate scintillation and transmission models. The infrared portion of the RED experiment was conducted on a propagation path approximately 10 km long connecting an IR broadbeam transmitter onboard R/V FLIP (13 m above waterline) and an IR telescope receiver at Malaekahana (3 m above ground) on the northeast coast of Oahu. The broadbeam source accommodated the pitch, roll, yaw, and translation of FLIP. The scintillation measurements were taken in the mid-wave (3.6 m m to 4.1 m m) every 15 minutes and at a 300 Hz sampling rate. Each measurement was 110 seconds long. Due to FLIP's tidal motion (± 200 m in the NE-SW direction), hourly telescope alignments were necessary to ensure proper signal reception. The data collection process occurred almost uninterruptedly for 16 continuous days. Concurrent environmental measurements were obtained from a buoy located at the mid-point of the propagation path and from a meteorological station mounted on top of a 30-ft mast outside the receiver station.

There are two components of the infrared signal analysis. First we will utilize the abundant meteorological and surface flux measurements made during the test. The data from the mid-path meteorological buoy will be an essential part of this effort. The second component is a signal processing effort. The signal levels recorded from the field test are unexpectedly weak and it is meaningless to perform C_n² calculations using raw, unfiltered data. An approach to separate signal from noise without destroying the turbulent nature of infrared signal is necessary, and we will describe our use of a wavelet filtering scheme to accomplish the task.

We will comment on the progress toward the ultimate goals for this test. First, what is the influence of ocean waves on the near-surface propagation environment? The second goal is a determination of the accuracy of current C_n² estimation models for various ocean-atmosphere interaction scenarios including stable, near-neutral, and unstable conditions, and possible upgrade paths for the models.