A Principal Component Analysis Noise Filter VAP to Remove Uncorrelated Noise from AERI Observations
| Turner, David | Pacific Northwest National Laboratory |
| Lo, Chaomei | Battelle, PNNL |
| Knuteson, Robert | University Of Wisconsin |
The temporal sampling stategy of the Atmospheric Emitted Radiance Interferometer (AERI) was chosen to optimize the requirements for clear-sky radiative transfer studies and profiling atmospheric water vapor and temperature. This resulted in a 3-min average of sky radiance every 8 minutes. However, this sampling strategy is inadequate for the study of cloud microphysical properties, as clouds can advect into/out of the instrument's field-of-view during the sky averaging period, resulting in a combination of clear and cloudy contributions to the observed radiance. The ARM program is in the process of improving the temporal resolution of the AERI to 10-30 s. However, the decrease in the averaging time for each scene results in a larger component of random noise in the observed spectra. There is high correlation in the observed radiance across the spectrum, and therefore a principal component analysis (PCA) can be used to separate and remove uncorrelated noise from the spectra. We have developed a new Value Added Procedure (VAP) to routinely apply this PCA noise filter. The VAP automatically determines the appropriate number of principal components to use in the reconstruction to eliminate as much random noise as possible. An analysis of several months of data from the AERI at the North Slope of Alaska will be presented to demonstrate the magnitude of the noise reduction for different spectral regions in different systems. We will also present an analysis of the "residual" that was removed from the data to demonstrate that it appears to be random error rather than atmospheric signal.
This poster will be displayed at the ARM Science Team Meeting.