Spectrometers for Earth Observation require inflight radiometric calibration, for which the sun can be used as a known reference. For wide field instruments, a diffuser is placed in front of the spectrometer, scattering incoming sun light into the entrance slit and ensuring a homogenous illumination. As drawback, the diffuser induces a specific radiometric error caused by interference, which is called Spectral Features.
The scattering of the incident light at the diffuser induces a random path difference yielding a specific interference pattern at the entrance slit, known as speckle pattern. These speckles are propagated through the disperser to the detector plane and further integrated by the detector pixels. The resulting feature can yield a significant signal error contribution, whose spectral variation is referred to as spectral features. The magnitude of this error is evaluated in terms of the Spectral Features Amplitude (SFA), the ratio of the signal standard deviation with its mean value over a specific wavelength range.
There have been several ways implemented to measure the SFA of a spectrometer, e.g. end-to-end measurements with representative instruments. Typically the measurement accuracy is not sufficient to isolate the SFA from other radiometric errors. As a consequence, the instrument layout can hardly be optimized to suppress Spectral Features.
We propose a novel characterization technique for Spectral Features based on the direct acquisition of monochromatic speckle patterns at the entrance slit. This allows the observation of Spectral Features below the level of the spectrometer spectral and spatial resolution. The Spectral Features are derived from various observed speckle patterns by properly mimicking the real spectrometer in the data analysis. With this measurement technique we are able to gain insight into the mechanism behind speckle induced Spectral Features. This insight will be used to develop a parameterized model facilitating the design of future space based spectrometers.