Analysis of Martian Spectral Modeling by Spot-Spectra Normalization

Mark Pultrone


We used a method called principal component analysis (PCA) to try to find meaningful surface spectral models in our Martian spectral image data that were uncalibratable due to the absence of a comparison star. The data are ground-based, near-infrared spectra images taken during the 19xx opposition. In an attempt to get something useful out of them, we normalized the uncalibratable data in three different ways: disc mean, disc median, and spot-spectra. Our research has found that the method that produced the best results was spot-spectra normalization. Building off these findings, we decided to analyze this method further.

Using only the spot-spectra normalized data, we performed four different PCA tests: using the full 1.5–4.1 μm wavelength spectrum and the full image; using the full spectrum and disregarding the Martian poles; omitting the 1.9–2.2 µm wavelengths from the spectrum but using the full image; and omitting the 1.9–2.2 μm wavelengths from the spectrum and disregarding the Martian poles. The 2 µm region omission tests were performed because that spectral region is dominated by the non-linear absorption features of the atmospheric gas, making the linear PCA modeling inaccurate. The polar region omission tests were performed because they contain large amounts of surface ices, which could affect the creation of a “standard surface’’ model spectrum.

We will present here our analysis of these tests and our assessment of which should be used in the next phase of the project—creation of a surface model in order to measure the water ice abundance in the Martian clouds.


Mars; Atmosphere; PCA

Full Text: PDF


  • There are currently no refbacks.