An Accurate Atmospheric Water Vapour Estimation based on Regression Model
Saqellari Likoka, Arlinda
National Technical University of Athens, GREECE

The interferometric phase is analyzed in many components such as the phase delay due to the local topography, atmospheric phase delay, surface deformation component due to minor shifts, and noise term mainly due to the unwrapping procedure. The atmospheric path delay component is caused by ionospheric and tropospheric effects. Although the impact of the ionosphere is usually limited, tropospheric disturbance can significantly affect the interferometric phase, producing a phase delay. Thus, the total water vapor column is associated with the atmospheric phase delay, which can be estimated through the proposed mathematics approach. The objective is the parametrization of water vapour estimation errors in terms of coherence values will be accomplished through regression analysis. Regression analysis will further allow the prediction of the error, in case that in situ data are not available. Introduction of a metric relative to matrices conditional number will make the prediction insensible from inherent mathematical errors. This approach based on mathematical theory will open the way observation and accurate assessment of atmospheric phase screens contributions in interferometry process.