On the value of high-resolution weather models for atmospheric mitigation in SAR interferometry

Atmospheric delay is one of the major error sources in In-SAR, hindering the accurate monitoring of ground motion. Here we use the WRF (Weather Research and Forecasting) weather model to hindcast atmospheric delays at SAR acquisition times over both mountainous and flat regions. The performance of the model is evaluated by comparing it to interferograms formed using acquisitions with short temporal baselines (¿4 months). Our results show that for flat regions the model not only misestimates atmospheric delay in magnitude and location but also largely underestimates the (horizontal) spatial variation (turbulent mixing) of the delay. In mountainous areas it can model the height dependent (vertical stratification) part of total delay correctly in some cases but not always. By removing the height dependent part we find again that the model may underestimate the spatial variation of the delay. Therefore, we conclude that the WRF weather model is in general not reliable for the operational mitigation of atmospheric delay in interferograms.