GNSS-RO Refractivity Bias Correction Under Ducting Layer Using Surface-Reflection Signal

Abstract

Due to its high vertical resolution and cloud-penetrating capability, GNSS-Radio Occultation (RO) remote sensing technique has been utilized to observe the vertical structure of the Planetary Boundary Layer (PBL) in recent years. However, the critical refraction, or ducting, caused by large refractivity gradients usually associated with the top of the stratocumulus clouds, can negatively bias the retrieved refractivity and humidity within the PBL. Previous research has shown that combining RO retrievals and the external information, such as collocated precipitable water (PW) estimates, can effectively reduce the negative bias and enhance the retrieval quality. Nevertheless, the requirement of collocated observations from other techniques limits the applicability of this reconstruction method in practice. Here, we describe an alternative approach that uses the coherent grazing signals from the same RO event that are reflected by the Earth’s surface to remove the bias due to ducting. Additional observations are no longer necessary in this approach because the reflected signals provide the extra constraint. A least squares framework is used to select the candidate from a family of solutions wherein reflected bending angle best matches the corresponding observation. This new method was validated by both multiple phase screen (MPS) simulation and the simulated RO bending angle via forward Abel transform, and it was tested with the actual GPS-RO measurements. While, in general, the reflected bending retrieved from the current mission was noisy, the results show that this approach can potentially reduce the negative bias and improve RO observation within the PBL without assistance by the external information, such as PW.