Retrieval of Moisture from GPS Slant-path Water Vapor Observations using 3DVAR and its Impact on the Prediction of Convective Initiation and Precipitation

Haixia Liu


The accurate prediction of convection initiation and associated precipitation in a cloud-resolving numerical model is highly dependent on the precise estimate of the three-dimensional moisture distribution because water vapor is directly involved in the formation of the cloud and precipitation. However, the water vapor is currently poorly characterized due to its high variation over space and time. A 3DVAR system is developed in this study to retrieve the moisture distribution from simulated ground-based GPS slant-path integrated water vapor (SWV) data which is in high quality with high temporal and spatial resolutions.

This 3DVAR system is based on a terrain-following coordinate. A non-negative water vapor weak constraint is included in the cost function. The background term and its associated background error covariance are considered in the system and modeled by spatial filters. Moreover, a direct way to estimate a flow-dependent background error covariance based on the idea of Riishøjgaard (1998) is proposed for the moisture analysis. Firstly, the explicit Gaussian-type filter is applied to this system with both isotropic and anisotropic options to perform the moisture retrieval. It is demonstrated that this system is robust on deriving mesoscale moisture structure and that the analysis is improved when the anisotropic background error covariance is considered. Secondly, the implicit recursive filter which is much more computational efficient replaces the explicit filters in the 3DVAR and similar set of water vapor retrieval experiments to those using explicit filters is performed. The analyses obtained with recursive filters are generally comparable to or better than those using corresponding explicit filters. It is feasible to systematically examine the sensitivity of the analyses to the spatial de-correlation scales of the background error using recursive filters due to their computational efficiency.

To examine the impact of GPS data on the prediction of convective initiation and subsequent evolution within an OSSE framework, a high-resolution numerical simulation is first conducted using Advanced Regional Prediction System (ARPS) for a case that occurred on 12 June, 2002 and involved multiple initiations of convection. A set of OSSEs is then performed based on this realistic numerical simulation for this case and using the 3DVAR system with recursive filters. The preliminary results suggest the storm initiation in the model without strong low-level mesoscale forcing is highly sensitive to the moisture initial condition and the use of GPS data and anisotropic background error improves the moisture analysis and thus the initiation and precipitation forecast.