Assimilating Cloud Observations in the High Resolution Rapid Refresh Data Assimilation System (HRRRDAS)

Successful data assimilation is of paramount importance for initializing numerical models, yet assimilating cloud observations can be particularly challenging. NOAA’s Global Systems Division (GSD) developed an ensemble Data Assimilation System to initialize the High Resolution Rapid Refresh (HRRR) version 4, known as HRRRDAS, that includes cloud assimilation. The HRRRDAS system with HRRRv4 is in the process of being transitioned to operations. Two areas of cloud assimilation with these systems will be presented.

Cloud observation assimilation development for the HRRRDAS has been underway to improve low-level cloud retention and cloud prediction overall. To facilitate these needed improvements, a new observation operator for cloud base observations was developed to translate cloud information into water vapor space. HRRRDAS experiments with the new operator leveraged in the Ensemble Kalman Filter framework within GSI show positive results. Single analysis and retrospective case study experiments examining the impact of cloud assimilation will be presented. The sensitivity to observation errors and the balance between cloud and moisture observation types will be discussed. Qualitative and quantitative comparisons with satellite observations and near surface conditions will be assessed.

Additional investigation of cloud data assimilation includes the interaction with the model subgrid cloud fraction. The existing cloud assimilation techniques focus on explicit clouds, which restricts the use of partial cloud observations (e.g. scattered clouds). Given the physical parameterization development to account for cloudiness within a grid cell, new analysis experiments are underway to update model cloud fraction data with cloud observations. Case analysis examples will be presented to highlight the development in this area.