HRD Seminar – Dr. Karina Apodaca, CSU/CIRA and AOML/HRD – 24 October 2018

Dr. Apodaca presented a seminar titled “Advances in new data assimilation methodologies and observations for the benefit of NOAA operations”.

In an effort to advance the operational weather forecasting enterprise a vital objective is having the ability to accurately predict severe weather. Integrating new methodologies to enhance data assimilation systems and incorporating new and future observations not previously used in operations can help in this quest.
Current data assimilation for numerical environmental prediction at operational centers is based on Bayesian statistics and probability, which are inherently tied to the Gaussian assumption of the probability density function of the observation and background errors. In reality, the errors involved in key variables linked to the prediction of high-impact weather (clouds and precipitation) exhibit probability density function distributions that are far from being Gaussian. Due in part to historic computing limitations, calculations had to be done under such assumptions, but the fast-paced advances in computer science are motivating us to explore new methodologies for improved data assimilation and to implement them in NOAA’s operationally used data assimilation software. 
Moreover, new and future remotely sensed observations useful for improving the initial conditions of clouds can give rise to nonlinear and non-Gaussian behaviors. Two examples include observations of all-sky radiances and lightning. In particular, to evaluate the benefit of the incorporation of the GOES-16,17/Global Lightning Mapper (GLM) into the operational data stream at NOAA/NCEP, two observation operators for the assimilation of lightning flash rate were recently included in the NCEP/Gridpoint Statistical Interpolation (GSI) system by following a variational methodology. The enhanced software includes two nonlinear observation operators, subsequent linearization, and development of adjoint models for lightning flash rate.
In this presentation, the first steps towards implementing a new non-Gaussian formulation in the variational solver of the GSI system in an application to all-sky radiance assimilation at Convection Allowing Model (CAM) resolutions and an overview of the new observation operators for the GOES-16,17/GLM sensor suitable for global and CAM applications will be covered. 

A copy of the presentation is available on the anonymous ftp site: