Implementation of the GOES-16 GLM Lightning Assimilation into the NCEP/GSI System for Improved Global and Convective-Scale Assimilation and Prediction

Blending the information content from observations with the output from numerical models to obtain the best possible representation of the true state of a system (i.e. an analysis) and consequently an improved forecast, are the goals of every data assimilation methodology. With the launch of new observing systems, like the recently launched GOES-R (now GOES- 16), more and more observations are becoming available and these have a tremendous potential for advancing the operational weather forecasting enterprise. In the case of measurements of lightning activity by the Geostationary Lightning Mapper (GLM) instrument aboard the GOES-16 satellite, one of the many important applications that will stem from their use, will be related to improvements to the initialization of numerical weather prediction models via more accurate representations of some of the fields associated with storms/lightning activity and consequently, improved analyses and prediction.

However, the success of new and future observing system missions will ultimately depend on the capabilities of data assimilation systems to efficiently process their measurements. To that end, a methodology to assimilate lightning flash rate observations, in preparation for the GLM instrument, was implemented in the National Centers for Environmental Prediction /Gridpoint Statistical Interpolation (NCEP/GSI) system, following a variational approach, with the required tangent-linear, and adjoint operators, as well as, a hybrid data assimilation methodology.

To evaluate the benefit of the potential incorporation of the GLM instrument into the operational data stream at NCEP, this project has been divided in two parts: (1) the incorporation of an observation operator for lightning flash rate suitable for the current Global Data Assimilation System/GFS, accounting for the system's intrinsic coarse-resolution and simplified cloud microphysics, in which convection cannot be resolved explicitly, therefore, evaluating the impact of lightning observations on the large-scale environment and (2) in preparation for the NOAA NGGPS FV3-based Unified Modeling system, a second forward operator for lightning flash rate is being incorporated into the hybrid GSI, with subsequent testing in a non-hydrostatic, cloud-resolving model that permitted the inclusion of precipitating and non-precipitating hydrometeors as analysis control variables. Results that highlight the benefit of both lightning flash rate observation operators in global and convective-scale data assimilation and prediction will be presented.