The WRF model (Michalakes et al., 2001) is a next-generation mesoscale atmospheric/chemistry model and data assimilation system model developed at National Center for Atmospheric Research (NCAR) and grass-roots collaboration with several institution and universities. While succeeding a number of physics scheme from the previous-generation Mesoscale Modeling 5 (MM5), the dynamic core of the numerics in the WRF has been re-invented to conduct high-resolution regional weather simulations. The WRF has been used in NASA for various applications such as storm simulations, operational weather forecasting model, land-atmosphere interactions, and regional aerosol transport by coupling various physics packages. Recently, different development efforts and groups in NASA have been integrated to create the NASA-Unified WRF (NU-WRF) within the NASA Modeling And Prediction (MAP) program. The NU-WRF feature
Physics modules: Goddard one-moment bulk microphysics, Goddard shortwave/longwave broadband radiation, GOddard Chemistry Aerosols Radiation Transport model (GOCART), and those physics coupling modules.
Externally coupled Modeling Systems to support observation-driven model framework: GSFC Land Information System (LIS) that includes various land-surface models, physical boundary conditions (land covers, greenness, soil types, etc…), data assimilation, and model evaluations. The Goddard Satellite Data Simulator Unit (G-SDSU) that predict satellite instrumental-level signals from the geophysical parameter.
Lateral and initial boundary conditions: In addition to the traditional global and regional models (e.g., ERA, NCEP reanalysis), the NU-WRF can be driven by the NASA GEOS-5 global forecasting simulations or the MERRA reanalysis, and the global GOCART simulations for atmospheric aerosols.
Various physics modules, external modeling framework, and boundary conditions have been integrated and hardwired to simulate fully coupled aerosol-cloud-precipitation-land surface processes at the satellite-resolvable scales. Our group mostly contributing the development of i) cloud microphysics, ii) radiation, iii) satellite simulator and satelliate datasets, and we also apply the NU-WRF to study various aspect of cloud-precipitation processes associated with aerosols, synoptic forcing, land-surface processes. 




NU-WRF Integrated Development Framework (NUWRF-IDF)
The NU-WRF will be designed for “observation-driven” framework, by incorporating the advanced satellite-based model evaluation framework and land and cloud-precipitation data assimilation system (NUWRF-DAS). To this end, we will develop unique development cycle: the NU-WRF Integrated Development Framework (NUWRF-IDF). First, evaluation of the NU-WRF physics with the statistical composite of satellite (in-situ) data will identify the strength and weakness of the NU-WRF physics, and guide us to better direction for their improvement. Second, less-biased physics packages will be used for the NU-WRF DAS to improve the forecasting skill (analysis) through spatio-temporal evaluation against the satellite data. With improved physics and forecasting skill, the NU-WRF can better answer science questions. Better analysis eventually feedback to the next-round physics improvement. 
During the previous years, we have compiled unique high-resolution L1B database (spatial map and statistical composite) from the A-Train (Aqua, CloudSat, CALIPSO), the TRMM, and the Geostational satellites as benchmark database for the NU-WRF evaluation. These unique “value-added” L1B satellite observations will improve NU-WRF physics modules as well as improve NU-WRF forecast skill through the coupled G-SDSU that translates model geophysical parameters into the satellite-observable signals.
This NU-WRF IDF will be capable to handle upcoming satellite missions: the Global Precipitation Measurement (GPM), Joint Polar Satellite System (JPSS), Soil Moisture Active Passive (SMAP), and Geostationary Operational Environmental Satellite-R (GOES-R) missions. The G-SDSU is the “official” GPM simulator that is capable of simulating GPM CORE instruments with detailed specs of radar and radiometer scanning patterns, channel frequency, sampling rate, and it will be capable to support the SMAP L-band radiometer, NPP/JPSS VIIRS, GOES-R sensors as well.



Currently, we are developing the fully coupled aerosol-cloud-precipitation-land surface process modeling on the satellite-resolvable scales (horizontal grid spacing of 1~5km). The following example is the simulation of Hurricane Earl and Saharan Air Layer accompanied with mineral dust. The left panel is satellite observations and the right panel is the NU-WRF simulation.
a)                                                           b)
(a) MODIS aerosol optical depth (lower color scale, white areas indicating missing data) and TRMM multi-satellite 24-h accumulated precipitation (upper color scale) for Aug. 29. Black contours are 700 hPa wind speeds at 4 m s-1 intervals from NCEP global meteorological analyses. (b) Same as (a) except for the NU-WRF model output.
More details can be found in the NU-WRF portal (