Data Assimilation and Modeling

The scientists at Atmospheric and Environmental Research (AER) are experts in weather prediction through the use of numerical weather models. They contribute to the overall improvement of weather prediction and understanding of atmospheric chemistry processes through improved models and better use of data, especially data from satellite remote sensors.

AER has experience in a wide range of numerical weather prediction (NWP) models and projects, from basic research to development of real-time information services. Because of our expertise and experience, private and public clients from around the world come to us for a variety of needs, including on-site, customized consulting services for particular atmospheric modeling needs. The majority of our staff members holds advanced degrees and is skilled in a broad range of computer programming, software engineering, and project management disciplines.

NWP models that we have experience with include (but are not limited to):

• Weather Research and Forecasting (WRF)
  • North American Mesoscale Model (NAM)
  • Hurricane WRF (HWRF)
  • Advanced Research WRF (ARW)
  • Advanced Hurricane WRF (AHW)
  • WRF model coupled with Chemistry (WRF-Chem)
  • WRF Data Assimilation system (WRFDA)
• Community Atmosphere Model (CAM)
• Whole Atmosphere Community Climate Model (WACCM)
• Data Assimilation Research Testbed (DART)
  • o Coupled to WRF (WRF-DART)
  • o Coupled to CAM/WACCM (CAM-DART)
• Global Forecast System (GFS)
• European Centre for Medium-Range Weather Forecasts (ECMWF)
• Regional Atmospheric Modeling System (RAMS) – including Brazilian variant BRAMS
• Rapid Update Cycle (RUC) and High-Resolution Rapid Refresh (HRRR)
• Geophysical Fluid Dynamics Laboratory (GFDL)
• Coupled Ocean / Atmosphere Mesoscale Prediction System (COAMPS)
• High Resolution Limited Area Model (HIRLAM)

Areas of expertise:

• Mesoscale model development
  • AER collaborated with NCAR on the development of the Weather Research and Forecasting (WRF) components of a four-dimensional variational analysis (4D-Var) data assimilation system.
  • AER implemented and evaluated its improved shortwave and longwave RRTMG radiation package (Iacono et al., 2008) in the WRF model (Iacono and Nehrkorn, 2010; Iacono et al., 2009). RRTMG is one of the most widely used longwave and shortwave radiative transfer codes within NWP models including ECMWF (Morcrette et al., 2009), GFS, CESM, CFS, GEOS-5, NAM, RUC, and WRF. New versions of the radiation models are being developed to improve performance through the use of GPGPU technology, and these are currently being tested in GEOS-5 and WRF (Iacono et al., 2014)
  • AER has developed an improved parameterization of dust lofting that is now included in the WRF-Chem model (Jones et al., 2012; Peckham et al., 2014), and that was adopted for all dust forecasting at AFWA.
  • AER developed several new inline model diagnostics that are part of the WRF v3.6.1 release. For example, a new hail diameter diagnostic incorporates a one-dimensional hail column model into WRF that was successfully tested at the 2014 NOAA Hazardous Weather Testbed Spring Forecasting Experiment (Adams-Selin et al., 2014).
  • AER has also integrated the Army FASST soil model with the NASA Land Information System (LIS).

• Real-time Applications of Weather Modeling
  • NOAA-University of New Hampshire (UNH) Joint Center for Ocean Observing Technology (JCOOT) - Real-time application of WRF supporting research and operational forecast applications for the New England region (Henderson et al., 2006)
  • 2004 Athens Olympic Games - AER developed and installed a weather nowcasting system for the Hellenic National Meteorological Service. (Henderson et al., 2005)
  • AER installed an operational mesoscale NWP model at the Thai Meteorological Department to support forecasting operations.

• Carbon-budget and trace-gas modeling
  • AER has provided transport modeling expertise for a large number of research projects that involve trace-gas modeling, in particular greenhouse gases.
• Data assimilation
  • AER assisted AFWA in the development of a unified 3D-Var and 4D-Var architecture, and was also involved in the transition to and the optimization of WRF for AFRL operational uses.
  • Feature Calibration and Alignment (FCA) - AER has developed a method to use FCA to correct for systematic feature displacement errors in the WRF variational analysis system (Nehrkorn et al., 2014, 2015).
  • AER, in collaboration with a commercial satellite business, is currently conducting a mesoscale Quick Observing System Simulation Experiment (OSSE) to evaluate the potential impact of GPS signal radio occultations (RO) from very large constellations of RO picosats (i.e., > 500 satellites).
  • AER is experienced with variational assimilation of scatterometer data (Atlas et al., 2011), oceanographic data assimilation using a Local Ensemble Transform Kalman Filter (LETKF, Hoffman et al., 2008a and b, 2009), and LETKF data assimilation for the atmosphere on Mars (Greybush et al., 2011, 2012)
  • Global Weather Control - AER conducted numerical assessments of the practicality of controlling weather systems such as hurricanes (Hoffman 2002).
  • Application of the MM5 adjoint for weather modification (Hoffman et al., 2006)
  • MM5 4D-Var - AER played a key role in the development of a 4D-Var application for distributed memory computer architectures using MM5.
• Homeland Defense
  • Homeland Defense – AER researchers characterized mesoscale weather prediction errors for dispersion modeling in support of the Defense Threat Reduction Agency (DTRA).
  • Modeling & Simulation – AER supports military training by processing NWP data into simulated at-the-sensor satellite image (multi-channel IR, visible, water vapor), METAR observation and TAF forecast messages, Google Earth overlays (in vector form), and atmospheric effects of military systems (IR weapons, Comms, Mobility) among other products.