GUI for Numerical Predictions in the public cloud (Graviton4) allows model users to manage different aspects of Numerical Weather Prediction (NWP), Air Quality Modeling (AQM) and coupled (NWP-AQM) simulations using a graphical environment. The AMI integrates optimized installations of the latest versions of CMAQ (v5.5), WRF (v4.7), and WRF-CMAQ along with pre & postprocessing tools mapped via a Geographic Information System (GIS). The integration of all these tools takes full advantage of the computational power offered by Graviton4 processors resulting in lower wall times and an outstanding cost/performance ratio. The GUI also allows model users to interact directly with other AWS services such as S3 and to download data for meteorological and air quality predictions in an accelerated fashion. The minimum memory for an instance sppining up from the AMI is 8 GiB but a more compehensive recommendation is to use instances with at least 32 GiB or higher for production jobs. As the GUI is developed to take advantage of high image quality displays and provides a breadth of options and information, the lowest recommended screen resolution is QHD (2560×1440) with UWQHD (3440×1440) or higher resulting in optimal performance. Note: Odycloud is currently working to increase the GUI functionality when using FHD displays.
GUI components:
WRF – The GUI allows model users to set up the simulation parameters, visualize the domain(s) in the chosen projection, download meteorological data (GFS, NAM, GDAS) and run the preprocessors (WPS). After it, the GUI can generate the initial and boundary condition files before running WRF using the instance available cores. Once the WRF simulation is complete, the postprocessing tools render the results using a GIS framework and interpolate & output the results as CSV files for an unlimted number of locations within the domain(s). Additionally, it is also feasible to directly transfer any of the generated files to S3 buckets.
CMAQ – The GUI allows model users to set up and run CMAQ computations based on the output from WRF simulations, which have been previously completed either via the GUI or other HW/SW combination. The GUI automatically picks up the WRF output, facilitates the selections of parameters (e.g. BTRIM), and sets up CMAQ. The latest version of the GUI admits to run the simulations either using CB5R7 or CRACMM2 as the mechanism. Furthermore, the GUI includes options to automatically time-shift or convert files from CB5R7 to CRACMM2 plus it accelerates the download of monthly NHEMIS datasets, which can be used for regridding purposes. The GUI also eases up the use of the common standard preprocessing tools such as MCIP, ICON or BCON and will run the CMAQ simulation using all available cores. The postprocessing tools allow the visualization of gases, aerosols, and AQIs. It also allows exporting hourly species concentration as CSV files or snapshots (PNG, SVG and PDF formats are available).
WRF-CMAQ – This option allows model users to perform coupled simulations with the option to incorporate aerosol feedback into the NWP simulations in order to gain a more accurate computation of the physical phenomena involved in the system. The set up of these simulations is a combination of WRF, CMAQ, and the parameters dictating the interaction between the solvers. The postprocessing tools help model users to visualize the WRF and CMAQ variables.
MPAS – The GUI now includes a new interface to perform global and regional meteorological predictions using MPAS. This new interface allows model users to complete all necessary steps from repositioning mesh refinements or creating regional meshes to the initialization of all required files and the completion of the computations. A world map aids achieving a more accurate positioning of refined and regional meshes and the visualization of the simulation output once this completes. Generation of snapshots in several formats (PNG, SVG, PDF) is also feasible and several enhancements to the postprocessing unit will be available in upcoming versions.