The Joint Numerical Testbed (JNT) is a collaborative facility within RAL that is connected to an international network of collaborators. The main goals of the JNT are to test and evaluate numerical weather prediction (NWP) systems to provide meaningful information about forecast performance to operational decision makers and to provide the research community with support in their development of these systems. Currently the JNT consists of four components: the Developmental Testbed Center (DTC), the Data Assimilation Testbed Center (DATC), the Tropical Cyclone Modeling Team (TCMT), and the Applied Statistics and Verification Research project (ASVR).
The DTC is focused on testing and evaluation of NWP systems and is part of the national Weather Research and Forecasting (WRF) program that is developing the next generation numerical forecasting systems in the US. It is an interagency program involving NSF, NOAA and DOD
The DATC is focused specifically on testing and evaluating various data assimilation systems. It has been funded by the Air Force Weather Agency (AFWA), NSF/OPP, Taiwan Civil Aeronautics Administration (CAA) and the Korean Meteorological Administration (KMA).
The TCMT is focused on testing and evaluation of NWP systems applied to the prediction of tropical cyclones. The funding for this testbed is currently provided through the NOAA Hurricane Forecast Improvement Project (HFIP).
The ASVR project is focused on developing statistically meaningful advanced verification tools for the assessment and comparison of the performance of NWP and other forecasts. In addition, the ASVR project provides statistical support for other projects in RAL.
It is expected that as funding opportunities become available, components may be added to the JNT to focus on other parts of the weather forecasting process.
Testing and evaluation undertaken by the developers of new NWP techniques from the research community is generally focused on case studies. Extensive testing and evaluation must be performed to ensure that these new techniques are indeed ready for operational consideration. Testing and evaluation by the JNT generally focuses on extended retrospective time periods. The cases selected for these retrospective tests encapsulate a broad range of weather regimes ranging from null, to weak and strong events. The exact periods chosen vary based on the type of phenomenon that is the focus of the test. For some test activities, these cases will be chosen from all four seasons (e.g., extra–tropical for general predictions), whereas for others the cases will come from a particular season (e.g., hurricane season, convective season). The JNT’s evaluation of these retrospective forecasts includes standard verification techniques, as well as new verification techniques when appropriate.
By conducting carefully controlled, rigorous testing, including the generation of objective verification statistics, the JNT is able to provide the operational community with guidance for selecting new NWP technologies with potential value for operational implementation. JNT testing also provides the research community with baselines against which the impacts of new techniques can be evaluated. The statistical results may also aid researchers in selecting model configurations to use for their projects.
Statistical verification of forecasts is a critical component of their development. Improvements can be made by evaluating forecast products throughout the development process as deficiencies in the algorithms are discovered. Verification is also beneficial to forecasters and end users because verification findings supply them with objective data about the quality or accuracy of the forecasts, which can feed into decision processes
Community code is a free and shared resource with distributed development and centralized support. Ongoing development of community codes is maintained under version control. Periodic releases, which include the latest in developments of new capabilities and techniques, are made available to the user community.
Community code is a free and shared software resource with distributed development and centralized support. Developers contribute new software capabilities to a shared software repository, and user support is provided via websites, on–line tutorials, on–site tutorials, workshops and helpdesk functions. Ongoing development of community codes is maintained under software version control, and includes extensive regression and pre–release testing. Periodic software releases, which include the latest in developments of new capabilities and techniques, are made available to the user community. The following codes are available to the community: