A History of Volcanic Ash Forecasting at NOAA Air Resources Laboratory

Given the Air Resources Laboratory’s (ARL) origins in the transport of nuclear material high in the atmosphere, one logical extension was to support volcanic ash forecasting for aviation safety. There are volcanoes erupting nearly every day throughout the world, though most are relatively small. Following the 1980 eruption of Mt. St. Helens, a Memorandum of Understanding between NOAA and the Federal Aviation Administration (FAA) became effective. Redoubt volcano in Alaska erupted in 1989-90. Near-realtime NGM 300 mb forecast trajectories were run, with dot-matrix printer graphics produced. ARL developed the Volcanic Ash Forecast Transport and Dispersion (VAFTAD) model. Details of the model will be described. The VAFTAD 8-panel graphical output showing ash location was developed in consultation with FAA. At that time, the principle was to avoid ash. Initially National Meteorological Center (NMC, now National Centers for Environmental Prediction, NCEP) 24x7 staff logged on an ARL research workstation to run VAFTAD, but in 1992 the model was transferred to NMC operations. Early verification consisted of manually overlaying graphics of satellite imagery showing observed ash clouds and the model footprint. In the 1990s, an ARL scientist was a member of the Office of the Federal Coordinator for Meteorology (OFCM, now Interagency Council for Advancing Meteorological Services (ICAMS)) Working Group on Volcanic Ash. The Group developed a National Volcanic Ash Operations Plan for Aviation and Support of the International Civil Aviation Organization (ICAO) International Airways Volcano Watch. In the decade of the late 2000s, ARL participated in a USGS-led effort on “eruption source parameters” (ESPs), which meteorological transport and dispersion modelers typically call the “source term”. Given a theoretical relationship between total eruption mass rate and eruption height, the eruption rate could be estimated, albeit with high uncertainty. In 2010, the eruption of Eyjafjallajökull in Iceland caused multi-day airspace closures in Europe, and led to many national and international meetings. The result was a change from the principle of ash avoidance to a risk-assessment approach. NCEP implemented new features, such as an option for time-varying source heights, the ability to modify the ash particle size distribution, and the ability to horizontally translate the ash cloud footprint to allow the 24x7 staff greater modeling flexibility. A new ARL contractor began to concentrate on the science of quantitative ash modeling and evaluation, which was clearly needed with the new risk-assessment approach, while another scientist handled the bureaucracy resulting from Eyjafjallajökull. ARL research on the source term included use of a cylindrical vs. linear eruption column and “data insertion” of quantitative satellite observations. Additional research included an observation-modeling inversion approach, in which the volcanic ash source - mass source strengths, vertical distribution, and temporal variations - were objectively and optimally estimated. The capability to obtain quantitative output was added to the ARL READY web page using the algorithm relating mass eruption rate to eruption height. In the late 2010s, ARL turned to the NCEP ensemble NWP to produce ensemble ash dispersion products. Moving towards the 2020s, ARL developed a prototype system to ingest the NESDIS/STAR automatically-processed quantitative satellite retrievals, couple them with HYSPLIT for source determination, produce probabilistic forecasts, apply bias correction, and include a verification system of the probabilistic ash forecasts. ARL also developed a system to automatically, in real-time, receive NESDIS/STAR VOLcanic Cloud Analysis Toolkit (VOLCAT) alerts, run deterministic HYSPLIT trajectories and dispersion using the quantitative satellite observations, and post the graphical results on an ARL web page. ARL has long followed the processes of research and development contributing towards improving volcanic ash forecasting for aviation safety, implementing capabilities on ARL systems which are as operational as they can be in a non-24x7 research setting, and then transferring them to 24x7 fully supported operations at the NWS, and is expected to continue this into the future.