ChEESE´s volcanic ash forecasting service is validated in exercise conducted with the Buenos Aires Volcanic Ash Advisory Center

14 December 2021

A real exercise was conducted on December 10, 2021 to test a potential service on volcanic ash forecasts developed by the EU-funded project, ChEESE Center of Excellence (CoE). The exercise, organised in cooperation with the Buenos Aires Volcanic Ash Advisory Center (VAAC), was made possible thanks to the supercomputing resources provided by the Barcelona Supercomputing Center (BSC), which allowed the delivery of higher-resolution ensemble-based deterministic and probabilistic products under the strict time constraints imposed by this urgent computing problem.

The new service used in this exercise, expected to be launched by late 2023, is called “Quantitative Volcanic Ash”. It was tested among National Agencies, Meteorological Watch Officers, air traffic Area Control Centers (ACCs), and the ICAO International NOTAM Office (the ICAO Office responsible for providing aeronautical information services). During the exercise, the ChEESE service was run using BSC´s MareNostrum 4 supercomputer. The objective of this exercise was to practice the response to reports of volcanic ash within the region of responsibility of the Buenos Aires VAAC in an operational environment and provide ensemble forecasts to the aviation community of ash cloud extent and movement.

Results have shown that ChEESE´s service, when compared to the current setup of the Buenos Aires VAAC, produced faster and more accurate volcanic ash forecasts. This validates the project´s “Quantitative Volcanic Ash” service as a functional and reliable tool that is on the way to being operational in the next few years.

Monitoring and forecasting volcanic clouds for aviation

Volcanic clouds jeopardise aerial navigation and operations. Volcanic aerosols and fine ash particles, angular in shape and highly abrasive, can damage turbine blades, airplane windscreens and fuselage, disrupt navigation instruments and, in the worst scenario, cause the clogging of cooling passages and potential engine stall.

In 1993, the International Civil Aviation Organization (ICAO) established a global network of 9 Volcanic Ash Advisory Centers (VAACs) with the mission of monitoring and forecasting the location and trajectories of volcanic clouds occurring under their respective areas of responsibility. In the event of an eruption, the international civil aviation arrangements state that the affected VAAC must issue periodic Volcanic Ash Advisories (VAA), consisting of text messages including the forecasted ash polygons delineating unsafe flight areas. The VAAs can eventually be issued also in graphic form in the so called Volcanic Ash Graphics (VAG). In the past decade, this qualitative (ash/no-ash) approach, combined with the increase and congestion of global air traffic routes, has yielded some undesired overreaction situations causing millions of euros in economic loss to airlines and stakeholders. At present, VAACs are under a profound to digital format for all volcanic ash information and provision of improved and more efficient volcanic hazard information services. This includes the development of the next generation of quantitative ash cloud forecasts and probabilistic (uncertainty) information as well as their integration into the System Wide Information Management (SWIM) environment. In fact, current (super)computers already allow solving this urgent computing scenario with the space-time model resolution requirements demanded by the aviation stakeholders.

The Center of Excellence for Exascale in Solid Earth (ChEESE; https://cheese-coe.eu) has developed a Pilot Demonstrator (PD) for volcanic ash clouds based on running an ensemble of FALL3D dispersal model simulations as a potential urgent computing service. Two types of products fulfilling the IAVW requirements can be generated from these ensemble-based model runs: (1) deterministic products giving a deterministic forecast based on some combination of the ensemble members (e.g. ensemble mean) and, (2) probabilistic products giving a probabilistic forecast based on the fraction of ensemble members that verify a certain condition, e.g. the probability of ash concentration exceeding a flight safety threshold. The routine Buenos Aires VAAC exercise in South American airspace has provided the opportunity to validate the ChEESE volcanic ash PD as a service (Technology Readiness Level 8-9) and test its future operational deployment.