The system is based on an estimate of the volcano's daily emissions gathered using OMI, GOME-2 and SCIAMACHY satellite observations, and is available for consultation over the Internet free of charge. These emissions vary on a daily basis, although, for forecasting purposes, the emissions observed on the satellites are assumed to be constant during the forecasting or model simulation period. This is the biggest source of uncertainty, as it is not known exactly how the volcano's emissions will evolve in the future.
The system combines information on the volcano's behaviour, gathered twice a day from the above satellites, with environmental information, like wind speed and direction, air humidity, etc., that influences the evolution of the volcanic ash cloud.
The system was developed based on a European-wide air quality forecasting system, also built at the Facultad de Informática. This system has been operational since 2008 and is based on MM5 (NCAR, US) and CMAQ (US EPA) models. For this new undertaking, satellite-gathered information on the emissions from the volcano has been added to the model.
Prognosis
Prediction is based on the assumption that the emitted volcanic ash cloud rises from 4.5 to 8 kilometres into the air. The system analyses the information as sulphur dioxide emissions (SO2). The daily calculations are made from the above satellite observations, and the results are visualized at three levels using Dobson units.
The Dobson unit (DU) is a way of expressing the quantity of ozone present in the Earth's atmosphere, specifically the stratosphere. In actual fact, it is a measure of the thickness of the ozone layer. In the case of the volcano, this measure is used to determine the density of the ash cloud.
The forecast is subject to some uncertainty because there is no way of ascertaining how the volcano will behave from one day to the next. However, the high-quality software model used has been providing forecasts of ozone and other pollutants stipulated by European Directives (and the respective national legislation) for several years based on MM5 since the year 2000 and CMAQ since 2007.
The system has been operating experimentally for several weeks and is run by the Environmental Software and Modelling Group (GMSMA) led by Prof. Roberto San José.