First large-scale intercomparison of satellite-based volcanic ash retrievals
January 2016 - Volcanic ash is a potential hazard for aviation, and serious damage to aircraft and their engines has been reported during past encounters. When the Icelandic volcano Eyjafjallajökull erupted in 2010, the ensuing volcanic ash plumes extended over a large area, and brought European air traffic to a halt. That event triggered the rapid development of a high-level competence in volcanic ash monitoring and forecasting within the Met Office.
The Met Office hosts one of the International Civil Aviation Organisation Volcanic Ash Advisory Centres (London Volcanic Ash Advisory Centre (VAAC)), with responsibility for issuing advisories for volcanic eruptions originating in Iceland and the North-Eastern Atlantic. We provide advice to the Civil Aviation Authority (CAA) on the location of volcanic ash, and provide forecasts using the Met Office Dispersion Model atmospheric dispersion model.
Satellite Applications are an essential tool to locate and quantify the volcanic plumes and to complement and verify the information given by the model. Several satellite products have been developed by different institutes, using imagers and sounders on several satellites, and one of these algorithms has been developed by the Met Office. The Met Office algorithm routinely uses the Spinning Enhanced Visible and InfraRed Imager (SEVIRI) which is on board the Meteosat Second Generation satellites (MSG), as well as other instruments such-as the Moderate Resolution Imaging Spectroradiometer (MODIS). See Figure 1 for an example of some of the Met Office satellite products.
In June 2015, Met Office experts took part in the first World Meteorological Organisation (WMO) Intercomparison of Satellite-based Volcanic Ash Retrieval Algorithms. Objectives of the intercomparison were (i) to establish a basic validation protocol, (ii) to quantify and understand differences in products and (iii) to standardise volcanic cloud geophysical parameters. Twenty-two algorithms from institutions and groups all over the world were compared, using passive satellite imagery from several satellite instruments such as SEVIRI, the MTSAT Imager, AVHRR, MODIS, IASI and many others.
The Met Office also provided one of the reference datasets used in the intercomparison, originating from six flights of the United Kingdom research aircraft. Although limited in time and space, this dataset was considered a good validation source by all groups taking part in the intercomparison (see Figure 2 for an example of the use of this dataset).
In general, the Met Office satellite detection and retrieval results showed good levels of agreement with other state-of-the-art schemes (an example of one such comparison is shown in Figure 3), and areas for future improvements have been identified. The report's conclusions highlight a threshold of ~0.1 g/m2 for the detection of volcanic layers using passive infrared imagery, and an error estimate of a factor between 2 and 4 on concentration estimates. Ash layer heights measured from satellite imagers were generally within ±2 km of aircraft observations.
The workshop recommended the full exploitation of the space-based Global Observing System, and underlined the importance of human expert analysis of the imagery (considered as the upper limit of detection sensitivity). Further collaboration of the VAACs through the recently enhanced WMO VAAC Best Practice workshops and the remote sensing community (primarily through the WMO/IUGG Volcanic Ash Scientific Advisory Group) has been encouraged, in order to form collaborative links for training and interpretation of events. The workshop also highlighted the importance of in situ measurements and ground-based remote sensing near the volcanic source, and the provision of airborne measurements of the distal plumes during forthcoming volcanic eruptions: in particular, the particle size-distribution was identified as a key piece of information.
Pending final analysis of results, a second intercomparison was recommended, using more focused, in-depth comparisons of algorithms, using well-understood case studies, with additional validation data/scenes available (including expert ash analyses, aircraft data and CALIOP analyses), and using the next-generation imagers/sounders.
The final report of the intercomparison, published in October 2015, can be downloaded by clicking on the following link: www.wmo.int/pages/prog/sat/documents/SCOPE-NWC-PP2_VAIntercompWSReport2015.pdf.