Radiant and mass fluxes in multi-platform, multi-payload satellite-based volcano monitoring

The enormous amount of remote sensing (RS) data available today at a range of temporal and spatial resolutions aid emergency management in volcanic crises. RS provides a technological solution for bridging critical gaps in volcanic hazard assessment and risk mitigation. Detection and measurement of high-temperature thermal anomalies enable eruption monitoring and new lava flow propagation forecasts, for example. The accuracy of such thermal estimates relies on the knowledge of input parameters, such as emissivity - the efficiency with which surfaces radiate thermal energy at various wavelengths and temperatures. Emissivity is directly linked to the measurement of radiant flux and therefore affects the mass flux estimate as well as any model-based prediction of lava flow behaviour. Emissivity is not commonly measured across the range of volcanic lava compositions and temperatures, and it is generally assumed to have a constant value between 1.0 and 0.80 for basaltic lava. There is a lack of field and laboratory-based emissivity data for robust, more realistic modelling. To address this deficit, experiments on ‘aa’ lava samples were performed using data from Mount Etna (Italy), representing the range of its eruptive behaviour. In three sequential stages, emissivity was measured over the widest range of temperatures (294 – 1373 K) and wavelengths (2.17 - 15.0 μm) executable in the laboratory environment. The results show that emissivity is temperature, composition and wavelength dependent. Measured emissivity increases non-linearly with temperature decrease (cooling), exhibiting significant variations above 900 K with values considerably lower than the typically assumed 0.80. The measured and modelled emissivity values were applied to various remote sensing applications as input parameters for physical modelling of lava flows. This new evidence has significant impact on the computation of radiant heat flux from spaceborne data, as well as on modelling of lava flow ‘distance-to-run’ simulations. Furnished with improved input parameters (multicomponent emissivity), the novel approach developed here can be used to test an improved version of an unsupervised multi-platform, multi-payload volcano monitoring system.