'Dynamic Emissivity-Temperature Trend' and its impact to Spaceborne Applications: Mount Etna case study

Abstract:

Detection and measurement of high-temperature thermal anomalies using satellite remote sensing (RS) enable volcanic eruption monitoring in terms of computation of radiant heat flux and mass flux (effusion rate), with application for lava flow propagation forecasting. The accuracy of such thermal estimates relies on the knowledge of emissivity, which affects the radiant heat flux and subsequent analyses that rely on the accuracy of this measurement. Here, we establish the implications that emissivity has in deriving lava temperatures on three major eruptions of Mt Etna, Italy (2001, 2002-2003 and 2017) from the spaceborne radiometers ETM+/OLI (Landsat 7/8), and MODIS (Terra and Aqua).

The emissivity-temperature trends were initially measured at a range of wavelengths (2.17-15.0 µm) and temperatures (773 K to 1373 K), using a Fourier Transform Infrared (FTIR) laboratory-based approach. The results were used as input parameters in RS applications (multicomponent emissivity approach) to provide means of computing more accurate apparent lava surface temperatures.

The combined laboratory-spaceborne analyses of the same target area confirm that emissivity is not only dependent on composition and wavelength, but also on temperature. Measured emissivity increases nonlinearly with decreasing temperature, exhibiting significant variation above 900 K with values considerably lower than the assumed for basalt (0.95-0.80), especially in MIR region. The largest and smallest increases in average emissivity between 2.17 µm and 15.0 µm during cooling were observed in the MIR and TIR regions (46% and 4% respectively), and the upper SWIR region showed an increase of 8%.

The influence of varying emissivity with lava temperatures on spaceborne data was assessed parametrically using both constant and multicomponent emissivity approaches. High-spatial (30 m) resolution results indicate that 40% variation in emissivity can produce ~20% difference in calculated total radiant heat flux. This suggests that multicomponent emissivity does play an important role in computation of more accurate radiant heat flux from spaceborne data, which would otherwise, especially in case of coarser spatial resolution data, produce a temperature overestimation, impacting both the effusion rate and lava flow distance-to-run estimates.