Temperature and Emissivity Separation ‘Draping’ Algorithm Applied to Hyperspectral Infrared Data

Abstract

In the presented work, the spectral emissivity of basaltic melts at magmatic temperatures was retrieved in a laboratory-controlled experiment by measuring their spectral radiance. Granulated bombs of Etnean basalts were melted and the radiant energy from the melting surface was recorded by a portable spectroradiometer in the short wavelength infrared (SWIR) spectral range between 1500 and 2500 nm. The Draping algorithm, an improved algorithm for temperature and emissivity separation, was applied for the first time to SWIR hyperspectral data in order to take into account the non-uniform temperature distribution of the melt surface and, at the same time, solving the two temperatures and the spectral emissivity. The results have been validated by comparing our results with the emissivity measured at a "lava simulator". Basalt spectral emissivity does not vary significantly at magmatic temperature, but shows an absorption feature in the range 2180–2290 nm, an atmospheric window pivotal for the IR remote sensing of active volcanoes.

Spectral Emissivity of Phonolite Lava at High Temperature

“…To the best of our knowledge, only a handful of studies have investigated the radiative properties of lavas at high temperatures: Lombardo et al [61] on Etna basalt up to 2300 K, Lee et al [8] on synthetic glasses up to 1673 K, Rogic et al [13] on Etna basalt up to 900 K, Pisello et al [10] on two series of alkaline et sub-alkaline synthetic glasses up to 873 K and Abtahi et al [5] on Kīlauea basalt up to 1323 K. The results of these works suggest that the emissivity of molten lava is significantly lower than that of its subsolidus and cooled counterpart, an observation at variance to ours. Fig.…”

High temperature spectral emissivity of glass and crystal-bearing basalts

“…To date, few laboratory works addressed this issue with relative success (Bouvry et al, 2017;Lee et al, 2013;Lombardo et al, 2020;Rogic et al, 2019a;Thompson and Ramsey, 2020) and their results highlight the non-linear relationship existing between emissivity and temperature (ε-T). However, most of these works are performed on crystals or synthetic glasses over a limited temperature and spectral range (typically the TIR region; Byrnes et al, 2007;Harris, 2013 and references therein;Lee et al, 2013;Pisello et al, 2019;Ramsey and Christensen, 1998;Rogic et al, 2019a).…”

Quantitative Thermal Emission Spectroscopy at High Temperatures: A Laboratory Approach for Measurement and Calibration

“…Previous work at temperatures <350 K used a temperature stabilized sample chamber to reduce these energy sources and produce highly accurate results (e.g., Ruff et al., 1997). Other more recent studies have developed methodologies to acquire emissivity of materials at high temperatures for a range of applications from volcanology to planetary science (Biren et al., 2020; Lombardo et al., 2020; Maturilli et al., 2019). These studies have been successful at measuring surfaces over a limited spectral and temperature range, however without addressing all sources of error, most notably downwelling radiance.…”

“…For example, high temperature emissivity spectra of mafic samples across glass transition temperatures were made in the laboratory by Lombardo et al. (2020). Their methodology does not provide a robust and easily repeatable approach to account for environmental and instrumental radiance contributions accurately.…”

“…These studies have been successful at measuring surfaces over a limited spectral and temperature range, however without addressing all sources of error, most notably downwelling radiance. For example, high temperature emissivity spectra of mafic samples across glass transition temperatures were made in the laboratory by Lombardo et al (2020). Their methodology does not provide a robust and easily repeatable approach to account for environmental and instrumental radiance contributions accurately.…”