About - MIROVA

MIROVA

MIROVA (Middle InfraRed Observation of Volcanic Activity) is an automatic hot-spot detection system developed to detect, locate and measure the heat radiation sourced from volcanic activity (Coppola et al. 2020).

The main goals of MIROVA are:

Support volcano observatories in evaluating the state of activity and in maintaining situational awareness during eruptive crises
Detect thermal unrest signals in potentially active volcanoes
Update a multi-year database of infrared satellite data at the most active volcanoes

Please read here about the use of the data provided in this website

Main Features

Automated detection of high-temperature volcanic activity in Near Real Time -NRT (within few hours from the satellite overpass);
Multi-sensor approach using data with different spatial, temporal and spectral features;
Quick overview of the latest thermal images;
Updated Volcanic Radiative Power (VRP) time-series in logarithmic and normal scale;
Last thermal map image of each volcano available for Google Earth overlapping;
Possibility to add any new target volcanoes in near real time observation, in any area of the globe (on request in few hours);

Sensors

MIROVA is developed to work in Near-Real-Time (NRT) by providing infrared images and thermal flux time-series within few hours from satellite overpasses. It was originally based on the analysis of Middle InfraRed Radiation (MIR) provided by MODIS sensor. Currently, the system integrates infrared data acquired from multiple sensors and elaborated by different algorithms:

MODIS (Moderate Resolution Imaging Spectroradiometer): On board the satellites Terra (Mar 2000) and Aqua (May 2002) it acquires in the Middle Infrared (MIR) and thermal infrared (TIR) regions, with a resolution of about 1 km. Image Cadence: approximately every 12 hours. Algorithm: Coppola et al. (2016).

VIIRS (Visible Infrared Imaging Radiometer Suite): On board the Suomi-NPP (Oct 2011) and JPSS-1 (Nov 2017) satellites it acquires in the Middle Infrared (MIR) and thermal infrared (TIR) regions, with a resolution of about 750 m (375 m for the imaging bands). Image Cadence: approximately every 12 hours. Algorithm: Campus et al. (2022).

MSI (Multispectral Instrument): On board the Sentinel 2A (Jun 2015) and Sentinel 2B (Mar 2017) satellites, it acquires in the Near Infrared (NIR) and Short Wave Infrared (SWIR) regions, with a resolution of about 20 m. Image cadence: approximately every 5 days. Algorithm: Massimetti et al. (2020).

OLI (Operational Land Imager): On board the Landsat 8 satellite (Feb 2013) it acquires in Near Infrared (NIR) short wave infrared (SWIR) regions, with a resolution of about 30 m. Image cadence: approximately every 16 days. Algorithm: Massimetti et al. (2020).

InfraRed Radiation

The ElectroMagnetic (EM) radiation with wavelengths longer that those of visible light is called InfraRed radiation (IR). This range of emissions spans from 0.74 micron to 250 micron and includes the thermal radiation emitted by the volcanic products as well as by objects near room temperature. However, the thermal emission from an object is attenuated by the atmospher resulting from absorption by gases and scattering by particles. This attenuation is wavelength dependent and may be so high that the radiation emitted from an object on the earth's surface is absorbed or scattered to a degree that is not detectable by satellite. However, there are certain parts of the infrared spectrum for which the atmospheric attenuation is low ("atmospheric windows", see Fig. 1) and allow the radiance leaving the surface to be tracked from space. The most useful regions for observing the thermal signature of volcanic bodies are those named SWIR, MIR and TIR (Fig. 1).

The ShortWave InfraRed region (SWIR), between 1.4 and 3 micron, it is very sensitive to very-high temperatures (> 700 C) and is particularly suitable for detecting active lava bodies, or small but very-high temperature fumaroles. MIROVA uses MSI and OLI data to detect and analyze SWIR radiation emitted from volcanic sources.
The Middle InfraRed region (MIR), between 3 and 5 micron, shows the lowest attenuation levels. MIR radiance is very sensitive to high to moderate temperatures (200 - 700 C) and is particularly suitable for detecting high temperature volcanic features such as lava flows, lava lakes, lava domes and open-vent activity. MIROVA uses MODIS and VIIRS data to detect and analyze MIR radiation emitted from volcanic sources.
The Thermal InfraRed region (TIR), between 8 and 13 micron, it is very sensitive to low-temperatures volcanic fetures (less 100 C) and is particularly suitable for monitoring low-temperature small fumaroles or intre-crter lakes. MIROVA does not currently use sensors to detect TIR radiation emitted from volcanic sources.

Fig. 1 - Typical transmission pattern of the atmosphere. The NIR, SWIR, MIR and TIR regions of EM are marked by the different background colors. Black lines represent Planck curves of typical volcanic sources.

Volcanic Radiative Power

The Volcanic Radiative Power (VRP) is a measurement of the heat radiated by the volcanic activity at the time of a satellite acquisition.

The VRP is calculated in Watt (W) and represents a combined measurement of the area of the volcanic emitter and its effective radiating temperature. MIROVA calculates the Volcanic Radiative Power (VRP) by using the "MIR method", an approach which was initially introduced by in order to estimate the heat radiated by active fires, using satellite data (Wooster et al. 2003).

This approach (also known as Middle InfraRed method) relies on the fact that whenever an hot emitter has an "effective radiating temperature" higher than 600 K, the "excess" of radiance detected in the MIR region (DLMIR), can be linearly related to the the radiative power (error about 30%). Hence, for any individual hot-spot contaminated MODIS pixels, MIROVA calculates the VRP as: VRP = 18.9 x APIX x DLMIR where 18.9 is a best-fit regression coefficient, APIX is the pixel size (1 km² for the MODIS pixels) and DLMIR is the "above background" MIR radiance of the pixel. When a hot-spot is detected in more than one pixel, the total VRP is calculated as the sum of all pixels detecting a hot-spot.