By Geoff Wadge
The atmosphere over an erupting volcano can be a very lively place. Lava bombs, ash, gas, lightning and torrential convective rainfall are commonplace in the immediate few kilometres from the vent. Such chaotic environments are not very well observed and modelled. Ash and aerosols can be dispersed thousands of kilometres with wide-ranging impacts (e.g. Pinatubo in 1991 and Eyjafjallajokull in 2010). But most of the time, the atmosphere over many non-erupting volcanoes behaves as it does over any mountain.
Figure 1. The Soufrière Hills Volcano on Montserrat; see text for explanation of colours
The image shown here is of the Soufrière Hills Volcano on Montserrat taken from the volcano observatory 5 km away. This volcano has been in intermittent eruption since 1995, but when this photograph was taken in 2012 no lava was being extruded. However, there was a plume of volcanic gases, including suphur dioxide with a flux of about 300 tonnes per day, from vents heated to 500˚C by the gas. This plume is bent over and is being advected from east to west (left to right here) by the trade winds.
The instrument in the foreground is a radar interferometer (with its designer, Charles Werner, at the controls). We had taken it to Montserrat to measure a rhythmic motion of the volcano’s surface that could constrain the dynamics of the magma rising through the volcano. Unfortunately, at this time there was no such magma-induced motion. The interferometer works by comparing the phase of the signal after its passage from the radar to the ground and back again with a similar measurement made later. Any change in the ground position (apparent motion along direction of the radar beam) is recorded as a change in phase. For this instrument a complete 360˚ cycle of phase corresponds to about 9 mm of positional change.
So what are those colours draped over the volcano? They are an image of phase change (an interferogram) over an hour, produced by averaging 60 such interferograms recorded every minute. Phase increases from magenta to yellow to green to blue to purple. But if there was no ground motion, what is causing the phase change recorded? The answer is water vapour, or rather the change in water vapour content along the radar line of sight. From the ground motion perspective this is noise. From the atmospheric perspective it is a potentially valuable source of data on the state of the boundary layer.
In the image, the phase generally increases with distance and elevation. This can be interpreted as a greater increase in water vapour at lower altitudes than at higher altitudes during this hour. The wider dataset collected shows a variety of interesting features: (i) A general pattern of lower, water vapour variance nights and higher variance days, as you might expect in a diurnal cycle of boundary layer behaviour. This suggests only trying to obtain the ground motion measurements at night. (ii) Sub-hourly jumps in variance. (iii) Quasi-stationary, high variance features in the lee of steep terrain. (iv) Possible examples of volcanic plume detection.
This is a new source of information on the variability of the water vapour content in the boundary layer. Its value would be enhanced by coupling with other data and models.
ACKNOWLEDGEMENTS Thanks to Tom Webb for the photoshopery.