UV cameras are a recent development in gas measurement techniques. They take 2-D images of the gas plumes at small time intervals in the UV. I work with a system that combines 2 CMOS cameras (each fitted with a bandpass filter) and a high-resolution UV spectrometer providing a full spectrum for one selected spot on each image. After calibration, the image represents a true snapshot of a moving plume, reducing the errors inherent to the movement of the plume during acquisition. Plume velocity can be determined directly from the time series of images, eliminating the need to rely on atmospheric stations that may or may not reflect the conditions at the altitude of the plume. With those improvements over the two principal sources of errors, the SO2 camera offers a significant advantage over classic measurement methods.This system can provide contextual information for a better interpretation of the results, and its temporal resolution is the highest available. Automated processing of the images is straightforward. However, analysis of the time series to extract emission rates is more difficult. I work toward that goal, in order to promote the use of the SO2 camera as a daily monitoring tool.
UVIcam is a software package I developed to operate the instrument and perform post processing analysis. The operating package is written in Labview and allows synchronised acquisition of images from up to 3 cameras + a spectrum at each time step. The post-processing software is written in Matlab, to perform image calibration, determine the 2D velocity field and calculate emission rates from time series of images. The conversion of both packages to Python is ongoing…
The movie below shows a time lapse of slant column density maps taken at Semeru volcano (Indonesia) in 2013.