Calibration of low light/near infrared cameras with satellite remote sensing to measure incandescence and thermal signatures for source parameter estimation

Volcanic activity ranges from low level lava effusion to large explosive eruptions, easily capable of ejecting ash up to aircraft cruise altitudes. Knowledge and understanding of precursory activity and thermal signatures are vital for monitoring volcanogenic processes. For many years, satellite remote sensing has been used to determine effusion rates and expensive ground based thermal cameras can calibrate these measurements. However, this investigation explores the use of webcams to image volcanic activity in the visible to near-infrared (NIR) portion of the spectrum, by comparison of webcam pixel brightness to temperatures obtained from co-located FLIR cameras. By determining this relationship, webcam imagery can be used to approximate volcanic eruption temperatures, indicating changes in activity. A field campaign, presented here, to Stromboli, June 2013, demonstrates the use of co-located cameras to determine temperatures from pixel brightness resulting from various Type 1, 2a and 2b eruptions. This method is ideal for monitoring (particularly in remote locations) as the cameras are cheap, consume little power, are easily replaced, provide near real-time data and the images can be compared to satellite observations. A plethora of webcam imagery also exists for past eruptions, that will be analyzed and initial results are presented. Preliminary investigations were conducted in laboratory settings to determine saturation levels of each camera (wavelength dependent) and the required temporal resolution to accurately detect thermal signatures and calculate rise rates. Combined together and coupled with other observations such as seismic, infrasonic and space-borne, this data analysis will provide an increased understanding into volcanogenic processes. Two pairs of time co-incidental images showing the progression of an eruption from Stromboli's southwest crater on June 23rd 2013 at 20:52:15 and 20:52:19 UTC. a) and c) show thermal infrared images taken using a FLIR S40 with b) and d) showing the same image taken with a low-light camera.