High Altitude Stationary Clouds Detection Algorithm as an Earthquake Precursor Using MODIS Terra/Aqua Satellite Data

Abstract

Prior to many major earthquakes, clouds appear over the future epicentral region. These clouds typically appear days before the major seismic events. They remain stationary for hours, sometimes days or have a stationary root. They seem to be often or always at very high altitude. This work documents algorithm that could recognize clouds and differentiate between stationary and moving clouds having different elevations. It is also able to address questions linked to ground fog, which often also lingers for a long time but is not causally linked to earthquake activity. Main objective is to locate high altitude stationary clouds from other clouds and fog. LST (Land Surface Temperature) and visible band images acquired by MODIS scanner installed on Terra/Aqua satellites are used. Top surface temperature measured on LST images have been used to classify fog, high and low altitude clouds. Images having good temporal resolution are picked to decipher dynamicity of clouds. LST images of MODIS scanners can be effectively used to classify fog and clouds of different elevations, as an earthquake precursor, which cannot be identified in visible band images. Although, threshold values to separate high altitude stationary clouds from other clouds is still a subjective topic.