Noise Reduction in Radon Monitoring Data Using Kalman Filter and Application of Results in Earthquake Precursory Process Research

Abstract

Monitoring the concentration of radon gas is an established method for geophysical analyses and research, particularly in earthquake studies. A continuous radon monitoring station was implemented in Jooshan hotspring, Kerman province, south east Iran. The location was carefully chosen as a widely reported earthquake-prone zone. A common issue during monitoring of radon gas concentration is the possibility of noise disturbance by different environmental and instrumental parameters. A systematic mathematical analysis aiming at reducing such noises from data is reported here; for the first time, the Kalman filter (KF) has been used for radon gas concentration monitoring. The filtering is incorporated based on several seismic parameters of the area under study. A novel anomaly defined as "radon concentration spike crossing" is also introduced and successfully used in the study. Furthermore, for the first time, a mathematical pattern of a relationship between the radius of potential precursory phenomena and the distance between epicenter and the monitoring station is reported and statistically analyzed.

Chaotic Behavior of Soil Radon Gas and Applications

“…The application is performed for data from Kozan and YakapÕnar regions near the East Anatolian Fault Zone, Turkey. In order to reduce the noise in the data at YakapÕnar region leading to a stationary time series, 10th order FIR filter is applied by Kaiser Window (Namvaran and Negarestani 2015). Non-linear time series analyses are used for extraction of important information about the underlying dynamic system behaviours.…”