Fault detection for salinity sensors in the Columbia estuary

Abstract

Sensors deployed in the Columbia River estuary gather information on physical dynamics and changes in estuary habitat. Of these sensors, conductivity sensors are particularly susceptible to biofouling, which gradually degrades sensor response and corrupts critical data. Several weeks may pass before degradation is visibly detected. Since the onset time of biofouling is unknown, an indeterminate amount of measurement data is corrupted. To speed detection and minimize data loss, we develop automatic biofouling detectors based on machine learning approaches for these conductivity sensors. We demonstrate that our detectors identify biofouling at least as reliably as human experts. In addition, these detectors provide accurate estimates of biofouling onset time. Real‐time detectors installed during the summer of 2001 produced no false alarms yet detected all episodes of sensor degradation before the field staff.