Detecting Abnormal Salinity Values in One-Dimensional Time-Series Data
Vincent Pham, Shuangquan Wang, Xiaohong Wang · 2024
Global warming has led to the rise in sea level and has resulted in saltwater intrusion accordingly. It poses a great threat to industrial and agricultural production and people’s daily lives in coastal regions. Real-time salinity value monitoring helps to make timely and appropriate decisions to minimize the impact of saltwater intrusion. However, it is not feasible to conduct continuous salinity monitoring relying on manual observation. This paper proposes a novel method to detect abnormal salinity values in one-dimensional time series data. The detection results might be used to automatically generate alerts for unusual salinity values and/or unusual changes. Experimental results on the simulated data in the Chesapeake Bay area show the proposed method is valid and promising.