E. Santos-Fernandez
Water monitoring plays a central role in achieving the Sustainable Development Goals. The motivation for real-time monitoring has led to the widespread adoption of in-situ sensors, which have proven invaluable in generating critical public information. However, the data collected from these sensors often contain spatial gaps and is susceptible to anomalies. In this presentation, we discuss recent advances in statistical modeling for high-frequency river data. Our focus will be on Bayesian spatio-temporal models that effectively capture spatial and temporal correlation in water quality parameters, enabling catchment-wide probabilistic predictions. Additionally, we will address the issue of anomalies that may arise from faulty sensors and introduce an efficient framework for automatic detection and identification of water events. The effectiveness of this framework will be demonstrated through a practical case study conducted in Australia.
Keywords: spatio-temporal models, anomaly detection, Bayesian statistics, water quality
Scheduled
Spatial and Spatio-Temporal Statistics
November 7, 2023 4:50 PM
CC3: Room 1