Observability‐Based Sensor Placement Improves Contaminant Tracing in River Networks. Issue 7 (21st July 2021)

• Record Type: Journal Article
• Authors: Bartos, Matthew; Kerkez, Branko
• Is Part Of: Water resources research. Volume 57:Issue 7(2021)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2021-07-21
• Subjects: sensor placement -- observability -- surface water quality -- networks -- contaminant source identification; Hydrology -- Periodicals; 333.91
• DOI: 10.1029/2020WR029551
• Languages: English
• ISSNs: 0043-1397

Abstract

Abstract: This study presents a new methodology for identifying near‐optimal sensor locations for contaminant source tracing in river networks. We define an optimal sensor placement as one that enables the best overall reconstruction of contaminant concentrations from observed data. To establish a physical basis for the problem, we first derive a linear time‐invariant (LTI) model for riverine contaminant transport using the one‐dimensional advection‐reaction‐diffusion equation. We then formulate an optimization problem to find the sensor placement that maximizes the observability of the modeled system and identify two heuristics for efficiently achieving this goal. By evaluating each sensor placement strategy on its ability to reconstruct initial contaminant loads from observed outputs, we find that the best sensor placement is obtained by maximizing the rank of the LTI system's Observability Gramian. This sensor placement strategy enables the best overall reconstruction of both magnitudes and distributions of nonpoint‐source contaminants. Our methodology will enable researchers to build sensor networks that better interpolate pollutant loads in ungauged locations, improve contaminant source identification, and inform more effective pollution control strategies. Plain Language Summary: Sensor networks are widely used to monitor water quality in surface water systems. However, there are few established guidelines for deciding where sensors should be placed. This study introduces a new algorithm for finding the sensor placement that produces the best overall estimate of contaminant concentrations when only a small subset of river branches can be monitored. By improving the design of water quality monitoring networks, our algorithm will help water managers to better identify where contaminants originate, and design pollution control strategies to mitigate downstream impacts. Key Points: A formal methodology for water quality sensor site selection in river networks is presented Drawing on a dynamical model of contaminant transport, this methodology identifies the sensor locations that maximize the observability of the system—enabling the best reconstruction of contaminant concentrations from observed outputs The resulting sensor placement improves contaminant source identification and estimation of pollutant loads in ungauged locations for a simulated river network