An automated low-cost monitoring station for suspended sediments and water level
Paulo V.R.M. Silva,
Robert B. James,
Kathryn L. Russell,
Tim D. Fletcher,
Maria F.S. Gisi,
Oldrich Navratil,
Frederic Cherqui,
Etienne Cossart
Affiliations
Paulo V.R.M. Silva
University of Lyon, INSA Lyon, DEEP, EA7429, 69621 Villeurbanne, France; WERG, SAFES, The University of Melbourne, Burnley, VIC 3121, Australia; University of Lyon, CNRS UMR 5600 Environnement Ville et Société, Lyon, France; Corresponding author at: University of Lyon, INSA Lyon, DEEP, EA7429, 69621 Villeurbanne, France.
Robert B. James
WERG, SAFES, The University of Melbourne, Burnley, VIC 3121, Australia
Kathryn L. Russell
WERG, SAFES, The University of Melbourne, Burnley, VIC 3121, Australia
Tim D. Fletcher
WERG, SAFES, The University of Melbourne, Burnley, VIC 3121, Australia
Maria F.S. Gisi
University of Lyon, INSA Lyon, DEEP, EA7429, 69621 Villeurbanne, France; WERG, SAFES, The University of Melbourne, Burnley, VIC 3121, Australia; University of Lyon, CNRS UMR 5600 Environnement Ville et Société, Lyon, France
Oldrich Navratil
University of Lyon, CNRS UMR 5600 Environnement Ville et Société, Lyon, France
Frederic Cherqui
University of Lyon, INSA Lyon, DEEP, EA7429, 69621 Villeurbanne, France; WERG, SAFES, The University of Melbourne, Burnley, VIC 3121, Australia; University of Lyon, Université Claude Bernard Lyon-1, F-69622 Villeurbanne, France
Etienne Cossart
University of Lyon, CNRS UMR 5600 Environnement Ville et Société, Lyon, France
The use of low-cost sensors, with open-source code, facilitates greater spatial resolution and flexibility of environmental monitoring, thus generating more information and overcoming limitations of traditional commercial sensors. Measurement of water turbidity using submerged sensors can be problematic in that rapid biofouling requires frequent site visits to remove, clean, calibrate and replace the sensor. We therefore designed an automated system using low-cost commercially-available sensors that pumps water from the stream, samples it for turbidity and purges remaining water, leaving the turbidity sensor dry between measurements, thus greatly reducing the biofouling problem and minimizing operation costs. Our station was able to estimate suspended sediment concentrations between 0 and 6 g/L with a root mean square error (RMSE) around 5 % of the total range, which meets typical research and operational study requirements. The results showed that the monitoring station is capable of monitoring water level and turbidity for long periods without the need of cleaning the turbidity sensor, due to its purge function. We demonstrated that spatially intense measurement of turbidity within catchments and drainage networks can be achieved at a relatively low cost, which allows a better understanding of the main sources of suspended sediments and their spatial and temporal variability.