Data analysis from a low-cost optical sensor for continuous marine monitoring

Abstract

Continuous monitoring of aquatic environments has increased in recent times and is likely to continue in the immediate future. To obtain the most desirable outcomes from in situ monitoring it would be ideal to assemble sensor webs of low-cost and robust sensors with a high spatial resolution that produce relevant, timely and accurate data. In this paper, data analytics techniques are employed on data retrieved from a low-cost and robust optical sensor, measuring the side scattering and transmission of light through water. The sensor was deployed in estuarine waters to generate useful evidence of bulk water parameter events. These techniques are easily scalable for use in real-time for sensor webs to give decision makers additional information (both spatially and temporally), to help inform grab sampling times and other environmental monitoring considerations. Results are presented which show the optical data can be used to observe opacity changes in the water and these changes are correlated with turbidity events. In addition estimates of sunrise and sunset times can be given, as well as indications of sunlight hours. Data from multiple long-term deployments in Dublin Bay are used in this study.