Ensuring reliable measurements in remote aquatic sensor networks

Abstract

During the last decades, private, governmental and non-profitable organizations have been developing information systems to monitor, alert and manage environment-related emergencies [1]. Platforms, such as flood emergency alert and warning systems [2], comprise an extensive network of sensors, a bundle of forecast simulations models, and decision-support modules that rely largely on a robust and reliable perception of the conditions of the physical monitoring environment. Sensors provide this insight of the real world, where the notion of continuous time and continuous values of implicated phenomena meet the computerized notion of discrete model of time and discrete estimation of the real data. Complex and powerful forecast systems are now able to predict water levels or to track storm events with low errors, but they depend on a continuous confirmation with data. Realtime monitoring data, such as surface water elevation, flow or precipitation depend solely on the sensor hardware deployed at the water bodies (oceans, river, lakes, etc…). The goal of this paper is to propose a framework to increase the validity of the information provided by these sensors.