A Neural Network Approach to Smarter Sensor Networks for Water Quality Monitoring

Abstract

Environmental monitoring is evolving towards large-scale and low-cost sensor networks operating reliability and autonomously over extended periods of time. Sophisticated analytical instrumentation such as chemo-bio sensors present inherent limitations because of the number of samples that they can take. In order to maximize their deployment lifetime, we propose the coordination of multiple heterogeneous information sources. We use rainfall radar images and information from a water depth sensor as input to a neural network (NN) to dictate the sampling frequency of a phosphate analyzer at the River Lee in Cork, Ireland. This approach shows varied performance for different times of the year but overall produces output that is very satisfactory for the application context in question. Our study demonstrates that even with limited training data, a system for controlling the sampling rate of the nutrient sensor can be set up and can improve the efficiency of the more sophisticated nodes of the sensor network.

Diverse Applications of Electronic-Nose Technologies in Agriculture and Forestry

“…Electronic-nose applications to detect plant pests in preharvest and postharvest crops and tree species continue to expand to include new insect [54–61] and disease [111,112,339,409–413] pests, primarily microbial plant pathogens, beyond those originally reported by Wilson et al [2,106,107]. In the macroenvironments adjacent to industrial plants and indoor working spaces within associated food- and fiber-production facilities, e-noses increasingly are being utilized to monitor air quality to detect hazardous chemicals [68–70,76,77,80,414–419], explosives and flammable gases [29,64], pollutants [420–422] and other VOCs that threaten human health. Likewise, malodorous gases produced from point sources, such as agricultural feedlots and paper-production facilities (pulp mills), increasingly are being monitored by e-nose devices to assure that release of gaseous odors and effluents are maintained below offensive and hazardous threshold levels [423–427].…”

“…Tests of soil health and microbiological activity will provide means of assuring that crop plants are grown in healthful growth environments and in soils free of harmful chemicals or microbes [429]. Finally, electronic-noses are having greater utility in indoor agricultural production within greenhouses, such as for environmental controls of air quality (pollutants) [104], relative humidity [102], fertigation metering [374], and irrigation water quality [80] to assure that ornamental and food crops remain free of biotic and abiotic diseases [110,430]. …”

Multi-modal sensor networks for more effective sensing in Irish coastal and freshwater environments

“…In the work outlined in [13] we describe a study whereby rainfall radar images and information from a water depth sensor are used as input to an Artificial Neural Network to dictate the sampling frequency of a phosphate analyser at the Lee Maltings site. Specifically we investigate a methodology for incorporation of pixel information from rainfall radar images and in-situ depth data into an ANN and the subsequent use of this network to predict average freshwater levels at a dynamic point of the river.…”

Investigation into the use of satellite remote sensing data products as part of a multi-modal marine environmental monitoring network

“…Image data was used to provide estimations of depth to complement a river monitoring network 7,8,9 and to detect ships at a port following the identification of a relationship between ship traffic and turbidity measurements 10 . We also investigated the optimisation of an in-situ monitoring network using neural network models which incorporate rainfall radar images 11 . Here the focus is on the use of satellite remote sensing data products as an additional data source in a marine monitoring network.…”