The Catchment That Measures Itself: A Wireless Sensor Network for High-Resolution Hydrological Measurement

Low cost, low power wireless sensors offer many potential benefits in the field of hydrology. Their viability is investigated through the deployment of a wireless sensor network to monitor a 0.7 square km. catchment in Southwestern British Columbia. The study catchment includes 3 V-notch weirs and 50 instrument clusters measuring air and soil temperature, humidity, rainfall, soil moisture, and groundwater head. Overland flow is measured at 16 locations. The foundation of each cluster is a data box containing an MDA300 data acquisition board and a MICA2 processor board (mote) from Crossbow Technologies, Inc. The 50 motes, each recording data every 15 minutes, communicate in a wireless sensor network. This allows for data to be transmitted from all instrument clusters to a base point at the catchment. Data is then communicated to a computer and broadcast directly on-line. Real time data collection from the wireless network enables viewing of data during storm events, allowing for on-line data quality control and on-demand maintenance. This improves the efficiency of collecting hydrological data, and proceeds towards integrating the steps of collecting and processing data. A sensor network of this level of spatial and temporal resolution provides many possibilities for data analysis. One of the primary data analyses was the interpolation of data values between measurement points using geostatistical analysis and subsequent development of animations. Full coverage of hydrological parameters allows for the viewing of patterns of small-scale spatial variability. Seasonal patterns and regimes in hydrological parameters, as well as the transitions between these patterns can be viewed due to the high spatial and temporal resolution. Wireless sensor technology makes the collection of data at the necessary resolution to undertake this type of analysis a much more reasonable task. The complement of sensors at each site made it possible to closely monitor the type and magnitude of response to rain events between different measurements and investigate the nature and timing of response from soil moisture, groundwater level and overland flow. Response of groundwater, soil moisture, and overland flow could also be compared directly to local throughfall, and outflow from the base of the catchment. The relationship between responses of soil moisture and groundwater level will be utilized to determine the presence of lateral stormflow in the study catchment.