Short noteMonitoring soil moisture and water table height with a low-cost data logger
Introduction
Comprehensive temporal data sets are often needed in geosciences to understand and model environmental phenomena. Time series data sets in the geosciences have traditionally been captured by costly commercial sensors and data loggers ($100–1000s). Dedrick et al. (2000) presented a less costly ($10 s) and publicly available device known as the Hobart and William Smith Data Logger (HWSDL) (Halfman and McKinney, 2001). Parts list, plans, schematics, manuals, software and other essential items to build your own HWS Data Loggers are available on the Internet to educators and researchers.1 Two recent projects required adaptations of the HWS Data Logger technology to: (1) record soil moisture by incorporating a dielectric aquameter, and (2) record subsurface water levels by reworking the circuit board layout and instrument housing to fit the logger and a pressure transducer sensor into a 5 cm (2 in) diameter well.
The HWSDL includes three independent components, a logger, a sensor and a reader. The original design utilized an 8-bit digital value and had a storage capacity of 4K. An upgraded version increased the data resolution to 12-bit values and memory capacity to 16K (McKinney and Halfman, 2002). This work incorporates these upgrades into two new sensor designs. The updated logger is based on Microchip's PIC16C773 microcontroller, which digitizes and records an analog voltage from a sensor at a programmable sample period. The unit is still powered by AA and 9 V batteries. The logger stores the data in a non-volatile EEPROM (24LC256, Microchip Technologies). The system interfaces to a PC compatible computer and communication is performed through the computers RS-232 serial port to transfer sample period information and collected data. Data sets are saved on the PC as delimited text files.
A variety of sensors exist for the HWSDL system, including devices for measuring temperature, light intensity, and water pressure (see footnote 1). All sensors have a number of features in common. To save battery life the sensor toggles on and off by solid-state relay connected to a control line from the loggers microprocessor. The control line is turned on (+Vss) approximately 0.5 s before sample time to provide power to sensor circuit. This allows the sensor to warm up, stabilize and then provide output to the logger. It is turned off (ground) after digitizing and storing sensor output. An optional LED illuminates whenever the control line is on (Dedrick et al., 2000). Sensors transmit an analog signal to the logger, and the logger digitizes this signal as the ratio of the signal to the reference voltage (usually 5 V). This digital value is calculated as where * is rounded to nearest decimal and + replaces 255 with 4096 when using the 12-bit logger.
This simplicity of Eq. (1.1) makes the HWSDL adaptable to a variety of sensors, but new sensors may require voltage amplification to maximize sensor resolution. Details of the sensor boards discussed in this research are provided in Fig. 1 and Table 1.
Section snippets
Soil moisture sensor
The Decagon Devices ECH2O Dielectric Aquameter (20 cm length) (Decagon, 2002) was identified as a potential cost-saving tool for monitoring soil moisture in a study of infiltration rates in urban soils within Syracuse New York. This probe is preferred over other possibilities because of minimized interference from salinity (Campbell, 2001a) and temperature (Campbell, 2001b) as well as its low cost. Adaptations described below generated the working prototype data logger plus sensor for less than
Groundwater monitoring sensor
The HWSDL was designed for sub-aqueous monitoring of surface water elevations (Riley and Halfman, 2001). However, the logger's casing diameter, pressure transducer, and 8-bit chip resolution prohibited its use in typical 5-cm (2-in) diameter groundwater monitoring wells. Additionally, the 8-bit resolution is inadequate to resolve mm-scale changes in water table elevation (Table 4). A narrower circuit board, the 12-bit HWSDL, and a new differential pressure sensor (model ASCX) from SenSym,
Conclusions
Modifications of the HWSDL presented in this paper expand upon the original logger design described by Dedrick et al. (2000). Increased digital resolution of the logger and sensor modifications allow for a greater variety of measuring and monitoring applications, including (1) soil moisture and (2) ground water elevation or piezometric head in a groundwater well.
- (1)
Decagon Devices suggest the ECH2O probe has a typical accuracy of ±3% (.03 m/m) and as great as ±1% with soil specific calibration
Acknowledgements
We appreciate the support from a National Science Foundation grant to Halfman, and grants to Endreny from the US-HUD and US Forest Service. Hobart and William Smith Colleges reserves the rights to the Data Logger upgrade and allows at no charge educational and research use of the data logger system. We are grateful to the thoughtful comments by the reviewers Quick and Heisler of an earlier version of this manuscript.
References (9)
- et al.
An inexpensive, microprocessor-based, data logging system
Computers & Geosciences
(2000) Single Supply, Micro-Power Instrumentation Amplifier
(1997)Response of the ECH2O Soil Moisture Probe to Variation in Water Content, Soil Type, and Solution Electrical Conductivity
(2001)Response of ECH2O Soil Moisture Sensor to Temperature Variation
(2001)
Cited by (10)
UAV-borne LiDAR revolutionizing groundwater level mapping
2023, Science of the Total EnvironmentCitation Excerpt :Another important limitation of in-situ methods is the accessibility to the observation points: the owner's authorization is required to access the private land and to obtain hydrogeological information at each point, which can be especially difficult when wells are not legalized. Some of these obstacles, nowadays, are being overcome by low-cost automated sensors designed to continuously measure and store high volumes of piezometric information as well as other variables such as conductivity or temperature (Johnson et al., 2005; Riley et al., 2006; Greswell et al., 2009; Alexander et al., 2011). However, these sensors require calibration and barometric correction processes and it is necessary to install one device in each of the observation wells, which can mean a large number of instruments in the case of dense monitoring networks.
Advances in understanding river-groundwater interactions
2017, Reviews of GeophysicsA developed computer-aided earthquake warning system using basic electronic components and integrated circuits
2017, 9th Congress of the Balkan Geophysical Society, BGS 2017A Computer-Aided Data Acquisition System for Multichannel Seismic Monitoring and Recording
2016, IEEE Sensors Journal