Automated soil water balance sensing: From layers to control volumes

Authors: Evett, Steven - Steve; Schwartz, Robert; Casanova, Joaquin; Oshaughnessy, Susan

Submitted to: Meeting Abstract
Publication Type: Abstract Only
Publication Acceptance Date: 8/16/2013
Publication Date: 9/8/2013
Citation: Evett, S.R., Schwartz, R.C., Casanova, J.J., Oshaughnessy, S.A. 2013. Automated soil water balance sensing: From layers to control volumes [abstract] Automated methods for continuous measurements agro climatalogy and forestry workshop, Ben Gurion University, Israel, September 7-12, 2013. p. 7

Interpretive Summary:

Technical Abstract: Continuous sensing of soil water status has been possible in some ways since the advent of chart recorders, but the widespread adoption of soil water sensing systems did not occur until relatively inexpensive dataloggers became available in the late 1970s and early 1980s. Early systems relied on pressure transducers used with tensiometers; the 0-10 VDC or 4-20 mA output of the pressure transducer was fed to a chart recorder for continous recording, but numerical data were then read from the chart. Early time domain reflectometry (TDR) systems for soil water sensing used the chart recorders available with the Tektronix TDR cable testers, and with automated toggling of waveform capture, data could be continuously recorded, although manual interpretation of waveforms was still necessary to convert essentially analog data to numerical data. With the advent of 8-bit microprocessors and analog to digital (A/D) conversion chips, it became possible to directly convert analog signals to numerica data. But this did not become widely feasible until the aforementioned dataloggers and the personal computer (PC) revolution of the early 1980s, during which A/D add-in cards for the PC backplane became widely available. Dataloggers and PC-based A/D systems were increasingly used in research and practical applications to digitize analog signals from pressure transducers (tensiometers), electrical conductivity sensors (e.g., the four-wire Wenner array), pH sensors, and TDR waveforms. Multiplexing systems for switching signals to A/D inputs also became available in the 1980s for using with dataloggers, making multi-channel digital data acquisition possible. Around 1990, the first computer programs became available for automatically acquiring TDR waveforms and interpreting them to determine pulse travel times and water contents in digital form. These were quickly followed by programs that determined both water content and soil bulk electrical conductivity. Several TDR multiplexing systems became available in the early 1990s and are still available today, making possible automatic acquisition of data from hundreds of locations. While TDR and tensiometers were and are very successful for measurements in individual soil layers, there were few applications that sensed water status in the entire soil profile relevant to growing plants such as has been routinely done using the neutron probe since the 1960s. The neutron probe cannot, however, be left unattended, so instances of automatic data acquisition using neutron thermalization were infrequent until the advent of the COSMOS system. COSMOS utilizes neutrons produced naturally from cosmic ray interactions with atomic nucleii in the atmosphere and surface soil. The frequency domain (FD) methods, which like TDR rely on electromagnetic soil properties, were developed for soil water sensing in the late 1970s and early 1980s, but for manual use. It was not until the 1990s that automated frequency domain sensing systems were made commercially available and routinely used. Since then a plethora of FD sensing systems has become available, either as individual local sensors used with dataloggers or as systems of sensors utlized to sense soil layer water status from within non-metallic access tubes, usually automatically. Until recently, the FD sensors were much less expensive than TDR systems, which drove their wide adoption. The use of FD sensors has been clouded, however, by fundamental problems related to sensed volume, the geometric factor that determines the permittivty value that is sensed and which is randomly variable in most soils, and interferences from soil bulk electrical conductivity, pore water conductivity, bound water and soil temperature effects on these. The history and current practice of automated soil water sensing will be discussed as well as innovations being developed for future sensing