Soil water sensing methods-Usefulness for evapotranspiration monitoring and links to remote sensing

Authors: Evett, Steven - Steve; Schwartz, Robert; OCHSNER, TYSON - Oklahoma State University; Cosh, Michael

Submitted to: Meeting Abstract
Publication Type: Abstract Only
Publication Acceptance Date: 8/1/2013
Publication Date: 11/25/2014
Citation: Evett, S.R., Schwartz, R.C., Ochsner, T., Cosh, M.H. 2014. Soil water sensing methods-Usefulness for evapotranspiration monitoring and links to remote sensing[abstract].

Interpretive Summary:

Technical Abstract: Soil water sensing methods are widely used to characterize the rhizosphere and below, but only a few are capable of delivering water content data with accuracy for the entire soil profile such that evapotranspiration (ET) can be determined by soil water balance with minimal error. One such is the neutron probe (NP), which when field calibrated and appropriately used can resolve ET over weekly or longer intervals with acceptable accuracy for most uses provided that there are not important unquantifiable soil water fluxes at the bottom and sides of the control volume used for calculation of ET. When used in networks of access tubes, the NP can be used to characterize field-scale ET, useful for remote sensing studies. The much less expensive, and non-radioactive, capacitance sensors used in non-metallic access tubes have been shown to be too inaccurate for such work or even for routine irrigation scheduling by soil water depletion calculations. Capacitance sensors used in field-scale networks of access tubes do not yield accurate field-scale ET. Indeed, the capacitance methods have been shown to suffer from the fundamental theoretical weakness that the geometric factor in the capacitance equation is undefined in routine soils work. And, capacitance sensors typically work at frequencies that render them susceptible to interference from soil temperature and bulk electrical conductivity effects as influenced by clay type and content, bound water and salinity. The time domain methods, when well implemented, are capable of better accuracy than the capacitance methods, partly due to the absence of a theoretical geometric factor and partly due to the fact that they utilize a fast rise time electrical pulse that is inherently comprised of much higher frequencies than are commonly used in capacitance sensors, which makes them relatively immune to the aforementioned interferences. Heretofore prohibitively expensive and difficult to use deeply, time domain methods are becoming reasonably priced due to circuit miniaturization and the use of inexpensive very high frequency components from the cellular telephone industry. Relatively inexpensive sensors using both time domain reflectometry and time domain transmissometry methods are on, or coming to, the market, including one that measures deeply enough for ET determination by soil water balance. Accurate ET from soil water balance methods is needed in measurement networks covering field-scale areas for testing of remote sensing and flux tower estimates of ET. While not always possible due to unquantifiable soil water fluxes into or out of the control volume, ET determination by soil water balance is often useful in tying lysimeter ET to field ET data. Related to the NP, but different in that it is a surface method, is the Cosmic Ray Soil Moisture Observing System (COSMOS), which responds to surface soil water content changes in a circular area of radius up to several hundred meters. Such data are of great interest to remote sensing based studies because they hold some potential to provide ground truthing or at least an opportunity for data assimilation in those studies. Data from a field and lysimeter test of COSMOS at Bushland, Texas, show that the depth of measurement is shallow, that response to subsurface upward water fluxes is variable and that response to surface wetting is influenced by crop biomass.