Skip to main content
ARS Home » Plains Area » Bushland, Texas » Conservation and Production Research Laboratory » Soil and Water Management Research » Research » Publications at this Location » Publication #323690

Title: Current developments in soil water sensing for climate, environment, hydrology and agriculture

item Evett, Steven - Steve
item Schwartz, Robert
item CLEWETT, CATHERINE - West Texas A & M University
item O`Shaughnessy, Susan
item Colaizzi, Paul

Submitted to: Meeting Abstract
Publication Type: Abstract Only
Publication Acceptance Date: 10/1/2015
Publication Date: 10/28/2015
Citation: Evett, S.R., Schwartz, R.C., Clewett, C., Oshaughnessy, S.A., Colaizzi, P.D. 2015. Current developments in soil water sensing for climate, environment, hydrology and agriculture. Seminar Series of Texas A&M University Departments of Biological and Agricultural Engineering, and Water Management and Hydrologic Sciences

Interpretive Summary:

Technical Abstract: Knowledge of the four dimensional spatio-temporal status and dynamics of soil water content is becoming indispensable to solutions of agricultural, environmental, climatological and engineering problems at all scales. In agronomy alone, science is severely limited by scant or inaccurate knowledge of soil water status and dynamics during studies of rooting, root water and nutrient uptake, nutrient and other soil physiochemical dynamics, soil-borne disease progression, plant physiological responses to drought, limited irrigation regimes and salinity, and many other problems in which soil water interacts with soil chemical and physical attributes and related plant responses. In climatology, mesoscale synoptic weather models have advanced to the point where knowledge of the availability of water in the near-surface and surface soil-plant-atmosphere continuum is not only usefully assimilated but is essential to further progress in accuracy of weather forecasting. Satellites such as the Soil Moisture Active Passive (SMAP) mission are being flown to provide near-surface soil moisture content for not only weather forecasting but a host of other modeling needs; but satellites like SMAP require ground truthing based on accurate in situ soil water content sensing. Excellent soil water sensing systems, such as the neutron probe, that served past research and management needs well, have given way to new in situ methods based on sensing the electromagnetic (EM) properties of soils, not because of inaccuracy but because of difficulty involved with acquiring data at required frequencies and spatial coverages. The EM methods involve sensing capacitance (frequency domain, FD, sensors), travel time of an EM pulse (time domain reflectometry and transmissometry, TDR and TDT, respectively) and nuclear magnetic resonance (NMR). The use of NMR for soil water sensing is advancing but remains a research problem. However, the FD, TDR and TDT methods have been used for frequent, unattended soil water monitoring at multiple depths and locations. The least expensive of these, the FD sensors, have been the most widely used but are the least accurate and most affected by interfering factors common in soils. Inaccuracy of FD sensors is linked to relatively small EM frequencies (typically <=100 kHz) and to the fundamental physical problem that the geometric factor in the physical equation describing capacitance is unknown and changeable in nature. The relatively small EM frequencies render sensors sensitive to bulk electrical conductivity, which varies in soils depending on clay content and type, water content, pore water chemistry and temperature. The geometric factor, which varies depending on the three-dimensional propagation of the EM field into the soil surrounding the sensor, is changeable depending on random small scale variations in bulk density, water content and BEC within the sensed volume and which affect the sensed volume and shape of the EM field. The TDR and TDT methods are relatively immune to the factors affecting FD sensors both because they propagate EM fields at much greater frequencies (about 1 GHz), considerably decreasing the effect of BEC changes, and because the physical equations describing EM pulse travel time do not involve a geometric factor. However, the high frequency circuits required for true TDR and TDT methods are expensive and systems based on these methods have in the recent past been too expensive and complicated for routine use. Earlier attempts to produce inexpensive TDR and TDT sensors have been problematic due to inaccurate determination of travel time (e.g., the so-called frequency domain reflectometers, FDR). A true TDR or TDT sensor records the reflected or transmitted EM pulse voltage over time and analyzes the resultant "waveform" for travel time, something most earlier attempts did not accomplish. Only