MONITORING TEMPORAL SOIL MOISTURE VARIABILITY
WITH DEPTH USING CALIBRATED IN-SITU SENSORS
Daniel C. Hymer*, M. Susan Moran and Timothy O. Keefer
USDA Agricultural Research Service, Southwest Watershed Research Center, Tucson, AZ * Corresponding author: USDA-ARS Southwest Watershed Research Center, 2000 E. Allen Road, Tucson, AZ 85719; email: dhymer@tucson.ars.ag.gov
I. INTRODUCTION
Studies have shown that soil moisture is the basic link between the hydrologic cycle and the energy budget of land surfaces (Houser, 1996). Therefore, soil moisture data is useful in many disciplines including agriculture, forest ecology, civil engineering, water resources, meteorology and soil science (Wetzel and Woodward, 1987). In semiarid rangelands, soil moisture is a major control of evapotranspiration and a significant factor in runoff generation. Consequently, soil moisture is one primary component of many hydrologic models. Unfortunately, long term soil moisture data sets are rare because conventional soil moisture techniques (e.g., gravimetric and Time-Domain-Reflectometry) have historically been procedurally and labor intensive. Electrical Resistance Sensors, however, are capable of collecting nearly continuous data with little maintenance using data-loggers. Like most soil moisture instruments, ERS require a calibration to convert a measured signal to a volumetric water content. In this experiment, we calibrated a network of in situ ERS sensors in the Walnut Gulch Experimental Watershed to create a long term, temporally continuous soil moisture data set.
Two primary goals of this research are:
1. To develop a 16 month, hourly soil moisture data set at 5, 15 and 30 cm depths under three bare and three shrub cover surfaces; and 2. To use these data to validate the Simultaneous Heat and Water (SHAW) model for use in storm and inter-storm hydrologic modeling.
2. APPROACH
Electrical Resistance Sensors (ERS) used in this study are similar to those described by Colman and Hendrix (1949) and identical to those of Amer et al. (1994). Original laboratory and site calibration procedures for ERS used in this experiment were described by Amer et al. (1994). Their research described the calibration of the ERS in representative soil samples extracted from the installation site. Measured resistance values from known volumetric water contents ([theta]v) were used to identify calibration parameters. Furthermore, gravimetric samples taken during final sensor installation were used to test the laboratory calibrations. After installation, soil moisture values calculated from laboratory determined calibration coefficients were excessive, indicating an incompatibility between in situ and laboratory calibrations.
Time Domain Reflectometry (TDR) has been established as a nondestructive method for measuring [theta v] (Young et al., 1997; Topp et al., 1980). Over the past decade, research has focused on the reliability of TDR [theta v] measurements. Results proved that TDR readings are only slightly influenced by changing temperature and salinity conditions, and are independent of temperature in soil textures finer than sand (Halbertsma et al., 1994). Furthermore, soils with electrical conductivity values less than 8 dSm-1 will not have any effect on TDR measured [theta v] values (Dalton, 1992).
During the initial installation, TDR probes were placed adjacent to ERS at each of the installation sites. One disadvantage of the TDR probes installed at these sites is that they require an on-site operator and post processing. Therefore, TDR measurements were sampled at intervals varying from daily to biweekly. Analysis revealed that the TDR measurements were accurate by " 2% [theta v] (L.Bach, personal comm). To increase our confidence in the ERS measurements, we decided to use an in situ calibration that would integrate high frequency ERS data with reliable TDR data.
3. EXPERIMENT
In 1990, USDA - Agricultural Research Service (ARS) scientists installed ERS and TDR sensors in the Walnut Gulch Experimental Watershed (31.72 N, 100.00 W). Eighteen pairs of ERS and TDR probes were installed horizontally into trench faces under three bare and three shrub covered surfaces in the Lucky Hills subwatershed at 5, 15 and 30 cm depths. Data loggers recorded hourly ERS values and ARS scientists collected TDR samples at varying time intervals. ERS readings were stored as a series of resistances (ohms) while TDR values were recorded as [theta v] (m3m-3).
ERS and TDR measurements taken at identical times between August of 1990 and December of 1991 were extracted for individual trenches and depths. Calibration parameters for each ERS sensor were derived using the expression TDR = aAERSb, where TDR represents [theta v] (m3m-3), ERS equals raw resistance (ohms) and a and b are calibrated parameters. Seyfried (1993) found a non-linear relationship between TDR measured [theta v] and ERS resistance values, as did Amer et al. (1994).
Parameters a and b were optimized with non-linear curve fitting techniques from a statistical software package. Once acceptable parameters were calculated for each sensor, the calibration expression was applied to the entire hourly ERS data set to produce 16 month, hourly [theta v].
4. RESULTS
Statistical tests indicated that our calibration expression worked well
for each ERS data set.
Table 1 provides parameter and coefficient
of determination (r2) values for all 18 sensors. Data presented in Figure
1 show a representative calibration curve with matched TDR and ERS values
(r2 = .88). Clearly, resistance values approach zero as [theta v] increases.
Calibration parameters were applied to each individual ERS. A portion of the resultant data set for one individual trench with sensors at 5, 15 and 30 cm depths is shown graphically in Figure 2.
At the time of this writing, we have not completed the comparisons of these soil moisture measurements with the Simultaneous Heat and Water (SHAW) model. The SHAW model simulates heat and water movement through a plant-residue-soil system. A vertical, one dimensional profile extending from the vegetation canopy to a specified depth within the soil is represented in this model. As demonstrated in 1990, the SHAW model was capable of simulating heat and water fluxes at both grass and shrub dominated sites (Flerchinger, 1996).
5. CONCLUDING REMARKS
These findings indicate that an in situ calibration of ERS sensors that integrated high frequency ERS data with reliable TDR data worked well. An hourly soil moisture data set for a 16 month period was composed for 6 replicate sites at 3 different depths. This simple calibration procedure also demonstrates the utility of the ERS sensor for future soil moisture studies.
This calibrated soil moisture data set will be used to validate the SHAW model for use in storm and inter-storm modeling in Walnut Gulch. Additionally, we plan to apply this calibration technique to more recent ERS data sets.
6. ACKNOWLEDGMENTS
The authors would like to thank Carl Unkrich and Gary Richardson for their patience. Also, we acknowledge Leslie Bach, Saud Amer, Bill Kustas, the Water Conservation Laboratory (Phoenix, AZ) and the Tombstone field office staff for their efforts. This research was funded, in part, by an EOS grant (NASA NAGW-2425), a grant from the Landsat Science Team (NASA S-41396-F) and the Monsoon >90 project (IDP-88-086).
7. REFERENCES
Amer, S.A., T.O. Keefer, M.A. Weltz, D.C. Goodrich and L.B. Bach., 1994: Soil moisture sensors for continuous monitoring. Water Resources Research. 30, 69-83.
Colman, E.A. and T.M. Hendrix. 1949. The fiberglass electrical soil-moisture instrument. Soil Sci. 67: 425-438.
Dalton, F.N. 1992: Development of time-domain reflectometry for measuring soil water content and bulk soil electrical conductivity. Advances In measurements of Soil Physical Properties: Bringing Theory into Practice. SSSA Special Publication Number 30. Pp. 153-159.
Flerchinger, G.N., W.P. Kustas and M.A. Weltz., 1996: Simulating surface energy fluxes and canopy temperatures for two arid vegetation communities using the SHAW model.
Halbertsma, J., E.V.D. Elsen, H. Bohl and W. Skeirucha., 1994: Temperature effects of TDR determined soil water content. Symposium Proceedings: Time-Domain Reflectometry Applications in Soil Science.
Houser, P.R., 1996: Remote-Sensing Soil Moisture Using Four-Dimensional Data Assimilation. PH.D Dissertation, University of Arizona. Pp. 20-38.
Seyfried, M.S. 1993: Field calibration and monitoring of soil-water content with fiberglass Electrical Resistance Sensors. Soil Science Soc. Of America., 57: 1432-1436.
Topp, G.C., J.L. Davis, and A.P. Annan. 1980: Electromagnetic determination of soil water content: Measurement in coaxial transmission lines. Water Resour. Res. 16:574-582.
Young, M.H., J.B. Fleming, P.J. Wierenga and A.W. Warrick., 1997: Rapid laboratory calibration of time domain reflectometry using upward infiltration. Soil Science Soc. Of America., 61:707-712.
Wetzel, P.J., and R.H. Woodward, 1987: Soil moisture estimation using GOES-VISS infrared data: A case study with a simple statistical method. J. of Climate and Appl. Meteor., 26: 107-117.
homepage