SALSA Poster:

Monitoring Bank Storage in the San Pedro Riparian National Conservation Area, Arizona; Whitaker et al.

Return to AMS Posters, Table of Contents

	Monitoring Bank Storage in the San Pedro Riparian National Conservation Area, Arizona
M.P.L. Whitaker (1), T. Maddock III (1), B. MacNish (1), D. Goodrich (2) and B. Goff (2)
1. The University of Arizona, Dept. Of Hydrology & Water Resources, Tucson, Arizona, USA 2. USDA-ARS U.S. Southwest Water shed Research Center, Tucson, Arizona, USA email: mplw@hwr.arizona.edu

1. INTRODUCTION This paper describes the instrumentation, data collection and preliminary results for monitoring bank storage at the Lewis Springs site within the San Pedro Riparian National Conservation Area (SPRNCA), located in southeastern Arizona (Figure 1). Bank storage is a critical variable for accurately modeling the water budget in semi-arid riparian systems, but is particularly difficult to assess and quantify. It is especially essential for understanding ground-water/surface-water interactions.

2. INSTRUMENTATION An intensive network of vadose zone monitoring equipment was installed: ten water content reflectometery (WCR) probes, and three nests of four tensiometers were installed on both sides of the San Pedro River at Lewis Springs (Figure 2).

The WCR probes were installed in trenches which were subsequently backfilled. The locations of these trenches and the monitoring transect are identified in Figure 3. The tensiometers and WCR probes were installed at the water table interface and in the capillary fringe to monitor diurnal, seasonal and flood-induced fluctuations in soil moisture and soil matric potential within the cottonwood-willow forest gallery (i.e near-stream riparian biome -see Figure 4). A submersible pressure transducer was installed in the river to measure stream stage elevation data and provide stream stage fluctuations which correspond to the soil moisture and soil matric potential data. The tensiometers and WCR probes were wired to a datalogger to collect measurements every 30 minutes. Additionally, six neutron probe access tubes (NPATs) were installed on either side of the San Pedro River to supplement the automated soil moisture data. The locations of these NPATs are identified in Figure

3. PRELIMINARY RESULTS WCR probe soil moisture data and tensiometer matric potential data show both immediate and delayed responses to increases in stream stage. The data for two high-flow events are shown: a 38 cm stream stage increase on Julian Day 219 (August 7, 1997), and a 90 cm stream stage increase on Julian Day 221 (August 9, 1997). Figure 5 shows changes in soil moisture measured using WCR probes from the west trench. Diurnal fluctuations and an immediate response to an increase in stream stage are evident. Although the stream stage decreases sharply within 24 hours, the riverbank maintains a higher soil moisture for several days. The diurnal fluctuations exhibited by the WCR probes may reflect the evapotranspiration cycles of the cottonwood and willow trees in the SPRNCA. However, it may also be an artifact of the probes' sensitivity to temperature. The WCR probe data has not yet been normalized to 20E C. Additionally, the probes have not yet been calibrated for the soils in which they are installed; a standard polynomial provided by the manufacturer was used in the interim to discern relative changes in soil moisture.

Figure 6 shows fluctuations in soil matric potential measured using tensiometer data from nests D, E and F located on the east bank of the San Pedro River (see Figure 2). Diurnal fluctuations are considerably more subtle than those measured with the WCR probes. At least two factors may account for these differences. First, the pressure transducers associated with the tensiometers are calibrated to account for non-isothermal conditions using the method described by Lacher (1996). While both WCR probes and tensiometers' pressure transducers are installed below the ground surface to minimize temperature fluctuations, temperature may still affect the calibration of both instruments (Lacher, 1996 and Warrick et al., submitted 1997). However, the instruments' proximity to different plant roots may reflect diurnal evapotranspiration in the cottonwood-willow forest gallery of the SPRNCA.

The response of the tensiometers to increases in stream stage appears slower than the WCR probes, but the two graphs cannot be compared directly because they represent data from opposing river banks. However, it can be noted from Figure 6 that the 6.0 feet-deep tensiometers (6.0 D and E) - regardless of their lateral proximity to the river - respond more quickly to the rises in river stage than the tensiometers at 6.5 feet depth (6.5 D and F). This may be a function of soil texture, which was recorded during the installation process, but has not yet been factored into the interpretation of the results. Note that tensiometer 6.0 D showed a response to the increased stream stage prior to 6.0 E, which is further from the river by approximately 1.5 meters. A similar response can be noted in tensiometers 6.5 D and 6.5 F, respectively.

4. SUMMARY The response of soil moisture and soil matric potential to increases in stream stage indicate a prolonged increase in bank storage following a high flow event that increases the stream stage by as little as 90 cm. Additionally, the apparent diurnal response of soil moisture and matric potential provide information about the evapotranspiration (ET) patterns in the cottonwood-willow forest gallery. Additional soil moisture data, matric potential data and hydraulic head data, collected in coordination with high flow events, will provide crucial information on the evapotranspiration cycles of the cottonwood and willow tree species within the SPRNCA. This data will be used as input to either UNSAT2 or SWMS_2D to estimate ET, and will improve models that consider bank storage an important variable in the hydrologic cycle.

5. ACKNOWLEDGMENTS Financial support from the US Bureau of Land Management, the USDA-ARS Global Change Research Program and NASA grant W-18, 997 is gratefully acknowledged. Special thanks are extended to the USDA-ARS staff in Tombstone, Arizona for their diligent efforts. We also wish to extend our sincere thanks to the many ARS and University of Arizona staff and students and local volunteers who generously donated their time and expertise to make this project a success.

6. REFERENCES Davis, L.A. and S.P. Neuman. 1983. Documentation and User's guide: UNSAT2 - Variably Saturated Flow Model. NRC FIN B7361. Goodrich, D.C. et al. 1998. An overview of the 1997 activities of the Semi-Arid Land Surface-Atmosphere (SALSA) program (AMS Proceedings, paper no. 1.1). Lacher, L. J. 1996. Recharge characteristics of an effluent dominated stream near Tucson, Arizona. Ph.D. Dissertation, Dept. of Hydrology & Water Res., Univ. Of Ariz., 229 p. Maddock, T., R.D. Mac Nish, D.C. Goodrich, D.G. Williams, W.J. Shuttleworth, B.A. Goff, R.L. Scott, M.S. Moran, D.I. Cooper, L.E. Hipps, A.G. Chebouni. 1998. The overview of atmospheric and surface water coupling to regional groundwater models in semi-arid basins. (1998 AMS Proceedings, paper no. 1.10). Šimçnek, J., T. Vogel and M. Th. van Genuchten. 1994. The SWMS_2D Code for Simulating Water Flow and Solute Transport in Two-Dimensional Variably Saturated Media. V. 1.21. U.S. Salin. Lab., USDA-ARS Riverside, California. Warrick, A., P.J. Wierenga, M.H. Young, and S.A. Musil. Submitted August 1997. Diurnal fluctuations of tensiometric readings due to surface temperature changes. Water Resources Research.

Return to AMS Posters, Table of Contents

return tohomepage