Water and Carbon Fluxes in a Semi-Arid Region Floodplain: Multiple Approaches to Constrain Estimates of Seasonal- and Depth Dependent Fluxes at Rifle, Colorado

Thursday, 18 December 2014: 5:30 PM
Tetsu K Tokunaga1, Jiamin Wan2, Wenming Dong3, Yongman Kim4, Kenneth Hurst Williams5, Mark E Conrad5, John Neil Christensen5, Markus Bill5, Boris Faybishenko5, Chad Hobson5, Richard Dayvault6, Philip E Long7 and Susan S. Hubbard5, (1)Lawrence Berkely Natl Lab, Berkeley, CA, United States, (2)Univ of California, Berkeley, CA, United States, (3)Lawrence Berkeley National Lab, Berkeley, CA, United States, (4)Lawrence Berkeley National Lab, Earth Science Division, Berkeley, CA, United States, (5)Lawrence Berkeley National Laboratory, Berkeley, CA, United States, (6)S. M. Stoller, Grand Junction, CO, United States, (7)Lawrence Berkeley National Lab, Chelan, WA, United States
The importance of floodplains as links between watersheds and rivers highlights the need to understand water and carbon fluxes within floodplain profiles, from their surface soil, through the vadose zone and underlying groundwater. Here, we present results of field and laboratory measurements conducted to quantify fluxes at a remediated uranium/vanadium mill tailings site on a floodplain at Rifle, Colorado. This semi-arid site has a vegetated, locally derived fill soil that replaced the original milling-contaminated soil to a depth of about 1.5 m. The fill soil overlies about 4.5 m of native sandy and cobbly alluvium containing the shallow aquifer. The aquifer generally drains into the Colorado River and is underlain by low permeability Wasatch Formation shale. Within this system, key issues being investigated include water and carbon fluxes between the vadose zone and aquifer, and CO2 fluxes through the vadose zone soil out to the atmosphere. Magnitudes of these fluxes are typically low, thus challenging to measure, yet increasingly important to quantify given the expansion of arid and semi-arid regions under changing climate.

The results of field investigations demonstrated that the annual water table rise and fall are driven by snowmelt runoff into the Colorado River in late spring to early summer. Tensiometer data indicate that net recharge from the deeper part of the vadose zone into groundwater occurs later in summer, after water table decline. The effectiveness of summer evapotranspiration in limiting groundwater recharge is reflected in water potentials decreasing to as low as -3 MPa within the upper 1.5 m of the vadose zone. Examination of the historical precipitation record further indicates that net recharge only occurs in years with above-average precipitation during winter and spring. These short intervals of net recharge also facilitate C transport into groundwater because of higher organic C concentrations in the vadose zone. Fluxes of CO2 measured at the soil surface are consistent with estimates based on season- and temperature-dependent diffusion and respiration within the vadose zone. Thus, fairly predictable seasonal variations in water table elevation, evapotranspiration, and temperature can help constrain estimates of water and carbon fluxes.