Estimating Riparian Zone Evapotranspiration from Groundwater Level Fluctuations: Implication of River Stage

Groundwater evapotranspiration (ET) by phreatophytes in riparian zones is an important component of the water budget of many arid and semiarid environments. Diurnal fluctuations in groundwater levels have been used to estimate groundwater ET, but interference between nearby surface water (e.g., rivers, streams) and groundwater levels complicates the ET estimate. The objective of this study is to examine the influence of surface water level changes on the ET estimates in a riparian environment, thereby elucidating the potential caveats of using groundwater level diurnal fluctuations to estimate groundwater ET in riparian zones. The objective is accomplished using spectral analyses and descriptive statistics of time series data from groundwater monitoring wells and nearby river stages. The groundwater level and river stage data were monitored in a stand of Tamarisk Ramossisima on the Colorado River near Blythe, California, USA. Tamarisk Ramossisima is an obligate phreatophyte that requires direct access to groundwater for its source of water. Recently, several federal agencies are focusing efforts on eradicating or otherwise controlling Tamarisk. Results show that the most significant diurnal signal from the measurement of groundwater levels of monitoring wells can be attributed to ET; however, distance of the groundwater level monitoring wells to the river directly impacts our ability to separate the different signals from ET and river stage changes. For groundwater levels monitored near the river, the spectral analysis in both time and frequency domains gives strong indications of sensitive variations of groundwater fluctuation to water stage in the river. For wells closer to the river, although diurnal fluctuations also exist during high ET season, other site conditions were found to interfere with the fluctuations. As a result, ET estimates based on diurnal fluctuations of groundwater levels from wells closer to the river would have large errors.