groundwater tracer experiment

5.9 Stream Tracer Methods

Tracer techniques can be used to identify the pathways, timing and quantities of exchange between streams and groundwater and/or hyporheic zones. Tracers can also be used to determine discharge in small streams (e.g., Kilpatrick and Cobb, 1985). Commonly, tracers are introduced into streams to characterize hyporheic exchange, i.e., the transfer of surface water in and out of the immediately adjacent bed, bank and floodplain (Figures 29 and 30 ).

Tracer studies introduce a known mass of tracer, for example, a salt such as sodium chloride (NaCl), an ion like bromide (Br) and/or an organic dye such as Rhodimine-WT at an upstream location and then, over time, monitor the arrival of the tracer at a downstream location. The method is most manageable when stream discharges are small (<0.5 m 3 /s). The larger the discharge the more tracer is needed to generate a measurable change in the tracer concentration in the stream. Results are typically shown as plots of concentration (or conductivity) versus time at various locations along the stream reach. The plots are referred to as breakthrough curves (Figure 80).

Tracer breakthrough curves collected at the downstream monitoring point are assessed using basic transport and storage theory expressed through analytical and numerical models as described in Box 8 . Observed curves are compared to theoretical transport conditions (Figure 80). The tracer transport can be impacted by in-channel delays, the input of groundwater, and temporary storage in the channel or hyporheic system. Observed tracer concentrations also reflect the natural spreading and diffusion that occurs in the stream. The data are typically evaluated using one-dimensional analytical models to determine the degree of exchange between the stream and/or the hyporheic zone and groundwater. The field breakthrough curves are matched with parameters that fit results to expected conditions without loss of mass or delay of breakthrough. Cardenas (2015) describes several methods used for data analyses. The OTIS (One-dimensional Transport with Inflow and Storage) software code is often used to assess breakthrough data sets and generate exchange values and components (Runkel, 1998). Data analyses are complicated by the possibility that more than one factor may influence the observed breakthrough curve.

In some studies, monitoring wells located in the floodplain and hyporheic zones and mini-piezometers in the stream channel are sampled during a channel tracer experiment to examine the locations, pathways and rates of circulation of stream water tracer into the hyporheic zone (e.g., Cardenas, 2015; Boana et al., 2014).

Designing a tracer test includes selecting the tracer and determining: the method of input; needed volume and concentration of tracer; and cost of the tracer analysis, and permits or permissions needed to introduce and monitor the tracer. Tracers and detection methods should be inexpensive; tracers should occur in low concentrations in background surface water and groundwater, be non-reactive (conservative), and non-toxic. When background concentrations are low, relatively inexpensive tracer tests include introducing: NaCl accompanied by monitoring of changes in specific conductance using a conductivity meter; Br measured with a specific ion probe; or organic dye tracers such as Rhodamine-WT monitored with a fluorimeter. It is useful to make an estimate of the amount of tracer that will be needed to create a measurable change in constituent concentration at the downstream site. Estimates of the stream concentration of the tracer can be computed using a mixing model that considers a slug or continuous-source input and the fully-mixed concentration once the dye is distributed in the streamflow. Some empirically derived equations have been suggested for slug input of a tracer (e.g., Kilpatrick, 1970). The best approach is to compute tracer inputs and then run a preliminary tracer test to see if concentrations and volumes need adjustment. Experience has shown that attempts at estimating tracer inputs are often poorly constrained.

Batchelor and Gu (2014) investigated how a conservative bromide tracer and a reactive nitrate tracer behaved in two small creeks and the corresponding hyporheic zones of a stream in a North Carolina, USA. At Winkler Creek (mean annual discharge 0.2 m 3 /s), a continuous injection of tracer designed to raise the concentrations at the downstream observation point by 1 to 2 mg/l of bromide and 2 to 3 mg/l of nitrate above background concentrations was initiated. The tracers were injected at 30 L/h, and breakthrough concentrations at a location 50 m downstream were reported (Figure 81). The OTIS code (described in Box 7) was used to examine storage in the stream system. Batchelor and Gu observed only small changes in the breakthrough curve and estimated that 1.05% of the streamflow over a 200 m reach of the creek entered transient storage in the channel/hyporheic zone based on analysis of the breakthrough curves observed at the downstream monitoring points (Figure 81).