4.4 Groundwater residence times and travel times

The definitions of travel time, residence time and groundwater age provided by other Groundwater Project books (e.g., “Isotopes and environmental tracers as indicators or water sources and flow rates”) and are discussed in detail there. Nevertheless, they are worth reintroducing here because the terms are commonly used interchangeably to describe the time a parcel of water spends in the saturated zone, and that time can be estimated as an advective travel time from the point of recharge to the sampling point, i.e., a time that depends on groundwater velocity. In the context of contaminant fate, transport and remediation, the term ‘residence time’ might also be used to refer to the time a parcel of water spends in a treatment zone within some limited volume of an aquifer. Regardless, the time that a parcel of water spends in the ground can be immensely important for the fate of its chemical constituents and is strongly affected by groundwater velocity. In keeping with the example of a pollutant as a groundwater constituent, this is true both from the standpoints of 1) pollutant abatement by biotic or abiotic degradation reactions that require a minimum residence time in the aquifer to progress toward completion, and 2) processes that affect the transport-and-storage of pollutants in variably permeable geologic materials. It is useful to discuss these concepts in the context of flow systems, which are — even in the simplest examples — composed of regions of relatively fast and slow flow rates (Figure 21).

The concept of residence time, as it relates to a flow system, can be further illustrated for an advection dominated flow system by recalculating the particle tracks in Figure 21d for a range of travel times and recording the ages of the particles at each final coordinate location. Contours of these ages are referred to as isochrons, and for the simplistic model in Figure 21d the isochrons reveal that the oldest water is deepest, and the youngest water is shallow. Moreover, these ages are layered horizontally (Figure 22).

The connection between residence times and chemical transformations exists for any substance that is chemically active, i.e., out of equilibrium. For example, consider a pollutant that undergoes a transformation with a half-life of 7.5 days and suppose that the path-lines in Figure 21d represent 30-day travel times, i.e. 4 half-lives of the pollutant. This residence time is sufficient to reduce the pollutant concentration to about 1/16th of its original value. For practical purposes, 7 half-lives provide a time period sometimes considered sufficient for ‘complete’ degradation because it reduces the concentration to less than 1% of the original value. Now imagine that four parcels of water containing the pollutant are released on the surface at various locations along the flow system. The track-lines shown in Figure 21d indicate that parcel 4 will travel more than 16 m horizontally and nearly 15 m vertically, reaching the right boundary before the pollutant is degraded to target levels. On the other hand, parcel 1 with about the same 15 m vertical distance travelled but covering no appreciable horizontal distance over the same time period, will experience full degradation by the time it reaches the boundary, about 135 days (~7 half-lives) after release. In this scenario, it might be concluded that there is greater risk associated with the movement of parcel 4 than parcel 1, largely because of its higher velocity. However, this conclusion may change for cases in which the pollutant has longer reaction times or is non-reactive. In that scenario, dissolved contamination from a short-lived spill (instantaneous source) would be flushed from the system more quickly along track 4 than track 1, attaching the greater risk to parcel 1 because of the persistence of the pollutant. Once again, the importance of a detailed characterization of the flow system, particularly through knowledge of groundwater velocities, is fundamental to assessing the fate, transport and risk associated with chemicals carried by groundwater.

A Cross-section of a simple, homogeneous flow system with a flow divide
Figure 21 – a) A Cross-section of a simple, homogeneous flow system with a flow divide (no water crosses a divide) on the left and an impermeable base layer. Water recharges from the surface and exits the system at the right boundary. b) Contoured water levels (equipotential lines as elevations in meters) in the aquifer; closer spacing of the lines indicates faster flow. c) Streamlines showing the paths of groundwater flow, and points (red dots) with velocity vectors (blue lines) showing direction of flow and scaled so lines are proportional to the water speed. d) Path lines of four parcels of water originating from different locations in the flow system and traveling for the same period of time. Note that because of its location in the flow system, parcel 4 travels a greater overall distance in the same period of time than parcel 1 — this is particularly evident in terms of the horizontal distances travelled.
Cross-sectional view of part of a 30 m by 30 m hypothetical aquifer
Figure 22 – Cross-sectional view of part of a 30 m by 30 m hypothetical aquifer with 30-day particle tracks and colored isochrons displayed. The calculations performed for this image are approximate, based on the four particles shown and six different times (5, 10, 20, 25, and 30 days), but the near horizontal layering of groundwater ages (or residence times) is also observed in more sophisticated models.

To gain insight into groundwater residence times, hydrogeologists have sometimes relied on the sampling and analysis of tracers. An especially useful family of tracers are the radioactive isotopes. When these are released into the groundwater, either by design or by accident, they serve as tracers with on-board clocks that can be used to estimate water residence times. Isotopes that might be used this way include tritium (3H, = 12.5 years), carbon-14 (14C, = 5730 years) or strontium-90 (90Sr, = 28.8 years) as discussed by Cook (2020). If there is knowledge of the isotope concentration at the time of its introduction to the ground then, in principle, any subsequent determination of its concentration will be enough to calculate how long it has been there. This assumes ideal conditions, in which other processes that contribute to declines in isotope concentration can be assumed minimal (e.g., dilution or sorption). Knowing the distance between the release location and the sampling location, an average velocity and total residence time (i.e., time between release and sampling) can be estimated. The applicability of this method is strengthened if isotope ratios of parent to daughter compounds are considered. Isotopes have been employed these ways as in situ indicators of plume-scale velocities. However, a problem arises when the aquifer being investigated is heterogeneous. Here, the relationship between residence time and transport-and-storage becomes important.

In heterogeneous media — we will consider the case of interbedded sands and clays here, but the following discussion is relevant to other combinations of geologic materials — groundwater will deliver pollutants to the more permeable sand zones first, because the water travels faster there, setting up concentration gradients at the boundaries between the sand and clay (Figure 23). Flow rates in clays are very small and so the transport mechanism that dominates in these materials is diffusion. As a pollutant, or isotope tracer, passes through the sandy material, its plume continuously loses mass to the clays by diffusion, effectively storing mass in the clays until the concentration gradients are reversed. The loss of pollutant mass from the sand-bound plume has the effect of slowing the forward movement of the plume, i.e., contours of a specified concentration advance less rapidly than they would in a homogeneous aquifer. This is important for at least two reasons: first, knowledge of the seepage velocity from Darcy calculations may overestimate the tracer velocities and predict arrival times at receptors that are unrealistically short (note: in cases of severe heterogeneity, the opposite error could occur — strong channelized flow might result in Darcy calculations that underestimate seepage velocities, as illustrated in Figure 23c); second, this mechanism can lead to greater than anticipated residence times for pollutants, potentially fostering greater degrees of degradation if the pollutant is reactive, or simply increasing its time in the ground if the pollutant has low reactivity. Either way, the pollutant mass that would have contributed to a growing plume instead collects in the ground and may be difficult and expensive (with respect time and/or money) to recover in a remediation program.

Conceptualization of plumes being transported in homogeneous and heterogeneous media
Figure 23Conceptualization of plumes being transported in a) a homogeneous sandy medium, b) a heterogeneous medium consisting of sand and clay, and c) a second heterogeneous medium consisting of gravel and clay. Compared to the homogeneous medium, the other two constitute transport in two regions: predominantly advective transport in the sand and gravel layers and mainly diffusive transport in the clay. The effect of the diffusive transport is to store mass and retard the plume compared to the homogenous case. If the clay is coupled with highly permeable sediment (perhaps not known to be present), such as gravel, the advective transport can proceed much faster, overwhelming any retardation that might be expected for the sand-clay coupling


Groundwater Velocity Copyright © 2020 by J.F. Devlin. All Rights Reserved.