Groundwater Age Distributions Research Articles

Using regional groundwater flow models for prediction of regional wellwater quality distributions

► Groundwater flow model interpretations imply groundwater age distributions. ► Taking an example model we compare implied ages and wellwater quality distributions. ► We find broad regional agreement, though some sub-regional disagreement. ► Model age distribution is complex locally and locally sensitive to model assumptions. ► Comparing implied model ages and quality distributions can enhance calibration. This paper evaluates the use of standard regional groundwater flow models in predicting regional patterns of water chemistry based on wellwater samples. The regional groundwater flow in a moderately sized sandstone aquifer has been represented using a time-variant, three layer numerical model typical of those developed for water resource management. Reverse particle tracking was used to determine the implied age distribution of the water in the aquifer and the age structure and recharge location of the wellwaters at the time of sampling. At the regional scale, the model age predictions are broadly consistent with a pre-existing interpretation of wellwater chemistry, though the model suggests a complexity far greater than could be mapped using the 150 wells from which analyses were available. The main features of the chemical distribution pattern are controlled to a significant degree by the pre-abstraction flows within the aquifer, even though heavy pumping has occurred over a period of more than 100 years. The model suggests that some modern recharge is expected almost everywhere across the aquifer, despite its cover of low permeability Quaternary deposits. At individual wells, wide ranges in water age are often predicted, suggesting that water quality prediction for individual wells is likely to be uncertain, providing an explanation for locations where apparently anomalous wellwater chemistry occurs. We conclude that particle tracking using the implied flow field of a standard regional groundwater flow model has proved very worthwhile in the case studied. It has highlighted uncertainties in both flow and regional qualitative water quality interpretations, has confirmed or revised various explanations associated with water quality distributions, indicates that joint chemical and head calibration is advantageous, and suggests that broad predictions of future water quality changes might be attempted using the same approach, though predictions of the chemistry of individual wells is highly uncertain. The aquifer studied is typical of many sandstone aquifers, and we suggest that particle tracking investigations using existing regional-scale multi-layer, transient models might also be undertaken with advantage elsewhere.

Research Spotlight: Modeling groundwater age in large aquifer systems

Groundwater age, the amount of time groundwater has been in an aquifer, is important because the length of time water spends in an aquifer can influence many geologic processes. However, determining groundwater age—which can be as short as a few days or as long as thousands of years—can be complicated because groundwater basins are large and complex and groundwater age distribution is affected by mixing and transport of different parcels of water.

Simultaneous rejuvenation and aging of groundwater in basins due to depth‐decaying hydraulic conductivity and porosity

The age of groundwater is a manifestation of the temporal scale of groundwater flow in basins, whose pattern was recently found to be influenced by depth‐dependent hydraulic conductivity (K). In this paper, we show through numerical simulations how well‐documented depth‐decaying K and porosity (θ) influence groundwater age. In the unit basin, depth‐decaying K and θ cause aging in deeper parts and rejuvenation near the discharge zones, and the size of rejuvenated zones decreases with the decay exponent (A). In the Tóth basin, the geometry and size of rejuvenated zones, which are generally located at the interfaces between flow systems in the mid to lower reaches of the basin, are sensitive to A. In both basins, the maximum relative age and the relative age of groundwater at the lowest discharge point are dependent on A. Therefore, the depth‐decaying K and θ cannot be ignored when interpreting groundwater age distribution.

Relationships among groundwater age, denitrification, and the coupled groundwater and nitrogen fluxes through a streambed

The relationships among coupled groundwater and nitrogen (N) fluxes, groundwater age, and denitrification were examined for a section of West Bear Creek, an agricultural stream in the coastal plain of North Carolina, United States. Simultaneous streambed measurements of hydraulic conductivity (K) and hydraulic head gradient (J) and the concentrations of NO3− ([NO3−]), dissolved gases, and chlorofluorocarbons in groundwater were interpolated, mapped, and (for water flux v = KJ and nitrate flux fNO3 = v[NO3−]) integrated over the streambed area. Nitrate and dissolved organic N accounted for 92 and 8% of N flux through the streambed, respectively. Streambed maps show a band of greater groundwater age, and lower [NO3−] and fNO3, running through the center of most of the study reach. Nitrate flux (fNO3) exhibits this “center‐low” pattern even though one of its controlling factors, groundwater flux (v), has on average the opposite “center‐high” pattern. An inverse relationship between [NO3−] and age is indicative of fertilizer as the primary source of groundwater NO3−. Denitrification reduced mean fNO3 by ∼50%, from 370 mmol m−2 d−1 (what it would have been in the absence of denitrification) to 173 mmol m−2 d−1 (what it actually was). Measurement of both groundwater age and v made possible a new method for estimating flow‐weighted mean groundwater age (τFWM), an important aquifer hydraulic characteristic related to groundwater storage and recharge rate. This method gives τFWM = 30 years, which, along with the overall distribution of groundwater ages, suggests the possibility of a significant time lag between changes in N fertilizer application rates and NO3− flux from groundwater to West Bear Creek. Differences in streambed groundwater chemistry between the left and right sides of the streambed suggest differences in agricultural practices on opposite sides of the stream.

Age-distribution estimation for karst groundwater: Issues of parameterization and complexity in inverse modeling by convolution

summary Convolution modeling is useful for investigating the temporal distribution of groundwater age based on environmental tracers. The framework of a quasi-transient convolution model that is applicable to twodomain flow in karst aquifers is presented. The model was designed to provide an acceptable level of statistical confidence in parameter estimates when only chlorofluorocarbon (CFC) and tritium ( 3 H) data are available. We show how inverse modeling and uncertainty assessment can be used to constrain model parameterization to a level warranted by available data while allowing major aspects of the flow system to be examined. As an example, the model was applied to water from a pumped well open to the Madison aquifer in central USA with input functions of CFC-11, CFC-12, CFC-113, and 3 H, and was calibrated to several samples collected during a 16-year period. A bimodal age distribution was modeled to represent quick and slow flow less than 50 years old. The effects of pumping and hydraulic head on the relative volumetric fractions of these domains were found to be influential factors for transient flow. Quick flow and slow flow were estimated to be distributed mainly within the age ranges of 0–2 and 26–41 years, respectively. The fraction of long-term flow (>50 years) was estimated but was not dateable. The different tracers had different degrees of influence on parameter estimation and uncertainty assessments, where

Discriminant analysis for estimation of groundwater age from hydrochemistry and well construction: application to New Zealand aquifers

Concentrations of tritium, chlorofluorocarbons and sulfur hexafluoride have been measured at over 100 groundwater monitoring sites across New Zealand, followed by interpretation of groundwater age distribution using the exponential-piston flow model. Interpreted mean residence times ranged from less than 1 year to more than 100 years, with the 25th, 50th (median) and 75th percentiles being approximately 10, 40 and 100 years, respectively. Classification functions derived from discriminant analysis and based on nine input variables (well depth, electrical conductivity and the concentrations of the ions Na, K, Ca, Mg, HCO3, Cl and SO4) allowed assignment of 71% of the sites to the correct of four age categories (mean residence time 10 years or less, 11–40 years, 41–100 years, or more than 100 years). The discriminant analysis classification functions were more effective than regression methods for estimating groundwater age from hydrochemistry and well depth, and can thus be used to predict the groundwater age category for any monitoring site in New Zealand.

Using Environmental Tracers in Modeling Flow in a Complex Shallow Aquifer System

Using environmental tracers in groundwater dating partly relies on the assumption that groundwater age distribution can be described analytically. To investigate the applicability of age dating in complex multiaquifer systems, a methodology for simulating well specific groundwater age distribution was developed. Using a groundwater model and particle tracking we modeled age distributions at screen locations. By enveloping modeled age distributions and estimated recharge concentrations, environmental tracer breakthroughs were simulated for specific screens. Simulated age distributions are of irregular shapes and sizes without being similar to the assumed age distributions used in the analytical approach. The shape of age distribution to some extent depends on sampling size and on whether the system is modeled in a transient or in a steady state, but shape and size were largely driven by the heterogeneity of the model and by topographical variations as well. Analytically derived groundwater ages are dependent on sampling time. This time dependence relates to the nonlinearity of recharge concentrations and the shape and size of age distribution that has no coherence with the simplified assumptions of traditional approaches. Accordingly, constraining flow models by “age observations” may lead to misrepresentations that are biased depending on sampling time. If environmental tracers are used directly in terms of concentrations instead of ages, spatial as well as temporal variations become useful in constraining models.

Radiocarbon age distribution of groundwater in the Konya Closed Basin, central Anatolia, Turkey

Annual abstraction of 2.6 × 109 m3 of groundwater in the 53,000 km2 Konya Closed Basin of central Turkey has caused a head decline of 1 m/year over the last few decades. Therefore, understanding the hydrogeology of this large endorheic basin, in a semi-arid climate, is important to sustainable resource management. For this purpose, the groundwater’s radiocarbon age distribution has been investigated along a 150-km transect parallel to regional flow. Results show that the groundwater ranges in age from Recent at the main recharge area of the Taurus Mountains in the south, to about 40,000 years around the terminal Salt Lake located in the north. In this predominantly confined flow system, radiocarbon ages increase linearly by distance from the main recharge area and are in agreement with the hydraulic ages. The mean velocity of regional groundwater flow (3 m/year) is determined by the rate of regional groundwater discharge into the Salt Lake. Calcite dissolution, dedolomitization and geogenic carbon dioxide influx appear to be the dominant geochemical processes that determine the carbon isotope composition along the regional flow path. The groundwater’s oxygen-18 content indicates more humid and cooler paleorecharge. A maximum drop of 5°C is inferred for the past recharge temperature.

Potential effects of regional pumpage on groundwater age distribution

Groundwater ages estimated from environmental tracers can help calibrate groundwater flow models. Groundwater age represents a mixture of traveltimes, with the distribution of ages determined by the detailed structure of the flow field, which can be prone to significant transient variability. Effects of pumping on age distribution were assessed using direct age simulation in a hypothetical layered aquifer system. A steady state predevelopment age distribution was computed first. A well field was then introduced, and pumpage caused leakage into the confined aquifer of older water from an overlying confining unit. Large changes in simulated groundwater ages occurred in both the aquifer and the confining unit at high pumping rates, and the effects propagated a substantial distance downgradient from the wells. The range and variance of ages contributing to the well increased substantially during pumping. The results suggest that the groundwater age distribution in developed aquifers may be affected by transient leakage from low‐permeability material, such as confining units, under certain hydrogeologic conditions.

Residence times and age distributions of spring waters at the Semmering catchment area, Eastern Austria, as inferred from tritium, CFCs and stable isotopes

The groundwater system in the mountainous area of Semmering, Austria, was studied by environmental tracers in several karst springs. The tracers used included stable isotopes (18O, 2H), tritium (3H) and chlorofluorocarbons (CFCs). The tracers provided valuable information in regard to (1) the mean altitude of the spring catchment areas; (2) the residence time and age distribution of the spring waters; and (3) the interconnection of the springs to a sinkhole. The combination of the stable isotopic data and the topography/geology provided the estimates of the mean altitudes of the catchment areas. Based on the stable isotopic data the recharge temperature of the spring waters was estimated. The smoothing of precipitation's isotopic signal in spring discharge provided information on the minimum transit time of the spring waters. Due to short observation time, 3H data alone cannot be used for describing the mean residence time of the karst waters. CFCs, though useful in recognizing the co-existence of young (post-1993) water with old (CFC-free) water, could not be used to resolve age distribution models. It is shown in this article, however, that the combined use of tritium and CFCs can provide a better assessment of models to account for different groundwater age distributions. In Appendix A, a simplified method for collecting groundwater samples for the analysis of CFCs is described. The method provides a real facilitation for fieldwork. Test data are given for this sampling method in regard to potential contamination by atmospheric CFCs.

Constraining the age distribution of highly mixed groundwater using 39Ar: A multiple environmental tracer (3H/3He, 85Kr, 39Ar, and 14C) study in the semiconfined Fontainebleau Sands Aquifer (France)

A multitracer (3H/3He, 85Kr, 39Ar, and 14C) approach is used to investigate the age structure of groundwater in the semiconfined Fontainebleau Sands Aquifer that is located in the shallower part of the Paris Basin (France). The hydrogeological situation, which is characterized by spatially extended recharge, large screen intervals, and possible leakage from deeper aquifers, led us to expect a wide range of residence times and pronounced mixing of different water components. Consequently, a large set of tracers with corresponding dating ranges was adopted. Commonly used tracers for young groundwater (3H, 3He, and 85Kr) can identify only those components with ages below 50 years. This approach is reliable if a large fraction of the water recharge occurs within this period. However, if a considerable fraction is older than 50 years, a tracer that covers intermediate age ranges below 1000 years is needed. We examine the use of 39Ar, a noble gas radioisotope with a half‐life of 269 years, to constrain the age distribution of groundwater in this timescale range. Recharge rate, depth of water table, and the age structure of the groundwater are estimated by inverse modeling. The obtained recharge rates of 100–150 mm/yr are comparable to estimations using hydrograph data. Best agreement between the modeled and measured tracer concentrations was achieved for a thickness of the unsaturated soil zone of 30–40 m, coinciding well with the observed thicknesses of the unsaturated zone in the area. Transport times of water and gas from the soil surface to the water table range between 10 and 40 and 1 and 6 years, respectively. Reconstructed concentrations of 85Kr and 3H at the water table were used for saturated flow modeling. The exponential box model was found to reproduce the field data best. Conceptionally, this finding agrees well with the spatially extended recharge and large screened intervals in the project area. Best fits between model and field results were obtained for mean residence times of 1–129 years. The 39Ar measurements as well as the box model approach indicate the presence of older waters (3H and 85Kr free). Using 39Ar to date this old component resulted in residence times of the old water components on the order of about 100–400 years. The 14C measurements provide additional evidence for the correctness of the proposed age structure.

Effects of intraborehole flow on groundwater age distribution

Environmental tracers are used to estimate groundwater ages and travel times, but the strongly heterogeneous nature of many subsurface environments can cause mixing between waters of highly disparate ages, adding additional complexity to the age-estimation process. Mixing may be exacerbated by the presence of wells because long open intervals or long screens with openings at multiple depths can transport water and solutes rapidly over a large vertical distance. The effect of intraborehole flow on groundwater age was examined numerically using direct age transport simulation coupled with the Multi-Node Well Package of MODFLOW. Ages in a homogeneous, anisotropic aquifer reached a predevelopment steady state possessing strong depth dependence. A nonpumping multi-node well was then introduced in one of three locations within the system. In all three cases, vertical transport along the well resulted in substantial changes in age distributions within the system. After a pumping well was added near the nonpumping multi-node well, ages were further perturbed by a flow reversal in the nonpumping multi-node well. Results indicated that intraborehole flow can substantially alter groundwater ages, but the effects are highly dependent on local or regional flow conditions and may change with time.

Comment on “Groundwater age, life expectancy, and transit time distributions in advective–dispersive systems: 1. Generalized reservoir theory,” by F. Cornaton and P. Perrochet

Comment on “Groundwater age, life expectancy, and transit time distributions in advective–dispersive systems: 1. Generalized reservoir theory,” by F. Cornaton and P. Perrochet

Reply to “Comment on groundwater age, life expectancy and transit time distributions in advective–dispersive systems: 1. Generalized reservoir theory” by Timothy R. Ginn

We thank T.R. Ginn for his interest in our recently pub-lished article [1] on the subject of groundwater age model-ling and reservoir theory. In his previous comment [4], T.R.Ginn expresses concern about some conceptual inconsis-tencies in the formulations presented in our work. We basi-cally agree with the fundaments of his comments, and wewish to continue the discussion.The first part of T.R. Ginn’s comment relates that ourdefinition of the groundwater age pdf, g

Groundwater age, life expectancy and transit time distributions in advective–dispersive systems: 1. Generalized reservoir theory

We present a methodology for determining reservoir groundwater age and transit time probability distributions in a deterministic manner, considering advective–dispersive transport in steady velocity fields. In a first step, we propose to model the statistical distribution of groundwater age at aquifer scale by means of the classical advection–dispersion equation for a conservative and non-reactive tracer, associated to proper boundary conditions. The evaluated function corresponds to the density of probability of the random variable age, age being defined as the time elapsed since the water particles entered the aquifer. An adjoint backward model is introduced to characterize the life expectancy distribution, life expectancy being the time remaining before leaving the aquifer. By convolution of these two distributions, groundwater transit time distributions, from inlet to outlet, are fully defined for the entire aquifer domain. In a second step, an accurate and efficient method is introduced to simulate the transit time distribution at discharge zones. By applying the reservoir theory to advective–dispersive aquifer systems, we demonstrate that the discharge zone transit time distribution can be evaluated if the internal age probability distribution is known. The reservoir theory also applies to internal life expectancy probabilities yielding the recharge zone life expectancy distribution. Internal groundwater volumes are finally identified with respect to age and transit time. One- and two-dimensional theoretical examples are presented to illustrate the proposed mathematical models, and make inferences on the effect of aquifer structure and macro-dispersion on the distributions of age, life expectancy and transit time.

Groundwater age, life expectancy and transit time distributions in advective–dispersive systems; 2. Reservoir theory for sub-drainage basins

Groundwater age and life expectancy probability density functions (pdf) have been defined, and solved in a general three-dimensional context by means of forward and backward advection–dispersion equations [Cornaton F, Perrochet P. Groundwater age, life expectancy and transit time distributions in advective–dispersive systems; 1. Generalized reservoir theory. Adv Water Res (xxxx)]. The discharge and recharge zones transit time pdfs were then derived by applying the reservoir theory (RT) to the global system, thus considering as ensemble the union of all inlet boundaries on one hand, and the union of all outlet boundaries on the other hand. The main advantages in using the RT to calculate the transit time pdf is that the outlet boundary geometry does not represent a computational limiting factor (e.g. outlets of small sizes), since the methodology is based on the integration over the entire domain of each age, or life expectancy, occurrence. In the present paper, we extend the applicability of the RT to sub-drainage basins of groundwater reservoirs by treating the reservoir flow systems as compartments which transfer the water fluxes to a particular discharge zone, and inside which mixing and dispersion processes can take place. Drainage basins are defined by the field of probability of exit at outlet. In this way, we make the RT applicable to each sub-drainage system of an aquifer of arbitrary complexity and configuration. The case of the well-head protection problem is taken as illustrative example, and sensitivity analysis of the effect of pore velocity variations on the simulated ages is carried out.

Groundwater ages in fractured rock aquifers

In fractured porous media, matrix diffusion processes mean that groundwater ages obtained with environmental tracers usually do not reflect the hydraulic age of the water. The distribution of groundwater ages within these heterogeneous systems will be related to the groundwater velocity within the fractures, but also to the size of the fractures and the geometry of the fracture network, and to the hydraulic properties of the aquifer matrix. In this paper, we present analytical and numerical simulations of environmental tracer concentrations in fractured rock aquifers to examine the effect of changes in aquifer parameters on the tracer distributions. In particular, we show that where horizontal fractures are strongly vertically connected, then it may be reasonable to use one-dimensional models of flow and transport through vertical fractures to represent flow through aquifers containing both horizontal and vertical fractures. The presence of large numbers of horizontal fractures will not cause flow to depart significantly from the one-dimensional approximation. Where a smaller number of horizontal fractures are present, then abrupt decreases in the vertical water velocity can occur, as water is intercepted and diverted laterally. Measurements of 14C, 3H, 36Cl, and chlorofluorocarbons within nested piezometers from the Clare Valley, South Australia, display a number of the features apparent in the generic simulations. The use of a number of different tracers appears to allow some fracture and matrix parameters to be constrained more tightly than might previously have been thought possible.

The spatial distribution of groundwater age for different geohydrological situations in the Netherlands: implications for groundwater quality monitoring at the regional scale

The spatial distribution of groundwater age is a key factor determining the distribution of dissolved contaminants in the subsurface when contamination loadings have increased in time. The effects of surficial drainage and aquifer heterogeneity on the spatial distribution of groundwater age in unconsolidated aquifers in flat areas were investigated, and consequences were presented for the monitoring of contaminants from diffuse sources. First, the effects were assessed using model simulations. Second, the groundwater age distribution was evaluated in the two regional monitoring networks of Noord-Brabant and Drenthe using tritium measurements. Theoretically, a simple spatial distribution of groundwater age is present in homogeneous aquifers with natural groundwater recharge, characterized by a horizontal pattern of residence time isochrones and a gradual increase of groundwater age with depth. The model simulations show that the isochrone pattern becomes distorted in areas with a surficial drainage network, resulting in relatively old groundwater at shallow depth and larger spatial variation in groundwater age at a specific depth. This drainage effect on the spatial distribution of groundwater age is relatively large compared with effects of regional scale aquifer heterogeneity or spatially varying groundwater recharge. The effects of surficial drainage on the spatial distribution of groundwater age were confirmed by tritium measurements made in the regional monitoring networks of two provinces in the Netherlands. At about 24 m depth, the proportion of post-1950 groundwater in drained areas was significantly less and the spatial variation of groundwater age was larger than in recharge areas that lack a drainage network. The age of the groundwater appeared to be related to the drainage network density and the water table regime. A preliminary survey showed that contamination patterns in the two networks agree well with the proportion of post-1950 groundwater.

Dispersion of groundwater age in an alluvial aquifer system

Interpretation of groundwater ages typically rests on assumptions of minimal mixing of different water ages in the water samples. The effects of three‐dimensional, geologic heterogeneity on groundwater mixing and tracer concentrations, however, have not been evaluated. In this study, we use a series of 10 detailed geostatistical realizations along with high‐resolution numerical groundwater flow and contaminant transport simulation to model distributions of groundwater ages and chlorofluorocarbon (CFC) ages at wells within a heterogeneous stream‐dominated alluvial fan aquifer system. Results show that groundwater reaching a well in the heterogeneous aquifer system typically consists of a wide distribution of groundwater ages (often spanning >50 years), even over short (<1.5 m) screened intervals. Additionally, simulated arithmetic mean groundwater ages do not correspond to mean ages estimated from simulated CFC concentrations. Results emphasize the potential ambiguity of “mean” groundwater ages estimated from environmental tracer concentrations in typically heterogeneous geologic systems. The significant dispersion of groundwater ages also implies that ultimate, maximum effects of nonpoint source, anthropogenic contamination of groundwater may not be reached until after many decades or centuries of gradual decline in groundwater quality.

Evaluation of mean residence time in subsurface waters using oxygen-18 fluctuations during drought conditions in the mid-Appalachians

Seasonal variations of the stable isotope, oxygen-18 ( 18O), were used to estimate mean residence times (MRTs) of water in two Valley and Ridge Province watersheds in central Pennsylvania. MRT was estimated by applying 18O input precipitation signatures to response function models that describe flow conditions in the subsurface system. Precipitation 18O is usually seasonally periodic; however, during this study period (March 1998 to June 1999), unusual weather conditions and severe drought caused an abnormal seasonal signature in precipitation 18O. The atypical input precipitation amounts and 18O signatures required that adjustments be made using recharge factors and an extended past input function to represent the varying fraction of input water that contributed to turnover within the watershed during the study period. Oxygen-18 variations were investigated in output waters from tension and zero-tension soil water lysimeters, shallow wells, and streamflow. Soil water 18O signatures were more responsive to precipitation 18O variations than streamflow 18O signatures. Theoretical models based on exponential piston-flow and dispersion flow processes fit data better than did other groundwater age distribution models (i.e. pure piston flow or pure exponential models). Analysis suggested that during drought conditions, larger portions of older water dominated streamflow 18O composition, and that the current year's precipitation 18O signature became more attenuated in the subsurface flow system. MRTs obtained for streamflow at each site were 9.5 and 4.8 months for a 14 and 100 ha watershed, respectively, and soil water MRT at 100 cm depth was on the order of a couple months; indicating a relatively rapid response of shallow groundwater to precipitation.