Introduction

Groundwater quality can be affected by both natural and anthropogenic factors, including evolution of climate. In order to quantify and evaluate these changes in groundwater quality appropriate methodology is required.

Groundwater is generally in perpetual motion and its physico-chemical composition evolves in space and time, adapting to changing geological environment as well as physico-chemical conditions along the flowpath. This changing geological environment can differ in terms of mineral and chemical composition, the presence of organic substances, etc., both at micro and macro scale. Groundwater movement

Faculty of Geology, Geophysics and Environmental Protection, AGH University of Science and Technology, al. Mickiewicza 30, 30-059 Krakow, Poland e-mail: [email protected]

K. Rozanski

Faculty of Physics and Applied Computer Sciences, AGH University of Science and Technology, al. Mickiewicza 30, 30-059 Krakow, Poland is usually slow enough (flow velocity ranges from few to hundreds of meters per year), to allow for chemical quasi-equilibrium controlled mainly by lithology and environmental conditions (e.g. pH, redox). This might not be true for karstic aquifers, where flow velocities of water could be much higher.

27.2 Hydrogeochemical Field

The variability of geological environment together with changes of physico-chemical conditions and the slow movement of groundwater are the main factors leading to formation of the spatially variable hydrogeochemical field in the groundwater reservoir (aquifer). The term "hydrogeochemical field" was introduced by Stilmark [1]. It defines the spatio-temporal distribution of physico-chemical properties and concentration of dissolved constituents in the given groundwater system.

Regions with sharp spatial changes of the hydrogeochemical field are usually associated with regions of strong changes of the redox conditions. Example of such changes is shown in Fig. 27.1 where iron concentration in the vertical profile representing shallow, unconfined aquifer in the Vistula River valley increases from less than 1 mg/L up to 50 mg/L across the distance of ca. 7 m. This abrupt change is clearly linked to changes of Eh and pH. The wide range of the measured changes of Fe concentrations is comparable with the concentration range of this element being observed in Quaternary aquifers in Poland.

In shallow, unconfined aquifers the variability of the hydrogeochemical field is mainly linked to anthropogenic influences of various nature. Figure 27.2 shows an

Aquifers Poland
Fig. 27.1 Changes of Fe, pH and Eh in vertical profile in the Quaternary deposits of the Vistula River valley [2]
Fig. 27.2 Modelled distribution of chlorides in groundwater along a road subject to deicing during winter (example of MT3D results)

example of the influence of anthropogenic activities on the surface (road deicing) on the concentration of chloride in near-by groundwater system.

Under natural conditions, the hydrogeochemical field is usually variable in space but stationary or quasi-stationary in time. Groundwater chemistry may reveal some seasonal variations but typically it does not show any persistent temporal trend. Example of the spatial variability of hydrogeochemical field is presented in Fig. 27.3 which shows the distribution of calcium concentration in small Neogene sandy aquifer in southern Poland. This kind of spatial variability of the hydrogeo-chemical field is typical for most of major ions in confined aquifers.

As a rule, contamination of groundwater or strong changes of the hydrodynamic field result in temporal evolution of the hydrogeochemical field. Example of long-term temporal variability of the hydrogeochemical field is shown in Fig. 27.4, which presents changes of Fe concentration in groundwater pumped from a well-field located in south-western Poland, over the period from 1965 to 2000. High withdrawal of water and lowering of water table in the late 1960s induced the transition from reducing to oxidizing conditions. The resulting oxidation of pyrite (iron sulphides) in this groundwater system, lead to increase of Fe concentration. Subsequently, due to depletion of pyrite reserves in the oxidised zone, Fe concentration gradually decreased and the system reached new, quasi-equilibrium stage with respect to this element.

556 560 565 570 575 580 562

556 560 565 570 575 580 562

Variogram model: sphericat o1® Sampling points

rr^"' Boundary of the MGWB 451

Fig. 27.3 Spatial distribution of Ca concentrations [mg/L] in the Bogucice Subbasin (MGWB 451) Poland. (The measured concentrations of calcium were interpolated using kriging technique)

1965 1970 1975 1980 1985 1990 1995 2000

Time (calendar years)

Fig. 27.4 Long-term changes of Fe content in water of the Zawada well-field, south-western Poland [3]

1965 1970 1975 1980 1985 1990 1995 2000

Time (calendar years)

Fig. 27.4 Long-term changes of Fe content in water of the Zawada well-field, south-western Poland [3]

27.3 Characterisation of Hydrogeochemical Field

The hydrogeochemical field is typically characterised by a wide range of concentrations of specific elements forming a distribution. It could be a simple Gaussian distribution or log-normal distribution or it may have a more complex shape. Consequently, characterisation of hydrogeochemical field can be performed in several ways. One way is via cumulative frequency curves of the concentrations of

- 1 1 1 ■/■■■>■■'

y rAïsJ 7

»Na ^ 8 &¥ .far ,f B

"SO, ff §

¥ / / / î

□ DOC o ^o//

/ â / L

- , .J/ //. /

Concentration (mg/L)

1000

0.001

0.01

Concentration (mg/L)

1000

Fig. 27.5 Cumulative frequency diagrams of major ion concentrations in the Bogucice Subbasin (MGWB 451) [4]

specific elements present in the system [5]. In the cumulative probability plots the normal or log-normal distribution of concentration appears as a straight line (Fig. 27.5). The quality of fitting of the measured concentrations varies from element to element. The value of 50% corresponds to median of the distribution. The slope of the straight line is a qualitative measure of the variability of the given component concentration in the studied system.

Sometimes, the observed distribution of the measured element does not fit the Gaussian or log-normal distribution function. This situation is presented in Fig. 27.6 where the distribution of SO4 concentration in the Kedzierzyn aquifer in southern Poland is presented. It is apparent in Fig. 27.6 that SO4 data obtained for this system, when presented in the form of cumulative probability plot, cannot be fitted by a straight line. In such situations a more appropriate solution could be a subdivision of the entire dataset into smaller groups, guided by careful analysis of the investigated groundwater system (regionalization of data).

A more complex shape of the distribution of dissolved constituents is usually caused by anthropogenic influence. Kunkel et al. [6] considered complex distributions as a superposition of distributions representing natural and anthropogenic components (Fig. 27.7). They proposed the separation of these two components using appropriate statistical tests.

Was this article helpful?

0 0

Post a comment