Groundwater early warning monitoring strategy - Conamexphi

United Nations Educational, Scientific and Cultural Organization

International Hydrological Programme

Groundwater Early Warning Monitoring Strategy A Methodological Guide

Edited by

Jaroslav Vrba Brian Adams

GEWMSnwnw:Mise en page 1


16/04/2008

11:13

Page 1


Groundwater Early Warning Monitoring Strategy A Methodological Guide

Edited by

Jaroslav Vrba Brian Adams


16/04/2008

11:13

Page 2

The designations employed and the presentation of material throughout the publication do not imply the expression of any opinion whatsoever on the part of UNESCO, in particular concerning the legal status of any country, territory, city or of its authorities or concerning the delimitation of its frontier or boundaries.

Published in 2008 by the United Nations Educational, Scientific and Cultural Organization 7, place de Fontenoy, 75352 Paris 07 SP Composed by Marina Rubio 93200 Saint-Denis Printed by UNESCO

SC-2008/WS/13 © UNESCO 2008 Printed in France


16/04/2008

11:13

Page 3

Groundwater Early Warning Monitoring Strategy A Methodological Guide ‘Prevention is Better than Cure’

Prepared for the International Hydrological Programme By the Project Working Group: Jaroslav Vrba (The Czech Republic) Chairman of the Working Group Brian Adams (United Kingdom) Editor of the report Edmund Gosk (Denmark) Daniel Ronen (Israel)


16/04/2008

11:13

Page 4

This book is dedicated to the memory of Edmund Gosk, a long time member of the International Association of Hydrogeologists and its Commission for Groundwater protection who died before this book was published


16/04/2008

11:13

Page 5

Preface International Hydrological Programme (IHP) is basically a scientific and educational programme, however, UNESCO has been aware from the beginning of a need to direct its activities toward the practical solutions of the world’s very real water resources problems. Accordingly, and in line with the recommendations of the 1997 United Nations Water Conference, the objectives of the IHP have been gradually expanded in order to cover not only hydrological processes considered in interrelationship with the environment and human activities, but also the scientific aspects of multipurpose utilization and conservation of water resources to meet the needs of economic and social development. Thus, while maintaining IHP‘s scientific concept, the objectives have shifted perceptibly towards a multi-disciplinary approach to the assessment, planning, and sustainable management of water resources. Since its very beginning, groundwater has always been important issue within all phases of the IHP. In nature groundwater is a key element in many geological processes, a geotechnical factor conditioning soil and rock behaviour and an ecological component which sustains spring discharge, river base-flow and many lakes and wetlands. Groundwater is the most valuable and safe source of drinking water in rural areas of developing countries, in arid and semi-arid regions and on islands. In some countries, such as Denmark and Austria, water supplies depend almost entirely on groundwater. Irrigation systems in many parts of the world depend on groundwater. The use of groundwater has increased significantly in recent decades due to its widespread occurrence, mostly good quality, high reliability during drought seasons and generally modest development costs. The idea that the geological environment protects groundwater from pollution and therefore groundwater is not vulnerable to human impacts prevailed for a long time and had serious consequences on groundwater quality. That is why the risk aspect became an important element in groundwater protection policy and management on international and national levels. UNESCO then considered to contribute to the preparation of a methodological guide on ‘Groundwater Early Warning Monitoringg Strategy’.

Alice Aureli

Jaroslav Vrba

Responsible for Groundwater Resources Activities Secretariat of the International Hydrological Programme Division of Water Sciences, UNESCO, Paris

Chairman of Groundwater Protection Commission of the International Association of Hydrogeologists


16/04/2008

11:13

Page 7

Acknowledgements This publication is the outcome of the International Hydrological Programme (IHP). Besides the listed authors, who contributed the implementation of the report, many other colleagues and members of the Groundwater Protection Commission of International Association of Hydrogeologists (IAH) provided valuable suggestions and remarks during the preparation of this report. Thanks are expressed also to the institutions, particularly to UNESCO for organization and budgetary support of the project, to the Ministry of Science, Culture and Sport of Israel and Israel Water Commission and to the Geological Survey of Denmark for hosting the meetings of Editorial Board of this report in Israel (2000) and Denmark (2001).


16/04/2008

11:13

Page 8

Contents Preface Acknowledgement Contents CHAPTER 1.

Introduction

13

Jaroslav Vrba and Daniel Ronen

CHAPTER 2.

Groundwater environments and their vulnerability

17

Brian Adams

2.1 2.2 2.3 2.4 2.5

Introduction Vulnerability to groundwater pollution The soil zone The unsaturated zone The saturated zone

CHAPTER 3.

17 17 18 18 19

Human impact on groundwater quality 20 Jaroslav Vrba

3.1 Introduction 3.2 Point pollution sources 3.2.1 3.2.2 3.2.3 3.2.4

Impact of industrial effluents on groundwater quality Impact of mining on groundwater quality Impact of uncontrolled waste disposal site on groundwater quality Impact of radioactive wastes on groundwater

3.3 Multipoint pollution sources 3.3.1 Groundwater pollution in urban areas 3.3.2 Groundwater pollution in rural areas

3.4 Diffuse pollution sources 3.4.1 Impact of nitrogen fertilizers on groundwater quality 3.4.2 Irrigation return flow 3.4.3 Impact of pesticides on groundwater quality

3.5 Line pollution sources 3.6 Areal pollution sources 3.7 Groundwater salinisation in coastal areas

20 22 22 25 25 26

27 27 28

28 29 30 30

31 32 32


16/04/2008

CHAPTER 4.

11:13

Page 9

Early warning groundwater quality monitoring programmes

34

Jaroslav Vrba

4.1 4.2 4.3 4.4

Introduction National groundwater quality monitoring programmes Regional groundwater quality monitoring programmes Site specific groundwater quality monitoring programmes

CHAPTER 5.

34 36 37 41

Some techniques used for early warning groundwater monitoring 43

5.1 Surface methods

43

Stanislav Mareš, Jan Švoma and Jaroslav Vrba 5.1.1 5.1.2 5.1.3 5.1.4

Introduction Geobotanical methods Photographic methods Geophysical methods

5.2 On-site methods

43 44 45 47

51

Daniel Ronen and Edmund Gosk 5.2.1 5.2.2 5.2.3 5.2.4 5.2.5 5.2.6

Introduction Suction cups Direct push Horizontal monitoring wells The separation pumping techniques The Multi Layer Sampler (MLS)

CHAPTER 6.

Data handling

51 51 53 56 57 58

64

Edmund Gosk

6.1 6.2 6.3 6.4 6.5

Introduction Data collection strategy Data analysis Modelling Forecasting

64 65 67 67 69


16/04/2008

CHAPTER 7.

11:13

Page 10

Early warning groundwater quality monitoring strategy

70

Jaroslav Vrba and Daniel Ronen

CHAPTER 8.

References

74

CHAPTER 9.

Case studies

78

9.1 Monitoring of groundwater quality problems

78

Petr Rasmussen and Edmund Gosk 9.1.1 9.1.2 9.1.3 9.1.4 9.1.5 9.1.6 9.1.7 9.1.8

Introduction Monitoring concept Instrumentation Groundwater sampling and chemical analysis Nitrate monitoring Pesticide monitoring Conclusion and recommendations References

9.2 Application of a multi layer sampler (MLS) for managerial decision-making regarding utilization of effluents for agricultural irrigation in the coastal plain in Israel

78 79 81 83 84 86 87 87

89

Daniel Ronen 9.2.1 Introduction 9.2.2 The unsaturated - saturated interface 9.2.3 Application of a MLS technique in an agricultural area irrigated with municipal sewage effluents 9.2.4 Conclusions 9.2.5 References

89 90 92 97 98

APPENDICES Abbreviations and acronyms Glossary

100 101


16/04/2008

11:13

Page 11

FIGURES 1.1 3.1 4.1 4.2

4.3 4.4

Schematic representation of pollutants (bold arrows) approaching a pumping well

16

Typical movement of LNAPLs (top) and DNAPLs (below) into the groundwater system

24

Comparison of pore-water profiles from unconfined and confined Triasic sandstones, south Yorkshire UK

38

Changes in hydrochemical profile of shallow fluvial aquifer in the period 1984 –1989, Monitoring well HP-65, Middle Elbe region in Central Bohemia, the Czech Republic

39

N-NO3 distribution in the vertical profile of unsaturated zone. Experimental station Samšín, The Czech Republic

390

Vertical distribution of N-NO3- concentrations following application of potassium nitrate with limestone in the non-vegetational (left) and vegetational (right) season. Experimental station Samšin, The Czech Republic

40

Scheme of the unsaturated zone and the uppermost part of the saturated zone with depth variation of the moisture content W and the water saturation Sw

48

Energy gamma-ray spectrum measured on the Earth’s surface in Prague, the Czech Republic, (1) after the Chernobyl accident on May 4, 1986, (2) compared with the natural gamma-ray spectrum of rocks

50

A) Location of suction cups, B) Cross section showing the installation of the section cups

52

Gravimetric water content (θ) in the capillary fringe of continuous-core boreholes CF6 to CF13, in the Coastal Plain aquifer of Israel

54

Chloride concentration in pore water of continuous-core boreholes CF2, CF3, CF4 and CF5

55

Installation of horizontal monitoring wells and section of the horizontal screen

56

5.7

Principle of separation pumping

57

5.8

Segment of a multi layer sampler (MLS) showing the dialysis cells spaced at 3 cm intervals and separated by flexible seals (left) and schematic representation of a segment of the MLS inside of a monitoring well (right)

59

5.9

Equilibration test of dialysis cells conducted at 22°C

60

5.10

Schematic representation of a research well for monitoring the water table region

61

Chemical profiles obtained in the water table region of research well 7 in the Veluve region, the Netherlands, in the forested area subjected to the input of acid rain and ammonia volatized from cultivated land and feed-lots

62

5.1

5.2

5.3 5.4 5.5 5.6

5.11

11


5.12

6.1

16/04/2008

11:13

Page 12

Profiles of xylene, xylidine (dimethyl aniline) and toluene obtained with the MLS in the water table region of Brook Haven National Laboratory, N.Y.

63

Principles of groundwater monitoring aimed at detecting groundwater problems

66

9.1

Location of the six experimental agricultural watersheds (LOOP1-LOOP6) 80

9.2

Schematic layout of groundwater nests and soil water station

81

9.3

Soil water sampler, construction details

82

9.4

Groundwater nests, construction details

83

9.5

Median annual nitrate concentration in shallow groundwater for 3 sandy and 3 clay till watersheds in the period 1990–1996

85

Nitrogen load and nitrate in shallow groundwater in the sandy watershed Barslund Bek

85

9.6 9.7

Findings of atrazine and two metabolites 5 m bellow surface in the Lillebek watershed Electrical conductivity and CL-, NO3- and SO42- concentrations found in the water table region of two monitoring wells, WT-2 and WT-3 and three production wells pumping from the depth (37–55m) bellow the water table

90

A comparison between dissolved oxygen and dissolved organic carbon profiles obtained in the water table region of the study area (Glil Yam) and the profiles calculated in a simulation model by Molz et al. (1986b)

92

9.10

Production of N2O in the water table region of wells WT-2 and WT-3

93

9.11

Schematic representation of porous media with biochemically produced gas bubbles and antrapped air bubbles

94

9.8

9.9

9.12 9.13 9.14 9.15 9.16

87

Dramatic decrease in the horizontal component of the specific discharge (q) in the water table region of well WT-3 Example of discrete water layers of varying CL- content as detected in the unsaturated zone of the monitoring wells WT-2 and WT-3

95

Schematic representation showing two alternative possibilities for the vertical build-up of micro scale water parcels of different salinity

96

Vertical cross-sections through micro scale isothermal water parcels of CL-, NO3-, SO42- and HCO3- in the water table region of well WT-3

96

Eulerian changes of chloride in consecutive profiles obtained in well WT-2

97

95


16/04/2008

1

11:13

Page 13

INTRODUCTION

Groundwater, a renewable and finite natural resource, vital for the life of man, economic and social development and a valuable component of the ecosystem, is vulnerable to natural and human impacts (e.g. pollution by agriculture and industry). The use of groundwater has increased significantly in the last decades as a result of its widespread occurrence, high reliability during drought seasons, mostly good quality, advances in drilling and pumping technology and generally modest development costs. According to available data (UNESCO, 1998) half of the world’s drinking water comes from groundwater. Historically, little attention was given to the protection of groundwater quality, mainly because people were unaware of the threats to this hidden resource. The idea that the geological environment protects groundwater from the impact of surface pollution and that therefore groundwater is not vulnerable to human activities, prevailed for a very long time. This approach had serious and long-term consequences on the groundwater quality in many countries. During the sixties and seventies , there developed a growing interest in the need to protect groundwater. This lead to the establishment of conceptual approaches to groundwater protection and quality conservation based on monitoring, mapping, modelling and vulnerability assessment. The concept of groundwater protection and quality conservation has become an important element in national water planning, policy making and management in the course of eighties. The holistic concept for water resources policy and management, as emphasized at the International Conference of Water and the Environment in Dublin (1992), significantly influenced the holistic approach to development and protection of water resources. This concept stresses the social, economic and ecological value of groundwater, the close connection between groundwater and surface water and the need to maintain the integrity of aquatic and terrestrial ecosystems. Moreover, it gives the same attention to both groundwater quantity and quality and is based on a participatory approach involving planners, policy and decision makers, managers, stakeholders and the general public. Groundwater quality monitoring plays an important role in groundwater protection and effectively supports sustainable management of groundwater quality. It provides a valuable base for assessing the current state of and trends in groundwater quality, helps to clarify and analyse the extent of natural processes and human impacts on groundwater systems in space and time, as well as address groundwater problems in relation to the economic development and social and economic needs. Credible, accurate and consistent groundwater quality monitoring data should be available and readily accessible to planners, regulators, decision and policy makers and managers through data management systems. The data should also helps to increase active public participation in the process of groundwater protection and quality conservation. However, at the present level of

Groundwater Early Warning Monitoring Strategy

13


16/04/2008

11:13

Page 14

knowledge the relationship between the amount of a pollutant released at the soil and rock environment by various human activities and its concentration in groundwater is highly uncertain. This is particularly owing to both the lack of knowledge and the scarcity of data concerning the physical, chemical biological and transport and transformation processes undergone by pollutants in the unsaturated zone, the capillary fringe and the aquifer. Groundwater quality monitoring networks have to be developed and monitoring of the recharge-soil-unsaturated-saturated groundwater system applied to demonstrate our ability 1/ to study and forecast processes which influence the quality and chemical composition of groundwater, and 2/ to implement relevant measures in groundwater quality management strategy and groundwater quality conservation policy. Groundwater quality monitoring programmes operate at the international, national, regional/ provincial and local levels. The objective of each of the above programmes governs the extent of the monitoring activities, such as the design of monitoring networks, construction of monitoring wells, frequency and methods of groundwater sampling and number of analysed variables. International and national groundwater quality monitoring programmes are typically background-monitoring activities, whereas regional and local monitoring programmes are directed towards solving specific problems. Both background and site-specific monitoring activities should also include an early warning monitoring systems. Background monitoring is usually a long-term activity focusing on systematic observation of groundwater quality of large transboundary and national groundwater basins and aquifers. Early warning monitoring systems are not yet a common part of background groundwater quality monitoring programmes. Site-specific monitoring of groundwater quality is a suitable tool for the identification of the impacts of pollution sources on groundwater quality, observation of groundwater quality changes due to excessive aquifer abstraction, protection of public groundwater supplies and the detection of the response of aquifers to remediation. In these activities various methods of early warning monitoring may be used. They help in the early identification and control of the movement of pollutants in both the unsaturated and saturated zones. Early warning monitoring of groundwater quality is an activity or a sequence of activities that makes it possible to identify and to foresee the outcome of a process leading to groundwater pollution with enough anticipation for measures to be taken in order to change or reduce the magnitude of the impact of the said process. Design of an early warning monitoring system depends on the time needed to take appropriate action with respect to the specific groundwater pollution problem. The penetration of a pollutant into the groundwater system may be either from land surface, from an adjacent area due to lateral movement of a pollution plume or through the bottom of the aquifer (Fig. 1.1). Therefore, different approaches, techniques and methods must be applied for early warning groundwater monitoring according to the specific characteristics of the studied groundwater system and pollution impact. The target zone for early warning groundwater monitoring may be a vulnerable region of the aquifer or selected extraction zones where pumping wells are located. The early detection of changes in groundwater quality requires application of various monitoring methods, which may facilitate the observation of pollutant migration, in the gaseous or liquid phases, through the unsaturated and saturated zone of the aquifer. These methods include mainly

14



16/04/2008

11:13

Page 15

photographic imaging, geobotanical and geophysical surveying, soil gas surveys, installation of lysimeters and retrieval of sediment, gas and water samples through specially located and designed monitoring wells, the separation pumping techniques, suction cups, direct push, the multi layer samplers. Existing monitoring strategies tends to focus attention primarily on the transport and transformation of pollutants within the saturated zone and, in many cases, monitoring the impact on groundwater quality depends primarily on the analysis of water samples from production wells. This strategy reflects a fatalistic approach which can lead to a series of ex post facto sequential managerial activities: a) the pollution of production wells is recognized; b) studies are conducted to find the sources of groundwater pollution and design possible (usually lengthy and costly) remedies; c) production wells have to be closed because groundwater pollution is found to be irreversible, and d) water quality standards are changed to accommodate the need for continued use of the groundwater source. Production wells are mostly designed to pump water from deep below the water table. Therefore, evidence of pollution build-up in a production well reflects the mixing process in the aquifer, which often took place several months or even years after the pollutant had arrived at the groundwater table. Generally, existing groundwater quality monitoring programmes are mostly concerned with the identification and control of the consequences of groundwater pollution and therefore do not address the preventive protection of groundwater quality. Clearly, an early warning monitoring strategy is needed that detects pollutants before they are diluted in the aquifer and significant deterioration of groundwater quality occurs. This strategy supports groundwater management and protection policy and helps in identifying human impacts on groundwater quality in the unsaturated zone and uppermost part of the aquifer while they are still controllable and manageable. This report is addressed particularly to policy and decision makers in the field of groundwater quality protection. The objective is to convey the message that, in relation to groundwater quality, ‘prevention is better than cure’. This is particularly the case when considering time constraints and available experience regarding the cost-effectiveness of groundwater quality restoration programmes.


15


16/04/2008

Figure 1.1

11:14

Page 16

Schematic representation of pollutants (bold arrows) approaching a pumping well

(A) vertical downward influx from the unsaturated zone (e.g., nitrates); (B) lateral movement of a pollution plume (e.g., trichloroethylene); (C) vertical upward influx through the bottom of the aquifer (e.g. chlorides), and (D) lateral movement of a polluted water body (e.g., seawater intrusion). For all pollutants concentrations will decrease along the pathway of the small arrows where bold dots denote the position of early warning monitoring devices (e.g., multi layer samplers in a monitoring well for A and C and a series of monitoring wells for B and D).

16



16/04/2008

11:14

Page 17

2

GROUNDWATER ENVIRONMENTS AND THEIR VULNERABILITY

2.1 Introduction In order to plan the installation of effective early warning systems for the detection of groundwater quality problems, it is necessary to have a basic understanding of the hydrogeological environment within which the groundwater is found, particularly regarding groundwater vulnerability to pollution. Generally, confined aquifers are much less vulnerable than unconfined aquifers due to the presence of an overlying impermeable stratum. However, as pointed out by some authors (US National Research Council, 1993), all groundwater is to some extent vulnerable. Thus early warning groundwater quality monitoring systems are primarily designed for unconfined aquifers.

2.2 Vulnerability to groundwater pollution As stated in Chapter 1, the focus of this book is the qualitative aspects of groundwater. Thus we are interested primarily in the vulnerability of aquifers to pollution. Vrba and Zaporozec (eds. 1994) defined groundwater vulnerability to pollution as an intrinsic property of a groundwater system that depends on the sensitivity of that system to human and/or natural impacts. Adams and Foster (1992) defined vulnerability as a function of a) the hydraulic inaccessibility of the saturated zone to the penetration of pollutants, and b) the attenuation capacity of the strata overlying the saturated zone as a result of physicochemical retention or reaction of pollutants; this definition is particularly relevant to the objective of this manual to provide a basis for an early warning monitoring system. The concept of groundwater vulnerability is based on the assumption that the physical environment may provide some degree of protection to groundwater against natural and human impacts. The extent of the attenuation capacity (the decrease in the concentration of a pollutant as it penetrates the aquifer system) depends on physical, chemical and biological processes in the soil, and in the unsaturated and saturated aquifer system. Pollutant attenuation is also affected by the transport mechanism in the groundwater system and the class and properties of the pollutant involved.


17


16/04/2008

11:14

Page 18

Two approaches to groundwater vulnerability should be considered: a) specific vulnerability that is related to a specific pollutant and/or human activity, and b) intrinsic vulnerability that solely considers the natural situation (e.g. net recharge, land slope, soil composition, and the nature of the groundwater system). Vulnerability of groundwater is a relative, non-measurable, dimensionless property. The accuracy of vulnerability assessment depends above all on the quantity, quality and reliability of available data.

2.3 The soil zone Soil is commonly regarded as one of the principal natural attributes in the assessment of groundwater vulnerability. Most of the processes causing elimination and/or attenuation of pollutants occur in the biologically active soil zone as a result of its higher clay mineral and organic matter content and large bacteriological populations. Texture, structure, composition and thickness of soil vary according to a number of factors including climate, topography and underlying geology. However, it is important to stress that the attenuation capacity of soils may be finite. Many point pollution sources (pits, trenches, lagoons, underground tanks etc.) release their pollutants below the soil zone. Thus soil attenuation capacity is generally more important in connection with diffuse pollution sources such as the leaching of nutrients and pesticides from agricultural land. However, the soil’s function as a natural protective filter for the retardation and degradation of pollutants can be significantly decreased when the dynamic stability of the soil organic matter and the carbon/nitrogen ratio are disturbed. Thus in the context of early warning systems, sensors or sampling devices should generally be placed below the soil zone to monitor those pollutants that have not been retained by it.

2.4 The unsaturated zone The unsaturated zone represents the first line of natural defense of unconfined aquifers against groundwater pollution due both to its strategic position between the land surface and the saturated zone and its potential for pollutant attenuation. However, it must be noted that the role of the unsaturated zone can be complex and its ability to attenuate pollutants difficult to predict. The degree of attenuation will depend upon the chemical nature of the pollutant, the release mechanism on the land surface (e.g. small spill vs. rupture of a big container), and the lithological and geochemical composition of the unsaturated zone, its thickness and the flow regime through it. In effect, for persistent, mobile pollutants the unsaturated zone will merely introduce a time lag before arrival at the water table, without significant attenuation. In many other cases the degree of attenuation will be highly dependent upon the flow regime and residence time which are both largely dependent on annual recharge. Under conditions of natural rainfall infiltration it is reasonable to assume that transit times in the unsaturated zone will be a function of the annual infiltration rate and a moisture content approaching the specific retention capacity following prolonged drainage. Since the latter varies little among soil and rock types compared with the climatic variations of the former, under natural conditions the transit time is essentially controlled by annual average infiltration rate. Water movement through the unsaturated zone of sedimentary deposits is relatively slow and

18



16/04/2008

11:14

Page 19

restricted to small pores. The chemical condition is mostly aerobic and frequently alkaline. Thus there is considerable potential for a) interception, sorption and elimination of pathogenic bacteria and viruses, b) attenuation of heavy metals through precipitation, sorption or cation exchange, c) sorption and biodegradation of many natural and synthetic hydrocarbon compounds. In consolidated fractured rocks having secondary porosity, the potential for rapid by-pass flow will render groundwater highly vulnerable to pollution. Rapid transmission of water and pollutants in sometimes unexpected directions and over large distances is typical of karstic environments, where the fractures can become enlarged by solution. Thus the geochemical and lithological composition, especially the grade of consolidation and degree of fissuring of the unsaturated zone and groundwater table below surface are key elements in the assessment of aquifer vulnerability to pollution. Sampling and measurement of solid, liquid and gaseous phases of pollutants throughout the unsaturated zone (see chapters 4 and 5), is an important factor in the development of early warning groundwater quality monitoring strategies.

2.5 The saturated zone The definition of unconfined, semi-confined and confined conditions of the saturated aquifer is very important in the process of assessment of groundwater vulnerability. Unconfined aquifers bounded above by an unsaturated zone formed by permeable layers, containing both air and water, are considered highly vulnerable. Confined aquifers bounded above and below by impermeable layers, containing water under pressure, are considered to have low or very low vulnerability. In general terms recharge to unconfined aquifers is from the land surface immediately above it whereas for confined regions recharge is laterally from generally higher areas where the aquifer is unconfined. Rock texture and permeability, groundwater flow direction, hydraulic gradient, hydraulic conductivity, the contaminant class in terms of its mobility and persistence and biological, physical and chemical reactions all control pollutant attenuation and transport in the saturated groundwater system. Therefore, construction and screen location of monitoring wells for early warning monitoring of groundwater and dedicated sampling techniques should take in to consideration both the groundwater system and the pollutant properties.


19


16/04/2008

11:14

Page 20

3

HUMAN IMPACT ON GROUNDWATER QUALITY

3.1 Introduction This chapter is primarily concerned with adverse human effects on groundwater quality. However, certain natural constituents in groundwater can also reach undesirable concentrations with consequent potential impact on human health and ecosystems. The composition of groundwater is primarily controlled by: the properties of the soil and rock in which groundwater moves, the contact time an contact surface of groundwater with geological materials along flow paths, the rate of geochemical, microbiological and physical processes within the soil-rock-groundwater system and the presence of dissolved gasses. Differences in chemical composition between lateral (recharge and discharge areas) and vertical (shallow oxidation zones and deep reduction zones) profiles in groundwater systems are recognised. Generally, shallow groundwater in recharge areas has a lower dissolved solids content than groundwater in discharge areas from deeper aquifers. An increasing content of total dissolved solids and an anion evolution sequence HCO3- - SO42- - Cl-, expressing the change from oxidising to reducing conditions, are usually observed in the vertical profile of groundwater systems. However, such hydrochemical profile can not be applied to shallow coastal aquifers where groundwater composition is under the influence of saline water. Biological processes enhance the extent and rate of geochemical processes. They are particularly intensive in the soil/root zone, where oxygen is available for both organism respiration and the breakdown of organic matter. Examples of the influence of natural constituents on groundwater quality include particularly high iron and manganese content, the presence of zinc, arsenic and other trace metals released into groundwater from ore-bearing deposits, excessive concentration of fluoride and arsenic, elevated concentration of chloride in coastal aquifers and the presence of organic compounds from peat deposits. The natural chemical composition of groundwater (natural background) needs to be known, when assessing the extent of human impacts on groundwater. Groundwater quality deterioration and pollution as a consequence of human activities are recognized as a serious worldwide environmental and socio-economic problems. Groundwater pollution

20



16/04/2008

11:14

Page 21

is a process whereby water gradually or suddenly changes its natural physical, chemical or biological composition and quality and ceases to meet the criteria and standards set for drinking water, irrigation and other purposes (Vrba, 2000). Various criteria have been used to classify groundwater pollution. The commonly used classification systems are based on the extent of pollution (point, multipoint, diffuse, line, regional), the source or origin of pollution (industrial, mining and military activities, waste disposal sites, urban and rural settlements, agriculture, transport networks, oil lines, streams, sewerage systems and acid depositions) and the types of pollutants (physical, chemical, biological, radioactive). Pollution classification criteria are given in Table 3.1. This chapter describes the impact of the most frequent pollution sources on groundwater quality and related monitoring activities.

Extent of pollution

Source of pollution

Main pollutants

Point

Industry

Heavy metals (Pb, Zn, Cd, Cr), arsenic, phenols, petroleum products and additives, high BOD, suspended solids, chloride, sulphide, alkaline effluents, low pH, chlorinated hydrocarbons, PAHs, synthetic organic and organometalic compounds

Mining

Heavy metals, salts (chloride, sulphate), low pH, high TDS, cyanide, PAHs, petroleum products

Waste disposal sites including deep disposal wells

Heavy metals, ammonium, sulphate, chloride, phenols, various biodegradable and non-biodegradable organics, feacal pathogens

Radioactive wastes

3H

Cattle – breading lots

High suspended solids, BOD, total nitrogen, chloride, feacal pathogens

Urban areas

Heavy metals (Pb, Zn), ammonia, chloride, sulphate, petroleum products, chlorinated hydrocarbons, surfactants

Multipoint

- Tritium, 90Sr, 137Cs, 239 Pu, 129I, 226Ra, toxic metals

settlements quality Ammonia, nitrate,monitoring chloride, sulphate, surfactants, iron, manganese, pollution sourcesRural on groundwater and related activities. feacal pathogens

Military areas

Petroleum products, heavy metals

Non-point (diffuse)

Agriculture Crop and root-crop farming, irrigation

Fertilizers (organic and inorganic): nitrate, ammonia, chloride, phosphate, natrium, potassium, feacal pathogens, salinity Pesticides: organochlorine compounds (aldrine, heptachlore), carbamate insecticides (atrazine), polyphosphate, organometalic compounds (fungicides)

Line

Roads

High suspended solids, salts, petroleum products, solvents

Railways

Petroleum products, organic chemicals

Oil pipelines

Petroleum products

Sewerage systems

High suspended solids, nutrients, chloride, high BOD, feacal pathogens

Streams

Nitrate, ammonia, iron, manganese, phenols

Areal

Acid depositions

Aluminium, low pH, nitrate, sulphate

Coastal areas

Salinisation

Sodium, magnesium, chloride, sulphate, high salinity and TDS

Table 3.1 Classification criteria of groundwater pollution sources


21


16/04/2008

11:14

Page 22

3.2 Point pollution sources The immediate impact of point pollution sources on groundwater quality is either local or sitespecific in extent. However, if a pollution source is not soon identified, groundwater pollution could be detected a considerable distance (several hundreds metres or even kilometres) from it. Pollution of public groundwater supply wells located a long distance from pollution sources have been reported from several countries (Van Dam, 1967, Csanády, 1968, Williams and Wilder, 1971, Jackson, 1980, Elek, 1980, Ku, 1980, and many others). The most common point pollution sources are those relating to industrial, mining and waste disposal activities. Special attention is given here to hydrocarbons, the most prominent pollutants of groundwater.

3.2.1 Impact of industrial effluents on groundwater quality Groundwater pollution by industrial waste has been reported from many areas of the world. Sources of pollution include uncontrolled leaks and spills from poorly designed and improperly located ponds, lagoons, pits, basins or ditches, used for the disposal of various kind of industrial liquid and solid wastes, many of them hazardous. Deep injection wells are also used in some countries for the disposal of liquid and semi-liquid industrial waste. Groundwater pollution problems related to waste disposal in deep wells described e.g. Aust and Kreysing (1985) and LaMoreaux and Vrba (1990). Disposal of hot water from cooling processes into aquifers is also widely practised. Remediation of polluted water supply wells, particularly where organic substances are involved, is generally a long term, technological demanding and costly process. Wells have to be temporarily closed or even abandoned where groundwater pollution is irreversible. Groundwater quality monitoring systems around industrial pollution sources are often missing. The design of monitoring networks and construction of monitoring wells require considerable expertise and experience in contaminant hydrogeology. The establishment of an early warning monitoring system focuses on the identification of pollution in the unsaturated zone and on the control of the lateral movement of pollutants in the aquifer outside of the industrial area should be emphasized as a significant tool of groundwater protection policy. In the following overview the main industrial pollutants (often toxic) of groundwater are identified. Metal plating technologies and surface finishing of metals produce mostly acid waste containing hexavalent chromium, cadmium, lead, zinc and other heavy metals, cyanides bound with heavy metals of different stabilities, phenols, abrasive salts, oils, benzene and thiosulphates. Wastes produced by tannery factories are rich in dissolved chlorides, sulfides and chromium. Textile industry waste contains heavy metals, dyes and orgnaochlorine compounds. The electrical industry produces waste with high mercury content. Groundwater pollution by lead and fluorine originating from enamelling processes in ceramic factories has been also identified (Pellergrini and Zavatti, 1982). Sugar refineries, breweries and other food and drink facilities produce effluents with high suspended solids and colloidal and dissolved organic substances. Wastes from the production of gas by the distillation of coal, contain high concentration of various kinds of aromatic hydrocarbons, phenols, thiofene and cyanide complexes. Several sites where gas has been produced but have been abandoned for several years still show high levels of pollution in both the unsaturated and saturated zones. 22



16/04/2008

11:14

Page 23

The petrochemical, metal (cutting and cooling emulsions) and the fat processing industries produce various kinds of oil wastes. Tenzides are bound with wastes from the textile, leather-tanning and food industry. The paper and pulp industries produce waste containing organic matter and chlorinated organic substances. The chemical and pharmaceutical industries generate a wide range of organic often toxic pollutants, (including chlorinated and aliphatic hydrocarbons, polychlorinated biphenyls, phenols, pesticides – organochlorine and organophosphate compounds, surfactants) and inorganic pollutants (heavy metals, cyanide).

Groundwater pollution by oil hydrocarbons Refined petroleum products (gasoline, petroleum, kerosene, diesel fuel, oil, lubricants and emulsions) form the largest class of point pollution sources of groundwater. Spillages and leaks occur during the production and handling (oil refineries, oil processing plants, filling stations), the storage (fuel storage facilities, underground fuel tanks) and the transport (tanker trucks, railway tankers, petroleum pipelines) of petroleum products. Accidental spillages are usually visible, occur suddenly, and the location and amount of the spilled product are generally known. On the other hand, petroleum leaks due to corrosion or weld failures of underground storage tanks or petroleum pipelines are latent, and the volume of leakage involved will be usually long term unknown. Because early warning monitoring systems are generally not installed, a long period may pass before such leaks are identified. However, oil hydrocarbons in gaseous or liquid stage are generally easily detectable by early warning monitoring systems located in the unsaturated zone. Subsurface lateral and vertical movement of petroleum products depends on the nature of the groundwater system (particularly its vulnerability) as well as on the quantity and physical and chemical properties of the petroleum products discharged. Their vertical migration will be stemmed on reaching impermeable strata or when the threshold of residual saturation is attained. In such cases a petroleum plume is immobilised in the unsaturated zone above the water table. This generally occurs when the volume of the discharge is small relative to the surface area of spill, or the unsaturated zone is thick and the permeability low. However, discharged petroleum products will often reach the capillary fringe and then form a thin film on the groundwater table while emulsions or the main body of petroleum will sink to different levels within the aquifer. Hydrodynamic dispersion will cause lateral pollution migration in the direction of the groundwater gradient. Viscosity and density affect contaminant penetration and migration in the subsurface and significantly influence the design (particularly well screen installation) and location of monitoring wells. Generally, light nonaqueous phase liquids (LNAPLs), such as gasoline, kerosene and light oils, have a lower density and higher viscosity than water. LNAPLs entering the unsaturated zone are partly volatilised (and move as a soil gas by molecular diffusion) and when they attain residual saturation are gradually dissolved by infiltrating water. When LNAPLs reach the saturated zone they float in the immiscible phase on the groundwater table surface over long distance depends on the hydraulic gradient (Fig. 3.1). On the other hand, dense nonaqueous phase liquids (DNAPLs e.g. asphalt, heavy oils, lubricants and also chlorinated sotvents) have low solubility and are markedly denser and less viscous than water. Large spills of DNAPL will quickly penetrate the full depth of the aquifer and accumulate on its bottom. Whilst lateral migration of LNAPLs is controlled by the groundwater flow direction, DNAPL movement follows the slope of the impermeable strata underlying the aquifer and can move in the opposite direction to the groundwater gradient (Fig. 3.1b). The viscosity and density of oil hydrocarbons therefore, significantly affect monitoring wells construction particularly with regard to screen location. Groundwater Early Warning Monitoring Strategy

23


16/04/2008

Figure 3.1

11:14

Page 24

Typical movement of LNAPLs (top) and DNAPLs (below) into the groundwater system (modified after Lawrence and Foster, 1987)

The solubility of petroleum products in water is generally low. However, some light hydrocarbon compounds are very soluble, particularly during the first 24 hours of contact with water. The solubility of hydrocarbons in water increases with decreasing carbon number in the molecule. The sorption capacity of petroleum products depends on product properties, capillary forces, moisture content and the physical (grain size) and geochemical composition of the rock material. Those polar components most readily sorped are naphtenic acids, resins and asphaltens. Selective sorption of non polar components occurs in the sequence olefines → aromatics → cyclanes → alkanes. According to Schwille (1969) the values of petroleum products sorption by the rocks range from 5 to 40 l.m3. Evaporation of petroleum products is an important early warning indicator of the underground extent of pollution. Petroleum products in the gaseous phase spread within the unsaturated zone by molecular diffusion and are easily detectable by early warning monitoring of the soil air. The degree of volatility of the petroleum products depends on their boiling point and decreases from gasoline to aviation fuel, diesel oil and oil. Therefore, soil gas monitoring for the delineation of pollution by different volatile organic compounds and water quality monitoring within the unsaturated zone are important part of early warning groundwater quality programme (see chapters 4 and 5).

24



16/04/2008

11:14

Page 25

Emulsification of petroleum products is a typical property of heavier fractions of oil hydrocarbons. Various microorganisms act as emulsifiers (Schwille 1969). Petroleum products in both gaseous and liquid phases retained in the groundwater system are affected by oxidation and reduction processes. Products of oxidation processes include phenols, carbonic acids, resins, alcohols, and ketones. Oxidation of petroleum products also occurs when there is oxygen deficiency in the rock-groundwater system (Schwille and Vorreyer 1969). Microbial processes are controlled by free oxygen, nutrients, carbon and temperature and have a significant influence on the rate of degradation of petroleum products. Biodegradation processes are intensive in aerobic environments, however they also occur in anaerobic conditions. Microbial decomposition particularly occurs on the petroleum product/groundwater interface.

3.2.2 Impact of mining on groundwater quality Mining activities which can impact on groundwater quality include the extraction of ore, coal, oil, salt and non-metallic deposits, ore washing and dressing, coal preparation and other post extraction processing of mining material and uncontrolled leakages from tailings, piles, evaporation ponds, pits and other disposal sites of extracted mine material. Groundwater monitoring around mines and processing plants is often missing. However, monitoring of discharged acid mine effluents is needed to control their impact on groundwater and surface water quality. Discharged mine effluents are mostly acidic (pH 4 and less). They contain various kinds of mobile soluble anion complexes of heavy metals released by the oxidation process of metal sulphides, particularly present in ore bearing deposits. Iron sulphide (pyrite) is frequently present in coal-measure units, and, in an air-water environment, oxidises to form ferrous sulphate and sulphuric acid. Secondary reaction of sulphuric acid produces high concentrations of aluminium, manganese, calcium, sodium, which along with iron and sulphate are sources of groundwater pollution. Brines discharged from salt and potash mines are also potential pollutants of groundwater. Groundwater pollution by oil brines is frequent in oil fields; such pollution is considered a serious environmental problem in the USA (Everet, 1980).

3.2.3

Impact of uncontrolled waste disposal sites on groundwater quality

Landfills, particularly abandoned disposal sites with unknown composition of disposed waste, form significant potential sources of groundwater pollution. Large numbers of landfills have been poorly located (in permeable sediments above shallow water table aquifers, close to surface water bodies), poorly constructed (without liners and other techniques to prevent uncontrolled leaks) and do not have groundwater quality monitoring systems. However, in many European countries and in the USA early warning groundwater quality monitoring along and beneath landfills is now obligatory under the relevant legislation. Monitoring wells should be designated during the construction of waste disposal facilities. In studying the impact of landfills on groundwater quality, Knoll (1969) demonstrated significant


25


16/04/2008

11:14

Page 26

long-term increase of organic substances (560%), sulphate (310%) and chlorine (520%) in the aquifer bellow the landfill after 40 years of refuse disposal. According to Freeze and Cherry (1979) landfills from the Roman epoch are still generating leachate. Leachate migration contaminate groundwater over areas much more larger than the landfill, extended particularly in the direction of groundwater flow. Groundwater pollution due to uncontrolled leakages from landfills has been described by many authors. The extent and intensity of chemical and biological processes in the waste mass depend primarily on climatic conditions (amount and infiltration rate of precipitation and temperature) and on the nature and age of the waste. Jackson (1980) pointed out that changes of leachate composition are dependent on the age of the waste (e.g. volatile fatty acids change to higher molecular weight substances such as carbohydrates). Household waste contains a wide spectrum of organic and inorganic material, of which a significant part is soluble and biodegradable. However, disposed household waste also contains hazardous items such as batteries, medicines, paints and oils. A high content of sulphate, chlorine, ammonia, total organic carbon, biodegradable and persistent organics and xenobiotics, emissions of methane and carbon dioxide and high biological oxygen demand are typical for leakages from disposed household wastes. However, elevated concentrations of organochlorine compounds and other organic solvents and residues, heavy metals, pigments, oil hydrocarbons and other hazardous compounds are also recorded. The content of soluble metals in waste leakage is reduced by their adsorption on organic matter and by precipitation in immobilised form as metal sulphides (UK DoE, 1992). Gas production due to biochemical decomposition of disposed organic matter is a typical process occurring in sanitary landfill. Methane (CH4) and CO2 among other gases are the most abundant. Gas ventilation and monitoring systems operate in recent landfills, to observe both gas production and its migration outside of disposal site.

3.2.4 Impact of radioactive wastes on groundwater Uranium mining, the nuclear power industry and some medical and military facilities and activities produce a wide range of radionuclides, which may enter the aquatic system by a variety of routes. However, the normal operation of nuclear facilities does not present a serious threat to groundwater quality. Uranium mining, milling, refining, enrichment and reprocessing and land disposal of radioactive waste are those activities which pose the greatest potential risk to groundwater quality. In situ leach-mining of uranium-rich sedimentary deposits, which depends upon the introduction of a leaching solution into the uranium-rich formation via injection wells and the removal of the enriched solution by production wells, poses a high risk to groundwater quality and groundwater pollution by the chemicals present in the leaching solution. In situ restoration techniques, surface treatment and forward and reverse recirculation of treated water, are implemented to remediate polluted groundwater. Serious impacts on groundwater quality occur when polluted groundwater breaches the hydraulic barrier formed by production wells bounding the area being mined. An example of such groundwater pollution by acid-leach solutions in a large sedimentary aquifer has been recorded in the Czech Republic. Early warning groundwater monitoring systems established around the hydraulic barrier of production wells play important role in groundwater protection against the impact of uranium leach-mining.

26



16/04/2008

11:14

Page 27

Uranium mining produce large amounts of rock-waste and tailings from milling, both containing natural isotopes of uranium, thorium, radium and radon gas. 226Ra poses the greatest risk to the aquatic system because of its long time of half life (1600 years) and high £ and γ radiation. In such areas concentrations in groundwater are frequently greater than the permissible concentration of 226Ra (10 -9 mg.l -1) for drinking water. The process of uranium refining and enrichment generates low-level radioactive wastes containing 226 Ra, 230 Th and 238 U. The most significant potential risk to groundwater quality is posed by the subsurface burial of radioactive reactor wastes which contain acid water having high concentrations of nitrate and aluminum and a wide range of radionuclides of different half-life, solubility, persistence, and type of emitted radiation. According to Matthess (1982) the mobility of radionuclides in the rock-groundwater system is controlled by the concentration of radionuclides and their isotopes, the pH of the groundwater, the kind and concentration of other dissolved solids and the physical and chemical properties of the rock environment. Generally, tritium which serves as a tracer for groundwater dating is more mobile than strontium and caesium, the long half-life insoluble plutonium isotopes showing particularly low mobility. Pollution of groundwater by radionuclides may also occur as a consequence of nuclear accidents. The release of radionuclides into the environment from the Chernobyl power plant disaster (1986) produced groundwater pollution by 137Cs and 90Sr up to a radius of a few kilometres from Chernobyl power plant (UNESCO, 1992).

3.3 Multipoint pollution sources Urban and rural areas contain numerous point sources of groundwater pollution. Inadequate handling, treatment and management of household wastes and waste waters, industrial effluents, uncontrolled waste disposal sites, rain and melt waters, salt water intrusion in coastal aquifers, are the main sources of groundwater multipoint pollution. Design and operation of groundwater quality monitoring systems is not yet usual in urban and particularly in rural areas.

3.3.1 Groundwater pollution in urban areas More than half of the current world’s population is living now in urban settlements. The resulting enormous concentration of human activities generates a wide range of impacts on the urban environment including groundwater. Untreated or poorly treated municipal wastewater is the main source of groundwater pollution in urban areas. Discharge of wastewater to infiltration ponds, lagoons, wells or even the surface water bodies is still practiced in some countries and has serious consequences for groundwater quality. Cracks and breaks in urban sewerage systems are frequent sources of uncontrolled leaks of untreated wastewater into shallow aquifers. The greatest potential threats to groundwater quality from household wastes include dissolved organic compounds (chloride, sulphate, ammonium, nutrients – nitrogen and phosphorus), pathogenic micro organisms (bacteria and viruses), trace elements, high BOD and TOC, and various types of household surfactants. Uncontrolled leaks from industrial and commercial activities located in urban areas also provide a significant source of groundwater pollution by heavy metals, oil hydrocarbons, phenols and others.


27


16/04/2008

11:14

Page 28

Discharge of industrial wastewaters into waste disposal wells located in urban or suburban areas is practiced in several countries. Here, corrosion of well casing and/or incorrect placing of well screens can result in uncontrolled pollution of groundwater (La Moreaux and Vrba, 1990). Waste disposal wells are also used for the injection of treated wastewater as artificial recharge to the aquifers or for the formation of hydraulic barriers to counter salt-water intrusion in coastal areas. Rain and melt waters are further potential source of groundwater pollution in urban areas. They can transport different pollutants (oil hydrocarbons, organic chemicals and seasonally salts in some regions) from road and street networks, parking lots, gasoline station areas and industrial zones to groundwater bodies. Cases of groundwater pollution in urban areas are described by Balke at al. (1973), Jackson at al. (1980), Everett (1980), Mathess (1982), Vrba (1985), Foster at al. (1987), RIMV (1992) and many others.

3.3.2 Groundwater pollution in rural areas IIn rural areas the most frequent sources of groundwater pollution are uncontrolled leaks from septic tanks, cesspools or latrines. Unsewered rural sanitation can affect groundwater quality in both domestic and public water supply wells. Chlorides, sulphates, nitrates, phosphorous, ammonia, household detergents and disinfectants and pathogenic micro organisms are the main pollutants of groundwater from such sources. High concentration of nitrates and pathogenic bacteria in drinking water in developing countries are a major cause of infections, illnesses and mortality of rural population, particularly infants. Industrial and agricultural activities are other sources of groundwater pollution in rural areas. Local industry generates similar pollutants as described above in urban areas, but to a lesser extent. Uncontrolled spillages of liquid wastes from manure and silage liquors and slurry and manure disposal sites are the most frequent point sources of groundwater pollution by farming activities in rural areas.

3.4 Diffuse pollution sources Diffuse or non-point pollution of groundwater is mostly related to agricultural activities, above all to the massive application of fertilizer and pesticide on arable land and irrigation return flow. Historically, agriculture formed closed, environmentally sound and sustainable systems with insignificant impact on the environment. However, methods of contemporary agriculture have changed from crop rotation to monoculture and from single to mass-scale animal breeding. The intensification of agricultural production has created serious impacts on the quality of groundwater.

28



16/04/2008

11:14

Page 29

3.4.1 Impact of nitrogen fertilizers on groundwater quality A big difference exists between continents and individual countries in the amount of fertilizers applied. The average application rate of nitrogen fertilizers on arable land per year in Western and Central Europe, North America and Eastern Asia is in the order of hundreds kg N/ha and in many countries in Africa it may be as low as a single kg N/ha. The most widely occurring class of groundwater pollutants with respect to the impact of agriculture are nitrates originating from organic and inorganic fertilizers applied to arable land. Potassium and phosphorous compounds derived from fertilizers accumulate in the soil and the unsaturated zone due to their lower solubility and mobility, and their potential threat to groundwater is generally low. Conditions for groundwater phosphate pollution may arise in regions with large livestock concentrations and a resultant high production of manure. Metals (cadmium, copper, and others) derived from certain inorganic fertilizers, accumulate in soils with a negative effect on their fertility, but their impact on groundwater quality is recorded only exceptionally. Research into the crop-soil-unsaturated zone-saturated zone system indicates that nitrate increase in groundwater is not a random, locally limited phenomenon, but a serious environmental, social and economic problem that affects aquifers of many countries. A statistical relationship has been found between the amount of nitrogen fertilizer applied to the land surface and the nitrate content in groundwater. Agriculture contributes particularly to nitrate pollution of shallow water table aquifers, with economically accessible groundwater resources used for water supplies. The Netherlands National Institute of Public Health and Environmental Protection (RIVM, 1992) calculated nitrate concentrations in the leachate from agricultural soils (at 1 metre depth) in Europe. Model computations indicate that over 85 % of the agricultural area in Europe has nitrate concentrations above 25 mg/l and that drinking water standard (50 mg/l) is exceeded bellow approximately 20% of the agricultural soils. Many European countries (mainly north-western and central European countries) indicated that they had serious nitrate groundwater pollution problem. In the USA, particularly in the Corn Belt states, nitrate pollution of groundwater can be found in several regions of Iowa, Illinois, and Ohio. Occurrence of nitrate concentration in groundwater greater than drinking water standards (10 mg/l NO3–N) was documented in several regions of Nebraska (Exner and Spalding, 1990). Man-induced mineralization of natural soil nitrogen has resulted in nitrate pollution of groundwater in large semiarid areas of Texas and Montana (150 mg/l NO3 – N), however, such concentrations decreased after several years of cropping (Miller et al., 1981). A rapid increase of nitrate content in alluvial aquifers with shallow depths to groundwater under irrigated soils is observed in California, Florida and other US States (Spalding and Exner, 1991). Vertical nitrate zonality of shallow aquifers is observed in many parts of USA (Hallberg, 1989). The extremely high content of nitrate in shallow rural wells identified in Florida and Indiana is a consequence of their location close to point pollution sources such as animal corrals, cattle feeding areas or domestic sewage disposal sites (Spalding and Exner, 1991). In the developing economies of Asia and Africa, nitrate pollution of groundwater is mostly a result of the effects of point pollution sources. The quality of water in many shallow public and domestic wells is affected by their poor construction, inappropriate sitting close to pollution sources (septic tanks, latrines, dumps of animal slurry) or resulting from animals using the well head as a watering


29


16/04/2008

11:14

Page 30

point. Nitrate in groundwater in many rural wells in Asia and Africa reaches several hundreds mg/l and can be a serious health hazard. The risk of methaemoglobianemia, also referred to as Blue-Baby Syndrome, is considerable in many rural settlements in developing countries. Sustainable management of groundwater quality beneath agricultural lands requires maintenance of the dynamic stability of soil organic matter and restriction of the processes that lead to mineralization of organic nitrogen. Monitoring of the carbon / nitrogen ratio as an important indicator of the soil organic matter’s state should therefore be part of an early warning groundwater monitoring strategy in regions where shallow water table aquifers underlay areas of intensive farming activities. Monitoring of groundwater quality in many agricultural regions has proved nitrate vertical zonality and movement of a nitrate plume through the unsaturated and saturated zones of the aquifer (see chapter 4). An overview of the threats of agriculture to groundwater quality in various countries is presented in the Proceedings of the International Workshop organized by UNESCO /CIHEAM/UPC in Spain (Candela and Aureli, 1998). However, non-point pollution of groundwater due to agriculture has been studied by many scientists and several international conferences, reports and books have been devoted to this topic.

3.4.2 Irrigation return flow Soil and groundwater pollution as a consequence of irrigation return flow has been reported from many countries. Increase of dissolved solids in groundwater occurs in agricultural areas with overirrigated soil not having adequate drainage. Leached salts from the soil are transported by irrigated water and degrade the quality of the underlying aquifers. Soil-groundwater degradation processes accelerate when groundwater with high content of salts from aquifers beneath an irrigated area is recirculated. Groundwater salinisation in arid regions is described e.g. by Bouwer (1987) and Chilton (1995). Desert soils contain natural salts, which are leached by irrigated water and penetrate into the aquifer and degrade its quality.

3.4.3 Impact of pesticides on groundwater quality The aquatic system has been exposed to an increasing number of types and quantity of agricultural pesticides throughout the world for several decades. Different types of pesticides (the term of pesticides is used in broad sense for insecticides, herbicides and fungicides) are transformed to degradable residuals (owing to biodegradation or chemical hydrolysis), many of them break down to toxic derivatives. In Europe, based on the model calculation (RIVM, 1992), the pesticides drinking water standard is exceeded in the leachate under 75 % and 60 % of the total arable and crop land in north-western and central-eastern countries, respectively. The application of pesticides reflects in pesticide pollution of groundwater in many shallow aquifers bellow arable land in Europe. In the USA, systematic monitoring of pesticides in groundwater commenced at the beginning of the 1980s. As a result of the application of pesticides on agricultural land in many American states, aldicarb, atrazine, DBCP, ethylene diobromide and other kinds of pesticides have been found in many wells

30



16/04/2008

11:14

Page 31

(Spalding and Exner, 1991). The US EPA prepared a background document on the strategy of groundwater protection against agricultural chemicals in 1986. Groundwater vulnerability to pesticides is high in shallow water table aquifers below coarse or sandy soils having a high moisture and low organic matter and clay content, low adsorption and cation exchange capacity and low-level bioactivity and biodegradability. There are differences in the persistence of various groups of pesticides in soil. In general, the ionic pesticides are more soluble than non-ionic groups. The most resistant organo-chlorine insecticides can persist in soil for years. On the other hand, organophosphate insecticides and aliphatic acid-herbicides degrade relatively rapidly in less than three months. The occurrence of toxic organochlorines in groundwater and their potential carcinogenic effects led many countries during the sixties to ban their use. Sustainable management of pesticide application and groundwater quality monitoring are particularly desirable in agricultural areas used for the cultivation of vegetables, fruits, vines and potatoes. Systematic monitoring of pesticides is particularly missing in both developed and less developed countries. In less developed countries the appropriate laboratory equipments are often not available.

3.5 Line pollution sources Road and railway transport, municipal and rural sewerage networks, oil and gas pipelines and surface streams are the main potential line pollution sources of groundwater. Polluted runoff water from roads, soil and groundwater acidification by transport emissions and spills of various substances due to road and railway accidents can all have a serious influence on groundwater quality. Runoff from roads contains oil hydrocarbons, various salts (NaCl, KCl, CaCl2) and solvents. Transport emissions contain fuel additives like lead (particularly in the past) and VOC (benzene). Highly soluble and mobile salts, applied in large amounts on the roads in winter, are a source of pollution for shallow aquifers, particularly where roads cross their recharge or vulnerable areas. The most dangerous risk for groundwater pollution is from traffic accidents involving truck tankers which can result in spills of several tonnes of various hazardous substances including flammable or even explosive ones. Downward penetration of such liquids through permeable soils into the groundwater body can have major effects on groundwater quality. However, a more important source of groundwater pollution is that of spills of large quantities of hazardous chemicals released from cisterns in train accidents. The consequences on groundwater quality are often very serious because one train usually transports a variety of chemicals having various properties and compositions. Uncontrolled leaks of wastewater and ballast water from defective sewerage networks form a major hazard for shallow aquifers in urban areas. Microbiological pathogens and synthetic surfactants are the most frequent pollutants. Seepage losses of wastewater from surface canals discharging household liquid wastes from rural settlements in less developed countries have similar impacts on groundwater quality. Accidental spills and uncontrolled leaks of petroleum products and liquid natural gas from buried pipelines are potential sources of soil and groundwater pollution by oil hydrocarbons and other petrochemicals. Spills of petroleum products due to the failure of welds in petroleum pipelines have


31


16/04/2008

11:14

Page 32

been recorded in many countries. Different monitoring systems are established along pipelines and different remote sensing monitoring methods are applied to control spills from pipelines (see chapter 5). The hydraulic gradient between a surface stream and a groundwater level controls the possibility of bank infiltration of polluted surface water into an underlying aquifer. Groundwater pollution may occur far from the pollution source, at locations where the river is a losing stream and the conditions of surface water infiltration exist.

3.6 Areal pollution sources Acid atmospheric emissions (sulphur dioxide-SO2, nitrogen oxides -NOX) transported hundreds of kilometres over continents and their chemically converted products (sulphuric and nitric acids) are major sources of regional transboundary pollution of soil and surface water. However, their influence on groundwater quality has also been recognised (Holmberg, 1987, Stanners and Bourdeau, 1995). The sources of such pollution include: 1/ sulphur dioxide emissions produced by the burning of fossil fuels (oil and particularly coal with content of sulphur), 2/ nitrogen oxides mostly generated by combustion engines, and 3/ ammonia emissions (originating from manure and sludge produced by mass scale animal breeding) converted by nitrification processes to nitric acids. The capability of soil to neutralise acid deposition depends on its chemical composition (which controls cation exchange and attenuation capacity), its physical condition and the land use. Generally, thin sandy soils with a coarse granular texture, lacking in nutrients and developed on acid crystalline rocks have a low neutralising capacity. Shallow aquifers beneath such soils are badly buffered and thus highly vulnerable to acidification. Leaching of sulphur (as sulphate), nitrogen (as nitrate), aluminium and heavy metals into groundwater as a consequence of soil acidification is observed in several regions in Europe. Rocks with a high content of calcium carbonate such as limestone, dolomite, marlite and calcareous marl, have great neutralising capability and therefore low vulnerability to the adverse effects of acid deposition is registered in groundwater within these rock environments. However, exhaustion of the neutralising attenuation capacity of the unsaturated zone leads to a lowering of pH values and subsequent gradual groundwater quality degradation.

3.7 Groundwater salinisation in coastal areas Groundwater salinisation in coastal areas is a specific category of groundwater pollution. Coastal aquifer salinisation occurs particularly when the rate of groundwater exploitation exceeds mean annual recharge, the interface between saline and fresh water is disturbed and thus conditions for the invasion of saline water into the groundwater body are created. However, the risk of saline intrusion also depends on location and density of abstraction wells and wells construction and depths. An interface between fresh and salt groundwater occurs naturally due to the high difference in

32



16/04/2008

11:14

Page 33

density between fresh water (0.99 g.cm-3) and sea water (1.026 g.cm-3). The best method for the identification of the interface is electrical logging carried out in monitoring wells. The interface between salt and fresh groundwater is in fact marked by a zone of mixed water (transition zone), the thickness of which is controlled mainly by the intensity of the ocean tides, the stream flow changes, the volume of flow towards the seashore and the aquifer’s physical and chemical properties. Saline water intrusion results in serious degradation of groundwater quality. High chloride, sodium, magnesium, and sulphate content and high salinity and TDS are typical for brackish water. Recuperation of groundwater quality is a long-term process and water supply wells have to be often temporarily or even permanently abandoned. The design and operation of an early warning monitoring network is a very effective tool for the prevention of groundwater quality deterioration in coastal and small islands areas.


33


16/04/2008

11:14

Page 34

4

EARLY WARNING GROUNDWATER QUALITY MONITORING PROGRAMMES 4.1 Introduction The main objective of an early warning groundwater quality monitoring strategy is to identify and to foresee the threats on groundwater quality still in the unsaturated zone and thus to timely define measures leading to protection of the aquifers until large volume of groundwater become polluted. However, early warning monitoring should also incorporate saturated aquifers, particularly their uppermost part. The study of the soil-rock-groundwater environment and pollutant movement between the ground surface and the water table has been mostly ignored in the past. At present, particularly when pollutant hydrogeology issues are investigated, monitoring and assessment of unsaturated zone properties (matric, osmotic and gravitational potential and hydraulic features) with respect to its ability to store, retain, remove and attenuate pollutants and delay their migration to the saturated aquifer, are crucial tools for groundwater quality management. Early warning monitoring of both the unsaturated and saturated zones generates data about groundwater quality and the movement and fate of pollutants and is an important element for groundwater pollution risk assessment and designation of groundwater protection policy. Institutional and technical capacities for the efficient development and implementation of early warning monitoring strategy need to be developed. Institutional capacity building includes: 1) the establishment of governmental institutions on national and local levels to formulate effective water policy, provide sustainable management of water resources and carry out the required administrative operations, 2) the establishment of a transparent and coherent legal framework, regulatory statutes and standards, 3) the creation of governmental water quality-control mechanism based on the polluter pays principle and on the implementation of repressive and stimulating instruments, 4) the recruitment of qualified and experienced human resources and 5) public awareness and information about groundwater protection and pollution issues. Technical capacity mainly includes: 1) groundwater system investigation and evaluation of the area to be monitored, 2) assessment of groundwater system vulnerability and pollution risk, 3) identification and inventory of potential and existing threats to which the groundwater system is exposed and evaluation of their nature, extent and impact on groundwater quality, 4) setting-up of field and laboratory monitoring equipments and procedures.

34



16/04/2008

11:14

Page 35

Early warning groundwater quality monitoring is technically demanding and also a financially expensive process in terms of capital, installation, operation and maintenance costs. However, the implementation of a groundwater early warning monitoring strategy is many times less expensive than the costs related to aquifer remediation and the investments required to overcome social and ecological impacts of groundwater pollution. It is worthwhile noting that, in spite of huge budgetary investments and long periods of aquifer remediation there are only few cases known when permanent restoration of aquifer conditions to acceptable groundwater quality levels has been achieved. Many times groundwater quality conditions in the aquifer improve for a short period of time, but pollution levels increase again after changes in the mode of aquifer management and /or raining seasons. Early warning monitoring is particularly important in the soil-root-unsaturated zone and the uppermost part of the aquifer. However, early warning monitoring is also effective when lateral movement of a pollution plume through the aquifer or upward penetration of pollutants through the bottom of an aquifer are observed and before they reach and affect the quality of groundwater supplies. At the same time, early monitoring is an appropriate method for the identification of pollution leaks from landfills, waste pits, lagoons and other point pollution sources. Changes in vegetation cover detected by geobotanical and photographic methods are also important indicators of threats to groundwater systems and supplement regional and particularly site-specific early warning monitoring systems. The important initial aspect in considering a early warning groundwater monitoring is the definition of objectives and information needs. Clearly defined objectives are essential to achieve the expected results and they have to be stated before the monitoring system is established and the first sample of groundwater is taken. Objectives control the extent and variety of monitoring technical activities (monitoring networks design, wells construction, sampling frequency, variables analyzed and others). Transformation of collected, processed and evaluated data (see chapter 6) into a usertailored information product is a second important step of early warning groundwater monitoring. It should be emphasized that the cost of the monitoring system design and operation will only be acceptable if the relevant information is transmitted on time and in an intelligible form to the users (planners, regulators, managers, decision makers, stakeholders, scientists, public) and effectively utilized. Simple, easy to understand information that can be translated into practical instructions is necessary for potential polluters (usually not professionals) and the general public. These same data should be transformed into more sophisticated information supported by GIS techniques to be presented to professionals, governmental authorities, planners and water managers. Integration and coordination of early warning groundwater quality monitoring activities with groundwater, surface water, meteorological and soil monitoring networks is recommended, because of the close relations between them. Different approaches are applied to national, regional and site-specific early warning monitoring programmes (Vrba, 2000): (a) National groundwater monitoring programmes are mainly related to data collection of natural background levels and the current state and trends of groundwater quality in time and space across the entire country. An elaboration of a conceptual model of the main country aquifers and assessment of their quality and vulnerability is essential before the design of a national monitoring network is proposed.


35


16/04/2008

11:14

Page 36

(b) Early warning monitoring systems on a regional level are focused on specific groundwater quality problems of selected country areas. The objective of regional monitoring system is to acquire statistically significant data sets which support regional management of groundwater resources quality, groundwater protection policy and decision making in regional integrated land use planning and agricultural practices. (c) For site specific problems, monitoring systems are established around potential point pollution sources to give an early warning of groundwater quality changes and deterioration. They are also located near existing point pollution sources to observe and control vertical and lateral movement of pollution plumes, and to detect the effectiveness of applied protection measures or the results of pollution remediation. Different methodological approaches for each of the above monitoring systems are desirable. However, comprehensive groundwater system analyses based on evaluation of the properties of aquifers and their vulnerability, identification of the threats to which groundwater is exposed and risk assessment of groundwater pollution are a basic premise of the design and effective operation of all early warning groundwater quality monitoring programmes.

4.2 National groundwater quality monitoring programmes The monitoring network of a national early warning monitoring programme would be composed of baseline and trend monitoring stations as classified by Maybeck (1985). Monitoring stations are preferably located outside of the influence of pollution sources or groundwater abstraction sites. The design of monitoring networks should permit separate measurement and testing of individual aquifers and their vertical hydrochemical profiling. Permeability and porosity of a rock medium involve significantly on design of monitoring wells. Design of monitoring wells in rocks having fissure permeability and secondary porosity is more complicated than monitoring of groundwater in a rock medium having primary porosity and diffuse groundwater flow system. In karst regions where groundwater flow is in conduits or in large open fissures, monitoring of springs and other discharge phenomena of the groundwater system (base flows of streams), allows more representative observation of groundwater quality than monitoring wells. The importance of groundwater as a drinking water source and the vulnerability of aquifers control the density of monitoring stations (e.g. wells, springs) and frequency of monitoring. However, monitoring methods are not yet standardized and vary in each country. Groundwater basin or aquifer system are the basic monitoring units of national groundwater monitoring programmes. Sampling frequency is generally low, in European countries mostly 2– 4 times per year or less; in fractured and fissured consolidated rocks and particularly in karst areas, with secondary permeability features, sampling is often more frequent. However, automatic in situ measurements of pH, temperature, electrical conductivity oxygen demand and other substances at daily or weekly intervals, along with groundwater level measurements, are carried out on many monitoring wells within national monitoring networks of several countries and give early warning of changes in chemical composition of groundwater. Until now, the early warning monitoring approach has had limited application in national monitoring networks, and monitoring is mainly related to the saturated zone. One such example is the

36



16/04/2008

11:14

Page 37

National Monitoring Network of the Netherlands (Duijvenbooden, 1987). In the shallow aquifers of the Netherlands monitoring wells were installed having three screened segments at different depths. The wells are used to monitor the chemical composition of groundwater and nitrate vertical movement in highly vulnerable aquifers that extend over large areas under intensively cultivated arable land.

4.3 Regional groundwater quality monitoring programmes Regional early warning programmes are in operation in several countries and they are generally designed to monitor both the unsaturated and saturated zone. The trend and impact stations dominate in regional monitoring networks. Monitoring of vertical movement of water and pollutants in the unsaturated zone (by lysimeters, extraction of interstitial water from core samples, suction caps, direct push sediment sampling or other methods, see chapter 5) and saturated zone (nested monitoring wells with single screened segment placed at different depths, small diameter piezometrs nest placed at different depths of a monitoring well, implementation of multilevel samplers in a single monitoring well, use of packers in monitoring wells in anisotropic aquifers, horizontal monitoring wells, separation pumping techniques, sampling under unaerobic conditions using a vaccuum technique and others, see chapter 5) is implemented particularly in areas affected by diffuse pollution of phreatic aquifers, in the catchment areas of public groundwater supply systems, in coastal aquifers and in wetlands regions. The extent and risk of human impacts on vulnerable groundwater systems control the design of early warning groundwater quality monitoring networks, monitoring wells construction and the range of variables analyzed for. Sampling frequency is higher than in national monitoring programmes. Sensors for automatic in situ measurements of selected substances in monitoring wells are often applied. Statistical techniques help to define an optimal sampling frequency (monthly, weekly or even daily). There follow some examples of regional early warning monitoring systems focused on the impact of agricultural activities on groundwater quality, particularly diffuse nitrate pollution and on the protection of catchment areas of groundwater supplies. (1) In Great Britain, deep profiles of pore water in the unsaturated zone were obtained from the British Chalk by Foster and Young (1980). They used microanalytical techniques on pore-water samples centrifuged from cores obtained by periodic destructive sampling. Many unsaturated zone profiles were obtained beneath long-standing arable farmland, dryland pastures converted to cereal cultivation and permanent dryland pastures. Unsaturated zone monitoring significantly improved the knowledge of complex processes and interactions between the quality of vadose zone water, soil moisture conditions, rainfall regime and infiltration rate, agriculture cropping regime and nitrogen fertilizer management (e.g. Foster et. al., 1982; Lawrence and Foster, 1986; Geak and Foster, 1989; Whitmore et. al., 1992; Foster and Chilton, 1998). Monitoring confirmed generally slow average rates of unsaturated zone transport of contaminants leached from cultivated soils; even for non reactive pollutants vertical fluxes did not exceeded 2 m/year beneath non irrigated land and 5 m/year under irrigated soils (Foster, 1976; Bouwer, 1987). Vertical stratification of groundwater quality bellow the outcrop with peak concentrations of major ions in the top of the saturated zone and N–NO3 progressive decrease with depth may be seen on the Figure 4.1 (Foster at al., 1986).


37


16/04/2008

11:14

Page 38

Figure 4.1

Comparison of pore-water profiles from unconfined and confined Triasic sandstones, south Yorkshire UK (Foster et al., 1986)

(2) A regional monitoring system for shallow vulnerable aquifers has been operated in the Czech Republic for the study of the areal distribution of nitrates in the soil-groundwater system (Pěkný, Skořepa and Vrba, 1989). The monitoring network was installed in a water table shallow aquifer in the fluvial deposits of the Elbe river valley covering an area of 3,000 km2 devoted to long-term crop farming on arable land. Monitoring wells having screens at three different depth intervals (Figure. 4.2) confirmed movement of a nitrate front from the upper part to the bottom of the aquifer within a five year period (Vrba and Pěkný, 1991). The transport and transformation processes of nitrogen compounds in the soil-groundwater system have been also studied. The intensity of the processes that take place in the nitrogen mobilization/immobilization cycle has been determined using the nitrification constants of soil samples. The analysis has proved the considerable importance of these processes for the total balance of nitrate nitrogen. The intensity of denitrification processes over the unsaturated zone’s profile was determined on the basis of the NO3-N and Cl ion content and their ratio in samples taken from the soil, deposits of the unsaturated zone and in groundwater (Figure 4.3). In the non-vegetational season in the recharge period, the NO3-N /Cl ratio was 1.5–3.5 over the 0.0–0.3 m soil profile. Higher nitrate contents were indicated in the upper, more aerated soil

38



16/04/2008

11:14

Page 39

layers. Over the unsaturated zone profile, due to the decreasing permeability and porosity, the concentration of nitrate nitrogen and chlorine declined with increasing depth. At the groundwater table level, 2,5 m under the surface, their content was about 60% lower than their relative average content in the soil layer (Beneš at al., 1989). Assuming that 60 % of nitrogen based on the NO3 -N/Cl ratio in the unsaturated zone are largely accounted for by the denitrification processes, the remaining 40 % of nitrogen flow away with the groundwater.

Figure 4.2

Changes in hydrochemical profile of shallow fluvial aquifer in the period 1984–1989. Monitoring well HP-65, Middle Elbe region in Central Bohemia, the Czech Republic

Figure 4.3

N -NO3 distribution in the vertical profile of unsaturated zone. Experimental station Samšín, The Czech Republic

The dynamics of nitrogen uptake by vegetation was systematically being determined with respect of vegetational and nonvegetational seasons. The Figure 4.4 indicate the time/space development of N -NO3 concentrations in the soil profile, with fertilizers applied to the soil


39


16/04/2008

11:14

Page 40

surface void of vegetation as well as at the time of the optimum growth of vegetation. Both figures suggest that fertilizer application in the nonvegetational season is not effective. During this period nitrates penetrate the soil profile to greater depths, thereby the risk of groundwater pollution is much more greater than if nitrogen uptake by vegetation and its root zone occurs.

Figure 4.4

Vertical distribution of N-NO3 - concentrations following application of potassium nitrate with limestone in the non-vegetational (left) and vegetational (right) season. Experimental station Samšin, The Czech Republic (Benes at al., 1989)

Study of the soil organic matter state is essential for gaining insight into the processes which control the amount of nitrogen leached into the saturated zone. A significant indicator of the soil organic matter state is the carbon/nitrogen ratio. When C:N is greater than 10, the free nitrate ions are immobilized by the microbial biomass; when C: N is less than 10 , the ammonia released during mineralization processes is utilized by heterotrophic microflora for protein synthesis, and its surplus is oxidized to NO3 by nitrification bacteria. The intensity of these processes depends mainly on: the soil’s hydrothermal conditions (temperature, initial and incubation moister), the composition of the soil’s organic and inorganic components, the CO2 content in the soil atmosphere, the amount of remineralized NH4 needed in nitrification processes as well as on the sowing procedures, the types and doses of organic and inorganic fertilizers applied and techniques of soil cultivation. When the stability of soil organic matter is disturbed, inorganic nitrogen compounds and organic carbon-nitrogen compounds are washed out into the unsaturated zone and groundwater. Perturbation of the organic carbon and nitrogen balance in soil, with consequences for the groundwater quality, occurs particularly when traditionally crop rotation is replaced by monoculture (over a monitoring period of 4 years the nitrate content increase in groundwater under monoculture was twice as high as for soil with crop rotation), or when organic fertilizers are replaced by inorganic fertilizers or the doses of inorganic fertilizers suddenly increased (both are reflected in the degradation of soil organic matter which leads to an increased influence of stochastic processes and loss of control over ground-

40



16/04/2008

11:14

Page 41

water quality). However, research has also revealed the significance of short-term changes for the nitrogen balance due the immobilization and denitrification processes in a year’s cycle. Generally, short –term cyclic changes in nitrate content occur in the aquifer upper parts, depending mainly on the climatic conditions. The long-term s and the upward trend in nitrate content in groundwater reflect human impacts, especially farming effects. Based on this early warning monitoring groundwater system, several changes of agricultural practices have been proposed to reduce conflict of interests between agricultural and water sectors. However, restoration of the soil-groundwater system’s quality is a complicated and longterm process requiring improvements in the management of nutrients and chemicals applied on the farm land, control over the extent and intensity of agricultural activities particularly in regions having vulnerable and valuable aquifers and consistent attention to natural conditions. (3) An early warning monitoring strategy on a regional scale is an important part of the protection policy and management of public groundwater supplies. Monitoring networks are established in outer or second degree protection zones encompassing usually the whole source catchment area. Water supply wells or springs and monitoring wells located in designated protection areas form monitoring networks. Sampling frequency of water supply wells is high; some variables which are sensitive to human impacts are analyzed for daily. Monitoring wells located in recharge and vulnerable areas of the supplies’ protection zones are observed less frequent. Groundwater early warning monitoring of water supply systems will alert managers to groundwater chemistry changes and/or quality degradation at an early stage, allowing them to protect and manage aquifer systems so that water is supplied according to available drinking water quality standards. Monitoring recharge and vulnerable areas of groundwater supplies is of paramount importance. In these regions land-use activities should be controlled and restricted under special guidelines or even prohibited under an established legal framework. Distribution of benefits and costs and related compensation measures must be implemented for the population, especially farmers, whose economic activity is affected by restrictions in groundwater protected areas. (4) Natural ecosystems of regional extent, particularly wetlands, can also benefit from the implementation of early warning regional groundwater quality monitoring. Wetlands are highly vulnerable to toxic pollutants however, they have a high microbiological capacity and an ability to degrade many organic pollutants. Wetlands are closely connected with aquatic systems. Monitoring and evaluation of the relation between wetland hydrological regime and adjacent groundwater system is therefore an important task. Early identification of pollution impacts on groundwater in wetlands, natural reservations and parks, is an important monitoring activity that may help maintaining the ecological sustainability of these delicate environments.

4.4 Site specific groundwater quality monitoring programmes Site specific early warning groundwater quality monitoring networks established around potential or existing point pollution sources (e.g. waste disposal sites, oil storage facilities, industrial and mining sites) operate independently of national or regional monitoring networks. Good knowledge


41


16/04/2008

11:14

Page 42

of groundwater system, aquifers boundaries and hydraulic conditions, groundwater flow nets and pollutant behaviour is desirable before early warning site specific monitoring is designated. The impact stations (Maybeck, 1985) are the core elements of site specific monitoring networks. A high density of monitoring wells (designated according to origin, magnitude and duration of pollution and mobility and extent of pollution plume), designation of suitable in situ and on-site remote sensing monitoring methods (selected with respect to the specific properties of pollutants), high sampling frequency, and analyses of variables chosen with respect to the type of specific pollutants are typical for site specific early warning monitoring programmes. Monitoring of the local climatic conditions is particularly important in regions repeatedly affected by catastrophic rains, storms and other catastrophic atmospheric phenomena. Installation of well screens and monitoring device selection and their placement with respect to pollution source properties is critical, particularly when the unsaturated zone is highly stratified. In such a case, lateral pollution migration prevails and a combination of vertical and horizontal monitoring facilities should be utilised both to avoid drilling through polluted soil-rock media and dissemination of pollution into not yet polluted aquifer and to ensure placement of the pollution sensors at the right depths. However, the design of monitoring wells of site specific early warning monitoring system is mainly adapted on observation of water and pollutant vertical movement at different depths of the unsaturated zone or on observation of the aquifer vertical hydrochemical profile. Implementation of sophisticated remote sensing methods in combination with in-situ monitoring of soil gas, soil solute and ground water in the chapter 5 supports early detection of groundwater quality problems, before the pollution reaches the groundwater level. Some of the monitoring wells that are located close to the potential pollution source can be used as recovery wells when groundwater pollution occurs. Reference monitoring wells should be also established as part of early warning monitoring network to observe the natural groundwater quality. Operation of site specific early warning monitoring networks is usually limited to a specific period (e.g. until the industry, mining facility or other potential pollution sources cease the operation, pollution plume is clean up, etc.). Early monitoring of pollutants in the unsaturated zone and pollutant fluxes at maximum concentration in the water table before they are diluted in the aquifer, gives ample time for remedial action to be undertaken before massive pollution of groundwater occurs. Such measures enable establishment of a quantitative relationship between the amount of a pollutant released on the ground and the amount that reaches groundwater body. In some European countries and the USA, early warning monitoring of the unsaturated zone and the saturated aquifer around and beneath potential pollution sources (e.g. municipal landfills or other facilities) is obligatory under the relevant legislation. Vertical and horizontal monitoring stations are designated during the construction of treatment facilities.

42



16/04/2008

11:14

Page 43

5

SOME TECHNIQUES USED FOR EARLY WARNING GROUNDWATER MONITORING Remote sensing and on-site methods are both widely used for the early detection of pollution in the soil, the unsaturated zone and the uppermost part of the saturated zone. Remote sensing techniques (geobotanical, airborne photographic and geophysical methods) are particularly helpful for the detection of pollution in the soil and the unsaturated zone. A large variety of on-site methods (core sampling and groundwater sampling) are applied in hydrochemical profiling of the unsaturated and saturated part of the groundwater system and for studying the vertical movement of a pollution plume towards the water table and in the saturated aquifer. The following text describes the most frequently used methods, which provide a considerable time between pollutant detection and its arrival into groundwater supply systems, recharge and vulnerable areas of aquifers, or as saline intrusion in coastal areas.

5.1 Surface methods 5.1.1 Introduction Experience has proved that geobotanical and photographic methods are most effective for the early detection of soil and shallow groundwater pollution when implemented together. Both methods are based on the response of vegetation cover to the presence of specific substances in the soil, surface water and groundwater. The detected substances causing pollution are not always harmful to vegetation; sometimes they could even be beneficial for plants and the biosphere. Therefore, 1) the state of plants health, shift of phenophases (i.e. blossoming and ripening), density of vegetation cover, and 2) the presence or absence of certain plant species and communities and the change of successions, size, habitat and their abundance, both are helpful tools to reveal the presence of pollutants in soil and rock environments and the hydrosphere. Other photographic methods are rare, except for recording the state and changes of rooted plants which is limited to sites where polluted groundwater or acid mining water discharge into the rivers or lakes. The detection of pollutants is then based on their physical and chemical properties which


43


16/04/2008

11:14

Page 44

cause changes of reflectance of surface water both in the visible and the near infrared spectral range. Unfortunately ultraviolet bands, which are theoretically suitable for revealing oil slicks on the water table, cannot be employed in photography because of the lack of special films; the use of necessary multispectral scanners lies beyond the range of photographical methods. Geophysical methods are successfully implemented if there is sufficient contrast between the measured parameters of the geological system, the groundwater body and the pollution plume. Generally, geophysical methods are more effective in detection of inorganic pollutants and less effective in the case of organic pollutants, particularly in the case of Dense Nonaqueous Liquids (DNAPLs). Continuous electromagnetic profiling measurements with high lateral resolution and resistivity methods with high vertical resolution are those geophysical methods which are most frequently used for detecting the extent and flow direction of a pollution plume. However, several others geophysical techniques are often applied in pollutant hydrogeology, such as magnetometry and electromagnetic induction for the location of pollution spills from damaged underground pipelines or tanks, radioactive spectrometers for the identification of aerial and site-specific radioactive pollution, and borehole logging methods for the detection of a groundwater-saline water interface.

5.1.2 Geobotanical methods Geobotanical detection of soil and shallow groundwater pollution depends on the site’s natural characteristics, the chemical and physical composition of the pollutants and particularly the properties of the vegetation cover. The principle of pollution detection is to compare the state of vegetation at the pollution site with a botanically similar unaffected area. When there are pre-pollution botanical data available, comparison should be made with the present state of vegetation affected by pollution. The species of vegetation within the investigated area and their variability, particularly the presence of those plants which are sensitive to and those plants which are tolerant to pollution, form the basis of geobotanical methodologies. The resistance of certain species, however, depends on several natural factors such as soil organic and mineral matter and moisture content, groundwater level and groundwater chemistry, and the local climate. The botanical indicative value is a useful parameter in this context which rises with the number of species and their diversity and depends positively on the presence of trees with deep roots systems (Pyšek et al., in press). The pollutants most easily detected by geobotanical methods are oil and oil products, natural gas, fertilizers, pesticides and heavy metals. Oil and oil products affect vegetation in different ways according their composition and concentration. While a small amount of oil hydrocarbons in water or soil may have no or even a beneficial effect on plants, strong concentrations can result in necrosis of leaves leading to the total death of vegetation in the most drastic events. The most frequent symptoms of oil pollution are, in case of individual plants, growth (nanism, gigantism), leaf colour change (change of green colour and its shades, yellowing, whitening), turgor (wilting), necrosis, defoliation, sterility and shift of phenophases (blossoming, ripening). In the case of plant communities the typical symptoms are presence, variety and diversity of species and the density of plants or canopy. The response of plants to pollution should be carefully analysed. It depends both on the vegetation itself and also on complex environmental factors, such as water and nutrient content, temperature, air pollution, etc. Nevertheless the general division of plants into two basic groups, sensitive and tolerant, has proved to be suitable for pollution detection by geobotanical methods.

44



16/04/2008

11:14

Page 45

Sensitive vegetation includes Scots Pine (Pinus sylvestris), Silver Birch (Betula pendula), Mugwort (Artemisia vulgaris), Fat Hen (Chemopodium album) and Meadow Fescue (Festuca pratensis). Resistant plants include Common Elder (Sambucus nigra), Stinging Nettle (Urtica dioica), Wood Small Reed (Calamagrostis epigejos), False Acacia (Robinia pseudoacatia), Cow Parsley (Anthriscus sylvestris) and others. These representative plants can be used as photoindicators of pollution due to their common occurrence in anthropogenically affected areas (Pyšek at al., in press). Pyšek A.(1983) formulated the following main geobotanical procedures for the detection of hydrocarbon pollution: 1/ gather information on all significant human activities in the studied area, 2/ evaluate the occurrence and vitality of the petroleo-sensitive species, 3/ observe the reaction of petroleo-tolerant species, 4/ record all growth abnormalities performed by species present in polluted area, 5/ observe the shift of flowering phenophases, 6/ try to detect the existence of the border zone of vigorous plant growth. Natural gas generally impacts indirectly on vegetation. Biodegradation of gaseous hydrocarbons eventually leads to carbon dioxide production which causes a reducing environment and the absence of oxygen needed for plant life. In the Czech Republic, field tests were carried out to find those plants which were either sensitive or tolerant to natural gas with a view to their cultivation as indicators of leaks on land above natural gas pipelines. The most sensitive were found to be sugar beet, feed cabbage, sunflower, barley, alfalfa, potatoes, maize and larch. The most resistive to leaking natural gas were beans and onions (Pyšek P. et al., 1988). The indicative symptoms of the impact of gas on vegetation are: deficiencies in growth (height reduction, lodging of straws), anomalies in development (shift of phenophases), physiological changes (colouring of leaves) and death of highly stressed plants which make very striking soil circles above large leaks in pipelines. Nitrates generally impact on vegetation in such a way as to be almost indistinguishable to the effect of other fertilizers. Nevertheless combined usage of nitrophyllic species which react with a loss of vitality to a lack of nitrogen in soil and the negative response of nitrophobic species to an abundance of nitrogenous matter, enable phytoindication. Generally, common species of corns show increased growth, cover density and have deep dark green colours with higher nitrates concentrations. The amount of some weeds declines with the rising doses of N-fertilizers: Persian Speedwell (Veronica persica), Field Pansy (Viola arvensis), Black Medick (Medicago lupulina) etc. Diversity of species is highest at plots having lowest nitrate content. (Pyšek P. et al., 1988). Heavy metals are generally toxic on vegetation, even in low concentration (Ernst, 1974). Toxic effects manifest in dwarf growth -Scots Pine (Pinus sylvestris), leaves necrosis-Silver Birch (Betula pendula), change of green colours- Norwegian Spruce (Picea abies).Strong pollution by heavy metals causes, at general, absence of trees and lower variety of herbs. Among resistant species against the impact of heavy metals may be mentioned Vernal Sandwort (Minuartia verna), Sea Pink (Armeria maritima), Bladder Campion (Silene vulgaris), (Pyšek A. et al., in press).

5.1.3 Photographic methods Photographic pollution detection primarily makes use of the state of and changes in vegetation caused by the presence of pollutants in the rhizosphere, which can be recorded on films at visible and near infrared wavebands. Photographic detection employs two causally and temporally


45


16/04/2008

11:14

Page 46

connected phenomena: 1) the state of and changes in plant health, 2) the density of and changes in vegetation cover (canopy) below ground. Infrared (IR) photography detects vegetation stress, which is manifested by loss of reflectance, chiefly at wavebands greater than 700 nm. The lost of cells turgor, which is the reason for the decline of near photographic reflectance, generally appears sooner and more distinctly than changes of colour in visible light. Of cause there are more reasons which may lead to the loss of near IR reflectance and to the worsening of state of plant health (mechanical damage, pests, drought, air pollution). Therefore close cooperation between specialists evaluating photographs and geobotanical data is vital. The optimal means for recording initial physiological changes is a combination of true colour and IR colour photography. The state of highly stressed dying vegetation is easily visible on both panchromatic and IR black and white film. Photographic detection of simulated escapes of natural gas from a pipeline at an experimental site gave very good results (Svoma 1990). Distinctive symptoms of vegetation stress were found using colour IR photography 10 to 21 days after the start of the experiment. The main diacritical symptom was loss of vegetation cover (canopy). The affected area varied from 13 to 36 m2 according to the sensitivity of the plants. The agricultural plants most sensitive to the gas proved to be winter wheat followed by alfalfa with regard to the time of the response and potatoes and sunflower with regard to the extent of the area affected. The plants were reacting to doses of 36 to 100 cubic metres of natural gas injected into subsurface holes. Photographic detection of groundwater pollution depends on the pollutant class and concentration, the sensitivity of the plants and the relation between root systems of sensitive plants (mostly trees) and the level of the groundwater table (GWT). An investigation carried out at geologically different regions of former Czechoslovakia proved that at many contaminated sites the GWT was at a depth (up to 4 metres) which enabled contact between the rhizosphere of trees and groundwater. Thus favourable hydrogeological conditions for pollution detection by photographical methods existed. A study of 2,201 boreholes used for pollution remediation showed that in 5,3% bore-holes the GWT was from 0-1 m below ground; in 15,2% between 1 and 2 m below ground; in 48% 2 to 4 m below ground; in 21,9% 4 to 8 m and in 9,6% the GWT was more than 8 m below. Diversity of plant species is not at all advantageous for pollution detection by photographic methods (in contrast to geobotanical methods). The difference in reflectance of species both in the visible and near IR bands, makes evaluation of photographs more difficult and may even mask the changes of reflectance caused by vegetation stress. The most suitable vegetation for this approach is that provided by monocultures (agricultural or forest). Over-fertilized monocultures on arable land with potential leakages of nitrogen into the soil and groundwater system are easily detectable by both true colour and IR colour films. Broad leaf trees enable better recognition of stress that needle-leaved trees because of the generally low IR reflectance of conifers. The best time for surveying depends on the kind of vegetation. Spring or early summer seems to be the most suitable. Autumn is not recommended because of the natural seasonal changes in leaf colour. Simultaneous use of vertical true colour and false colour (colour IR) photography has proved to be the optimum combination of photographic techniques for the detection of pollution impact on vegetation. The best results are given by Kodak Aerochrome IR film with a dark yellow Wratten filter 70. For the early detection of stress in conifers, additive colour

46



16/04/2008

11:14

Page 47

compensating magenta filters should be used. Aerial photography used for pollution detection should sometimes be replaced by more the practical and cheaper method of photographing from radio-operated models or moveable platforms (Švoma, 1990). The use of video cameras does not provide any additional detail to still photography. Nevertheless, it can be very useful to have immediate information regarding the state of pollution recorded in real time. Continuous (overlapping) aerial photographic surveys provide an extremely useful form of photographic pollution detection as they allow a simultaneous comparison of polluted areas and adjacent unpolluted areas and also provide a record of changes in the state of the vegetation with time. Photographic records can also be used as legal evidence because they show the state and extent of pollution at any given time. Photographic pollution detection for groundwater without the combined use of geobotanical methods is limited to sites where polluted water discharges into surface water bodies. The possibility of direct photographic detection is limited by the physical properties of the dissolved or floating pollutants transported by the discharging groundwater after their dilution and/or interaction with the surface water. The change of reflectance of the surface water affected by yellowish coloured mining water and landfill water with compounds of oxidized ferrous ions is mostly detectable in visible light and may be easily recorded on true colour films. However, the detection of oil slicks, films or layers in discharging polluted groundwater is the most common application of this methodology. It is even possible to estimate the thickness of a floating oil layer by consideration of the colour changes seen (Estes, 1972). Toxic substances discharged by groundwater into surface waters may kill a large percentage of the fish population whose floating bodies are then detectable on black and white and colour films. Such discharges may also result in a decline in algal growth which is easily detectable by IR film which is generally more efficient than true colour film in this case. On the other hand a massive increase in algal growth in ponds or lakes owing to discharged groundwater polluted by phosphates may be distinctly detected in the green or near IR wave bands using both kinds of photography.

5.1.4 Geophysical methods Geophysical methods provide relatively limited possibilities for the detection of polluting substances in the unsaturated zone. This is particularly due to the inhomogeneity of the vadose zone (which includes the soil layer close to the ground, the capillary fringe above the water table and the proper unsaturated zone in between – see Fig.5.1). The water content W of the rock medium and the water saturation Sw of the pore space both change with depth below the surface and with time (due to changes in precipitation, dry and wet seasons etc.). Therefore, the measurable physical properties (resistivity, electric permittivity, chargeability, density and the hydrogen index) of the unsaturated zone are also subject to change with depth and with time. Industrial fluids – the main sources of pollution of the soil and rock medium and consequently also groundwater – can be divided into three groups according to their chemical composition, physical properties and capacity for detection by geophysical techniques.


47


16/04/2008

Figure 5.1

11:14

Page 48

Scheme of the unsaturated zone and the uppermost part of the saturated zone with depth variations of the moisture content W and the water saturation Sw

Inorganic compounds (acids, alkalis, salts) are usually easily soluble in water and dissociate to positive and negative ions. The conductivity γwc of polluted water, is proportional to the total dissolved chemicals and exceeds by one to two orders the conductivity of fresh water. It has a great influence on the conductivity of the geological formation γt. The electric permittivity εr of inorganic solutions, however, is comparable to the permittivity of fresh water (εr = 80). Inorganic chemicals drastically change the electrical properties of the rock medium. The electric conductivity (γ t) of the upper part of the unsaturated zone (soil with suspended capillary water) will be particularly influenced. Under these conditions, inorganic chemical compounds can be detected in the soil by electromagnetic conductivity meters (e.g. EM-31 product of Geonics Com., Canada or CM-31 product of Geofyzika Com., Czech Republic), by standard resistivity arrays with short electrode spacing (0.5 m) or by multielectrode resistivity survey. This sophisticated geoelectrical method, if repeated over different time intervals, enables the estimation of the speed of spill-penetration through the unsaturated zone. Inorganic chemicals also change the charge of the soil and the induced polarization method can be very helpful in deliminating the extent of the contaminated area. Seawater intrusion in coastal areas causes a similar geophysical effect to those of inorganic compounds. However, it can also be recorded by repeated remote sensing images in the IR range (3 to 20 um). Organic compounds – typically hydrocarbons – are usually characterized by a very high resistivity (practically insulators) and very low electric permittivity (ε r < 5), approaching the permittivity of the air (ε r = 1). Hydrocarbons have generally low solubility in water and can be considered practically immiscible with water. Their transport and accumulation in geological media is controlled by their density.

48



16/04/2008

11:14

Page 49

Direct detection of organic pollutants in the unsaturated zone by geophysical methods is extremely difficult. The best way is to combine resistivity survey together with atmogeochemical methods using e.g. special equipments on detection of hydrocarbons in the soil-air, based on the photoelectric effect. Anomalies of hydrocarbon content in soil-air correlate with the areas of higher soil resistivities only if artificial or natural disturbing materials in the soil (stones, concrete pieces, PVC products etc.) are not present. Then the existence and thickness of a Light Nonaqueous Phase Liquid (LNAPL) layer on the ground water table (at the bottom of the unsaturated zone) can be determined in unconsolidated clastic sediments by means of cone penetration tests combined with geophysical logging (gamma-ray, neutron-neutron, permittivity logging). The hydrogen index of hydrocarbons is comparable to the hydrogen index of water. The top of an aquifer saturated by an LNAPL is indicated on the neutron-neutron log, the top of the 100% water saturated aquifer is indicated on the permittivity log. Under favorable conditions, the water saturation Sw and the hydrocarbon saturation Sh of the transition zone at the groundwater level can be calculated. Both groups of pollutants, inorganic and organic, have characteristic electrical properties and so their detection will generally depend on geoelectrical methods. Radioactive materials contaminate the soil as a result of incidents in their transportation and/or processing. The source of pollution can be uncontrolled spills of the transported radioactive material following a vehicle accident (resulting in a polluted area of small extent) or radioactive rains carrying radioactive elements to the ground after an accident at an atomic power plant (large extent of contaminated area, such as the radioactive pollution of large regions of Eastern and Central Europe after the Chernobyl accident). These industrial radioactive elements usually have different emitting spectra of gamma rays in comparison to natural radioactive elements and can be detected by the use of radiospectrometric methods (Fig. 5.2). The properties of the transported radioactive material spilled as a result of a vehicle accident are usually known and consequently the type of radiation and its parameters (energy and half-life) are as well. Therefore appropriate radiometric equipment can be used for detecting the extent of the contaminated area (e.g. radiometers measuring total gamma activity, or gamma spectrometers in the case of isotopes, which are sources of gamma radiation). After the Chernobyl accident, the character of the distributed isotopes following the breakdown of the atomic power plant was also known (see Fig. 5.2), gamma spectrometers were thus very useful in the delineation of the extent of the polluted area. The first two groups of pollutants (inorganic and organic compounds) have characteristic electrical properties and this is the reason for the predominant use of geoelectrical methods in their detection. Pollution identification by geophysical methods usually precedes drilling of investigation boreholes and monitoring wells and provides the first overview of the pollution problem (Mares et al., 1997; Iliceto and Mares, 2000). Delineation of a pollution plume by geophysical methods is faster and less expensive than identification of the extent of pollution by the drilling of a large number of investigation boreholes. When the polluted area is defined by geophysical measurements, the number of monitoring wells needed to delineate the borders of the pollution plume is significantly reduced. To summarize, electromagnetic and resistivity methods are widely and successfully applied in con-


49


16/04/2008

11:14

Page 50

taminant hydrogeology, particularly for the delineation of the boundaries of a pollution plume. However, other geophysical methods should also be implemented, such as gravimetry, for sitespecific studies of polluted sites, thermal measurements to identify the extent of higher temperatures around landfills, or radiometric methods for delineation of polluted area by radioactive material. For the determination of pollution type and intensity in the vertical profile of the saturated zone (e.g. hydrocarbons floating on groundwater table, fresh-salt water interface) various electrical, electromagnetic and nuclear logging methods are often utilised in investigation or monitoring boreholes.

Figure 5.2

50

Energy gamma-ray spectrum measured on the Earth’s surface in Prague, The Czech Republic, (1) after the Chernobyl accident on May 4, 1986 (2) compared with the natural gamma-ray spectrum of rocks



16/04/2008

11:14

Page 51

5.2 On-site methods 5.2.1 Introduction This work does not intend to give a comprehensive review of all available hydrological sampling methods that can be employed for developing early warning monitoring systems in the unsaturated and saturated zones. In reality many of the methods described in the literature can be used for such a purpose provided that the scientists are willing to apply the groundwater early warning monitoring concept. Moreover, due to the wide range of possible scenarios where an early warning monitoring method could be applied (e.g., unsaturated zone, water table region, contact between two overlying aquifers, salt water-fresh water interface, Fig. 1.1); it is difficult to give a general methodological approach. There is, however, a basic fundamental principle: in an early warning monitoring system the sampling and/or in-situ detection technique employed should allow a considerable time lag between pollutant detection in the sampling (detection) system and pollutant arrival into protected target area (e.g., aquifer recharge area, pumping well, water supply region). The desired time lag should be established according to available response methods to overcome the problem. For example, the required time lag will be different if in response to a slug of organic pollutants it is necessary to build a water treatment unit or to operate an existing one.

5.2.2 Suction cups Pore water in the unsaturated zone will not flow into a well as long as the rock volume considered is under tension – which happens when the pores are partially filled with water and partially with air. In this situation the sampling is normally done using suction cups. The cups are made of porous material for which capillary forces are higher than the tension in the partially filled pores in the soils. Suction cups can be placed in the sampling location in several ways. The important principle should be, that the procedure applied during placement of the cups in the soil does not alter the soil conditions along the pathway of infiltrating water from the land surface to the location of the cup. An example of a procedure applied for sitting of suction cups is shown in Fig. 5.3 (Lindhardt, 2001). Suction cups were installed in such a way, that the soil above them was not disturbed. The cups were installed from excavation pits in two layers one meter apart. Suction cups, when installed at various depths, will support studies about the rate of transformation of potential pollutants during movement through the unsaturated zone. However, great care has to be taken to avoid any boundary influence on the readings from the cups. This can be achieved by sitting the cups away from the edges of the field and away from places where agricultural machines have to make specific manoeuvres such as turning; doses of fertilizers and pesticides in such places are different from doses applied at the remaining part of the field. The holes for installation of the suction cups were drilled to the desired depth minus 20 cm; the last 20 cm being completed using a steel rod with a diameter corresponding to the diameter of the suction cup. A thick slurry of silica flour was poured to the bottom of the hole immediately before


51


16/04/2008

11:14

Page 52

inserting the suction cup. The hole was then sealed with 20 cm of clay pellets and back-filled with bentonite. Each suction cup is connected to an individual sampling bottle, placed at the edge of the field being observed, via tubing running at a depth not accessible by ploughing. There is a large variety of suction cups available and careful selection of this equipment is advisable. Compatibility of the grain size of the monitored material and the pore size of the suction cups has to be assured. Use of silica slurry is recommended for the initial phase of the measurements. Refrigeration or some other form of sample preservation should be considered if the monitored pollutant requires it.

Figure 5.3

52

A) Location of suction cups, B) Cross section showing the installation of the suction cups (modified from Lindhardt, 2001)



16/04/2008

11:14

Page 53

5.2.3 Direct push The chemical composition of pore water in the unsaturated zone is frequently a precursor of changes in the chemical composition of groundwater. Scarcity of programmes monitoring pore water in the unsaturated zone may be the result of the technical difficulty in sampling this region. Drilling methods to obtain sediment and water from the unsaturated zone are relatively expensive and difficult to use. Continuous in-situ sampling techniques (e.g., porous cups) can only be used in regions of the profile where pore-water content is relatively abundant (e.g., the root zone after rain or irrigation; the capillary fringe – see for example, EPA, 1986). A convenient sediment probing technique which is cost-effective and produces extensive sampling rates is based upon the ‘direct push’ of sampling tools into the ground without the use of drilling to remove sediment to provide a pathway for the tool. An example of such an instrument is the Geoprobe®. The Geoprobe® instrument relies upon a relatively small amount of static weight (the weight of the carrier vehicle) combined with percussion as the energy for advancing the tool string. The probing tools do not remove cuttings from the probe-hole but depend on compression of soil or rearrangement of soil particles to permit advancement of the tool string (Christy and Spradlin, Geoprobe® Systems; Geoprobe® Systems Catalogue 1998-1999). With the direct push technique it is possible to obtain, for example, continuous sediment-cores or discrete core samples (Ronen et al., 1998), groundwater samples, and gas samples from the unsaturated zone (Affek et al., 1998). Soil probing equipment is typically employed for site investigations to depths up to 20 m, but has been used by the ‘direct push’ method up to depths of 30 m (Affek et al., 1998). Generally, direct push methods can not penetrate boulders or consolidated rock layers (e.g., limestone). In a study conducted in the Coastal Plain aquifer of Israel a Geoprobe® was used to sample the unsaturated zone in two sampling studies. Over two days, 13 continuous-core boreholes were drilled to sample a sandy and sandy-loam unsaturated zone within a radius of 5 m, in a region where the water table lay at a depth of 7 m. Three to five cores, each 1.2 m long, were collected from each borehole. The cores were collected within transparent plastic sleeves (having an internal diameter of 5 cm), which were immediately capped and subdivided into 10 cm subsections. The sediment content of each subsection was mixed inside a plastic bag and a sample was transferred into a glass vial for the determination of water content and particle density. Core samples, 20 cm long, were capped inside the plastic sleeve for the determination of bulk density and saturation percent. All samples were preserved in the field in iceboxes and were transferred to a refrigerator (5 °C) at the end of each sampling day. While drilling, there was almost no compaction of the sediment inside the plastic sleeve, and recovery (ratio between penetration of the corer to the length of the core) was almost 100%. To avoid the possibility of collecting displaced samples, the first 5 cm at both ends of each core were discarded. In the course of this study, devoted to the analysis of transport phenomena in the capillary fringe of granular sediments, approximately 550 samples were obtained from the unsaturated zone. For each sample, the gravimetric water content, the bulk density, the particle density and the saturation percent were calculated (Black, 1965). Figure 5.4 shows the gravimetric water content in the entire unsaturated profile and below the water table for 8 continuous-core boreholes obtained with the Geoprobe® (Ronen et al., 2000). Figure 5.5 depicts chloride concentration in pore water at the same study site. Pore water samples were obtained by the water addition-extraction method (Ronen et al., 1997).


53


16/04/2008

Figure 5.4

54

11:14

Page 54

Gravimetric water content (Θ) in the capillary fringe of continuous-core boreholes CF6 to CF13, in the Coastal Plain aquifer of Israel. The insert shows the gravimetric water content in the entire unsaturated profile and below the water table.



16/04/2008

Figure 5.5

11:14

Page 55

Chloride concentration in pore water of continuous – core boreholes CF2, CF3, CF4 and CF5. These boreholes were located in the same location as boreholes CF6 to CF13 (Figure 5.3), in the Coastal Plain aquifer of Israel. Note the heterogenity in the chloride concetration between profiles and within profiles. The study area is subjected to replenishment of rain in winter only. The gravimetric water content (Θ) is also shown.The shaded area in CF3 denotes the water content and chloride concentration in CF1 made solely below the water table (after Ronen at al., 1997).


55


16/04/2008

11:15

Page 56

5.2.4 Horizontal monitoring wells Pollutant transport through clayey till, sandy clays and other sediments characterized by a large variation of hydrogeological properties within short distances, is primarily governed by preferential flow pathways associated with fissures and occurrences of sandy lenses. It is obvious that sampling should be done within the preferential flow rather than within an isolated, ‘dead’ volume of clay not participating in the transport of the pollutants. Interception of the preferential flow pathways is much easier with horizontal rather than vertical monitoring wells. Horizontal wells can be divided into separated screened sections providing samples from different parts of the monitored field. An example of installation of a horizontal monitoring well (Lindhardt, 2001) is shown on Fig. 5.6.

Figure 5.6

56

Installation of horizontal monitoring wells and a section of the horizontal screen (Lindhardt, 2001)



16/04/2008

11:15

Page 57

The installation of a horizontal monitoring well takes place in three stages: drilling of a horizontal open hole is done by flushing with water (drawing A), reaming and placing of the outer casing (drawing B) and completion of the installation by placing of the screened intervals and removal of the outer casing (drawing C). The screened intervals, monitoring different parts of the well, are separated by bentonite seals preventing short-cuts between the various intervals. Each screened interval is equipped with two tubes connected to a peristaltic pump on the surface. All the tubes from the screened intervals are placed inside the inner pipe going to the ground surface.

5.2.5 The separation pumping techniques Sampling from wells will often provide samples consisting of a mixture of water with distinctly different concentrations of the pollutant considered. For an early warning system it is of paramount importance to collect samples representing the most adverse conditions in the aquifer, i.e. the highest concentration of the pollutant. The separation pumping technique helps to delineate the zones contributing the highest concentrations of the pollutant and allows the design of a remediation procedure aimed at these zones. The unpolluted groundwater zones are delineated as well and water from these zones can be discharged without treatment. Treatment cost can be reduced and efficiency increased significantly if the amount of polluted water requiring treatment is reduced. The separation pumping technique is one of the methods used for taking level-determined groundwater samples. This technique allows the collection of samples from a required depth interval and/or the definition of chemical profiles by analysis of a series of samples collected in either screened wells or open boreholes. The principle of separation pumping, illustrated in Fig. 5.7 (Gosk et al., 1992), is based on the creation of a water divide in the borehole by diverting flow to two pumps, one near the

Figure 5.7

Principle of separation pumping (Gosk et al., 1992)


57


16/04/2008

11:15

Page 58

water surface and one near the bottom of the sampling zone. By adjusting the rates of the two pumps, the position of the water divide can be adjusted to the required sampling level. If the diameter of the borehole is sufficiently large, a sampling set-up consisting of three pumps, and a heat-pulse-flow-meter can be applied, while for boreholes of smaller diameter an alternative system consisting of two pumps can be used. Two pumps, preferably with adjustable flow rate, are placed at the top and at the bottom of the well while the third pump (typically with smaller flow rate) is placed around the middle of the saturated (screened) depth with a heat-pulse-flow-meter, a device used to determine the direction of flow in the borehole, attached to it. Simultaneous pumping from the bottom and the top of the well creates a draw down in the well and inflow to the well from the surrounding formation. One part of the in-flowing water will go to the upper pump and the other part to the lower pump. The share of water delivered by each pump will be determined by the valves controlling the flow rate of the top and bottom pumps. The intervals in the geological formation supplying the upper and the lower pump respectively will be governed by permeability distribution at the face of the well screen or face of the open bore hole. In a set-up like this a zone of (almost) stagnant water will develop between the interval with a downward flow going to the bottom pump and an upward flow going to the top pump. The location of this stagnant zone is then determined by means of the heat-pulse-flow-meter. The third, smaller pump connected to the heat-pulse-flow-meter is turned on as soon as it is positioned within the stagnation zone. After the sampling is completed the procedure may be repeated with a different combination of the flow taken by the upper and the lower pump. The new position of the stagnant zone for the different combination of flow rates then has to be determined. For the purpose of early warning the set-up providing the highest concentrations of pollutant in the samples provided by the sampling pump should be maintained. The chemical profile obtained by this method will reflect the trends of pollutant concentration variation with depth rather than provide an exact chemical profile comparable to the profile obtained by analyses of pore water extracted from soil samples collected during drilling. In screened wells completed with a sand and/or gravel filter there will be a tendency to smear-out the chemical concentration profile existing in the formation, particularly if the permeability of the filter material is order(s) of magnitude higher than the permeability of the geological formation. There are several advantages associated with the separation pumping technique: existing boreholes can be used, inflow zones for polluted water can be localised and remediation actions requiring treatment of smaller amount of polluted water can be designed. The pumps do not need to be moved between repeat sampling events. It is possible to carry out multi-step separation pumping using only two pumps: one placed on the top and the other at the bottom of the investigated interval.

5.2.6 The Multi Layer Sampler (MLS) A major problem in hydrochemical studies is that of obtaining small-interval, representative groundwater samples of the undisturbed system of flow. Generally, samples are collected by pumping or by samplers lowered to different depths in observation boreholes. These procedures disturb chemical gradients and can yield mixed water samples from different horizons of the aquifer. Devices to obtain profile samples reported in the literature have been used primarily for shallow aquifers. In most of these devices the water sample is pumped to the surface and the sampling

58



16/04/2008

11:15

Page 59

intervals are greater than 0.5 m (e.g., Hansen and Harris, 1974; Oberman, 1982; Pickens et al., 1978; Harrison and Osterkamp, 1981; Molz et al., 1986a). The challenge is to sample a flowing system with minimal perturbation due either to the introduction of the sampler or the retrieval of the sample. With respect to the water table region, fluctuations of the water table necessitate a variable sampling system (not fixed with regard to the aquifer) enabling the adjustment of sampling depth according to variations in the water table. In order to sample the water table region without disturbing the chemical profile, a sampling system was developed composed of a Multi Layer Sampler (MLS) and a research well (Ronen et al., 1987a). The MLS enables water sampling at predetermined depth intervals by use of the dialysis cell technique (Mayer, 1976; Hesslein, 1976). The sampler consists of a rod with crisscrossed holes, which accommodates dialysis cells filled with distilled water (Fig. 5.8 - right). Sampling intervals (distance between dialysis cells) may be as small as 3 cm. The cells are separated in the well by seals which prevent mixing by vertical currents and diffusion (Fig. 5.8 - left). The sampler is built in a modular fashion so that several rod segments may be connected to attain any desired length. The equilibration time between the dialysis cell and a solution is determined empirically in the laboratory. To obtain a sample for the analysis of major ions (e.g., Cl-, NO3- and SO42-) the equilibration time is about 48 hours (Fig. 5.9). However, a minimum sampling period is established which enables reequilibration of the well-aquifer system to ‘normal’ hydrochemical conditions following the lowering of the sampler into the well (e.g., about 7 days in a sandy aquifer). After retrieving the sampler from the well the sample is extracted from the dialysis cell for chemical analysis. The MLS has also been used to obtain: (a) detailed vertical profiles of the horizontal component of the specific discharge, employing both a modified point-dilution technique under natural gradient flow conditions, and a mathematical diffusion model (Ronen et al., 1986), (b) samples of dissolved gases both in the water table region and the capillary fringe (Ronen et al., 1988a), and (c) samples of groundwater colloids (Weisbrod et al., 1996).

Figure 5.8

Segment of a Multi Layer Sampler (MLS) showing the dialysis cells spaced at 3 cm intervals and separated by flexible seals (left) and, schematic representation of a segment of the MLS inside a monitoring well (right). V1, C and V2, S designate the volumes and solute concentrations in water contained in the dialysis cell and sampling segment, respectively. For simplicity slots are only shown on the lower part of the PVC screen.


59


16/04/2008

Figure 5.9

11:15

Page 60

Equilibration test of dialysis cells conducted at 22 °C. To conduct this test pairs of dialysis cells (14 ml each) filled with distilled water were submerged in separated baths of a 800 ml solution (Cl - = 200 mg /l ; NO3- = 100 mg /l ; SO4 2 - = 60 mg/l). The cells were overed with nylon membranes with a pore size of 0.2 μm (Versapor V-200, Gelman Sciences). The water was continuously stirred at 126 RPM to eliminate spatial heterogeneities.

The sampler must be utilised in a screened well. It is recommended that the well is drilled without the addition of water or drilling mud. The main requirement of the research well (Fig. 5.10) is that screens be installed both above and below the water table to enable long term observations considering both short- and long-term fluctuations (monthly to yearly) of the water table.

60



16/04/2008

Figure 5.10

11:15

Page 61

Schematic representation of a research well for monitoring the water table region. Note that screens are installed both above and below the water table to enable long-term observations considering both short- and long-term fluctuations (monthly to yearly) of the water table. The PVC-coated stainless steel wire, mounted on one of the screens, was used to determine (with an ohmmeter) the exact position of the screen in relation to the water table. The well in this figure is located 15 km north of Tel Aviv (WT-2) and was used to monitorthe water table region under land irrigated with sewage effluents. The lithologicalprofile at the study site is also shown.


61


16/04/2008

11:15

Page 62

Figure 5.11 depicts typical chemical profiles obtained with the MLS in the water table region of the Veluwe region, of the Netherlands, in a forested area subjected to the input of acid rain and ammonia volatilised from cultivated land and feed-lots (Krajenbrink et al., 1988). The water table is at a depth of 7 m and the aquifer is composed of gravels, coarse sand, and clay loam layers. Figure 5.12 shows profiles of xylene, xylidine (dimethyl aniline) and toluene obtained with the MLS in the water table region of Brookhaven National Laboratory, N.Y. (BNL, Kaplan et al., 1991). The phreatic aquifer at BNL, composed of sand and gravel, has been polluted by a spill of mixed liquid fuels. Benzene, toluene and xylene have been identified in local wells at a low mg/l level. In samples taken with a pump, from depths greater than 200 cm below the water table (from the same research well into which the MLS was subsequently lowered), the concentration of toluene and xylene was 0.43 mg/l and 2.3 mg/l, respectively.

Figure 5.11

62

Chemical profiles obtained in the water table region of research well 7 in the Veluwe region, the Netherlands, in a forested area subjected to the input of acid rain and ammonia volatilized from cultivated land and feed-lots. Note that the oxygen concentration found in the dialysis cells located in the capillary fringe (empty circles just above the water table) reflects the expected concentration of atmospheric oxygen dissolved in water, at the prevailing temperature (9°C).



16/04/2008

Figure 5.12

11:15

Page 63

Profiles of xylene, xylidine (dimethyl aniline) and toluene obtained with the MLS in the water table region of Brookhaven National Laboratory, N.Y. (Kaplan et al., 1991). In samples taken with a pump, at depths greater than 200 cm below the water table, the concentration of toluene and xylene was 0.43 mg/l and 2.3 mg/l, respectively.


63


16/04/2008

11:15

6

Page 64

DATA HANDLING

6.1 Introduction An early warning monitoring system can be established to monitor different types of stress applied to the groundwater system under consideration. There are few specific properties of an early warning monitoring which make data handling for this system different from, for example, data handling for an ordinary groundwater monitoring network. While monitoring data are collected with a view to analysis some time in the future, early warning monitoring data have to be used immediately to enable important management decisions to be made. When the warning arrives there is no time for evaluation, lengthy investigations and discussions of remedial strategies. All this should be done before the emergency situation occurs. Therefore, it is necessary to make realistic simulations of the various scenarios and strategies, preferably with the help of a computer model of the aquifer system concerned. In principle all aquifer pollution cases can be regarded as irreversible within our time scale (tens to hundreds years) and so it is necessary to follow the pollutant long before its entry to the saturated zone of the aquifer. Therefore the typical early warning monitoring systems will consist of sampling stations situated in the unsaturated as well as the saturated zone of the groundwater system. Data requirements and data handling will vary for different types of monitoring systems and therefore the various early warning systems should be designed specifically for an existing stress situation or for situations where an immediate threat to the aquifer (water supply) exists. Data requirements for an early warning monitoring designed for a water supply system threatened by a chemical pollutant will mainly depend on:

• • •

Type of pollutant source and pollution scenario, Type and importance of the considered water supply, Complexity of the system and degree of knowledge of the hydrogeological situation.

Typically, an early warning monitoring system will be designed either for a point pollution or for a non-point (diffuse) and line pollution situation. Data requirements for these two types will be different. While the design of the network of observation points for point pollution cases will depend

64



16/04/2008

11:15

Page 65

heavily on the extent of the pollution plume, the network for surface pollution situations has to take account of the variability of the system in its design. The network density and frequency of data collection will be proportional to the importance of the well field or other object for which the monitoring system is established and to the complexity of the system.

6.2 Data collection strategy A proper strategy for detecting groundwater quality problems should enable detection of problems before groundwater pollution occurs. Therefore the data collected within a properly designed early warning monitoring system established for groundwater should:

• • •

Provide a detailed coverage of those zones currently with high concentration of pollutants, Assure that there are no undetected pollution zones, Provide information about the transport and transformation of pollutants along flow path from the ground surface to the water table and within the saturated zone.

In all cases it is necessary to make a compromise between the three objectives specified above. There will always be a trade off between the necessity to have detailed point information (necessary for determination of the health risks caused by the pollutant) and integrated volume information which assures that the whole groundwater aquifer volume is being monitored. Each of the objectives specified above require a specific approach with respect to the modelling. Therefore the concepts of point monitoring, line monitoring and volume monitoring techniques have been developed (Fig. 6.1). While point can be established both in saturated and unsaturated zone the line and volume monitoring refer mostly to the saturated zone. All above monitoring techniques correspond to the category of regional monitoring programmes Point monitoring technique consists of a network of sampling points collecting small water samples from the unsaturated and the saturated zones. Point monitoring sites provide information from a very limited volume of rocks extending from ground surface to the sampling point. The sampling point for this type of monitoring can be situated within the unsaturated zone, within a perched aquifer or just below water table in an observation well. Information obtained from a network of point monitoring points is closely related to the conditions on the land surface immediately above the point monitoring site. Line monitoring technique consists of a network of sampling points situated within the saturated zone along the groundwater flow path. Line monitoring sites provide information about the changes of groundwater chemistry occurring with time and the distance travelled. Line monitoring wells sample at rates which do not significantly disturb the existing natural flow field. Volume monitoring technique consists of wells pumping at a rate sufficient to affect the natural flow field. Volume monitoring sites provide information from a large surface area and a large aquifer volume. The information obtained from the volume monitoring sites will always be averaged both in space and in time. If the purpose of our groundwater monitoring system is detection of future groundwater quality problems, we need to know the averaged trend in groundwater quality for the aquifer under consideration and also the extreme values of selected parameters occurring locally.


65


16/04/2008

11:15

Page 66

Figure 6.1

Principles for groundwater monitoring aimed at detecting groundwater problems

6.3 Data analysis Handling of the data collected within early warning monitoring system can be done using:

• •

Isolated treatment (groundwater chemistry time series for a specific well or groundwater chemistry situation at a specific time) or Comprehensive treatment utilising the existing relationship between the different components of the system.

For the isolated treatment approach no knowledge about the dynamic behaviour of the system is needed and data handling may consist, for example, of a standard XY-plot of concentration versus time or a standard contouring of concentration distribution within the aquifer. The ease of data treatment is the main advantage in this approach. The main disadvantage of this approach is associated with the inability to make predictions about the future development of the situation due

66



16/04/2008

11:15

Page 67

to a lack of knowledge of the dynamics of the groundwater system. Most commonly, the amount of information gathered for the early warning system will be insufficient to make a meaningful prediction based on extrapolation of the existing data. Within this approach a lot of existing (and essential) information will not be used in the assessment Comprehensive treatment involves use of not only all the relevant information on the parameters (concentrations, heads) but also use of the relationship governing the dynamics of the aquifer system. To put it another way both the knowledge of parameter values and the governing equations such as flow and mass balance equations are incorporated in this approach. This approach makes it possible to verify our understanding of the system during the calibration procedure and to use the calibrated model for prediction of the future situation in the aquifer under different scenarios. The comprehensive treatment requires, as a rule, that computer modelling using monitoring data is carried out. Furthermore, several efficient GIS-software packages have been developed during the last decade which offer the opportunity of combining the various types of surface-related information (geology, hydrogeology, pollutant distribution etc.) in an efficient way.

6.4 Modelling Model in this context is understood to refer to a set of consistent rules established for a specific groundwater system regarding

• • •

Composition of the system, Geometry of the system, Interactions within the system and between the systems and surroundings.

While the composition and geometry of a system are in general time independent the interaction of the system with the surroundings will normally depend on time.

Composition of the system Properties of the various rocks, occurring within the considered groundwater system, such as lithology, permeability and groundwater chemistry should be described. Furthermore, properties relevant for transport and attenuation of relevant pollutants should be addressed. Groundwater chemistry: both background and anthropogenically influenced, should be described.

Geometry of the system Spatial distribution of the different types of rocks should be defined. In cases where no head distribution within the aquifer system is defined, a flow field based on the measured head values should be constructed.

Interactions within the system and between the systems and surroundings Definition of the interactions within the system is typically made using the basic laws governing groundwater flow and contaminant transport: such as Darcy’s law, Fick’s law etc. Definition of the interactions between the system and the surroundings involves an evaluation of: infiltration,


67


16/04/2008

11:15

Page 68

exchange with surface water bodies and intensity and distribution of the pollutant source. Establishment of the proper boundary conditions for the groundwater system is the most important and frequently the most difficult task during construction of conceptual and digital modelss.

Types of models Both the design of an early warning system and the subsequent data analysis should be done using models. The basic model necessary in all cases is the so called conceptual model. In most cases in addition to the conceptual model it will be feasible to develop a so called digital model. A conceptual model consists of a description of the hydrogeology of the groundwater system and definition of the boundary conditions. Establishment of this model is the most important part of the analysis of a groundwater system. To prepare a useful conceptual model it is necessary to have an idea about: direction and magnitude of groundwater flow, exchange of water between the different aquifers and between groundwater and surface water bodies, infiltration rate prevailing in the system and type and intensity of the existing pollution sources. A digital model, which is an extension of the conceptual model, requires strong discipline from the user because all points within the modelled system have to be defined as far as properties of the system are concerned. Depending on the amount and type of information available for the establishment of the digital model, it can either be of steady state or transient type. Input to the digital model consists of the definition of system properties within the solution domain and the definition of boundaries and initial conditions of the system. The primary output from the models is in the form of head (water level) distribution for flow models and pollutant distribution for pollutant transport models. Some simple techniques exist which allow simplified calculations of pollutant distribution to be made using the flow model with an assumption requiring that the pollutant behaves as a conservative tracer. While it is absolutely necessary to establish a conceptual model for the system being considered, construction of a digital model is optional and will mainly depend on the amount of information available and complexity of the system. It is a common misunderstanding, that sparse hydrogeological information excludes the use of digital models. Digital models should always be used for systems characterised by complex geometry and in cases where complicated boundary conditions make it impossible to apply analytical solutions. A digital model can be used to prove the validity of a conceptual model by successful calibration of the model against the existing data or to analyse the system by geometry and interactions which are too complicated for treatment without the aid of the computer. It is recognised, that analytical solutions for groundwater flow can only be obtained for a very limited number of cases with unrealistic geometry and highly idealised initial and boundary conditions.

Sensitivity of the system One of the most important tasks which can be solved with models is an evaluation of the sensitivity of the system to changes of the different parameters within the limits specified by the user of the model. Sensitivity analysis will provide information on which parameters are crucial for the performance of the system and will help to design the data collection campaign and field work in the

68



16/04/2008

11:15

Page 69

most effective manner. As a typical result of such analysis a prioritisation of the different parameters and conditions should be established and a cost-effective data-collection campaign launched.

6.5 Forecasting Important function of an early warning system is the forecasting of the development of the pollution plume under different conditions. A well-calibrated digital model is an suitable tool for the calculation of different scenarios covering a wide range of situations. Various remedial actions including different pumping strategies within the existing well fields, establishment of hydraulic barriers with the help of remedial wells, re-circulation of the contaminated water etc. can easily be investigated using digital models. This ease of forecasting is the best argument for the development of digital models for aquifers observed by early warning monitoring system.


69


16/04/2008

11:15

Page 70

7

EARLY WARNING GROUNDWATER QUALITY M O N I T O R I N G S T R AT E G Y

Early warning groundwater quality monitoring programmes can be regarded as a widely accepted concept of preventive care, helping to avoid serious outbreaks of groundwater pollution. In most countries, we are at best monitoring fully developed groundwater quality problems rather than attempting to monitor earlier stages of aquifer deterioration process. This is the contradictory philosophical approach fluctuating between post mortem analysis and a diagnosis of an in status nacendi problem by means of an early warning groundwater monitoring strategy. The main objective of an early warning groundwater quality monitoring strategy is therefore, to identify and to foresee the threats which may cause adverse effects on groundwater quality while they are still in the unsaturated zone and thus to timely define measures leading to protection of the aquifers. Early warning monitoring generates data about groundwater quality in space and time and movement and the fate of pollutants and is an important element of groundwater protection and quality conservation programmes. However, institutional and technical capacities for the development and implementation of monitoring strategies have to be available or to be established. Early warning, as an integral part of a groundwater quality monitoring programme, is broad in nature, has different objectives, requires a progressive approach and supports both short-term (specific monitoring) and long-term (background monitoring) groundwater protection policy and sustainable management of groundwater resources. Establishing a groundwater early warning monitoring programme is a scaling process, that advances step by step and which has to be funded and implemented within national and river or groundwater basin water management plans. Subtle changes in groundwater chemical composition may be observed many times before groundwater level decline or other hydraulic phenomena become evident. Therefore, early warning groundwater quality monitoring, particularly on regional and site specific levels, should be implemented and targeted with respect to the specific groundwater quality problem caused by pollution impact, land use changes, excessive aquifer exploitation or well failure. Establishment of early warning groundwater monitoring programmes supports also data for prevention, preparedness and mitigation of natural disasters and management of groundwater resources in emergency situations. The frequency and magnitude of disastrous events is increasing

70



16/04/2008

11:15

Page 71

worldwide. Therefore, the operation of integrated early warning monitoring programmes, inclusive of groundwater is desirale for 1/ the timeous recognition and better understanding of risk and impact of natural disasters on the population, 2/ identification, delineation and evaluation of groundwater resources resistant to natural hazards and suitable as a source of drinking water for emergency situations and 2/ formulation effective groundwater disastrous protection and mitigation policy. An early warning groundwater quality monitoring strategy supports:

• • • • • • •

Evaluation of the chemical composition and chemical evolution of groundwater in space and time. Identification of groundwater pollution risks and assessment of the threats to which the groundwater system is exposed. Assessment of groundwater vulnerability to both human impacts and natural disasters. Conservation of groundwater quality. Solution of groundwater quality problems while there are still at a controllable and manageable stage. Groundwater preventive protection policy and sustainable management of groundwater quality. Decision making taking into account pollution risks, potential water conflicts and sustainability between social and health requirements on groundwater sources, economic development and good status and functionality of groundwater dependent ecosystems

A priori it could be envisaged that it should not be necessary to convince politicians, decision makers and professionals about the necessity of having groundwater early warning monitoring systems. Such an approach could be considered to be parallel to the well-known slogan suggesting that ‘prevention is better than cure’. However, there is an intrinsic and not always spelled out psychological problem related to the exploitation and protection of groundwater. Since groundwater cannot generally be seen (as opposed to the case for a lake or a river) people do not feel much responsibility for this hidden resource and its quality. Moreover, since groundwater systems are known to have a delayed response, particularly with respect to quality changes, decision and policy makers know that reactions to impacts on the system will be felt years after they have resigned from their political and administrative responsibilities (this is a ‘benefit’ resulting from the very large residence time of groundwater as compared to the political residence time of bureaucratic officers). A possible reaction when phased with the ‘to-know-now’ possibility, offered by early warning monitoring systems, could be the ‘what do I need this for now’ approach. In other words, early warning groundwater monitoring systems need to overcome the inertia of systems that, in relation to groundwater, are accustomed to postponing action and investment of human and financial resources for its protection All the above is related also to the necessity of thinking in a non-conventional way. A manager accustomed to long term planning may regard the knowledge that a groundwater supply source will be polluted within a decade as information of great economical and social value. For another professional worried by daily problems, this same information can be considered as an anachronistic and undesired noise.


71


16/04/2008

11:15

Page 72

The establishment of early warning groundwater monitoring systems requires, in many cases, extra budgetary allowances. For example, for professionals to learn and get used to new field methods and new analytical approaches, for acquiring field equipment (e.g., multi layer samplers, porous caps), for the installation of new specifically designed monitoring wells and for obtaining periodic field profiles. These costs are generally not large when compared to the cost of developing an alternative water source due to groundwater supply pollution, or the cost needed for groundwater quality restoration. However, the dilemma of a contemporary investment as compared to a possible future one is always present. A administrative, legislative, and political framework must also be developed to efficiently utilize the information gained from groundwater quality early warning monitoring. In general, the administration and legislation will need to analyse and react to future scenarios on the basis of current information supplied by the groundwater monitoring network for what may in some cases be remote parts of the system (e.g., results obtained in the water table region suggesting that pumping wells downstream will be polluted within a five year period). Early warning groundwater monitoring systems generate data that enable assessment of groundwater quality of transboundary aquifers too. Special regard must be given to the transboundary aquifers whose recharge area is in one country and the discharge area is in neighbouring country or countries. Transboundary aquifers boundary, aquifers geometry, groundwater flow conditions from recharge to discharge areas and groundwater quality have to be therefore studied and known before transboundary groundwater monitoring network is established. The UN Convention on the Protection and Use of Transboundary Watercourses and International Lakes (Helsinki, 1992) endorsed the harmonization of rules and methods for establishment and operation of transboundary monitoring networks, correlation of monitoring programmes, standardization of monitoring procedures and analytical laboratory techniques and methods of data processing, evaluation and transfer among neighbouring countries. Monitoring of groundwater quality and establishment, operation of early warning groundwater monitoring systems around potential and existing pollution sources located in transboundary aquifers and data management, assessment and reporting will provide valuable information for a joint assessment and sustainable management of shared aquifers, coordination of groundwater protection policy and prevention of water related transboundary conflicts. However, internationally coordinated effort to collect, process and evaluate data in a standardized manner and thus support early detection of transboundary groundwater quality problems, is generally missing till this time. Establishing a conceptual model of the groundwater system is an important initial stage in the development of early warning monitoring network and programme. Monitoring of the unsaturated zone with respect to its ability to store, retain, remove and attenuate pollutants and delay their migration to the saturated aquifer, is a crucial tool for groundwater protection policy and strategy and groundwater quality management. Data collection by early warning groundwater quality monitoring is a technically demanding, time consuming and costly process. However, with growing industrialization, urbanization and intensification of agriculture, threats to groundwater quality are increasing and implementation of an early warning monitoring strategy is justified from social, environmental and economic points of view. Governmental institutions, water supply organizations and other water stakeholders in many countries of the world may not yet be prepared to accept the need for an early warning groundwater quality monitoring strategy. However, the reality in the field and the costs of restoration of polluted

72



16/04/2008

11:15

Page 73

aquifers suggest that early warning monitoring may be an important cost-benefit approach for preserving the quality of groundwater as a strategic source of drinking water and a valuable component of environment.


73


16/04/2008

11:15

Page 74

8

REFERENCES

Adams, B. and Foster, S.S.D. 1992. Land Surface Zoning for Groundwater Protection. J. Inst. Water Environmental Management, 6, pp. 213-312. Affek, H.P., Ronen, D. and Yakir, D. 1998. About Production of CO2 in the Capillary Fringe of a Deep Phreatic Aquifer. Water Resources Research, 34, pp. 989-996. American College Dictionary. 1964. Random House, New York. Aust, H. and Kreysing, K. 1985. Hydrogeological Principles for the Deep - Well Disposal of Liquid Wastes and Waste Waters. Federal Institute for Geosciences and Natural Resources, Hannover, Germany, 102 pp. Beneš, V, Pěkný, V., Skořepa, J., Vrba, J. 1989. Impact of Diffuse Pollution Sources on Groundwater Quality – Some Examples from Czechoslovakia. Environmental Health Perspectives, Vol. 83, pp. 5–24. Balke, K.D. 1973. Chemische und Thermische Kontamination des Grundwassers durch Industrieabwasser. Z. Ditsch, Geol. Ges., pp. 447-460. Black, C.A. et al. 1965. Methods of Soil Analysis, Part I. American Society of Agronomy, Inc., Madison, Wisconsin, USA, 770 pp. Bouwer, H. 1987. Effect of Irrigated Agriculture on Groundwater. J. Irrig. Drain. Eng., 113, pp. 4-15. Candela, L. and Aureli, A. 1998. Agricultural Threat to Groundwater Quality. Workshop Proceedings. UNESCO, CIHEAM, UPC, Zaragoza. Chilton, P. J. 1995. Salinization of Soils and Aquifers. BGS Technical Report WD/95/26, pp. 54-64. Christy, T.M. and Spradlin, I. (1998).The Use of Small Diameter Probing Equipment for Contaminated Site Investigation. Geoprobe® Systems, Salina, Kansas. Csanády, M. 1968. Ausbereitung der Schwermetall und Cyanverun Reinigung im Grundwasser. Hydrol. Jour., 48, Budapest, pp. 32-38. Duijvenbooden, W. 1987. Groundwater Quality Monitoring Networks: Design and Results. In: Proceedings and Information No. 38, Vulnerability of Soil and Groundwater to Pollutants, RIVM, The Hague, pp. 179-191. Elek. T. 1980. Hydrocarbon Contamination of an Alluvial Aquifer Beneath and Oil Refinery, Czechoslovakia. In: Aquifer Contamination and Protection, UNESCO-IHP, pp. 318-328. EPA. 1986. Permit Guidance Manual on Unsaturated Zone Monitoring for Hazardous Waste Land Treatment Units. Environmental Monitoring Systems Laboratory, Las Vegas EPA/530-SW-86040, 1986, 112 pp. EPA. 1986. Pesticides in Groundwater: Background Document, Washington, 72 pp. Ernst W., 1974. Schwermetallvegetation der Erde. Stuttgart. Estes J.E. and Senger L.W. 1972. The Multispectral Concept as Applied to Marine Oil Spils. Remote Sensing of Environment, 2, pp. 141-163. Everett, L.G. 1980. Groundwater Monitoring. General Electric Company. Schenectady, New York, 440 pp.

74



16/04/2008

11:15

Page 75

Foster, S.S.D. 1976. The Vulnerability of British Groundwater Resources to Pollution by Agricultural Leachates. MAFF Tech. Bull. B2, pp. 68-91. Foster, S.S.D. and Chilton P.J. 1998. As the Land so the Water the Effects of Agricultural Cultivation on Groundwater. In: Agricultural Threats to Groundwater Quality, Workshop Proceedings, UNESCO, CIHEAM, UPC, pp. 15-43. Foster, S.S.D. and Young, C.P. 1980. Groundwater Contamination Due Agricultural Land-Use Practices in the United Kingdom. Unesco-IHP Studies and Reports Hydrol. Series 30, pp. 268-282. Foster, S.S.D.,Ventura, H. and Hirata, R. 1987. Groundwater Pollution: An Executive Overview of the Latin Americal – Caribbean Situation in Relation to Potable Water Supply. Am. Centre for San. Eng. Env. Sciences. Lima, 38 pp. Foster, S.S.D., Cripps, A.C. and Smith-Carington, A.K. 1982. Nitrate Leaching to Groundwater. Phil. Trans. Royal Soc. London B 296, pp. 477-489. Freeze, A.R. and Cherry, J.A. 1979. Groundwater. Prentice-Hall, Inc., 604 pp. Geak, A.K. and Foster, S.S.D. 1989. Sequential Isotope and Solute Profiling of Unsaturated Zone of the British Chalk. Hydrol. Sciences J., 34, pp. 79-95. Geoprobe® Systems Catalog 1998-1999. Ghassemi, J.R., Jakeman, A.J. and Nix, H.A. 1995. Salinisation of Land and Water Resources. CAB International, Wellingford, UK. Gosk, E., Bishop, P.K., Burston, M.W. and Lerner, D.N. 1992. Field Investigation of Chlorinated Solvent Pollution of Groundwater in Coventry, UK. In: Weyer (ed.), Subsurface Contamination by Immicible Fluids. Balkena, Rotterdam, pp. 44-449. Hallberg, G.R. 1989. Nitrate and Groundwater in the United States. In: Foll. RF (ed). Nitrogen, Management and Groundwater Protection. Elsevier, Rotterdam, 35 pp. Hansen, E.A. and Harris, A.R. 1974. A Groundwater Profile Sampler. Water Resources Research 10, 375 pp. Harrison, W.D. and Osterkamp, T.E. 1981. A Probe Method for Soil Water Sampling and Subsurface Measurements. Water Resources Research 17, pp. 1731-1736. Hesslein, R.H. 1976. An in situ Sampler for Close Interval Pore Water Studies. Limnol. Oceanogr. 21, pp. 912-914. Holmberg, M. 1987. Assessing Aquifer Sensitivity to Acid Deposition. In: Proceeding and Information, 38, Vulnerability of Soil and Groundwater Pollutants. TNO, RIVM, The Hague, pp. 373-380. Iliceto V. and Mares S. 2000. Recommendations for Application of Geophysical Methods in the Phase of Groundwater Prevention and Contamination, Carolinum, Praha, (in Italian), 86 pp. Jackson, R.E. et al. 1980. Aquifer Contamination and Protection. UNESCO-IHP Programme, 440 pp. Kaplan, E., Banerjee, S., Ronen, D., Magaritz, M., Machlin, A., Sosnow, M. and Koglin, E. (1991). Multi-level Sampling in the Water Table Region of a Sandy Aquifer. Ground Water 29, pp. 191198. Knoll, K.H. 1969. Hygienissche Bedeutung Natürlicher Selbstreinigungsvorgange für die Grundwasser Beschaffenheit in Bereich von Abfalldeponien. Müll Abfall, 1, 2, pp. 35-41. Krajenbrink, G.J.W., Ronen, D., Duijvenbooden van,W., Magaritz, M. and Wewer, D. 1998. Monitoring of Recharge Water Quality under Woodland. Journal of Hydrology, 98, pp. 83 – 102. Ku, H. 1980. Groundwater Contamination by Metal - Plating Wastes, Long Island, New York, U.S.A. In: Aquifer Contamination and Protection, UNESCO-IHP, pp. 310-317. La Moreaux, P. and Vrba, J. 1990. Hydrogeology and Management of Hazardous Waste by Deep-Well Disposal. IAH, Vol. 12, Verlag Heinz Heise, Germany, 136 pp. Lawrence, A.R. and Foster, S.S.D. 1986. Denitrification in a limestone aquifer in relation to the security of low nitrate groundwater supplies. J. Inst. Water Eng. Sci., pp. 40, 159-172.


75


16/04/2008

11:15

Page 76

Lindhardt, B. (ed.). 2001. The Danish Pesticide Leaching Assessment Programme, Site Characterization and Monitoring Design, Geological Survey of Denmark and Greenland. Mares, S., Kelly, W.E. and Mazač, O. 1997. Applied Geophysics in Environmental Engineering and Science, Carolinum, Praha, 98 pp. Matthess, G. 1982. The Properties of Groundwater. John Wiley, New York, 406 pp. Maybeck, M. 1985. The GEMS Water Programme 1978-1983. Water Quality Bull., Volume 10, No. 4, Canada, pp. 167-177. Mayer, L.M. 1976. Chemical Water Sampling in Lakes and Sediments with Dialysis Bags. Limnol. Oceanogr. 21, pp. 909-912. Miler, M.R., Brown, P.L., Donovan, J.J., Bergatino, R.N., Sonderegger, J.L., Schmid, F.A. 1981. Saline Deep Development and Control in the North American Great Plains – Hydrogeological Plans. Agricultural Water Management 4, 115 – 141 pp. Molz, F.J., Gouven, O. Melville, J.G. and Keely, J.F. 1986a. Performance and Analysis of Aquifer Tracer Tests with Implications for Contaminant Transport Modelling. EPA/600/2-86/062, 88 pp. Obermann, P. 1982. Moeglichkeiten der Anwendung des Doppelpackers in Beobachtungsbrunnen bei der Grundwassererkundung. Bohrtechn.-Brunnenbau-Rohrleitungsbau, 27, pp. 93-96. Pěkný, V., Skořepa, J. and Vrba, J. 1989. Impact on Nitrogen Fertilizers on Groundwater Quality – Some Examples from Czechoslovakia. Journal of Contaminant Hydrology. Vol. 4. Elsevier, pp. 51-67. Pellegrini, M. and Zavatti, A. 1980. Lead Pollution in the Groundwaters of the Modena Alluvial Plain, Po Valley, Italy. In: Aquifer Contamination and Protection, UNESCO-IHP, pp. 305-309. Pickens, J.F., Cherry, J.A., Grisak, G.E., Merritt, W.F. and Risto, B.A. 1978. A Multilevel Device for Groundwater Sampling and Piezometric Monitoring. Ground Water 16, pp. 322-327. Pfannkuch, H.O. 1990. Elsevier’s Dictionary of Enviromental Hydrogeology. Elsevier’s Science Publishers, Amsterdam, 332 pp. Pyšek, A. 1983. Role of Geobotany in Interpretation of Aerial Photography. CTVS Yearbook, Stavební geologie Praha, (in Czech), pp. 167-173. Pyšek, A., Švoma, J., Pyšek, P. and Murický, E. (in press). Photographic Detection of Rock Pollution. (In Czech). Pyšek, P. and Pyšek, A. 1988. Veranderungen der Vegetation durch Experimantelle Erdgasbehandlung. Weed Res., Oxford, 29, pp. 193-204. RIVM. 1992. The Environment in Europe: a Global Perspective. Report No. 48150, Bilthoven, 119 pp. Ronen, D., Magaritz, M., Paldor, N. and Bachmat, Y. 1986. The Behavior of Ground-water in the Vicinity of the Water Table Evidenced by Specific Discharge Profiles. Water Resources Research 22, pp. 1217-1224. Ronen, D., Magaritz, M. and Levy, I. 1987a. An in situ Multilevel Sampler for Preventive Monitoring and Study of Hydrochemical Profiles in Aquifers. Ground Water Monitoring Review 7, pp. 6974. Ronen, D., Magaritz, M. and Almon, E. 1988a. Contaminated Aquifers are a Forgotten Component of the Global N2O Budget. Nature 335, pp. 57-59. Ronen, D., Scher, H. and Blunt, M. 1997. On the Structure and Flow Processes in the Capillary Fringe of Phreatic Aquifers. Transport in Porous Media 28, pp. 159-180. Ronen, D., Scher, H. and Blunt, M. 1998. Transport Phenomena in the Capillary Fringe of Granular Sediments. Final Report to the Feldman Foundation, First Research Year. Department of Environmental Sciences and Energy Research, Weizmann Institute of Science, Israel, 21 pp. Ronen, D., Scher, H. and Blunt, M. 2000. Field Observations of a Capillary Fringe Before and After a Rainy Season. Journal of Contaminant Hydrology, 44, pp. 103-118. Schwille, F. 1969. Das Verhalten von Mineralöl im Untergrund Dargestellt and Hand von Modellver-

76



16/04/2008

11:15

Page 77

suchen und Schadensfüllen. In: Ochrana vod pred znecistenim ropou a ropnymi produkty. SVS a VUVH, Bratislava. Schwille, F. and Vorreyer, C. 1969. Durch Mineralöl ‘Reduzierte’ Grundwasser. Gas - Wasserfach, 110, 44, Munich, pp. 1225-1232. Spalding, R.F. and Exner, M.E. 1991. Nitrate Contamination in the Contiguous United States. In: Nitrate Contamination, NATO ASI Series G, Vol. 30, pp. 13-48. Stanners, D. and Bourdeau, P. 1995. Europe’s Environment. European Environment Agency, Copenhagen, 676 pp. Švoma J., 1990. Remote Sensing Methods for the Detection of Rock Medium and Groundwater Contamination. Proceedings International Symposium on Remote Sensing and Water Resources, Enschede, the Netherlands, 1, pp. 465-471. US National Research Council. 1993. Groundwater Vulnerability Assessment. National Academy Press, 204 pp. US Geological Survey. 1989. Subsurface-Water Flow and Solute Transport. Federal Glossary of Selected Terms. U.S.G.S., Reston, 38 pp. UK DoE. 1992. Digest of Environmental Protection and Water Statistics. UK Department of the Environment. HMSO, No. 15, London. United Nations. 1992. Convention on the Protection and Use of Transboundary Watercourses and International Lakes. Helsinki. UNESCO International Hydrological Programme. 1992. Hydrogeological Considerations in Relation to Nuclear Power Plants. Proceedings of the International Workshop, 409 pp. UNESCO International Hydrological Programme. 1998. Water: a Looming Crisis? Proceedings of the International Conference on World Water Resources at the Beginning of the 21st Century. Technical Documents in Hydrology, No. 18, 536 pp. UNESCO/WMO. 1992. International Glossary of Hydrology. UNESCO/WMO, Paris, 561 pp. Van Dam, J. 1967. The Migration of Hydrocarbons in a Water - Bearing Stratum. In: The Joint Problem of the Oil and Water Industries. Institute of Petroleum, London, 55 pp. Vrba, J. and Zaporozec, A. (ed.) 1994. Guidebook on Mapping Groundwater Vulnerability. International Association of Hydrogeologists. International Contribution to Hydrogeology, Vol. 16, 131 pp. Vrba, J. 1985. Impact of Domestic and Industrial Wastes and Agricultural Activities on Groundwater Quality. In: Memories of the 18th Congress. Hydrogeology in the Service of Man. Vol. XVIII, Part I, pp. 91-117. Vrba, J. 2000. Groundwater Protection Policy and Management. In: Proceedings of the Conference Evaluation and Protection of Groundwater Resources, Wageningen, pp. 15-25. Vrba, J. and Pekný, V. 1991. Groundwater-Quality Monitoring – Effective Method of Hydrogeological System Pollution Prevention. Environ. Geol. Water. Sci.,Vol. 17, No. 1, pp. 9-16. Weisbrod, N., Ronen, D. and Nativ, R. (1996). A New Method for Sampling Groundwater Colloids under Natural Gradient Flow Conditions . Environmental Science and Technology 30, pp. 3094-3101. Whitmore, A.P., Bradbury, J.N. and Johnson, P.A. 1982. Potential Contribution of Ploughed Grassland to Nitrate Leaching. Agric. Ecosystems and Environ., 39, pp. 221-233. Williams, D.E. and Wilder, D.G. 1971. Gasoline Pollution of a Groundwater Reservoir, A Case History. Groundwater, 9, pp. 50-56.


77


16/04/2008

11:15

Page 78

9

CASE STUDIES

9.1 Monitoring of groundwater quality problems caused by agriculture in Denmark 9.1.1 Introduction Serious problems with groundwater quality are occurring all over the world. These problems will accelerate with increasing demand for potable water and for irrigation water. Unlike polluted streams and rivers, where the proper efforts and removal of the source of pollution can restore good water quality within a few years, restoration of groundwater qualityis a long term process. Therefore avoiding groundwater pollution is much wiser than aquifer remediation. In order to avoid groundwater quality problems it is necessary to have a strategy that facilitates detection of possible future pollution problems long before these problems are encountered in water supply wells. In most cases when the water supply wells show signs of pollution the affected volume of aquifer is so large that no strategy can help to save the aquifer. This logic appeals to the Danish politicians and protection of groundwater resources is one issue that almost all of them can agree upon. This case study refers to selected aspects of the Danish Environments Monitoring Program in general and to the monitoring of agricultural pollution in particular. In this context no attempt has been made to present the whole Danish program, rather just those aspects relevant to groundwater monitoring strategies are emphasized Denmark is situated in the Temperate Zone. The annual precipitation varies from less than 650 mm in the central and eastern part of the country to over 900 mm in western part. The annual net precipitation amounts to 280 mm. The highest point in Denmark is less than 180 metres above sea level. In Denmark about 65% of the total area is used for intensive agricultural production.

78



16/04/2008

11:15

Page 79

In Denmark approximately 99% of the water supply comes from groundwater. Until 20 to 30 years ago the groundwater was often withdrawn from quite shallow, phreatic aquifers. However, due to the increasing pollution of these vulnerable aquifers by nitrates, pesticides and other organic micropollutants, an increasing number of supply networks are now getting their groundwater from deeper aquifers. Practically all of Denmark is covered by Quaternary, glacial-influenced deposits such as tills and melt-water sand and gravel. The central and eastern parts of Denmark are dominated by lime-rich clayey till, whereas glacial sand and gravel dominate the northern part. The landscape is undulating. In the western part of Denmark ‘hill islands’ and alluvial ‘heath plains’ of glacial sand and gravel dominate the upper sediments and landscape. In the central and western parts of Denmark shallow aquifers consist mainly of dilluvial sand, whereas in the eastern part shallow chalk aquifers are found. Growing problems with nitrate pollution of groundwater and surface water during the 1980s prompted the Danish government to initiate a nation wide Environment Monitoring Programme in 1988. One of the elements of this programme was the implementation of an integrated, comprehensive monitoring system in 6 agricultural watersheds called LOOPs. For each LOOP nutrient input, nutrient transport through the different parts of the environment and nutrient removal from the area as crops and leakage were either measured or calculated. The monitoring system was later extended to include pesticides. One of the main objectives of the LOOPs was the development of a monitoring system allowing the determination and prediction of the magnitude and extent of groundwater pollution caused by agricultural production. The monitoring of shallow and vulnerable groundwater aquifers in Denmark, carried out within the Agricultural Watershed Monitoring Programme, has has previously been described in the papers of Gosk (1988) and Rasmussen (1986, 1998). The following description is based upon those papers.

9.1.2 Monitoring concept Monitoring of shallow groundwater is being carried out in 6 agricultural watersheds varying in size from 5 to 15 km2 (Fig. 9.1). These watersheds are called LOOP1, LOOP2, ... LOOP6. These watersheds represent the range of agricultural land use and drainage, the most important soil types, shallow groundwater, recharge areas, top of stream systems and climatic variation found in Denmark (Rasmussen, 1996). The aims specified for the LOOPs were:

• • •

To evaluate the effects of measures taken by the Danish government to reduce the input of nutrients and chemicals to the environment. To obtain fast and field related measurements of changes in agricultural practice (early warning). To measure nutrient flow and pesticide leaching under ‘farming as usual’ conditions.

To fulfil the objectives specified above, water, nutrient and pesticide fluxes were measured in different parts of the hydrological cycle. A network of stations for monitoring climate, groundwater


79


16/04/2008

11:15

Page 80

levels, stream flow and nutrient content together with other water quality parameters in soil water, drainage water, groundwater and streams was established during 1988 and 1989. Sampling frequencies for the different parameters varied from once per week to once every second month, depending on the type of monitoring station, water quality parameter and time of year.

Figure 9.1

Location of the six experimental agricultural watersheds LOOP1 - LOOP6

In every LOOP annual surveys of agricultural practice were conducted for all the farms in order to determine the magnitude of nutrient input and output associated with farming. The surveys were carried out on a field scale and covered all the fields within the watershed. Detailed soil survey and description of the geological and hydrogeological conditions in the watersheds were included in the monitoring programme (Grant and Andersen 1996). Each county, responsible for an individual LOOP, prepared yearly reports summarising the results of the sampling programme and sent all the data to the National Environmental Research Institute (NERI) and the Geological Survey of Denmark and Greenland (GEUS). NERI and GEUS made an evaluation of the results on a national scale. Administratively the Danish Environmental Protection Agency made an overall evaluation of the results from the whole monitoring programme. A unique data transport system, STANDAT, was developed for the transfer of monitoring data from the counties to NERI and GEUS.

80



16/04/2008

11:15

Page 81

9.1.3 Instrumentation A variety of monitoring stations is present in every LOOP. Two kinds of monitoring stations were specially developed for the LOOPs: groundwater nests and soil water stations, while other stations like drainage and stream flow stations were of standard type. Both the groundwater nests and the soil water stations were constructed in such a way, that no change of agricultural practice was necessary in spite of large number of instruments present at varying depths in the field. The samples, originating at spots located far away from the edges of the fields, i.e. where the conditions are representative for the agricultural practice, were transported to the edge of the field using either a vacuum or compressed nitrogen (or air) and a special tubing system (Fig. 9.2). This system permits normal cultivation of fields while allowing the collection of samples from the unsaturated zone and from the top of the saturated zone.

Figure 9.2

Schematic layout of groundwater nests and soil water station

A typical LOOP contains:

• • • • •

6 to 8 soil water stations, each station containing 10 samplers, 21 to 25 groundwater nests with a total of 50-72 screens, situated in 15-17 fields, Piezometers for groundwater level measurement drilled down to 5 to 7 meters below surface next to all fields with groundwater nests, Several drain stations and stream gauges and 1 to 2 rain gauges.

Soil water stations, collecting water from the unsaturated zone immediately below the root zone, can be regarded as a first stage of a system monitoring transport and transformation of contaminants. Water samples from the soil water stations will give the first warning about possible problems with groundwater quality in the saturated zone. Due to a large spatial variation in the transport velocity within the unsaturated zone, each soil water station has to contain several sampling points in order to give meaningful information.


81


16/04/2008

11:15

Page 82

In the Danish LOOPs each soil water station consists of ten sampling points (samplers) arranged as a letter ‘V’, see Fig. 9.2. The samplers, made of Teflon or ceramics, are placed about one metre below the soil surface in sandy areas and 1.2 metres in clayey areas. The cells were installed from a 0.4 m wide trench and the cells are inserted at an angle of 45° (Fig. 9.3). This procedure allows collection of samples of water infiltrating through practically undisturbed soil. Each cell was connected to an individual bottle using two 1/8-inch polyethylene tubing. The bottles were stored in a thermo-box placed below the ground surface at the edge of the investigated field.

Figure 9.3

Soil water sampler, construction details

At the beginning of each one-week sampling period and after each sampling event the bottles were subjected to a 0.7 bar vacuum. A tube system, extending from the sampling stations to the edge of the field, allows the sampling procedure to be carried out without interference with the usual agricultural routines. Sampling at the soil water stations is normally possible only during winter, when there is a surplus of infiltrating water. Usually, samples of groundwater from groundwater nests (Fig. 9.4), if compared to samples of soil water from the unsaturated zone, are characterised by significantly larger concentrations of the different chemical compounds present in the infiltrating water when averaged over time and space. Normally such an averaging is advantageous but, when strategies for detecting groundwater quality problems are concerned, the extreme values of different pollutants in soil water and groundwater are important. When we are looking for the future threats to our groundwater we do not want to unnecessarily dilute a potential pollutant before sampling. Therefore a special system of groundwater nests has been developed for the LOOPs where it is possible to collect the youngest groundwater from the very top of the unconfined aquifer or from the perched, small aquifers separated from the main unconfined aquifer by varying thickness of unsaturated zone. In a similar way to the soil water stations, the groundwater nests, see Fig. 9.4, permit normal cultivation of fields, while allowing for collection of groundwater samples from the top of the saturated zone. Monitoring of nitrate and pesticides in shallow groundwater 2 to 5 meters below the surface for clayey areas and 1 to 3 metres below the water table for sandy areas provides an early warning for deeper regional aquifers. The nests are located 20-25 meters from the edge of the field

82



16/04/2008

11:15

Page 83

(Fig. 9.2). Tubes extend from the top of the wells, one metre below the ground surface, to the edge of the fields, where the samples are collected by the montejus pump system (Andersen, 1990). Compressed air or nitrogen is used as a driving agent.

Figure 9.4

Groundwater nests, construction details

In the clayey till watershed it is often difficult and time consuming to determine water level in the upper layers. Therefore at these locations the wells are placed at fixed depths: 1½, 3 and 5 metres below surface. Due to the more predictable groundwater level in the sandy watersheds it is easier to relate the filter location to the measured water level. In the LOOPs situated in sandy areas the screens are placed at 1 and 2 metres below the measured groundwater table. The screens are 30 cm long and are made of PVC.

9.1.4 Groundwater sampling and chemical analysis As the frequency of sampling in groundwater nests is typically once per month, it is necessary to remove the old water prior to sampling. In clayey watersheds, where the hydraulic conductivity is low and water movement is slow, it is necessary to pre-empty the wells 1 to 3 days before sampling. In the sandy watersheds where the hydraulic conductivity is higher and re-filling of the nests is fast, the wells are pre-emptied at least three times immediately before sampling. The amount of groundwater obtained from the nests is typically 0.5 to 4 litres. During sampling, measurements of pH and conductivity are made. Samples are analysed by certified laboratories adhering to Danish Standards.


83


16/04/2008

11:15

Page 84

All farmers within the entire watershed participated in an annual interview carried out in order to determine the input and output of nutrients at the field level. The survey includes among other things crop type, time of sowing and harvest, yield, amount and time of application of fertiliser and manure, as well as type of manure. From 1993 the use of pesticides in selected fields was included in the survey. Since 1998 all fields within the LOOPs are included in the survey. Nitrate monitoring was carried out from the very beginning of the LOOP project (1989-90). The groundwater wells were sampled between 4 and 10 times per year and analysed, for 13 additional parameters beside nitrate: Ammonium, Total Nitrogen, Ortho-phosphorus, pH, Conductivity, Potassium, Alkalinity (or Acidity), Sodium, Chloride, Sulphate, Calcium, Magnesium and Iron. Since 1993 pesticide monitoring has been carried out in the six LOOP-areas. Groundwater samples have been taken from 1.5 to 5 metres below surface in clayey LOOPs and 0.5 to 4 metres below groundwater table in the sandy areas. Samples were analysed for up to 44 different pesticides and metabolites 1 to 6 times a year. Trend and seasonality analysis for nitrate content in groundwater samples were carried out for quarterly averages of nitrate concentrations. The Kendall Tau Test was used for trend analysis and the Kruskal-Wallis Test was used for seasonality analysis of the data. In order to get a better picture of seasonality, the linear trend was removed using the Ordinary Least Squares method. Similarly, to improve the estimation of trends it was necessary to remove the seasonal variations from the data set by subtracting the mean of all observations in a given quarter from each observation in that quarter (Phillips et al., 1988).

9.1.5 Nitrate monitoring The nitrate content in the shallow groundwater is generally high and clearly influenced by agricultural production. During the very dry spring/summer of 1992 the harvest was poor and a large surplus of nitrogen was seen as increasing nitrate content in the shallow groundwater wells. A dry autumn/winter 1995-96 with low groundwater recharge may explain the decreasing nitrate content in 1995-96 (Fig. 9.5). Changes in groundwater nitrate content due to changes in agricultural practice should first be detected in the shallow groundwater. A trend analysis has been carried out on 111 nitrate time series from wells located between 1.5 metres and 7 metres below surface in the 6 LOOPs (Phillips et al., 1988). Only time series containing not less than 20 quarters within the first six years of monitoring have been evaluated. Out of the 111 time series analysed, 25 series showed a significant decrease of nitrate concentration, 11 time series a significant increase of nitrate concentration and 75 time series showed no significant trend in the nitrate concentration. ‘No trend’ was defined when the annual change of nitrate concentration was less then ±1.0 mg NO3 / l. The trend test was made at a 10% level of significance. Fig. 9.6 shows how large changes in the amount of nitrogen applied (graph A) influence the nitrate concentrations in the shallow groundwater (graph B). The example shows that the response in two down-gradient wells is fast and well correlated with the changing input.

84



16/04/2008

Figure 9.5

Figure 9.6

11:15

Page 85

Median annual nitrate concentrations in shallow groundwater for 3 sandy and 3 clayey till watersheds in the period 1990 –1996

Nitrogen load and nitrate in shallow groundwater in the sandy watershed Barslund Bæk; Graph A: annual net amount of nitrogen applied; Graph B: nitrate content in two down-gradient wells sampled 2 and 3 metres below surface


85


16/04/2008

11:15

Page 86

9.1.6 Pesticide monitoring Investigation shows that out of 15 cases with pesticide (and metabolites) pollution of shallow groundwater in 1996, it was possible in 10 cases to determine that the discovered pesticides were used up-gradient of the wells in the period 1993 to 1995. Due to the fact that from 1998 all fields are included in the annual interview of farmers, the possibility of relating pesticide use and pesticide in groundwater has been improved. In one case the degradation product of atrazine, atrazine-desethyl, was found at a concentration of 0,12 μg/l, exceeding the drinking water limit of 0,1 μg/l (Fig. 9.7). The result of repeated analysis and findings of atrazine and the 2 metabolites, atrazine-desethyl and atrazine-desisopropyl, in groundwater samples taken 5 metres below the surface in the clayey Lillebæk watershed are shown in Fig. 9.7. 11 analysis for atrazine and 2 for atrazine-desethyl and atrazine-desisopropyl were taken in the period 1990-95. The last known use of atrazine on the surrounding fields was from 1990 to the Spring of 1993. Fig. 9.7 illustrates that: (1) there is a time lag between the use of the pesticide in the field to the occurrence of pesticides and metabolites in groundwater, (2) the concentration of pesticides at a given well may oscillate between values exceeding drinking water standards (0,12 μg/l) and the detection limit of 0,01 μg/l and (3) the concentration of a metabolite might be higher than the parent product.

Figure 9.7

86

Findings of atrazine and two metabolites 5 metres below surface in the Lillebæk watershed; the five measurements plotted on the horizontal dotted line (detection limit) correspond to clean samples (no pesticide); the last known use of atrazine on the surrounding fields was in the period 1990 to 1993



16/04/2008

11:15

Page 87

9.1.7 Conclusions and recommendations Development of strategies for detecting groundwater quality problems for important aquifers is necessary in many industrialised and developing countries. Often a compromise between the interests of the agricultural and water supply sectors will be needed. Such a compromise can be achieved if a reliable methodology and adequate monitoring equipment for tracking of the pollutant’s path to the aquifer is available. The groundwater nests and soil water stations developed for nitrate and pesticide monitoring within the Danish Environment Monitoring Programme perform well under different soil conditions and can be recommended as a part of a system designed for early detection of the future groundwater quality problems. Monitoring nitrate and pesticides in shallow groundwater nests provides a fast field related early warning system for deeper regional aquifers. The system of groundwater nests permits normal cultivation of fields while allowing the collection of samples from the top of the saturated zone. An annual survey of farming practice is carried out as part of the monitoring programme. The nitrate concentration in the shallow groundwater is strongly affected by climatic variations during the year and from year to year. Based on Danish experience and conditions, 6 to 10 groundwater analyses a year are needed to describe the annual variation in nitrate concentration. The monitoring has shown that large, short-term changes in nitrogen load cause significant changes in nitrate concentration within one to three years in shallow groundwater sampled down gradient. But although a (small) overall improvement in the utilisation of fertiliser and manure was observed in the 6 watersheds, only 25 out of 111 shallow wells show a significant decrease in nitrate concentration during the 6 years of monitoring. Based on monitoring groundwater quality over several years it was concluded that short-term cyclic change in nitrate content is influenced by the climate, whereas long-term trends reflect man-made impacts (Vrba and Pekný, 1991). The same system of groundwater nests has been used for analysing pesticide leaching. Water samples were analysed for between 8 to 44 different pesticides and metabolites. It was found that, for regular use, both pesticides and metabolites cause concentrations exceeding the official sustainable leaching concentration of 0.1 μg/l. Of 15 findings in 1996 the survey showed the use of 10 of the same pesticides up-gradient the wells. These were detected between 6 months and 5 years after the application of the pesticides.

9.1.8 References Andersen, L.J. 1990. Botesam, Separation Pumping and Capillary Barrier. A Remedial-Action Concept Applicable to Point Pollution. In: Proceedings, First USA/USSR Joint Conference on Environmental Hydrology and Hydrology. Minneapolis: American Institute of Hydro-logy, pp. 271-279. Gosk, E. 1988. Monitoring Programme for the Agricultural Watersheds. In: Vækst 6/88 (in Danish). Grant, R., Jensen, P.G., Andersen, H.E., Laubel, A.R., Deibjerg, C., Rasmussen, H. and Rasmussen, P.1996. ‘Agricultural Watershed Monitoring. Nation-Wide Monitoring Programme’. Danish Environmental Research Institute, Technical Report No. 175 (in Danish).


87


16/04/2008

11:15

Page 88

Phillips, R.D., Hotto, H.P. and Loftis, J.C. 1988. ‘WQStat II. A Water Quality Statistics Program’. User’s Guide. Colorado State University. 42 pp. Rasmussen, P. 1996. Monitoring Shallow Groundwater Quality in Agricultural Watersheds in Denmark. In: Environmental Geology, Vol. 27, No. 4, pp. 309-319. Rasmussen, P. 1998. Early Warning by Monitoring Shallow Groundwater. In: IAH International Groundwater Conference. Proceedings. Groundwater: Sustainable Solutions. University of Melbourne, Australia 8-13 February, 1998. pp 551-556. Vrba, J. and Pekný, V. 1991. Groundwater-Quality Monitoring - Effective Method of Hydrogeological System Pollution Prevention. In: Environ. Geol. Water Sci., Vol. 17 (1), pp. 9-16.

88



16/04/2008

11:15

Page 89

9.2 Application of a Multi Layer Sampler (MLS) for managerial decision making regarding utilization of effluents for agricultural irrigation in the coastal plain of Israel 9.2.1 Introduction Management of groundwater resources quality requites an understanding of pollutant behaviour in subsurface environments. At the present level of knowledge the quantitative relationship between the amount of pollutant released at the soil surface, e.g. by agricultural and industrial activity, and its concentration in groundwater is highly uncertain. This is primary due to both the lack of knowledge and the scarcity of data concerning the physical, chemical, biological and transport processes undergone by pollutants in the unsaturated zone, the capillary fringe and the uppermost saturated part of the aquifer – the water table region. Groundwater quality monitoring networks have been developed as a result of our inability to forecast the cause-effect processes which influence the chemical composition of groundwater. Groundwater pollution is usually a complex and long-term process which can be divided into four schematic stages: A. B. C. D.

Surface disposal of pollutants Transport through the unsaturated zone Arrival at the groundwater table surface Transport within the saturated zone

Existing monitoring techniques tend to focus attention primarily on events occurring at stage D and, in many cases, monitoring of groundwater quality depends primarily on the analysis of samples obtained from active pumping wells. Generally, production wells are designed to pump water from deep below the water table. Therefore, evidence of pollution build-up in a production well reflects the mixing process in the aquifer which often took place years after the pollutant had arrived at the water table. An illustration of such delay is presented in Fig. 1.1 The study area depicted in Fig. 9.8 had been under irrigation with sewage effluents for 22 years (Ronen and Magaritz, 1985) and the vertical flow rate of anions through the 30 m thick unsaturated zone was determined to be about 1.4 m/yr (Gvirtzman et al., 1986). The very high concentration of solutes in the water table region (monitoring wells WT-2 and WT-3; Fig. 9.8) is related to the influx of sewage effluents. However, the concentration of solutes in the deep production wells (e.g., Glil Yam B) is not yet affected by sewage irrigation. Since the residence time in the active part of the aquifer (upper two thirds of the saturated thickness of the aquifer) is about 25 years, the quality of groundwater in the pumping wells reflects the replenishment conditions

* The water table region is here defined as a water layer of about 3 m thick immediately below the surface along which the hydrostatic pressure is equal to the atmospheric pressure.


89


16/04/2008

11:15

Page 90

some 45 years earlier. Eventually, the pollutants detected in the water table region will be dispersed through the groundwater body, and their concentration in the production wells will increase. It is worth noting that a similar phenomenon will be observed in a monitoring well where the screen is located deep below the water table.

Figure 9.8

Electrical conductivity (E.C. μmho/cm) and Cl-, NO3- and SO4= concentrations (mg /l) found in the water table region of two monitoring wells, WT-2 and WT-3 and three production wells pumping from deep (37-55 m) below the water table. The average water table depths in WT-2 and WT-3 are 27 and 30 m, respectively. The vertical flow rate of anions through the unsaturated zone was determined to be about 1.4 m/yr (Gvirtzman et al., 1986). The black arrow shows the regional direction of groundwater flow. The coordinates denote distance (km).

Clearly a different monitoring approach is needed. Much as the Monitor (a lizard of the genus Varanus) gives warning of the approach of crocodiles (American College Dictionary, 1964) a monitoring system is needed which will alert concerned parties before massive pollution of groundwater occurs. A MLS technique introduced here, which monitors the arrival of pollutants at the water table, was described in the paper of Ronen et al. (1987a). In this paper importance of MLS technique to the management of groundwater resources is discussed.

90



16/04/2008

11:15

Page 91

9.2.2 The unsaturated-saturated interface The interface between the unsaturated and saturated zones of a phreatic aquifer is characterized by the change from a three phase (rock-water-gas) system to a two phase (rock-water) system. Saturated conditions already exist in the capillary fringe above the water table (Davis and DeWiest, 1966; Bear, 1972; Ronen et al., 1997). In this zone water is held by surface tension forces (at pressures lower than atmospheric) and the ‘real’ unsaturated-saturated interface has an irregular shape (Ronen et al., 1997). In the unsaturated zone, water is in close contact with the gaseous phase which fills the pore space. Moisture content increases in the capillary fringe, and at the water table (or more strictly when the pores are totally saturated with water), groundwater is isolated from the atmosphere above. If a solute (e.g., Cl -, NO3- or dissolved organic carbon [DOC]) originating in the top-soil, is transported downwards through the unsaturated zone, and the input concentration of the solute is higher than the background concentration in the aquifer (the common situation in areas with a high surface load of pollutants), it would be expected that its highest concentration in groundwater would occur in the water table region. The variation with time in concentration of a conservative solute (e.g., Cl -), in the water table region, can be defined by the rate of transport controlling processes (in the capillary fringe and in the water table region). The variation in concentration of a non-conservative component (like nitrogen and DOC) may also be influenced by chemical and biological processes. Since the solubility of oxygen in water is low (8.7 mg/l at 22°C) and because oxygen replenishment in subsurface environments is limited, the biodegradation of only a small amount of organic matter can lead to anoxic conditions in the water table region (Ronen et al., 1987b). Therefore, processes such as denitrification (Ronen et al., 1987c) may control the influx of nitrate from the unsaturated to the saturated zone. From the above arguments it is reasonable to assume that large chemical variabilities can be expected to be found in the water table region. Chemical heterogeneity in the water table region can also result from recharge of varied chemical composition (e.g., rain water and irrigation water; Gvirtzman et al., 1986). In the water table region the concentration of a chemical in the liquid phase is the result of a mass balance between: I - (a) its rate of supply from the unsaturated zone, (b) the rate at which it is produced in situ (e.g., nitrification), and II - (a) the rate at which it is dissipated into the bulk aquifer by transport processes, (b) the rate at which it is transformed into other chemical species by chemical and biological reactions (e.g., denitrification). All the above processes eventually determine the influx of solutes to the groundwater body. Therefore, the water table region should be monitored both to alert against initial arrival of a pollutant (‘early warning’ or ‘preventive monitoring’) and to measure its influx from the unsaturated zone (stage B).


91


16/04/2008

11:15

Page 92

9.2.3 Application of a MLS technique in an agricultural area irrigated with municipal sewage effluents Sewage effluent is foreseen as the only economic water source for agricultural use in many arid to semiarid countries around the world. When use of this ‘free’ new source of water is considered, the eventuality of groundwater pollution by sewage irrigation is often overlooked. In other cases, discussions about the potential of pollution continue long after pollutants have arrived in the water table region. The MLS (see Section 5.2.6) has been used to study and monitor the water table region of a 30 m deep sandy, phreatic aquifer in Israel in an agricultural area irrigated with municipal sewage effluents. The following are the major findings of this project.

Figure 9.9

92

A comparison between dissolved oxygen (O2) and dissolved organic carbon (DOC) profiles obtained in the water table region of the study area (Glil Yam) and the profiles calculated in a simulation model by Molz et al. (1986b). Note the sharp decrease of DOC in the capillary fringe (shaded area) and the development of anoxic conditions in the water table region. The persistence of DOC under unsaturated conditions for very long time periods (about 20 years in the study area) indicates that moisture content may be a major controlling factor in biodegradation.



16/04/2008

11:15

Page 93

The study site (Glil Yam), located 15 km north of Tel Aviv, is devoted to the cultivation of citrus trees and cereals. Municipal sewage effluents from the city of Herzlia have been used for irrigation of 75 ha during the summer season, May to November. In the early 1960s sewage was treated in an oxidation pond; in 1977 an extended aeration treatment plant replaced the old facility. The study area is under a high organic carbon load. About 140 kg Corg /ha are annually applied each summer, in contrast to the negligible amount added to fields irrigated with fresh water. DOC is not completely biodegraded as it percolates from the topsoil to groundwater (Amiel et al., 1990). A DOC mass balance suggests that the unsaturated zone still contains about 50% of the total DOC input. DOC persistence for more than 20 years under unsaturated conditions suggests that moisture content may be a major controlling factor for biodegradation (Amiel et al., 1990). The average DOC flux to the water table region has been calculated to be at least 3.1 x 10-2 mg Corg/cm2 per year. The high concentrations of DOC found in the water table region (up to 8 mg/l), the anoxification process (DO < 1 mg/l) resulting from the biodegradation of the DOC in this zone (Fig. 9.9; Ronen et al., 1987b), and the high concentrations of N2O (up to 400 μg /l; Fig. 9.10) and CO2 (2% to 5%; Magaritz et al., 1990) are evidence of both: (a) DOC mobility through the unsaturated zone and, (b) DOC biodegradability as the water content of the system changes in the capillary fringe/water table region. This observation seems to contradict commonly used models which suggest a relatively large retardation factor for organic components in relation to water. Moreover, the results obtained differ significantly from the findings on the removal of most of the DOC during infiltration from effluent ponds (Rettinger et al., 1991). This difference is related to the

Figure 9.10

Production of N2 O in the water table region of wells WT-2 and WT-3 (Fig. 9.8). The N2 O molar fraction in the gaseous phase of the capillary fringe (empty squares and circles) can reach 12,000 μg/l. This is forty times higher than the atmospheric concentration. The inverted triangles denote the position of the water table.


93


16/04/2008

11:15

Page 94

conditions that exist during infiltration to groundwater. While saturated conditions prevail under the infiltration ponds, unsaturated conditions exist under land irrigated with the sewage effluents. The gases produced during the biodegradation of the DOC and/or air entrapped in the pore space during groundwater recharge accumulate as a distinct gas phase – bubbles – down to a depth of less 1 m below the water table (Ronen et al., 1989). Large bubbles (radius 200 μm) reduce volumetric water content and, hence, hydraulic conductivity. Small bubbles (radius 50 μm) clog pores without significantly decreasing the volumetric water content (Fig. 9.11). In the studied area, and at a depth of less than one metre, the pressure at a point in the moving fluid (10-1 atm) is at least one order of magnitude smaller than that required to both initiate the movement of bubbles through a pore space and to overcome the resistance to flow offered by detached gas bubbles and liquid drops in capillary conduits. Thus, under natural gradient flow conditions, the presence of gas bubbles significantly reduces the flow, leading to the development of an almost stagnant water layer in the water table region (Ronen et al., 1986, Fig. 9.12). It is expected that stagnant water layers which are the result of biochemical activity will develop mainly in regions under high organic loads such as sanitary landfills, feedlots, or areas where dissolved organic carbon has been mobilized from natural sources by anthropogenic activity. Air bubbles will preferentially develop in aquifers where: (a) the fluctuations of the water table are rapid, and (b) the rate of groundwater replenishment is large.

Figure 9.11

94

Schematic representation of porous media with biochemically produced gas bubbles and entrapped air bubbles. The bubbles are present below the water table as seen in the monitoring well (right hand side). Note the small bubbles clogging the pore conduits without significantly reducing the volumetric water content. For the studied area it was calculated that the critical depth at which bubbles are most likely to be found is of about 1 m (Ronen et al., 1989). This estimate coincides with the depth of 0.60 m of an almost stagnant water layer found at the study site under natural gradient flow conditions.



16/04/2008

11:15

Page 95

Figure 9.12

Dramatic decrease in the horizontal component of the specific discharge (q) in the water table region of well WT-3 (Fig. 9.8). The decrease is attributed to (a) gas bubbles produced during biodegradation of organic matter (e.g. CO2, N2O), and (b) air bubbles entrapped during inhibition. The vertical axis denotes depth from the water table.

The intermittent input of rain and irrigation water of varying chemical composition (e.g., 20 mg/l Cl- from rain and about 200 mg/l Cl- from sewage effluents) along with the periodic input of other chemicals such as fertilizers creates a vertical profile composed of discrete water layers of varying composition in the unsaturated zone (Fig. 9.13; Ronen et al., 1988b). Evapotranspiration reduces the amount of water recharge mainly during summer, thus further increasing the difference between the salinity input of sewage effluents and that of rain. The replenishment of groundwater by this

Figure 9.13

Example of discrete water layers of varying Cl- content as detected in the unsaturated zone of monitoring wells WT-2 and WT-3 (Fig. 9.8). Also shown is the calculated water density (σ0 = ρ0 [kg / m3] – 1,000, where ρ0 is density at 0 oC; Ronen et al., 1988b). The vertical axis denotes depth from the land surface. The inverted triangle denotes the water table.


95


16/04/2008

11:15

Page 96

influx and the almost stagnant conditions prevailing in the water table region lead to the development of micro scale parcels of water having vertical and horizontal length dimensions of less than 1metre (Fig. 9.14) and varying chemical composition (e.g., in some cases by more than 50% in the concentration of Cl-, NO3- and SO4=) and density (Fig. 9.15). Micro scale water parcels may also be formed in situ in the water table region by chemical or biochemical processes (e.g., denitrification, Fig 9.15). These processes determine the influx of pollutants into the main groundwater body. The sharp boundaries between overlaying water parcels, which may have in some cases persisted over long periods of time (months) and the calculated range of the Peclet number at the study site (6.25 x 10-2 to 6.25 x 10-3) implies that mechanical dispersion by advection is not an important

Figure 9.14

Schematic representation showing two alternative possibilities for the vertical build-up of micro scale water parcels of different salinity (S). In alternative ‘b’ groundwater velocity V2 is greater than V1. See vertical cross sections through such parcels in Fig. 9.15.

Figure 9.15

96

Vertical cross sections through micro scale isothermal water parcels (Fig. 9.12) of Cl-, NO3-, SO4= and HCO3- in the water table region of well WT-3 (Fig. 9.8). Profile 13 was obtained 30 days after profile 12. Note denitrification in the upper 50 cm of the profiles and concomitant increase of HCO3-. The concentration of HCO3- was measured in the laboratory and is given as the relative amount in relation to the HCO3- content of the standard.



16/04/2008

11:15

Page 97

mixing mechanism in the water table region. The observed rapid destruction of the boundaries between the overlaying water parcels, which may occur within one month (Fig. 9.16), suggests haline convection. The overlaying water parcels are at times gravitationally unstable due to destabilizing salinity-density differences. The critical density difference which overcomes viscous drag forces at the study site is in the range of 0.230 to 0.281 kg/m3. This value fits the estimated free convection parameter or modified Rayleigh number for porous media (Bear, 1972). Haline convection should overcome stratification, which would develop under very slow laminar flow conditions, and therefore greatly influence the influx of pollutants to bulk groundwater (Ronen et al., 1988b).

Figure 9.16

Eulerian changes of chloride in consecutive profiles obtained in well WT-2 (Fig. 9.8). Note the development of sharp interfaces between water parcels. Gravitational instability between the parcels (after the density difference between them increased up to 0.230 kg/m3 on September 20) triggers convective flow. Note that the Cl - profile on November 1 below a depth of 120 cm has an average concentration which can be ascribed to the convective mixing of both water parcels observed on September 20. Haline convection is suggested to be a mechanism that influences the influx of pollutants to bulk groundwater.

9.2.4 Conclusions and recommendations The potential of the sampling methodology presented in this case study should be recognized for the development of groundwater early warning systems in phreatic aquifers. Sampling and measuring the actual pollutant fluxes reaching the water table through the unsaturated zone, before they are diluted in the groundwater body, has several advantages: (a) it increases the detection sensitivity of the monitoring system as pollutants arriving from the unsaturated zone will be found at maximum concentration in the water table region; (b) it gives ample time (decades) for remedial action to be undertaken before the onset of massive groundwater contamination, and (c) it enables the establishment of a quantitative relationship between the amount of a pollutant released on the top soil and the amount that actually reaches groundwater.


97


16/04/2008

11:15

Page 98

For the monitoring example presented above, available results suggest that long term and continuous irrigation with sewage effluents over the recharge area of a water table aquifer will lead to groundwater quality degradation. Three questions must be answered to quantify this phenomenon: (a) what is the net influx of pollutants to the saturated zone?; (b) what is the new steady-state concentration of the pollutant in the aquifer?; (c) what is the time that will elapse before wells produce water in violation of standards? The proposed monitoring system was designed especially to answer the first critical question. For example, if decision-makers utilize the proposed warning scheme as a real ‘Monitor’ they will be able to take one of the following managerial decisions: (a) permit the irrigation with effluents until a preestablished concentration of pollutant is reached, (b) re-allocate the water pumped from the aquifer according to its predicted quality, (c) design a dual pumping system that uses the upper part of the aquifer (the polluted region) for agriculture and continue to pump water from deeper strata for general use. The benefits of this last alternative are: (1) continued use of existing pumping wells (which pump deep below the water table), and (2) reduction in the quantity of required added fertilizers when taking advantage of effluent nutrient content in the upper water layer.

9.2.5 References American College Dictionary, 1964. Random House, New York. Amiel, A.J., Magaritz, M., Ronen, D. and Lindstrand, O. 1990. On the Mobility of Dissolved Organic Carbon in the Unsaturated Zone under Land Irrigated by Sewage-Effluents. Journal Water Pollution Control Federation, 62, pp. 861-866. Bear, J. 1972. Dynamics of Fluids in Porous Media. American Elsevier Pub. Co. Inc., New York, 764 pp. Davis, S.N. and DeWiest R.J.M. 1966. Hydrogeology. John Wiley, New York, 463 pp. Gvirtzman, H., Ronen, D. and Magaritz, M. 1986. Anion Exclusion During Transport through the Unsaturated Zone. Journal of Hydrology, 87, pp. 267-283. Kaplan, E., Banerjee, S., Ronen, D., Magaritz, M., Machlin, A., Sosnow, M. and Koglin, E. 1991. Multi-Level Sampling in the Water Table Region of a Sandy Aquifer. Ground Water, 29, pp. 191-198. Magaritz, M., Brenner, I. and Ronen, D. 1990. Ba++ and Sr++ Distribution at the Water Table: Implications for Monitoring Groundwater at Nuclear Waste Repository Sites. Applied Geochemistry, 5, pp. 555-562. Molz, F.J., Widdowson, M.A. and Benefield, L.D. 1986b. Simulation of Microbial Growth Dynamics Coupled to Nutrient and Oxygen Transportation in Porous Media. Water Resources Research, 22, pp. 1207-1216. Rettinger, D., Ronen, D., Amiel, A.J., Magaritz, M. and Bischofsberger, W. 1991. Tracing Sewage Influx from a Leaky Sewer in a Very Tthin and Fast-Flowing Aquifer. Water Research, 25, pp. 75-82. Ronen, D. and Magaritz, M. 1985. High Concentration of Solutes at the Upper Part of the Saturated Zone (Water Table) of a Deep Aquifer under Sewage-Irrigated Land. Journal of Hydrology, 80, pp. 311-323. Ronen, D., Magaritz, M., Paldor, N. and Bachmat, Y. 1986. The Behavior of Ground-water in the Vicinity of the Water Table Evidenced by Specific Discharge Profiles. Water Resources Research, 22, pp. 1217-1224. Ronen, D., Magaritz, M. and Levy, I. 1987a. An in situ Multilevel Sampler for Preventive Monitoring and Study of Hydrochemical Profiles in Aquifers. Ground Water Monitoring Review, 7, pp. 69-74. Ronen, D., Magaritz, M., Almon, E. and Amiel, H. 1987b. Anthropogenic Anoxification (‘Eutrofica-

98



16/04/2008

11:15

Page 99

tion’) of the Water Table Region of a Deep Phreatic Aquifer. Water Resources Research, 23, pp. 1554-1560. Ronen, D., Magaritz, M., Gvirtzman, M. and Garner, W. 1987c. Microscale Chemical Heterogeneity in Groundwater. Journal of Hydrology, 92, pp. 173-178. Ronen, D., Magaritz, M. and Almon, E. (1988a). Contaminated Aquifers are a Forgotten Component of the global N2O budget. Nature, 335, pp. 57-59. Ronen, D., Magaritz, M. and Paldor N. (1988b). Microscale Haline Convection - A Proposed Mechanism for Transport and Mixing at the Water Table Region. Water Resources Research, 24, pp. 1111-1117. Ronen, D., Berkowitz, B. and Magaritz, M. 1989. The Development and Influence of Gas Bubbles in Phreatic Aquifers under Natural Flow Conditions. Transport in Porous Media, 4, pp. 295-306. Ronen, D., Scher, H. and Blunt, M. 1997. On the Structure and Flow Processes in the Capillary Fringe of Phreatic Aquifers. Transport in Porous Media, 28, pp. 159-180.


99


16/04/2008

11:15

Page 100

APPENDICES Abbreviations and acronyms BOD

Biological Oxygen Demand

Bull.

Bulletin

Conf.

Conference

DNAPLs

Dense Nonaqueous Phase liquids

ed(s).

Editor(s); also: edition

e.g.

for example (Latin exempli gratia)

EPA

Environmental Protection Agency

IAH

International Association of Hydrogeologists

IHP


IHP-V

Fifth phase of the IHP

Jour.

Journal

LNAPLs

Light Nonaqueous Phase Liquids

NAPLs

Nonaqueous Phase Liquids

No.

Number

pp.

Pages

RIVM

Rijksinstituut voor Volksgezondheid en Milieuhygien

TDS

Total Dissolved Solids

UK

United Kingdom

UNESCO


Vol.

100

Volume



16/04/2008

11:15

Page 101

Appendix 2: Glossary A conscious effort has been made to write in clear language, keeping technical jargon to a minimum. Even so, some terms are in such common use that their inclusion is considered to be justified. However, some terms are used with rather different meanings by individual authors and in different countries. For these terms a particular meaning has been utilized in this publication as is explained in the glossary. The following terms and definitions were compiled from various sources (American College Dictionary, 1964; Phannkuch, 1990; UNESCO/WMO, 1992).

Adsorption: The attraction and adhesion of ions from an aqueous solution to the solid soil or rock surfaces with which they are in contact. Advection: The process by which solutes are transported with and at the same rate as moving groundwater. Aeration: The process by which air becomes dissolved in water. Alkalinity: The ability of the salts contained in water to neutralize acids. Anaerobic: Describing a process conducted in the absence of oxygen. Aquifer: A geologic unit that is capable of yielding a significant amount of groundwater to a well or spring. Aquifer, confined: An aquifer bounded above and below by confining beds of distinctly lower permeability than that of aquifer itself. Aquifer, unconfined: An aquifer in which there are no confining beds between the zone of saturation and the ground surface. Attenuation: The intrinsic ability of earth materials and groundwater to reduce, remove, dilute or retard contaminants by the complex of physical, chemical, and biological processes acting in the soil–rock–groundwater system. Base flow: That component of the flow of streams composed solely of groundwater discharge. Biodegradation: The breakdown of chemical constituents through the biological processes of naturally occurring organisms. Capillary fringe: The zone immediately above the groundwater table in which water is drawn upward by capillary attraction and the voids are filled with water under pressure less than atmospheric. Cation exchange capacity: A measure of the availability of cations that can be displaced from sites on solid surfaces and that can be exchanged for other cations. Cone of depression: A depression in the groundwater table or potentiometric surface that develops around a well from which water is being withdrawn. Contaminant (Pollutant): A naturally–occurring or human-produced physical, chemical biological, or radiological substance that renders water unfit for a given use. In this publication pollutant is used as a synonym. Contamination (Pollution): The introduction into water of any undesirable chemical, biological or radiological substances which render the water unfit for its intended use. In this publication pollution is used as a synonym.


101


16/04/2008

11:15

Page 102

Contamination (Pollution) plume: The spreading of a contaminant (pollutant) in the direction of groundwater flow. Density: The mass or quantity of a substance per unit volume. Detergent: Any material with cleaning powers, including soaps, synthetic detergents, alkaline materials, and solvents. Diffusion: The process by which both ionic and molecular constituents move under the influence of their kinetic activity in the direction of their concentration gradient. Diffuse contamination (pollution) source: A source of contamination (pollution) in which a contaminant (pollutant) entering the receiving water can not be attributed to a single outlet. Dispersion: A process of contaminant (pollutant) transport that occurs as a result of mechanical mixing and molecular diffusion. Dissolution: The process of dissolving. Dissolved solids: The weight of inorganic and organic matter in true solution in a stated volume of water. DNAPL: Acronym for dense, nonaqueous phase liquid. Effective porosity: Ratio of the volume of interconnected pore space to the total volume of a porous material. Eh – Redox potential: A measure of the electron balance in an water sample. Electrical earth resistivity: A surface geophysical method in which a direct or low frequency current is applied to a pair of electrodes into the ground and the resulting voltage is measured at a second set of electrodes. Electric logging: Methods oriented towards estimation of physical and/or chemical properties of formation, fluids filling the borehole or the formation voids and fractures, and the determination of geometrical parameters of the borehole and encountered layers. Flow net: A set of intersecting equipotential lines and flow lines representing a two- dimensional steady flow field in porous media. Flow path: The direction of movement of groundwater (and contaminants) as governed principally by the hydraulic gradient. Fracture: A break in a rock formation due to mechanical failure by stress; includes cracks, joints (fissures), and faults. Groundwater: Subsurface water in the saturated zone. Groundwater flow: The movement of water through openings in sediment and rock that occurs in the saturated zone. Groundwater runoff: That portion of precipitation which is absorbed by soil to the groundwater body and later discharged to surface streams. Groundwater protection zone: An area of land within which activities liable to contaminate (pollute) groundwater are restricted or prohibited. Halogenated hydrocarbons: Organic compounds containing one or more halogens. Hazardous waste: Any waste that poses a substantial present or potential hazard to human health or living organisms. Hydraulic conductivity: The quantity of water that will flow through a unit cross-sectional area of a porous material per unit of time under a hydraulic gradient of 1.00. Hydrodynamic dispersion: The process by which groundwater containing a solute is diluted by uncontaminated groundwater and spread out from the flow path because of mechanical mixing during fluid advection and molecular diffusion due to the thermal- kinetic energy of the solute particles. Longitudinal dispersion is usually much stronger than lateral dispersion. Hydrolysis: The chemical decomposition of a compound by reaction with water.

102



16/04/2008

11:15

Page 103

Imaging methods: Remote sensing methods in which data are displayed in the form of an image, as in a set of photographs along a flight path. Immiscible: Fluids that are not significantly soluble in water. Infiltration: Passage of water downward from the land surface into and through the soil and rock layers. Injection well: A well used for injecting fluids into an underground stratum. Ion exchange: A process by which an ion in a mineral lattice is replace by another ion that was present in an aqueous solution. Karst (karst topography): A topographic area that has been formed by the dissolution of carbonate rocks and that is characterized by sinkholes, caves, caverns, and lack of surface streams. Landfill: A disposal facility in which waste is placed in or on the land. Leachate: A solution produced by water or other liquid percolating through soil or solid waste and the subsequent dissolution of certain constituents in the water. Line source: A linear source of contamination that can spread contaminants over large distances, e.g. leaking pipelines or contaminated streams. Lithology: Description of rocks in terms of mineral composition and texture. Losing stream: A stream in which water flows and infiltrates from the streambed into the ground. Lysimeter: Unsaturated (vadose) zone sampling device used to collect soil pore water via suction or gravity drainage; is capable of retaining the accumulated water within the sampling vessel. Magnetometry: A surface geophysical method for measuring the total intensity of the earth’s magnetic field by means of high resolution proton magnetometers. Mobilize: To accelerate the movement of a contaminant in the groundwater system by changing the prevailing chemical conditions. Molecular diffusion: The process whereby ionic or molecular constituents in solution move under the influence of their kinetic activity in the direction of their concentration gradient. Monitoring well: A well that is designed for the purpose of extracting groundwater samples for testing, or for measuring groundwater levels. Nested wells: A series of single-cased monitoring wells that are closely spaced, but with screens at different depths. Neutralization: The inorganic reaction of an acid and a base to create a salt and water. Non-point contamination (pollution) source: See diffuse source. Oxidation: A chemical reaction in which there is an increase in valence resulting from a loss of electrons. Percolation: Downward movement of water under gravity or hydrostatic pressure through earth materials. Piezometer: A small-diameter well installed to measure the elevation of the groundwater table or potentiometric surface, or to permit collection of groundwater samples from discrete horizons. Permeability: The ability of a soil or rock medium to transmit a fluid. Point contamination (pollution) source: Any discrete, well-defined source of contamination (pollution). Pollutant: See contaminant. Porosity: The ratio of volume of void spaces in a rock or sediment to the total volume of the rock or sediment. Potable water: Water that is suitable for human or animal consumption. Precipitation: The formation of solids out of constituents that were once dissolved. Recharge: The addition of water to the groundwater system by natural or artificial processes. Recharge area: An area in which there are downward components of hydraulic head in the aquifer.


103


16/04/2008

11:15

Page 104

Reduction: A chemical reaction in which there is a decrease in valence as a result of gaining electrons. Remote sensing: The process of obtaining information about the surface of the upper layer of the earth’s crust from aircraft, satellites or by other above-ground techniques. Retardation: Preferential retention of contaminants or slow down of their travel in the subsurface by physical, chemical, or biological processes. Root zone: The zone from the land surface to the depth penetrated by plant roots. Saturated zone: The zone in which the voids in the rock or soil are filled with water at a pressure greater than atmospheric. Soil moisture content: The amount of water in the soil expressed as a fraction of the total porous volume of the soil. Solute transport: The net flux of solute through a hydrogeologic unit controlled by the flow of subsurface water and transport mechanism. Solution: A homogenous mixture of two or more components. Sorption: The combined effect of adsorption and absorption. Spring: A discrete place where groundwater flows naturally from a geologic formation onto the land surface or into a body of surface water. Texture: The interrelationship between the size, shape, and arrangement of minerals or particles in a rock or soil. Total dissolved solids (TDS): The total concentration of dissolved constituents in solution. Transmissivity: The rate at which water is transmitted through a unit width of an aquifer under unit hydraulic gradient. Unconfined aquifer: An aquifer that has a water table forming a free upper surface. Unsaturated zone: The zone between land surface and the water table that contains both water and air. It includes the root zone, intermediate zone and capillary fringe. Viscosity: The property of a fluid describing its resistance to flow. Volatile constituents: Solid or liquid compounds that are relatively unstable at standard temperature and pressure, and undergo spontaneous phase change to the gaseous state. Vulnerability (groundwater): An intrinsic property of a groundwater system that depends on the sensitivity of that system to human and/or natural impacts. Water table: The surface in a groundwater body at which the pore water pressure is atmospheric. Water table region (defined for this report): a water layer of about 3 m thick immediately below the surface along which the hydrostatic pressure is equal to the atmospheric pressure. Well screen: A filtering device, typically steel or plastic, that allows groundwater to flow freely into a well from the adjacent formation. Zone of saturation: See saturated zone.

104
