Environmental Management of Water Systems under...

Environmental Management of Water Systems under Uncertainty

Christian Baresel

Doctoral Thesis Stockholm, Sweden 2007

TRITA-LWR PHD 1034 ISSN 1650-8602 ISRN KTH/LWR/PHD 1034 -SE ISBN 978-91-7178-670-8

TRITA-LWR PHD 1034 ISSN 1650-8602 KTH Mark- och Vattenteknik ISRN KTH/LWR/PHD 1034 -SE SE-100 44 Stockholm ISBN 978-91-7178-670-8 SWEDEN Akademisk avhandling som med tillstånd av Kungl Tekniska Högskolan framlägges till offentlig granskning för avläggande av teknologie doktorsexamen torsdagen den 7 juni 2006 klockan 10.00 i Sal D2, Lindstedtsv 5, Kungl Tekniska Högskolan, Stockholm. Christian Baresel, May 2007 Tryck: Universitetsservice US AB


ABSTRACT

Hydrological drainage/river basins constitute highly heterogeneous systems of coupled natural and anthropogenic water and pollutant flows across political, national and international boundaries. These flows need to be appropriately understood, quantified and communicated to stakeholders, in order to appropriately guide environmental water system management. In this thesis, various uncertainties about water and pollutant flows in drainage/river basins and their implications for effective and efficient water pollution abatement are investigated, in particular for mine-related heavy metal loadings in the Swedish Dalälven River basin and for nitrogen loadings in the Swedish Norrström drainage basin. Economic cost-minimization modeling is used to investigate the implications of pollutant load uncertainties for the cost-efficiency of catchment-scale abatement of water pollution.

Results indicate that effective and efficient pollution abatement requires explicit consideration of uncertainties about pollution sources, diffuse contributions of the subsurface water system to downstream pollutant observations in surface waters, and downstream effects of different possible measures to reduce water pollution. In many cases, downstream load abatement measures must be used, in addition to source abatement, in order to reduce not only expected, but also uncertainties around expected pollutant loads. Effective and efficient environmental management of water systems must generally also consider the entire catchments of these systems, rather than focusing only on discrete pollutant sources. The thesis presents some relatively simple, catchment-scale pollutant flow analysis tools that may be used to decrease uncertainties about unmonitored water and pollutant flows and subsurface pollutant accumulation-depletion and diffuse loading to downstream waters. Key words: Uncertainty; environmental management; water pollution; cost effectiveness; pollution abatement; input-output analysis; nitrogen; mining.

i


ii

ACKNOWLEDGEMENTS

This thesis was made possible by a KTH Honor Grant (“Excellenstjänst”) and by a grant from the EU project ERMITE (“Environmental regulation of mine waters in the EU”), funded within the European Commission’s Fifth Framework Programme under Contract No EVK1-CT-2000-0078; the support by KTH and the ERMITE project is gratefully acknowledged. Participation in the ERMITE programme has also allowed interdisciplinary problem formulation and collaboration, and contacts with various researchers in a number of different research fields. Here I would like to name Karin Larsén whose master thesis served as a starting point for parts of my research work.

This dissertation would not have been possible without the exceptionally valuable guidance and support I have received from my main supervisor, Georgia Destouni (Professor in Hydrology, Hydrogeology and Water Resources at the Department of Physical Geography and Quaternary Geology, Stockholm University). She has broadened and deepened my understanding of hydrogeology in general, while helping me to develop the critical perspective necessary for my research field. This dissertation also has benefited enormously from the critical input and review of my co-supervisors Ing-Marie Gren (Professor in Natural Resource and Environmental Economics at the Department of Economics, Swedish University of Agricultural Sciences, Uppsala) and Vladimir Cvetkovic (Professor in Water Engineering at the Department of Land and Water Resources Engineering, Royal Institute of Technology, Stockholm). Both have greatly improved my understanding of other research fields important for this thesis such as economics and helped me with fruitful discussions.

I thank everyone at the Department of Land and Water Resources Engineering, some for the help with different matters and some for simply socializing; I have greatly enjoyed my time at the department. Special thanks to the former members of the P&P group and our most loyal guest and my personal friend Tomo. Much thanks also to Aira for all help with not only administrative matters. I thank Vladimir and all others involved in activities related to the undergraduate course Environmental Dynamics for the great team work. Working with the course often provided a way to escape my research in times when I was lacking motivation.

Finally, a few but important personal acknowledgements: I want to thank Steffi and Lisa for supporting me not the least by distracting me from my research whenever necessary. I thank the rest of my family for their love and for sharing this moment with me.

Christian Baresel Stockholm, May 2007


iii

LIST OF PAPERS APPENDED

(I) Baresel, C., Larsén, K., Destouni, G., Gren, I.M. 2004. Economic Analysis of Mine Water Pollution Abatement in a Catchment. In: Mining Impacts on the Fresh Water Environment: Technical and Managerial Guidelines for Catchment-Focused Remediation. Younger, P.L., Wolkersdorfer, C. (Eds). Mine Water and the Environment, Suppl. Issue 1. Springer Berlin, pp 57-76, doi:10.1007/s10230-004-0028-0.

(II) Baresel, C., Destouni, G., Gren, I.M. 2006. The Influence of Metal Source Uncertainty on

Cost-Effective Allocation of Mine Water Pollution Abatement in Catchments. J. Environ. Manage. 78(2), 138-148, doi:10.1016/j.jenvman.2005.03.013.

(III) Baresel, C., Destouni, G. 2007. Subsurface Water System Contributions to Surface Water

Zinc Loads in Mining Areas. (manuscript to be submitted). (IV) Baresel, C., Destouni, G. 2005. Novel Quantification of Coupled Natural and Cross-

Sectoral Water and Nutrient/Pollutant Flows for Environmental Management. Environ. Sci. Technol. 39(16), 6182-6190, doi:10.1021/es050522k.

(V) Baresel, C., Destouni, G. 2006. Estimating Subsurface Nitrogen Accumulation–Depletion

in Catchments by Input–Output Flow Analysis. Phys. Chem. Earth 31(17), 1030–1037, doi:10.1016/j.pce.2006.07.007.

(VI) Baresel, C., Destouni, G. 2007. Uncertainty-Accounting Environmental Policy and

Management of Water Systems. Environ. Sci. Technol. 41(10), in press, doi:10.1021/es061515e.

http://dx.doi.org/10.1007/s10230-004-0028-0

http://dx.doi.org/10.1016/j.jenvman.2005.03.013

http://dx.doi.org/10.1021/es050522k

http://dx.doi.org/10.1016/j.pce.2006.07.007

http://dx.doi.org/10.1021/es061515e


LIST OF RELATED PUBLICATIONS NOT APPENDED

Baresel, C., Destouni, G. 2005. Quantitative Cross-Sectoral Analysis of Water and Pollutant Cycling in Catchments. Proceedings from the 15th Stockholm Water Symposium – Drainage Basin Management - Hard and Soft Solutions in Regional Development, August 21-27, Workshop 3, 98-99.

Baresel, C., Destouni, G. 2005. Input-Output Analysis as a Quantitative Tool for Integrated

Water Management. Geophysical Research Abstracts, Vol. 7, 04390, European Geosciences Union, General Assembly, Vienna 24-29 April 2005.

Baresel, C., Destouni, G. 2004. Implications of Metal Load Randomness for Mine Water

Pollution Abatement. Eos Trans. AGU, 85(46), Fall Meeting Suppl., Abstract H33D-0492.

Baresel, C., Destouni, G. 2004. Cost-Effective Abatement of Stochastic Metal Loading to Water Recipients. Proceedings of the Symposium: Mine Water 2004 - Process, Policy and Progress, Volume 1, September 19-23, Newcastle upon Tyne, ISBN: 0-9543827-4-9, pp 101-106.

Destouni, G., Baresel, C. 2003. Cost-Effective Remediation of Mine Waste Sites on a

Catchment Scale. EGS-AGU-EUG Joint Assembly, Nice, European Geophysical Society.

Baresel, C., Destouni, G. 2003. Cost-Efficient Mine Water Pollution Abatement on a Catchment Scale. Proceedings from the 13th Stockholm Water Symposium – Drainage Basin Security-Balancing Production, Trade and Water Use, August 11-14, Workshop 5, pp 271-274.

Amezaga, J., Baresel, C., Destouni, G., Göbel, J., Gren, I.M., Hannerz, F., Larsén, L., Loredo,

J., Malmström, M., Nuttall, C., Santamaría, L., Veseliè, M., Wolkersdorfer, C., Younger, P. in: ERMITE Consortium. 2004. Mining Impacts on the Fresh Water Environment: Technical and Managerial Guidelines for Catchment-Focused Remediation. In Younger PL, Wolkersdorfer C (eds): ERMITE Report: D6, The European Commission Fifth Framework Programme, Energy, Environment and Sustainable Development, Contract No EVK1-CT-2000-0078, University of Oviedo.

Baresel, C., Larsén, K., Destouni, G., Gren, I.M. 2003. Economic Analysis of Mine Water

Pollution Abatement on a Catchment Scale. ERMITE Report: D5, The European Commission fifth Framework Programme, Energy, Environment and Sustainable Development, Contract No EVK1-CT-2000-0078, University of Oviedo.

iv


CONTENTS

1 INTRODUCTION 1 2 MATERIALS AND METHODS 3

2.1 Case study areas 3 2.2 Catchment-Scale Economic Analysis of Pollution Abatement 3 2.3 Input-Output Analysis of Water and/or Pollutant Flows 5

3 CATCHMENT-SCALE COST-EFFICIENT ABATEMENT OF WATER POLLUTION IN MINING AREAS (PAPER I) 6

4 POLLUTION SOURCE UNCERTAINTY EFFECTS ON COST- EFFICIENT ABATEMENT OF WATER POLLUTION IN MINING AREAS (PAPER II) 8

5 SUBSURFACE WATER SYSTEM CONTRIBUTIONS TO SURFACE WATER POLLUTION IN MINING AREAS (PAPER III) 10

6 QUANTIFICATION OF COUPLED NATURAL AND CROSS- SECTORAL POLLUTANT FLOWS IN CATCHMENTS (PAPERS IV AND V) 12

7 UNCERTAINTY-ACCOUNTING ENVIRONMENTAL POLICY AND MANAGEMENT OF WATER SYSTEMS (PAPER VI) 14

8 GENERAL DISCUSSION 16 8.1 Pollutant Emission and Pollutant Transport-Attenuation Uncertainties 16 8.2 Subsurface Water System Contributions to Pollution Loadings 16 8.3 Economic Analysis of Pollution Abatement 17 8.4 Accounting for Uncertainty in Environmental Policies 17

9 CONCLUSIONS 19 10 REFERENCES 20

v


1

1 INTRODUCTION

Water is as essential to sustainable development as it is to life and ecosystem status and dynamics (UN World Water Assessment Program (2003)). Over the last century, human water use has increased at least 6-fold (with human population increasing “only” 3-fold), in total and in the individual household, agricultural, and industrial sectors (Cosgrove and Rijsberman 2000; Gleick 2003). This increase of human water quantity use has also been accompanied by increasing water pollution and deterioration of water systems due to various anthropogenic pollutants, which may be one of the most serious current environmental problems (Turner and Rabalais, 1994; Nixon 1995; Conley 2000; European Commission, 2000; Kavanaugh et al., 2003; ERMITE Consortium; 2004). Aquatic ecosystem pollution is caused, for instance, by the polluted water drainage from mine wastes, and other active or abandoned mine ground and mining facilities (Herlihy et al., 1990; Younger et al., 1997; Iribar et al., 2000), and nitrogen loadings from different types of known and yet unknown point and diffuse sources (e.g., HELCOM 1993; Nixon 1995; Howarth and Marino, 2006). International and national legislations and agreements (e.g. European Water Framework Directive (WFD; European Commission, 2000), HELCOM (1993) and Swedish national environmental goals (Government Bill, 2000)) aim at maintaining or improving the quality of the aquatic environment for humans and ecosystems. This requires catchment-scale implementation of integrated water resource management (IWRM) principles including an improved scientific basis for coupled natural and anthropogenic water system cycling and cause-effect links between water systems and various water using/impacting stakeholders in catchments. Environmental policies and the management for mitigating and protecting water systems from pollution aim further at finding a

balance between maintaining or achieving certain environmental standards and minimizing possible impairments to economically and socially beneficial uses of the water environment. Objective investigation of where this balance may lie or efficient achievement of already decided standards requires quantification of expected costs for the environmental management and protection (European Commission, 2000; Kavanaugh et al., 2003; ERMITE Consortium, 2004; Fleischer et al., 1991; Gren 1993, 1995; Gren et al., 1997). The environmental standards may, for instance, be expressed as maximum concentration levels (MCLs), maximum pollutant loads (MPLs) (Kavanaugh et al., 2003), minimum aquatic ecosystem status (European Commission, 2000), or minimum pollutant load reductions (HELCOM 1993). However, many studies have pointed out inherent uncertainties of complex environmental systems that make it difficult or impossible to reliably identify all significant sources of water pollution (Younger et al., 1997; Wood et al., 1999; Younger 2003). Furthermore, other recent studies of pollutant transport and attenuation downstream of pollution sources show large physical and biogeochemical pollutant spreading in natural, heterogeneous ground-water systems (Berglund et al., 2003; Malmström et al., 2004, 2007; Lindgren et al., 2007). This leads to additional, diffuse and long-term sources of previously immobilized water pollutants that are later slowly released and transported from their sorption/ precipitation sites and more or less immobile water zones throughout aquifers to surface waters. This further implies that it is difficult or impossible to reliably identify costs and optimal allocation of pollution abatement measures within catchments without accoun-ting for such uncertainties (Kavanaugh et al., 2003; Beavis and Walker, 1983; McSweeny and Shortle, 1990; Shortle et al., 1998;

Christian Baresel TRITA-LWR PHD 1034

Costanza 2000; Gren et al., 2000a, 2002; Andersson and Destouni, 2001; Popper et al., 2005). A general objective of this thesis is to contribute to the understanding of the significance of uncertainty aspects in managing water systems and to a methodological development for quantitative consideration of such uncertainty in environmental policy and management of water systems. In addition, each of the appended papers that together constitute this thesis has also different specific objectives. The main objectives of the thesis, however, may be summarized to be the investigation of and answers to the following questions:

Do uncertainties that are related to pollutant emissions and pollutant transport-attenuation in catchments have

significant impacts on the quantification of water pollutant loadings and on the costs and allocation of measures for cost-efficient water pollution abatement?

How important is the role of diffuse subsurface contributions for pollution loadings to surface waters and are there relatively simple quantification tools that may be used to identify such contributions in absence of any direct measurement data?

Is it possible to account explicitly for risk and uncertainty in environmental management of water systems and thereby identify environmental policies that are robust over a range of different real and future states and requirements of the environment?

2


3

2 MATERIALS AND METHODS

Specific case studies have been used to investigate and answer the different research questions addressed in the appended papers. Two Swedish drainage basins have been selected because of their importance for the water environment in both the basins themselves and their pollutant-receiving coastal waters of the Baltic Sea. Furthermore, two general methods have been applied in the different studies. In Papers I-II and VI, a general cost minimization model for achieving a pre-specified maximum pollutant load at a specific compliance boundary (CB; associated with some essential water recipient) is used as investigation tool. The pollutant transport from the various pollutant sources to the CB is described using a simple catchment-scale solute transport-attenuation model. In Papers III-V, the input-output analysis (IOA) methodology is used to study water and/or pollutant mass flows among the different natural and anthropogenic water using/impacting sectors and water systems in a catchment. The following sections present briefly the two case study areas and the main research methods used. They provide also information about why these methods have been selected for the different involved investigations. For details on general and case-study specific model setups, model quantifications and parameterizations, the reader is referred to the appended papers, especially Papers I-II and Paper IV for a thorough presentation of the economic analysis methodology and the input-output analysis methodology, respectively.

2.1 Case study areas

Papers I-III and VI consider heavy metal and especially zinc, copper and cadmium loadings to the Swedish Dalälven River from mining activities within its two main areas of both historical and currently active mining importance, Falun and Garpenberg (Figure 2.1). The water pollution of most general

public interest is the metal leaching to the Dalälven River itself (Svensson 1988, Hartlén and Lundgren 1990, Lindeström 1999) and through the river to the Baltic Sea (HELCOM 1993). But also the possible pollution of local water bodies, in the near vicinity of mine waste sites, is interesting to investigate for local long-term sustainability and also according to the WFD (e.g. European Commission, 2000). Papers IV-V address the problem of nitrogen flows across the different natural and anthropogenic sectors and water systems of the Swedish Norrström drainage basin (Figure 2.1). The investigation of nitrogen flows in the Norrström drainage basin is interesting because the basin suffers from excess nutrient loading to its own inland waters and not least to Lake Mälaren (Lake Mälaren Society (Mälarens Vattenvårdsförbund, MVF), 2004), which provides drinking water for the Swedish capital Stockholm and other areas. The basin further constitutes one of the main Swedish contributors of nitrogen loads to the Baltic Sea (HELCOM, 2004).

2.2 Catchment-Scale Economic Analysis of Pollution Abatement

In Papers I-II and IV, a cost minimization model is used to investigate the cost-effective abatement of water pollution from mine wastes and other active or abandoned mine ground and mining facilities in the Dalälven River basin. In the cost minimization approach, a given water quality target, for example given by political decisions or by law, determines the required water quality improvement. The aim is to achieve this pre-specified water quality target (or targets) in chosen water recipients at minimum cost (Gren et al. 2000a, 2000b). The economic optimization used in this thesis is static and considers average annual pollutant flows. Therefore, temporal aspects, such as when specific abatement measures should be


Figure 2.1. The Swedish Norrström drainage basin with Lake Hjälmaren and Lake Mälaren, and the Swedish Dalälven River basin and its main mining areas, Falun within the sub-catchment of Lake Runn and Garpenberg within the sub-catchment of Lake Gruvsjön.

applied and when pre-specified pollutant load reductions should and will be achieved, are not addressed. Basically, an economic optimization model consists of an objective function that is subject to defined constraints. For the studied problem of reducing metal loadings to the Dalälven River, the objective function defines the total cost of allocating different pollutant emission and load reduction measures in the relevant catchment. Thus, the cost functions of the different possible abatement measures must be known. The main constraint to the cost function is the requirement to meet a pre-specified load reduction at a specific compliance boundary (here the water recipient). The pollutant load

to the recipient (or into a downstream abatement measure near the recipient) is determined by the natural pollutant attenuation that may occur along the pollutant transport pathway from the source to the recipient. The attenuation effect is quantified by a pollutant mass delivery fraction (also referred to as a delivery coefficient) for each sub-catchment. In addition to the constraint of achieving a pre-specified pollutant load reduction, the objective function may be subject to other constraints such as the technical feasibility of abatement measures at specific sites. The different cost functions and constraints imply that minimizing the total cost of water pollution abatement in the Dalälven River

4


5

basin is a nonlinear and mixed integer optimization problem that has here generally been solved using various global optimization algorithms of the General Algebraic Modeling System (GAMS; Brook et al., 1996).

2.3 Input-Output Analysis of Water and/or Pollutant Flows

The input-output analysis (IOA) metho-dology was originally developed by Leontief (1936) and has mainly been used for the quantification of capital flows in economic systems but also for the quantification of material flows in various political and socioeconomic systems (municipalities, cities, countries) (Schrøder 1995; Xikang 2000) and ecological systems (Wulff et al., 1989; Hinrichsen and Wulff, 1998). Furthermore, Gren and Wulff (2003) and Gren and Elofsson (2004) have demonstrated the suitability and usefulness of IOA for quantification of nitrogen flows in the marine basins of the Baltic Sea and associated investigation of cost-effective solutions for nitrogen load reductions.

In general, input-output analysis (IOA) uses a matrix representation of flows among the different subsystems/sectors of the system considered. The IOA method enables tracing of system outputs back to associated system inputs (backward flow analysis (Leontief, 1936)), or conversely tracing the fate of inputs through a system to associated outputs (forward flow analysis (Augusti-novics, 1970)). The mathematics of the method is straightforward, but the data requirements are large because the inflows and outflows of each subsystem/sector have to be quantified. The robustness of the approach lays in the locking effect of various system and subsystem (sector) mass balances that must be honored simultaneously in the

IOA matrix. This effect also enables quantification (at least to some extent) of uncertain or unknown mass flows in the system considered. Due to the lack of available data and the varying data quality, only static analyses of average annual flows have been performed in this thesis. Further, present input-output analyses have been limited to water and/or pollutant flow studies; previous studies, however, have shown the possible extension and coupling to economic modeling (Gren and Wulff, 2003; Gren and Elofsson, 2004).

In Papers IV and V the IOA is used for quantifying different nitrogen flow scenarios for the water system of the Swedish Norrström drainage basin. The analysis considers the main natural ground- and surface water subsystems, and the engineered-economic (or anthropogenic) subsystems/sectors in the basin that use/ impact these waters. The IOA quantifies average annual flows of total nitrogen, based on independently reported data of nitrogen fluxes, nitrogen concentrations in combi-nation with water flows, and physically reasonable estimates, through and between these systems/sectors. In addition, external flows describe the nitrogen mass transfer between the internal hydro-technosphere (waters and water using/affecting systems and sectors) of the basin and influencing or influenced system/sectors outside the Norrström basin. In Paper III, the IOA is applied to a meta-analysis of reported zinc flow estimates in the Dalälven River basin, in order to investigate the role of diffuse subsurface contributions to zinc loadings in surface waters. Reported zinc flow estimates considered in this study have also been used in the economic analysis of Papers I-II and VI.


3 CATCHMENT-SCALE COST-EFFICIENT ABATEMENT OF WATER

POLLUTION IN MINING AREAS (PAPER I)

Paper I presents the cost-minimization model for determining cost-effective allocation of emission and/or load reduction measures within the Dalälven River basin, in order to achieve pre-specified zinc, copper or cadmium load reductions to selected recipients, including the Dalälven River itself and the local recipients Lake Runn and Lake Gruvsjön. The water pollution by water flowing through mine wastes and abandoned mines is in this paper (and also in Paper II) generally referred to as mine water pollution (ERMITE Consortium, 2004). Different scenarios are used to consider various, practically feasible remediation measures and designs, including soil or water covering of the mine waste deposits, and downstream wetland construction close to or at the compliance boundaries (CBs) that are associated with the different selected water recipients. Further, these scenarios include varying technological efficiency, cost and lifetime of applied remediation/abatement measures and natural metal attenuation. Cost-efficient measures and their allocation, and associated total and marginal annual costs for minimum-cost compliance to different environmental targets (ETs; in terms of metal load reduction) and CB locations (recipients), are then determined. In addition, the paper includes a general presentation and discussion of economic decision rules for choosing the measure allocation for waste site remediation and water pollution abatement within a catchment.

Figure 3.1 shows results for minimum costs for different targeted zinc load reductions to the recipients Lake Runn, Lake Gruvsjön and the Dalälven River, for the example scenario where in addition to soil and water covering of mine waste site, wetland construction is applied as a feasible abatement measure for the Dalälven River. Figure 3.1 also exemplifies the possible cost range that results from various uncertainties

in the technological efficiency, cost and lifetime of applied abatement measures (here considered as different scenarios) for the Dalälven River case.

In general, the results show that the total abatement cost for achieving a certain load reduction may be as high for a local water environment, as for the entire catchment scale, thus implying much higher marginal costs for the former, local compliance. Furthermore, the cost-efficient abatement measure allocation solution for local compliance may be very different from that for the entire catchment scale. We note that the European Union Water Framework Directive may be interpreted to allow for the possibility to use designation of heavily modified waters, which may for instance be water recipients close to main pollutant sources, as pollutant sinks. Thereby, remediation would be more focused on achieving good water quality in downstream, more practically restorable water bodies. The active choice of compliance boundary location is then of outermost importance for the costs and the cost-efficient catchment-scale allocation of measures for mine-related water pollution abatement.

The results also show that discontinuity in the technical feasibility of certain remediation measures (here soil and water covering of mine waste sites) implies that relatively low chosen load reduction targets may not be achievable at relatively low cost. In general, local minima in costs may occur only at certain, discrete load reduction target levels, which must be identified and quantified for achieving economic efficiency. Wetland construction, or other possible abatement measures in the direct vicinity of recipients, may offer an alternative (to the discrete mine waste covering measures) continuous abatement measure possibility, which may be an important (or even the only, as shown in the paper for copper and cadmium load

6


7

abatement for the Dalälven River) part of the cost-efficient solution for abatement measure allocation within a catchment.

0

500 000

1 000 000

1 500 000

2 000 000

2 500 000

3 000 000

0 10 20 30 40 50 60 70 80 90 100

Compliance target [zinc load reduction in %]

Dalälven RiverLake RunnLake Gruvsjön

Tota

l aba

tem

ent c

osts

for c

ompl

ianc

e of

targ

eted

zin

c lo

ad re

duct

ion

[SEK

yr-1

]

0

2

4

6

8

10

12

14

16

0 10 20 30 40 50 60 70 80 90 100

Compliance target [zinc load reduction in %]

0

50

100

150

200

250

300

350

Dal RiverLake RunnLake Gruvsjön

Mar

gina

l aba

tem

ent c

osts

for c

ompl

ianc

e of

targ

eted

zi

nc lo

ad re

duct

ion

to th

e D

aläl

ven

Riv

er a

nd L

ake

Run

n [S

EK y

r-1 k

g-1]

Mar

gina

l aba

tem

ent c

osts

for c

ompl

ianc

e of

targ

eted

zi

nc lo

ad re

duct

ion

to th

e La

ke G

ruvs

jön

[SEK

yr-1

kg-1

]

(a)

(b)

Figure 3.1. Compliance costs for different targeted zinc load reduction at Lake Runn (dotted lines), Lake Gruvsjön (dashed lines) and the Dalälven River (solid lines), in terms of: (a) total annual abatement costs, and (b) associated annual marginal abatement costs. Vertical error bars (solid lines) indicate the resulting cost range for the Dalälven River from different considered scenarios of technological efficiency, cost and lifetime of applied abatement measures (for more detailed results see Paper I).


4 POLLUTION SOURCE UNCERTAINTY EFFECTS ON COST-EFFICIENT

ABATEMENT OF WATER POLLUTION IN MINING AREAS (PAPER II)

The main aim of this paper was to assess important general effects and implications of pollutant source uncertainty on cost-effective abatement solutions, with the Dalälven River basin as the specific application example. Abatement of mine-related water pollution (in the paper referred to as mine water pollution (ERMITE Consortium 2004)) has so far commonly focused on discrete, known mine waste sites (e.g. Swedish Environmental Protection Agency (SEPA), 1998; Salmon and Destouni, 2001; European Commission, 2003) rather than on a catchment-scale water quality perspective. Such a focus does not consider possible water impacts from all different sources of polluted mine water within a catchment. Recent studies, however, indicate the possibility of considerable contribution to mine water pollution from other sources than the known mine waste sites, for instance from abandoned mine voids, which Younger et al. (1997), Wood et al. (1999) and Younger (2003) have shown to constitute around (or more than) 90% of total mine water pollution loads in some cases. Furthermore, other recent studies of mine water transport and attenuation downstream of mine waste sites show large physical and biogeochemical pollutant spreading in natural, heterogeneous ground-water systems (Berglund et al., 2003; Malmström et al., 2004, 2007). This leads to additional, diffuse and long-term sources of previously immobilized mine water pollutants that are later slowly released and transported from their sorption/precipitation sites and more or less immobile water zones through the aquifers to downstream surface waters.

This paper extends the model analysis of cost-effective abatement of mine-related water pollution in the Dalälven River basin presented in Paper I. Specific objectives were to investigate the effects of uncertainty in: (a) metal source distribution between known mine waste sites and other possible unknown

point and diffuse sources of metal loadings to surface waters; and (b) total metal discharges from all sources, which in the particular Dalälven River basin have been quantified to range within about ±30% of the average discharge estimate (Lundgren and Hartlén, 1990).

The study considers the same reported natural metal attenuation for the sub-catchments and abatement measures as the study of Paper I. Abatement measures thus include mine waste remediation measures that are commonly used in Sweden (Elander et al., 1998) and practically feasible in the sub-catchments, and wetland construction close to the outlet of the sub-catchments to the Dalälven River as potential additional/ alternative downstream abatement measures. The efficiency of constructed wetlands was calculated to decrease linearly as pollutant loading decreases by abatement at upstream mine waste sites. Effects of source distribution and discharge uncertainty were quantified such that either the total discharge (sum of diffuse and point sources) or the relation of diffuse and point sources to the total discharge remained the same as the diffuse and point sources changed in the different investigation scenarios.

Effects of source distribution and discharge uncertainty on total annual abate-ment costs for compliance to targeted zinc load reductions from the two sub-catchments Falun and Garpenberg to the Dalälven River are illustrated in Figure 4.1. The figure shows that, for the specific measure efficiencies and costs considered in this study, total abatement costs are generally smaller if downstream abatement measures are feasible than if they are not. Further, total abatement costs for feasible downstream measures appear relatively insensitive to zinc source distribution uncertainties (Figure 4.1a) but significantly more sensitive to discharge

8


0

20 000

40 000

60 000

80 000

100 000

120 000

140 000

160 000

10 30 50 70 90 10 30 50 70 90 10 30 50 70 90

Downstream measures notpossibleDownstream measurespossible

(a) Complia nce target

[Load reduction in % ] Complia nce target

[Load reduction in % ]Complia nce target

[Load reduction in % ]

Base case scenario Increase of DSF to 90 percent

Increase of DSF to 50 percent

Downstream measures not feasible Downstream measures feasible

Compliance not poss ible without downstream

measures

0

20 000

40 000

60 000

80 000

100 000

120 000

140 000

160 000

10 30 50 70 90 10 30 50 70 90 10 30 50 70 90

Downstream measures notpossibleDownstream measurespossible

(b) Complia nce target

[Load reduction in % ] Complia nce target

[Load reduction in % ]Complia nce target

[Load reduction in % ]

Base case scenario Increase of TLA by 30 percent

Decrease of TLA by 30 percent

Downstream measures not feasible Downstream measures feasible

Tota

l com

plia

nce

cost

s [U

SD y

r-1]

Tota

l com

plia

nce

cost

s [U

SD y

r-1]

Figure 4.1. Total annual abatement costs for compliance to targeted zinc load reductions from the two sub-catchments Falun and Garpenberg to the Dalälven River for (a) scenarios with different diffuse source fractions (DSF); and (b) for changed total zinc leakage amount (TLA) by ±30% (Figure modified from Paper II).

uncertainties (Figure 4.1b). If downstream measures are not feasible, total abatement costs for a given load reduction target are instead quite insensitive to discharge uncertainties (Figure 4.1b), but very sensitive to distribution uncertainties (Figure 4.1a).

The results shown in Figure 4.1 imply that, with an increasing fraction of diffuse sources, the source measures at mine waste sites will abate a smaller fraction of the total zinc leakage amount, thus implying higher marginal costs per unit abated load fraction and thereby higher total costs. In addition, compliance may not even be achievable without downstream measures for relatively high zinc load reduction targets (Figure 4.1a).

In general these results imply that, if downstream measures are not practically feasible, knowledge of the correct pollution source distribution between point and diffuse sources becomes critical for both cost-effective measure allocation and associated abatement costs. In contrast, if downstream abatement measures are feasible, uncertainty in the magnitude of total pollutant discharge will affect neither the cost-effective measure allocation nor the associated costs. Thus, the paper shows that whether downstream abatement measures, such as constructed wetlands, are practically feasible or not is a critical factor for uncertainty effects on the efficient allocation of measures for mine-related water pollution abatement in catchments.

9


5 SUBSURFACE WATER SYSTEM CONTRIBUTIONS TO SURFACE WATER

POLLUTION IN MINING AREAS (PAPER III)

Different studies have proposed that abandoned subsurface mine voids may yield downstream water pollution of similar or greater magnitude than mine wastes over long time periods (Younger et al., 1997; Wood et al., 1999; Younger 2003). Other studies of subsurface pollutant transport downstream of mine wastes indicate that large pollutant spreading and mass transfer, with an associated delay in pollutant delivery may occur in the groundwater system due to both geochemical reactions and flow and transport variability (Berglund et al., 2003; Malmström et al., 2004, 2007; ERMITE Consortium, 2004). These spreading and delay effects imply a subsurface pollutant accumulation and subsequent delayed release, which may constitute a diffuse, long-term source of pollution for downstream ground- and surface water. This paper investigates the possible subsurface water system role for zinc mass loading into the Dalälven River, which has been unaccounted for in previous

studies (including Paper I and II; Hartlén and Lundgren, 1990; Fällman and Qvarfort, 1990; Tröjbom and Lindeström, 2005).

The study revisits and reinterprets available zinc load data for the Dalälven River, by considering the previously unaccounted for, yet still realistic possibility of significant diffuse subsurface zinc loading into the river using an input-output flow analysis (IOA) methodology. The paper focuses on three surface water sections in the main mining areas of the basin, including the Dalälven River and the local recipients Lake Runn and Lake Gruvsjön (Figure 5.1).

The results of the study show the possibility of significant contributions from the subsurface water system to the surface water zinc loads (Table 5.1), which may have important implications for the abatement effectiveness and cost-efficiency of water pollution discussed in Papers I and

Urban areasObservation sites

Garpenberg

Falun

Lake Runn

Lake Gruvsjön

Dalälven

Figure 5.1. The case-study area located in the Swedish Dalälven River basin with the two mining regions Falun and Garpenberg within the sub-catchments of local recipients Lake Runn and Lake Gruvsjön, respectively, and the approximate location of main water and zinc flow observation sites: a-a (Torsång/Borlänge); b-b (Långhag/Fäggeby); c-c (Näs bruk); d-d (Runn); e-e (Garpenberg); f-f (Forsån); * (Gysinge; only partly used in this study) for exact locations, see Lindeström (1991) and SMHI (1995)).

10


11

Table 5.1. Quantifications of average annual zinc inflow and outflow ranges (for the different interpretation scenarios) and resulting relative zinc attenuation for each river section considered. These quantifications either neglect (underlined) or account for (italic) any possible subsurface water contributions to the river zinc loads.

II. These contributions are more consistent with available surface water data, in particular as they yield more consistent relative zinc attenuation magnitudes in different river sections (Table 5.1) than the total neglect of subsurface pollutant loading to the Dalälven River. This has so far been common practice in the Dalälven River basin and more generally in Swedish environmental monitoring and management of water systems. In addition to these site-specific findings, the input-output flow system analysis methodology used in Paper III may also be generally useful for identifying and quantifying significant unmonitored and uncertain subsurface water system contri-butions to surface water pollutant loading.

The study further notes that the Swedish groundwater quality monitoring is generally

so sparse that each metal concentration measurement in the national groundwater monitoring grid is supposed to be representative for conditions in an average surrounding area of about 6700 km2. Moreover, there is no particular clustering of monitoring wells existing for instance in important Swedish mine sites. This lack of subsurface water quality and flow data makes it impossible to directly prove or disprove any chosen quantification of subsurface water system contributions to the Dalälven River as presented in the paper. Nevertheless, the present quantification summarized in Table 5.1 shows at least that significant diffuse subsurface zinc load contributions are possible and may be more consistent with available surface water data than the so far common neglect of such contributions.

Surface water section

a-a/e-e/b-b


f-f/g-g


b-b/g-g/c-c

186-378 226-418

7-8 7-8

156-267 Input zinc flow to river section [t yr-1]

196-307 147-260 187-300

4-6 4-6

153-260 Output zinc flow to river section [t yr-1]

153-260 Resulting relative zinc 21-31

21-31 25-42 25-42

0-3 attenuation in river section [%] 15-22


6 QUANTIFICATION OF COUPLED NATURAL AND CROSS-SECTORAL

POLLUTANT FLOWS IN CATCHMENTS (PAPERS IV AND V)

In recent years, one important aim of hydrological science application and research has been the development of concepts and tools for integrated water resource management (IWRM) of hydrological drainage/river basins as main water management units (e.g., the Water Framework Directive (WFD; European Commission, 2000). Such basins, however, constitute highly heterogeneous systems of coupled natural and engineered water sub-systems. The physical extension and/or influence zones of these systems may cross several political national and international boundaries. The flows of water and pollutants between such coupled water sub-systems and across political boundaries need to be properly understood, quantified and communicated to stakeholders, in order to appropriately guide monitoring, management and regulation decisions for efficient and sustainable water resource use, development, protection and remediation (Kavanaugh et al., 2003; ERMITE Consortium, 2004; Hannerz et al., 2005).

Especially the transport in and attenuation of nutrients in drainage basins determine nutrient loading to inland and coastal waters and are therefore important research topics in many scientific studies (e.g., Arheimer and Brandt, 1998, 2000; Alexander et al., 2000, 2002; Lindgren and Destouni, 2004; Darracq and Destouni, 2005; Darracq et al., 2005; Destouni et al., 2006; Lindgren et al., 2007). Such nutrient budget modeling commonly focuses on forward description of nutrient transport and transformation processes in natural water systems, subject to various inputs. To address the problem of coupled feedback mechanisms between natural and cross-sectoral anthropogenic flows of water, nutrients and pollutants in drainage basins, Paper IV suggests and demonstrates the possible application of the relatively simple, compact and powerful input–output flow

analysis (IOA) to basin-scale water quality management. Paper V investigates further some of the uncertainties associated with the subsurface nitrogen accumulation–depletion results of Paper IV, by use of scenario sensitivity analysis.

Results of site-specific application of IOA to water and nitrogen flows within and from the Norrström drainage basin indicate considerable nitrogen load contributions to surface and coastal waters from slow groundwater flow paths and legacies of accumulated nitrogen in subsurface and immobile water pools. The results are robust even for extreme assumptions of nitrogen discharges and transport pathways from agriculture to surface and groundwater in the basin. This implies that effective nitrogen load abatement cannot focus only on active sources but must also include downstream measures, which can capture and abate nitrogen/pollutant loading from different types of known and yet unknown point and diffuse sources within the relevant catchments.

Papers IV and V further show that IOA enables relatively simple, compact and fruitful quantification of: i) integral catchment behavior of water flow and nitrogen load responses to changes in the various natural and engineered water subsystems; ii) water flow and nitrogen load interactions between the various catchment sub-systems; and iii) external water flow and nitrogen load interactions of the catchment and its subsystems. The IOA quantification is highly transparent and readily communicated to stakeholders, and identifies clearly important information and data gaps that require improved monitoring and more detailed dynamic water system modeling. Furthermore, the IOA tool does not only quantify human impacts on water environments, but also environmental target impacts on the engineered-economic systems

12


13

and sectors that use/affect water for meeting various human needs. Such quantification of catchment-scale cause-effect relations and feedback mechanisms between natural water and engineered-economic systems and

sectors is necessary for efficient and sustainable IWRM, but commonly not handled by traditional hydrologic-biogeochemical flow and pollutant transport modeling.


7 UNCERTAINTY-ACCOUNTING ENVIRONMENTAL POLICY AND

MANAGEMENT OF WATER SYSTEMS (PAPER VI)

Commonly, environmental policies for water quality and ecosystem management do not require explicit stochastic accounts of uncertainty and risk associated with the quantification and prediction of waterborne pollutant loads and abatement effects. However, many studies (including Papers I-IV) have pointed out inherent uncertainties of complex environmental systems that make it difficult or impossible to reliably predict minimum costs and optimal allocation of pollution abatement measures (Beavis and Walker, 1983; McSweeny and Shortle, 1990; HELCOM, 1993; Shortle et al., 1998; Costanza 2000; Gren et al., 2000a, 2002; Andersson and Destouni, 2001; Kavanaugh et al., 2003; Popper et al., 2005). Therefore, Paper VI formulates and investigates a possible environmental policy (in the paper referred to as stochastic uncertainty-accounting policy) that does require an explicit stochastic uncertainty account. The environmental and economic resource allocation performances of such an uncertainty-accounting environmental policy are then compared with the corresponding performances of deterministic, risk-prone or risk-averse environmental policies under a range of different hypothetical, yet still possible, scenarios. The case of zinc loadings to the Dalälven River from mining activities is used for site-specific application and quantification in this investigation.

The study considers pollutant loads to a recipient as random variables (expressed and quantified in terms of their statistics, such as expected value and variance) in order to account for the randomness and uncertainty of stochastic pollutant loads in the cost minimization model. Environmental policy

may (stochastic uncertainty-accounting policy) or may not (deterministic policies) then consider the variability and randomness of pollutant loads to the compliance boundary. The general consequence of that variability and randomness is that the policy and management assumed or the chosen pollutant load variabilities and the targeted load reductions may differ from their actual real values and necessary pollutant load reductions for achieving desired environ-mental and ecological effects, respectively. Therefore, the paper investigates the abatement and economic performances of the two deterministic and the alternative stochastic uncertainty-accounting policies for a range of different possible scenarios of actual, real pollutant load variability and environmentally-ecologically necessary pollu-tant load reduction.

The comparison indicates that a stochastic uncertainty-accounting policy may perform better than deterministic policies over a relatively wide range of different scenarios (Figure 7.1). This is of course only true if there is indeed some significant uncertainty about pollutant loads; otherwise, there is no meaning or reason at all in an uncertainty-accounting policy. If pollutant loads are uncertain, as is often the case, reported literature values of pollutant load statistics appear to be useful for capturing the essential uncertainty in pollutant load quantifications and predictions, even in the absence of reliable site-specific data. Explicit uncertainty accounting therefore appears to be practically feasible for identifying environmental policies for water systems with good environmental and economical performance over a range of different scenarios.

14


15

Policy I - Policy II - Policy III -Deterministic risk-prone Deterministic risk-averse Stochastic uncertainty-accounting

0,9 0,90,8 0,80,7 0,70,6 0,60,5 0,50,4 0,40,3 0,30,2 0,20,1 0,1

0,9 0,90,8 0,80,7 0,70,6 0,60,5 0,50,4 0,40,3 0,30,2 0,20,1 0,1

0,00 0,25 0,50 0,75 1,00 1,25 1,50 0,00 0,25 0,50 0,75 1,00 1,25 1,50 0,00 0,25 0,50 0,75 1,00 1,25 1,50CV[L] CV[L] CV[L]

(a) Abatement performance (b) Economic performanceactual probability of abatement success relative difference between allocated and actually required costs

- < 0.50 - very wasteful (≥ 0.8; allocated resources are much higher than required resources) - 0.50 to 0.80 - wasteful (0.4 to 0.8; allocated resources are higher than required resources) - 0.80 to 0.95 - good, slightly wasteful (0.1 to 0.4; all. res. equal or are slighly higher than req. res.) - 0.95 to 0.99 - optimal (-0.1 to 0.1; allocated resources equal required resources) - ≥ 0.99 - good, slightly insufficient (-0.4 to -0.1; all. res. equal or are slighly lower than req. res.)

- insufficient (-0.8 to -0.4; allocated resources are lower than required resources) - very insufficient (≤ -0.8; allocated resources are much lower than required resources)

1-L m

ax/L

'

1-L m

ax/L

'1-

L max

/L'

1-L m

ax/L

'

(a)

(b)

Figure 7.1. (a) Abatement (in terms of the probability of meeting the environmentally necessary reduction in zinc load) and (b) economic performance of different policies in reducing zinc loads for a desired probability of success for different possible realizations of the environmentally necessary zinc load reduction and of the zinc load variability. The economic performance is measured as the relative difference between policy-specific minimum costs of abatement and the minimum abatement costs for meeting the environmentally necessary zinc load reduction and the zinc load variability of a specific scenario. Policy-specific reductions in zinc load and load variability are marked with a cross in the relevant parameter combination cell for each policy (Figure modified from Paper VI).


8 GENERAL DISCUSSION

8.1 Pollutant Emission and Pollutant Transport-Attenuation Uncertainties

In order to quantify effects of pollutant source uncertainty, the studies of Papers I and II investigate different possible scenarios for metal leakage, including the so-called base case scenario and various alternative scenarios. In Paper II, the different scenarios differ in terms of their diffuse source fractions and total metal leakage amounts. For these different scenarios, cost-effective abatement solutions are investigated for the two cases that downstream measures (i.e. wetland construction) are or are not practically feasible. For metal load reductions to the local recipients Lake Runn and Lake Gruvsjön, only one scenario was investigated considering water and soil covering of major mine waste sites.

The unpredictable and uncertain variation of pre- and post-abatement pollutant loads, pollutant emissions, emission and load reductions, and pollutant attenuation have been discussed in Paper VI. Because the aim was not to resolve the various variability, randomness and uncertainty contributions of different parameters to the total pollutant load statistics, but to account for the possible latter statistics in the stochastic constraint of achieving maximum pollutant loads at a compliance boundary, only the resulting pollutant loads are considered as random variables. Further, for simplicity and clarity, only a normal distribution expression was considered in the study. Similar cost-minimization expressions, however, have for instance also been formulated for log-normally distributed pollutant loads (Gren et al., 2002).

8.2 Subsurface Water System Contributions to Pollution Loadings

Because of the highly heterogeneous and complex nature of the subsurface flow

system, possible diffuse subsurface contributions to surface water pollution loading are difficult to quantify even if subsurface monitoring exists. Therefore, the different studies included in this thesis had also to deal with a general lack of subsurface water flow and water quality data in both case study areas. Also, subsurface contributions to surface water pollutant loading had to be quantified using simplifications and physically reasonable estimates.

With regard to a more general lack of groundwater quality data, we note for instance that the Swedish groundwater quality monitoring is very sparse (see Paper III). A possible explanation for this subsurface water system neglect may be a traditional focus on point sources directly into surface waters which dominated when environmental awareness arose initially. A general lack of subsurface water quality and flow data makes it impossible to directly prove or disprove any chosen quantification of subsurface water system contributions to surface water systems.

The analysis of pollutant flows and especially the quantification of subsurface pollutant flows in the Norrström drainage basin (Papers IV and V) and the Dalälven River basin (Paper III) aim therefore at providing a possible range of implications for different water pollutant loads. In the case of the Norrström drainage basin, this is done by considering two different extreme assumptions of nitrogen discharges and transport pathways from agriculture to the surface and groundwater systems of the basin. In the Dalälven River basin, consistent relative zinc attenuation values in all river sections have been the basis for defining diffuse subsurface zinc contributions to the surface water system. Different assumptions of discharge and transport of pollutants to and through different water systems of a drainage basin may of course impact internal flows and external flows to and from the

16


17

basin. However, flow data that are completely or partly based on independent reports have been kept unchanged in order to focus the flow analysis on the effects of different possible subjective assumptions of pollutant discharge distribution to and from downstream surface and subsurface waters.

8.3 Economic Analysis of Pollution Abatement

Besides the cost minimization model used in this thesis, one may also choose pollution abatement solutions that yield or maximize net benefits (ERMITE Consortium, 2004; Paper I). Consideration of net benefits, however, requires monetary valuation of environmental benefits with water quality improvement. In view of the major difficulties and uncertainties of such valuation, the cost minimization approach has here been considered as more fruitful for the present research stage of effective mine-related water pollution abatement in catchments.

The cost-effective abatement solutions presented in this thesis consider only soil and water covering of mine wastes and constructed downstream wetlands as possible pollution abatement measures. Other abatement measures may as well be practically feasible and economically viable. Practical applications must consider all possible abatement measures, which will influence site-specific results and cost-effective solutions. This thesis does not discuss whether and which solutions are practically feasible and acceptable by regulators and stakeholders in the Dalälven River basin. It is also outside the scope of the thesis to compare technological perfor-mances, advantages and drawbacks of different remediation and abatement measures. For such comparison and discussion, the reader is referred to, e.g. Elander et al. (1998) and ERMITE Consortium (2004). Furthermore, the long time scales that are involved in mine-related water pollution will affect cost-effective pollutant abatement solutions. Metal turnover times in mine wastes may be

hundreds or thousands of years (SEPA, 1986), with remediation measures such as water and soil covering only prolonging these times.

Further, groundwater versus surface water contributions have not been explicitly addressed in the economic analysis of pollution abatement. The used reported data on metal concentrations and loads in surface water implicitly include also contributions from groundwater. Wetlands located at a sub-catchment outlet will take care of all groundwater contributions to streams within the sub-catchment, but not of direct discharges of groundwater to the recipient. However, considering the studies of this thesis, direct groundwater discharges to the recipients are neither included in the surface water data on pre-abatement metal loading that constructed wetlands must abate. Additional groundwater catchments will have to be analyzed in similar ways as done for surface water sub-catchments, in order to find cost-effective abatement solutions that also include direct groundwater discharges to a recipient.

Finally, the economic analysis discussed here does not answer the question on how total abatement costs should optimally be divided among different stakeholders (liable parties, regulators, tax payers). This is an important clarification to make, because a stakeholder-specific benefit-maximization or cost-minimization analysis, for instance, may tacitly or explicitly exclude some of the cost components and effects that may be important to include in a total identification and quantification of efficient abatement of mine-related water pollution for different possible target formulations on a catchment scale.

8.4 Accounting for Uncertainty in Environmental Policies

Paper VI suggests that a stochastic uncertainty-accounting policy may perform better than deterministic policies over a range of different scenarios even in the absence of reliable site-specific data. The pre-requisite for good performance of a stochastic


uncertainty-accounting policy is of course that there is some significant uncertainty about pollutant loads. If uncertainty prevails, it may also be possible to reduce it, and thereby facilitate better performance of deterministic policies. This may done by gathering additional information at some additional cost, so that any policy-specific assumptions of pollutant load variability and necessary pollutant load reduction become closer estimates of reality (Dakins et al., 1996; Dakins 1999). The methodology and types of results presented in Paper VI may then indicate whether an additional reduction in uncertainty is economically justified. The additional cost of such a decrease in uncertainty would only be economically justified if it was lower than the cost reduction (and possible other additional

benefits) associated with choosing a deterministic policy instead of the stochastic uncertainty-accounting policy.

The case study applications of Papers I, II and VI show how it may be possible to identify cost-effective abatement solutions for mine-related water pollution in a catchment on basis of given water quality targets. Besides defining pollutant load reductions, the necessary pre-specified water quality target formulation for identifying a minimal-cost solution also requires a clear specification of compliance boundaries/ limits/scales in space and in time. In general, different possible target formulations are possible and need to be analyzed, in order to identify one or more relevant targets, which are practically possible to achieve at acceptable cost.

18


19

9 CONCLUSIONS

This thesis has investigated the role of various uncertainty aspects related to the pollutant emission, pollutant transport and attenuation, and efficient pollutant abatement in catchments for environmental manage-ment of water systems. Specifically, it has investigated uncertainties in the identification and/or quantification of known and unknown pollutant sources from observed pollutant measurements in downstream surface waters, and the uncertainty implications for cost-efficient water pollution abatement. Further, the particular role of diffuse subsurface water system contributions to downstream surface water pollution and associated source measure implications have been studied in two examples of the most significant water pollution problems in Sweden, i.e. water pollution caused by present and past mining activities in the Swedish Dalälven River basin, and by anthropogenic nitrogen loadings in the Norrström drainage basin.

The following general conclusions summarize main findings of the thesis:

Water pollution assessment and efficient pollution abatement require explicit consideration of various uncertainty aspects. These include uncertainties about pollution sources, diffuse contributions of the subsurface water system to downstream pollutant observations in surface waters, and abatement effectiveness and economic efficiency of different possible measures to reduce water pollution. In many cases, in addition to source abatement, efficient water pollution abatement may have to include downstream load reduction measures in order to reduce not only

expected, but also uncertainties around expected pollutant loads.

Effective and efficient environmental management of water systems must consider the entire catchments of these systems, rather than focusing only on discrete pollutant sources. Relatively simple flow analysis tools, such as the input-output flow approach, may through their mass balance constraints be used to obtain at least some important information about unmonitored water and pollutant flows, subsurface accumulation-depletion and associated uncertainties.

Explicit account of prediction uncer-tainties may make environmental policy and management of water systems more robust under different possible present and future conditions, even in the absence of reliable site-specific data to quantify such uncertainties.

The different investigations presented in this thesis all include various limitations discussed in the previous section, which may constitute important future research directions. Generalizations of the mostly case-study specific results in this thesis have to be handled with care, considering all the different model simplifications, parameter assumptions, data gaps, etc., that have been part of the different studies. In spite of these constraints, however, the thesis does identify some important uncertainty aspects, which should be worthy of consideration in environmental policy and management of water systems.


10 REFERENCES

Alexander, R.B., Smith, R.A. & Schwarz, G. E., (2000) Effect of stream channel size on the delivery of nitrogen to the Gulf of Mexico. Nature 403, pp 758–761.

Alexander, R.B., Elliott, A.H., Shankar, U. & McBride, G.B., (2002) Estimating the sources and transport of nutrients in the Waikato River Basin, New Zealand. Water Resour. Res. 38, pp 1268–1290.

Andersson, C. & Destouni, G., (2001) Groundwater transport in environmental risk and cost analysis: role of random spatial variability and sorption kinetics. Ground Water 39, pp 35–48.

Arheimer, B. & Brandt, M., (1998) Modelling nitrogen transport and retention in the catchments of Southern Sweden. Ambio 27, pp 471–480.

Arheimer, B. & Brandt, M., (2000) Watershed modelling of nonpoint losses from arable land to the Swedish coast in 1985-1994. Ecol. Eng. 14, pp 389–404.

Augustinovics, M., (1970) Methods of international and intertemporal com-parison of structure. In Contributions to Input-Output Analysis. Carter, A. P., Brody, A., Eds., North-Holland, Amster-dam, pp 249–69.

Beavis, B. & Walker, M., (1983) Achieving environmental standards with stochastic discharges. J. Environ. Econ. Manag. 10, pp 103–111.

Berglund, S., Malmström, M.E., Jarsjö, J. & Destouni, G., (2003) Effects of spatially variable flow on the attenuation of acid mine drainage in groundwater. Sudbury 2003 Mining and the Environment, May 25-28 2003, Sudbury, Canada. Paper No. 94 (CD-ROM).

Brooke, A., Kendrick, D. & Meeraus, A., (1992) GAMS. A User’s Guide. The Scientific Press, San Francisco. 276 p.

Conley, D.J., (2000) Biogeochemical nutrient cycles and nutrient management strategies. Hydrobiologia 410, pp 87–96.

Cosgrove, W.J. & Rijsberman, F.R., (2000) World water vision, Making water everybody’s business. World Water Council, Earthscan Publications: London.

Costanza, R., (2000) Visions of alternative (unpredictable) futures and their use in policy analysis. Conserv. Ecol. 4 (1), 5, http://www.consecol.org/vol4/iss1/art.

Dakins, M.E., (1999) The value of the value of information. Hum. Ecol. Risk Assess. 5 (2), pp 281–289.

Dakins, M.E., Toll, J.E., Small, M.J. & Brand, K.P., (1996) Risk-based environmental remediation: Bayesian MonteCarlo analysis and the expected value of sample information. Risk Anal. 16 (1), pp 67–79.

Darracq, A. & Destouni, G., (2005) In-stream nitrogen attenuation: model artifacts and management implications for coastal nitrogen impacts. Environ. Sci. Technol. 39, pp 3716–3722.

Darracq, A., Greffe, F., Hannerz, F., Destouni, G. & Cvetkovic, V., (2005) Nutrient transport scenarios in a changing Stockholm and Mälaren valley region. Water Sci. Technol. 51, pp 31–38.

Destouni G., Lindgren G. & Gren I.M., (2006) Effects of inland nitrogen transport and attenuation modeling on coastal nitrogen load abatement, Environ. Sci. Technol. 40, pp 6208–6214, doi: 10.1021/es060025j.

Elander, P., Lindvall, M. & Håkansson, K., (1998) MiMi - prevention and control of pollution from mining waste products. State-of-the-art report, MiMi 1998:2, 29 p.

ERMITE Consortium, (2004) Mining Impacts on the Freshwater Environ-ment: Technical and Managerial Guidelines for Catchment-Focused Remediation. Younger, P. L., Wolkers-dorfer, C., Eds., Mine Water Environ. 23, Suppl. Issue 1, 80 p.

20


21

European Commission, (2000) Directive 2000/60/EC of the European Parliament and of the Council establishing a framework for the Community action in the field of water policy. Official Journal of the European Communities, L 327/1.

European Commission, (2003) Proposal for a Directive of the European Parliament of the Council on the management of waste from the extractive industries. COM(2003) 319 final, Luxembourg, 2003/0107 (COD).

Fleischer, S., Leonardsson, L. & Stibe, L., (1991) Restoration of wetlands as a means of reducing nitrogen transport to coastal waters. Ambio 20, pp 271–272.

Gleick, P.H., (2003) Water in crisis. Oxford University Press, 473 p.

GovernmentBill, (2000) Svenska miljömål -delmål och åtgärdsstrategier (Swedish environmental objectives - intermediate goals and action plans). (in Swedish), 2000/01:130, Swedish Government: Stockholm.

Gren, I.M., (1993) Alternative nitrogen reduction policies in the Mälar region Sweden. Ecol. Econ. 7, pp 159–172.

Gren, I.M., (1995) Costs and benefits of restoring wetlands: two Swedish case studies. Ecol. Eng. 4, pp 153–162.

Gren, I.M., Söderqvist, T. & Wulff, F., (1997) Nutrient reductions to the Baltic Sea: economics and ecology. J. Environ. Manage. 51, pp 123–143.

Gren I.M., Sharin, H. & Destouni, G., (2000a) Cost effective management of stochastic coastal water pollution. Environ. Model. Assess. 5, pp 193–203.

Gren, I.M., Turner, K. & Wulff, F., (2000b) Managing a sea: ecological economics of the Baltic Sea. Earthscan, London, 176 p.

Gren, I.M., Destouni, G. & Tempone, R., (2002) Cost effective policies for alternative distributions of stochastic water pollution. J. Environ. Manage. 66, pp 145–157.

Gren, I. M. & Elofsson, K., (2004) Cost of reducing nitrogen loads to Baltic Proper. Vatten 60, pp 85–94.

Gren, I. M. & Wulff, F., (2003) Cost-effective nutrient reductions to coupled heterogeneous marine water basins: An

application to the Baltic Sea. Regional Environ. Change [online], doi: 10.1007/s10113-003-0063-6.

Hannerz, F., Destouni, G., Cvetkovic, V., Frostell, B. & Hultman, B., (2005) A flowchart for sustainable integrated water management following the EU Water Framework Directive. Eur. Water Manage. Online [online], 04, http://www.ewaonline .de/journal/online.htm.

Hartlén, J. & Lundgren, T., (1990) Gruvavfall i Dalälvens Avrinningsområde - Metallutsläpp och åtgärdsmöjligheter. (in Swedish), Swedish Geotechnical Institute (SGI), Report no. 39, Linköping, 127 p.

HELCOM, (1993). The Baltic Sea Joint Comprehensive Environmental Programme. Baltic Sea Environmental Proceedings No. 48, Helsinki, 110 p.

HELCOM, (2004) 30 Years of Protecting the Baltic Sea-HELCOM 1974- 2004. Helsinki Commission, Baltic Marine Environment Protection Commission, Helsinki, 28 p.

Herlihy, A.T., Kaufmann, P.R., Mitch, M.E. & Brown, D.D., (1990) Regional estimates of acid mine drainage impact on streams in the Mid-Atlantic and Southeastern United States. Water, Air, Soil Pollut. 50, pp 91–107.

Hinrichsen, U. & Wulff, F., (1998) Biogeochemical and physical controls of nitrogen fluxes in a highly dynamic marine ecosystem - model and network flow analysis of the Baltic Sea. Ecol. Model.109, pp 165–191.

Howarth, R.W. & Marino, R., (2006) Nitrogen as the limiting nutrient for eutrophication in coastal marine eco-systems: Evolving views over three decades, Limnol. Oceanogr. 51, pp 364–376.

Iribar, V., Izco, F., Tames, P., Antigüedad, L. & da Silva, A., (2000) Water contamination and remedial measures at the Troya abandoned Pb-Zn mine (The Basque Country, northern Spain). Environ. Geol. 39, pp 800–806.

Kavanaugh,M.C., Rao, P. S. C. K., Abriola, L., Cherry, J., Destouni, G., Falta, R., Major, D., Mercer, J., Newell, C., Sale, T., Shoemaker, S., Siegrist, R., Teutsch, G.


& Udell, K., (2003) The DNAPL remediation challenge: is there a case for source depletion? Expert panel report

EPA/600/R-03/143, Environmental Protection Agency (EPA): U.S.A., available online at http://www.epa.gov/ ada/download/reports/600R03143/600R03143.pdf.

Leontief, W., (1936) Quantitative input–output relations in the economic systems of the United States. Rev. Econ. Stat. 18, pp 105–25.

Lindgren, G. & Destouni, G., (2004) Nitrogen loss rates in streams: scale-dependence and up-scaling methodology. Geophys. Res. Lett. 31, L13501, doi: 10.1029/2004GL019996.

Lindgren G., Destouni G. & Darracq A., (2007) The inland subsurface water system role for coastal nitrogen load dynamics and abatement responses, Environ. Sci. Technol. (in press).

Lindeström, L., (1991) Miljöbedömning av metallsituationen i Dalälven och Bottenhavet -Konsekvenser av att åtgärda gruvavfall. (in Swedish), Rapport för Dalälvs-delegationen, Miljöforskargruppen Fryk-sta, 145 p.

Lindeström, L., (1999) Dalälvens Vattenvårdsförening DVVF - Metaller i Dalälven. Miljöforskargruppen. (in Swedish), Rapport för Dalälvens vattenvårdsförening, Rapport 1999:16, Miljöforskargruppen Fryksta, 79 p.

Mälarens Vattenvårdsförbund (Lake Mälaren Society), (2004) Mälaren - en sjö för miljoner. (in Swedish), Västerås, 20 p.

Malmström, M.E., Destouni, G. & Martinet, P., (2004) Modeling expected solute concentration in randomly hetero-geneous flow systems with multi-component reactions. Environ. Sci. Technol. 38(9), pp 2673–2679.

Malmström, M., Berglund, S. & Jarsjö, J., (2007) Combined effects of spatially variable flow and mineralogy on the attenuation of acid mine drainage in groundwater. Appl. Geochem. (accepted).

McSweeny, W.T. & Shortle, J.S., (1990) Probabilistic cost effectiveness in

agricultural nonpoint pollution control. South. J. Agr. Econ. 22, pp 95–104.

Nixon, S.W., (1995) Coastal marine eutrophication - A definition, social causes, and future concerns. Ophelia 41, pp 199–219.

Popper, S.W., Lempert, R.J. & Bankes, S.C., (2005) Shaping the Future. Scientific American, April 2005, pp 48–53.

Salmon, S. & Destouni, G., (2001) National Case Studies-3. Sweden. ERMITE Report: D1, the European Commission Fifth Framework Programme, Energy, Environment and Sustainable Develop-ment, Contract No EVK1-CT-2000-0078, University of Oviedo, 26 p.

Schrøder, H., (1995) Input management of nitrogen in agriculture. Ecol. Econ. 13, pp 125–140.

SEPA, (1986) Gruvavfall. Naturvårdsverket informerar. (in Swedish), Swedish Environmental Protection Agency, Stockholm, 11 p.

SEPA, (1998) Gruvavfall-Miljöeffekter och behov av åtgärder. (in Swedish), Rapport 4948. Swedish Environmental Protection Agency, Stockholm, 153 p.

Shortle, J.S., Horan, R. & Abler, D., (1998) Research issues in nonpoint pollution control. Environ. Resource Econ. 11, pp 571–585.

SMHI, (1995) Svenskt vattenarkiv. Vattenföring i Sverige, D. 2, Vattendrag till Bottenhavet. (in Swedish), Swedish Meteorological and Hydrological Institute, vattenförings-serier tom 1990, Norrköping.

Svensson, K., (1988) Helhetssyn på Dalälven. (in Swedish), In Föreningen för samhällsplanering, 1947- (42:6), pp 277–283.

Tröjbom, M. & Lindeström, L., (2005) Ämnestransporter i Dalälven 1990-2003 –Mängder, ursprung & trender. (in Swedish), Dalälvens Vattenvårdsförening (DVVF), Report no. 2004:22, 34 p.

Turner, R.E. & Rabalais, N.N., (1994) Coastal eutrophication near the Mississippi river delta. Nature 368, pp 619–621.

22


23

World Water Assessment Programme (WWAP), (2003) The UN World Water Development Report: Water for People, Water for Life. http://www.unesco.org/water/ wwap/wwdr/index.shtml.

Wood, S., Younger, P.L. & Robins, N., (1999) Long-term changes in the quality of polluted minewater discharges from abandoned underground coal workings in Scotland. Q. J. Eng. Geol., Part 1 32, pp 69–79.

Wulff, F., Field, J.G. & Mann, K.H., (Eds), (1989) Network Analysis in Marine Ecology. Coastal and Estuarine Studies Series, Springer-Verlag, Berlin, 284 p.

Xikang, C., (2000) Shanxi Water Resources Input-Occupancy-Output Table and its Application in Shanxi Province of China. Thirteenth International Conference in Input-Output Techniques, Macerata, Italy, 18 p.

Younger, P.L., (2003) Waste or voids: which is the greatest pollution source. Mining Environmental Management 11 (4), pp 18–21.

Younger, P.L., Arnesen, R.T., Iversen, E.R. & Banks, S.B., (1997) Mine water chemistry: The good, the bad and the ugly. Environ. Geol. 32(3), pp 157–174.

Date post:	20-Jun-2020
Category:	Documents
Upload:	others
View:	1 times
Download:	0 times

Environmental Management of Water Systems under...

Documents