terrestrial biotic ligand model (tblm) for copper, and nickel toxicities to plants, invertebrates, and microbes in soils

217 Table B-1 Soil and soil solution properties and the bioassay data for Cu toxicity to Barley Root Elongation BRE With permission from Dr.. 224 Table B-2 Soil and soil solution propert

TERRESTRIAL BIOTIC LIGAND MODEL (TBLM) FOR COPPER, AND

NICKEL TOXICITIES TO PLANTS, INVERTEBRATES,

AND MICROBES IN SOILS

by Sagar Thakali

A dissertation submitted to the Faculty of the University of Delaware in partial fulfillment of the requirements for the degree of Doctor of Philosophy in Civil Engineering

Spring 2006

Copyright 2006 Sagar Thakali All Rights Reserved

UMI Number: 3221133

3221133 2007

UMI Microform Copyright

All rights reserved This microform edition is protected against unauthorized copying under Title 17, United States Code.

ProQuest Information and Learning Company

300 North Zeeb Road P.O Box 1346 Ann Arbor, MI 48106-1346

by ProQuest Information and Learning Company

TERRESTRIAL BIOTIC LIGAND MODEL (TBLM) FOR COPPER, AND

NICKEL TOXICITIES TO PLANTS, INVERTEBRATES,

AND MICROBES IN SOILS

by Sagar Thakali

Conrado M Gempesaw II, Ph.D

Vice Provost for Academic and International Programs

I certify that I have read this dissertation and that in my opinion it meets the academic and professional standard required by the University as a

dissertation for the degree of Doctor of Philosophy

Signed:

Herbert E Allen, Ph.D

Professor in charge of dissertation

I certify that I have read this dissertation and that in my opinion it meets the academic and professional standard required by the University as a

dissertation for the degree of Doctor of Philosophy

Signed:

William R Berti, Ph.D

Member of dissertation committee

I certify that I have read this dissertation and that in my opinion it meets the academic and professional standard required by the University as a

dissertation for the degree of Doctor of Philosophy

Signed:

Ronald T Checkai, Ph.D

Member of dissertation committee

I certify that I have read this dissertation and that in my opinion it meets the academic and professional standard required by the University as a

dissertation for the degree of Doctor of Philosophy

Signed:

Dominic M Di Toro, Ph.D

Member of dissertation committee

I certify that I have read this dissertation and that in my opinion it meets the academic and professional standard required by the University as a

dissertation for the degree of Doctor of Philosophy

Signed:

Donald L Sparks, Ph.D

Member of dissertation committee

ACKNOWLEDGMENTS

I express my profound gratefulness to my advisor and mentor Professor Herbert E Allen without whose extraordinary patience, guidance, and advice the completion of this work was not possible I am equally thankful to Professor Dominic

M Di Toro who provided me with extensive assistance, guidance, and advice in the course of this study I am grateful to the Center for Study of Metals in the

Environment (CSME) at the Universidy of Delaware, the International Copper

Association (ICA), and the Nickel Producers Environmental Research Association (NiPERA) for funding this study

I am also grateful to Dr William R Berti, Dr Ronald T Checkai and Professor Donald L Sparks, for their service in my dissertation committee and for providing me with comments and suggestions which have greatly improved the quality of this study I thank my collaborators, Dr Alexander A Ponizovsky, Ms Corrine P Rooney, Dr Fang-Jie Zhao, Dr Steve P McGrath, Ms Peggy Criel, Ms Hilde Van Eckout, Professor Collin C Janssen, Dr Koen Oorts and Professor Erik M Smolders They have been very gracious in providing me with the data necessary for this study

I want to express my sincere appreciation to Ms Dana M Crumety for offering me help always, to Mr Douglas J Baker, Mr Michael J Davidson, Mr Eric

C Eckman and Ms Lucille Z Short for all their help, and to my colleagues, Dr Zhenqing Shi, Mr David M Metzler, Mr Akash Sondhi and Mr Sammy Lin for their

I would like to thank my family for all the love and support that they gave

me I am grateful to my grand parents, Mr and Mrs Gyan B Thakali, and parents,

Mr Gopal S Thakali and Mrs Kamala Thakali, for all the sacrifices they have made

so I can do what I do I am grateful to my siblings, Mrs Prabha S Subba, Ms Muna Thakali and Mr Deep S Thakali, who have always been there to support in my

struggles and rejoice in my happiness Finally, I would like to express the most sincere gratefulness to my wife, Jasmeen Hirachan, for putting up with me and my ways and loving me back even when I do not deserve Without her all my effort will have no meaning

TABLE OF CONTENTS

LIST OF TABLES xi

LIST OF FIGURES xv

ABSTRACT xxii

Introduction 1

1.1 Background 1

1.2 The overall objective 4

1.3 The structure of the dissertation 4

1.4 References 6

Literature review 9

2.1 Bioavailability and ecotoxicity of metals in soils 9

2.2 Understanding metal speciation in soils 17

2.2.1 Total metal content of soils 18

2.2.2 Soil organic matter (SOM) 19

2.2.3 Oxides, carbonates and clays 20

2.2.4 Dissolved organic carbon (DOC) 21

2.2.5 pH 22

2.2.6 Dissolved cations 23

2.3 Modeling metal speciation 24

2.4 Summary and Conclusions 29

2.5 References 30

The terrestrial biotic ligand model 39

3.1 Introduction 39

3.2 The Windermere Humic Aqueous Model (WHAM VI) 41

3.2.1 Sub-model for humic substances 43

3.2.2 Sub-model for oxides 46

3.2.3 Sub-model for cation exchange 46

3.2.4 Limitations in modeling with WHAM VI 47

3.3 The toxicity model 48

3.4 The estimation of model parameters 51

3.5 References 52

Material and methods 55

4.1 Characterization of the soils 55

4.2 Experimental procedure for Cu and Ni partitioning and speciation 56

4.3 The bioassay experiments 60

4.3.1 Plant bioassays 60

4.3.1.2 Tomato shoot growth 62

4.3.2 Invertebrate bioassays 62

4.3.2.1 Redworm cocoon production 62

4.3.2.2 Springtail juvenile production 63

4.3.3 Microbial bioassays 64

4.3.3.1 Potential nitrification rate 64

4.3.3.2 Glucose induced respiration 65

4.4 References 65

Modeling copper speciation 68

5.1 Introduction 68

5.2 Materials and Methods 69

5.3 Results and Discussions 70

5.3.1 The whole soil approach 70

5.3.1.1 Active organic matter 70

5.3.1.2 Phases besides SOM 74

5.3.1.3 Competitive binding of Fe 3+ and Al 3+ 75

5.3.1.4 Sensitivity of calculated pCu to DOC 76

5.3.1.5 Sensitivity of calculated pCu to dissolved cations and P CO2 77

5.3.1.6 Final calculations using the whole soil approach 80

5.3.2 The solution approach 80

5.3.3 Predicting the dissolved Cu concentration 84

5.4 Summary and Conclusions 86

5.5 References 87

Modeling nickel partitioning in soils 90

6.1 Introduction 90

6.2 Materials and Methods 92

6.3 Results and Discussions 93

6.3.1 Active soil organic matter 93

6.3.2 Competitive binding by Fe3+ and Al3+ 94

6.3.3 Prediction of Ni2+ activity 98

6.3.4 Sensitivity of the calculations to dissolved cations 99

6.3.5 Final Calculations using WHAM VI 99

6.3.6 Assessment of soils with OC content < 1% 101

6.4 Summary and Conclusions 105

6.5 References 105

Copper, and nickel toxicities to barley root elongation 109

7.1 Introduction 109

7.2 Experimental and Data Analysis Methods 111

7.2.1 Selection of soils 111

7.2.2 Bioassays 111

7.2.4 Toxicity equations 113

7.2.5 Estimation of model parameters 113

7.3 Results and Discussion 114

7.3.1 WHAM VI speciation 114

7.3.2 Estimation of the TBLM parameters 117

7.3.3 Dose-response relationships 125

7.3.3.1 Nickel toxicity 125

7.3.3.2 Copper toxicity 131

7.3.4 Interaction of cations with the biotic ligand 131

7.3.5 Prediction of the EC50 total metal concentrations 135

7.4 Summary and Conclusions 137

7.5 References 138

TBLM for plants, invertebrates, and microbes 142

8.1 Introduction 142

8.2 Experimental and Modeling Methods 143

8.2.1 Bioassays 143

8.2.2 Estimating the EC50 values for individual soils 144

8.2.3 Whole soil metal speciation using WHAM VI 145

8.2.4 Estimation of model parameters 146

8.3 Results and Discussion 147

8.3.1 Estimation of model parameters 147

8.3.2 Dose-response relationships 153

8.3.3 Interaction of cations with the biotic ligand 178

8.3.4 The extent of the protective effect of the competing cations 180

8.3.5 Prediction of EC50 183

8.4 Summary and Conclusions 186

8.5 References 187

Copper, and nickel toxicities in calcareous soils 190

9.1 Introduction 190

9.2 Bioassay and soil solution composition data 193

9.3 Modeling approach 194

9.4 Results and Discussion 196

9.4.1 Dose response relationships for the BRE bioassays 196

9.4.2 Prediction of the EC50 values 201

9.5 Summary and Conclusions 204

9.6 References 204

Final conclusions and recommendations 207

Appendix A 211

Appendix B 223

Appendix C 289

Appendix D 293

Appendix E 306

Appendix F 309 Appendix G 311

LIST OF TABLES

Table 2.1 The Coefficient of Determination for the regression between Zn

uptake rate of plants, Zn uptake by an isopod, and various

bioavailability estimations in field soils (extracted from Koster et

al (2005)) 12 Table 2.2 Comparison of the models for cation binding by humic substances

(Extracted from Kinniburgh et al (1996)) 28 Table 4.1 Properties of the soils used in the Cu study (Dr E Smolders and

Dr K Oorts, personal communication) 57 Table 4.2 Properties of the soils used in the Ni study (Dr E Smolders and

Dr K Oorts, personal communication) 58 Table 5.1 Effect of the input variables on the RMSE of the predicted pCu,

the linear regression, and Coefficient of Determination (R2)

between the predicted and measured pCu 79 Table 5.2 A comparison between speciation based on the whole soil

approach and the soil solution approach 83 Table 7.1 Model fit summary and the optimum parameters associated with

the three models 126 Table 8.1 The non-calcareous soils excluded for the analyses in this study

for each of the bioassays 145 Table 8.2 The dose-response parameters (EC50 and β) for the three models

considered: Total Cu Model (Total Cu), the Free Ion Activity

Model (FIAM) and the Terrestrial Biotic Ligand Model (TBLM)

The binding constants are associated with the TBLM 151 Table 8.3 The dose-response parameters (EC50 and β) for the three models

considered: Total Ni Model (Total Ni), the Free Ion Activity

Model (FIAM) and the Terrestrial Biotic Ligand Model (TBLM)

The binding constants are associated with the TBLM 152

Table 9.1 The distribution of Cu and Ni in 5 calcareous soils (Moral et al.,

2005) Only the approximate ranges are listed 191 Table 9.2 Summary of the model fits to the barley root elongation data for

Cu and Ni 196 Table A-1 Data for copper partitioning and speciation study (DOC and DIC

refer to dissolved organic and inorganic carbon, respectively) 212 Table A-2 Data for nickel partitioning and speciation study (DOC and DIC

refers to dissolved organic and inorganic carbon, respectively) 217 Table B-1 Soil and soil solution properties and the bioassay data for Cu

toxicity to Barley Root Elongation (BRE) (With permission from

Dr S P McGrath (Rothamsted Research, UK)) 224 Table B-2 Soil and soil solution properties and the bioassay data for Ni

toxicity to Barley Root Elongation (BRE) (With permission from

Dr S P McGrath (Rothamsted Research, UK)) 230 Table B-3 Soil and soil solution properties and the bioassay data for Cu

toxicity to Tomato Shoot Yield (TSY) (With permission from Dr

S P McGrath (Rothamsted Research, UK)) 235 Table B-4 Soil and soil solution properties and the bioassay data for Ni

toxicity to Tomato Shoot Yield (TSY) (With permission from Dr

S P McGrath (Rothamsted Research, UK)) 241 Table B-5 Soil and soil solution properties and the bioassay data for Cu

toxicity to E fetida Cocoon Production (ECP) (With permission

from Dr C Janssen (Ghent University, Belgium)) 246 Table B-6 Soil and soil solution properties and the bioassay data for Ni

toxicity to E fetida Cocoon Production (ECP) (With permission

from Dr C Janssen (Ghent University, Belgium)) 251 Table B-7 Soil and soil solution properties and the bioassay data for Cu

toxicity to F candida Juvenile Production (FJP) (With permission

from Dr C Janssen (Ghent University, Belgium)) 256 Table B-8 Soil and soil solution properties and the bioassay data for Ni

toxicity to F candida Juvenile Production (FJP) (With permission

from Dr C Janssen (Ghent University, Belgium)) 262

Table B-9 Soil and soil solution properties and the bioassay data for Cu

toxicity to Potential Nitrification Rate (PNR) (With permission

from Dr E Smolders (Katholieke Universiteit Leuven, Belgium)) 267 Table B-10 Soil and soil solution properties and the bioassay data for Ni

toxicity to Potential Nitrification Rate (PNR) (With permission

from Dr E Smolders (Katholieke Universiteit Leuven, Belgium)) 273 Table B-11 Soil and soil solution properties and the bioassay data for Cu

toxicity to Glucose Induced Respiration (GIR) (With permission

from Dr E Smolders (Katholieke Universiteit Leuven, Belgium)) 278 Table B-12 Soil and soil solution properties and the bioassay data for Ni

toxicity to Glucose Induced Respiration (GIR) (With permission

from Dr E Smolders (Katholieke Universiteit Leuven, Belgium)) 284 Table C-1 An example of WHAM VI input file for the computation of cation

activities for Ni toxicity to BRE In addition to the following the

temperature = 293K and the partial pressure of CO2 = 10-3.5 atm 290 Table D-1 EC50 total metal concentrations estimated by TRAP and the

interpolated solution properties for Cu toxicity to Barley Root

Elongation (BRE) 294 Table D-2 EC50 total metal concentrations estimated by TRAP and the

interpolated solution properties for Ni toxicity to Barley Root

Elongation (BRE) 295 Table D-3 EC50 total metal concentrations estimated by TRAP and the

interpolated solution properties for Cu toxicity to Tomato Shoot

Yield (TSY) 296 Table D-4 EC50 total metal concentrations estimated by TRAP and the

interpolated solution properties for Ni toxicity to Tomato Shoot

Yield (TSY) 297 Table D-5 EC50 total metal concentrations estimated by TRAP and the

interpolated solution properties for Cu toxicity to E fetida Cocoon

Production (ECP) 298 Table D-6 EC50 total metal concentrations estimated by TRAP and the

interpolated solution properties for Ni toxicity to E fetida Cocoon

Production (ECP) 299

Table D-7 EC50 total metal concentrations estimated by TRAP and the

interpolated solution properties for Cu toxicity to F candida

Juvenile Production (FJP) 300 Table D-8 EC50 total metal concentrations estimated by TRAP and the

interpolated solution properties for Ni toxicity to F candida

Juvenile Production (FJP) 301 Table D-9 EC50 total metal concentrations estimated by TRAP and the

interpolated solution properties for Cu toxicity to Potential

Nitrification Rate (PNR) 302 Table D-10 EC50 total metal concentrations estimated by TRAP and the

interpolated solution properties for Ni toxicity to Potential

Nitrification Rate (PNR) 303 Table D-11 EC50 total metal concentrations estimated by TRAP and the

interpolated solution properties for Cu toxicity to Glucose Induced

Respiration (GIR) 304 Table D-12 EC50 total metal concentrations estimated by TRAP and the

interpolated solution properties for Ni toxicity to Glucose Induced

Respiration (GIR) 305 Table E-1 Concentrations of dissolved major cations in soil solutions of

different soils 307 Table F-1 WHAM VI input file for speciation of Cu at EC50 for Barley Root

Elongation (BRE) bioassays for calcareous soils 310 Table G-1 An example file for the response, the total Ni content and the

WHAM VI model output data for speciation computation for Ni

toxicity to barley root elongation bioassay in non-calcareous soils

with OC content > 1.0% 312 Table G-2 An example spreadsheet for the calculations involving the three

models for Ni toxicity to barley root elongation in non-calcareous

soils with OC content > 1.0% 314

LIST OF FIGURES

Figure 3.1 A schematic of the interactions considered in a TBLM for the

ecotoxicity of metals in soil systems (Modified from Di Toro et

al (2005)) 42 Figure 5.1 The predicted pCu values using WHAM VI based on the whole

soil approach and the measured pCu values The solid line represents the 1:1 ratio and the dashed lines are a unit above and below the 1:1 line 81 Figure 5.2 The predicted pCu values using WHAM VI based on soil solution

composition and the measured values of pCu The solid line represents the 1:1 ratio and the dashed lines are a unit above and below the 1:1 line 82 Figure 5.3 Predicted vs measured dissolved Cu concentration The solid line

represents 1:1 ratio, and the dotted lines represent a half log unit above and below the 1:1 ratio 85 Figure 6.1 (a) Comparison of predicted log [Ni]dis values using the whole

soil approach and the measured values for all non-calcareous soils; The solid line represents a 1:1 ratio and the dotted lines are

a log unit above and below the 1:1 line (b) The residual error (measured-predicted) as related to the total Ni content in the soil samples The residual error for the spiked samples of Montpellier and Aluminusa soils are shown in a box 97 Figure 6.2 Predicted vs measured log [Ni]dis for non-calcareous soils with

OC content > 1% The solid line represents a 1:1 ratio and the dotted lines are a half log unit above and below the 1:1 line 100 Figure 6.3 Relationships between the average residual error (measured –

predicted) of the predicted dissolved Ni concentrations and (a) the

OC content, and (b) the clay content of the soils The two soils (Montpellier and Aluminusa) are identified in the figures The dotted lines represent the no residual error (Note that there appears only 9 soils in (b) because the point (21% Clay and

Figure 6.4 The predicted vs measured dissolved Ni concentrations based on

the inclusion of clay The solid line represents the 1:1 line and the dashed lines are a half log unit above and below the 1:1 ratio The legends “No Clay”, “Default”, “50%” and “30%” represent the calculations without any clay, with all the clay fraction assumed

to be the default clay in WHAM VI, with 50% of the clay to be montmorillonite, and with 30% of the clay to be montmorillonite, respectively 103 Figure 7.1 Comparisons between dissolved metal concentrations predicted

by using the whole soil approach, (a) dissolved Ni concentration and (b) dissolved Cu concentration The solid line represents the 1:1 ratio and the dotted lines represent a half log unit above and below the 1:1 line 116 Figure 7.2 The RMSE of the predicted % root elongation as a function of log

KNiBL (y-axis) and various TBLM parameters (x-axis): (a) f50, (b) log KCaBL In the calculations, the value of β is fixed at 2.60 in all

cases None of the competitive cations were included in (a) and only Ca was included in (b) 119 Figure 7.2 The RMSE of the predicted % root elongation as a function of log

KNiBL (y-axis) and various TBLM parameters (x-axis): (c) log KHBL, and (d) log KHBL In the calculations, the value of β is fixed

at 2.60 in all cases Only H+ was included in (c) and both Ca2+ and

H+ were included in (d) with log KCaBL fixed at 1.5 120

Figure 7.3 The root mean square error (RMSE) of the predicted % root

elongation as a function of log KCuBL (y-axis) and various TBLM parameters (x-axis): (a) log f50 with no competitive cations and β

= 1 and (b) log KHBL with only protons as the competing cation

and β = 1 123 Figure 7.3 The root mean square error (RMSE) of the predicted % root

elongation as a function of log KCuBL (y-axis) and various TBLM parameters (x-axis): (c) log KCaBL with only Ca2+ as the competing cation and β = 1 and (d) log KMgBL with only Mg2+ as the

competing cation and β = 1 124 Figure 7.4a Dose-response relationships for Ni toxicity to barley root

elongation (BRE): (i) the TMM, (ii) the FIAM, (iii) the TBLM

Figure 7.4b Predicted vs observed barley root elongation (BRE) for Ni

toxicity models: (i) the TMM, (ii) the FIAM, (iii) the TBLM The dotted lines represent the 1:1 ratio The equations and the R2values for the linear regression between the predicted and the observed % BRE (solid lines) are also shown 128 Figure 7.5a Dose-response relationships for Cu toxicity to barley root

elongation (BRE): (i) the TMM, (ii) the FIAM, (iii) the TBLM;

The lines represent the models 129 Figure 7.5b Predicted vs observed barley root elongation (BRE) for Cu

toxicity models: (i) the TMM, (ii) the FIAM, (iii) the TBLM The dotted lines represent the 1:1 ratio The equations and the R2values for the linear regression between the predicted and the observed % BRE (solid lines) are also shown 130 Figure 7.6 The relationship between the Cu2+ and H+ activities at EC50 for

barley root elongation The observed data are based on WHAM

VI speciation and the model predictions are based on the FIAM (dotted line) and the TBLM (solid line) 133 Figure 7.7 The fraction (f) of the total biotic ligand sites on the barley roots

bound by various cations (Ni2+, H+, Ca2+ and Mg2+) at EC50 for

Ni toxicity in 8 soils 135 Figure 7.8 The predicted and the observed EC50 values for Ni and Cu

toxicity to barley root elongation The solid line represents the 1:1 ratio and the dotted lines are a factor of 2 above and below 137 Figure 8.1a Dose response relationships for tomato shoot yield (TSY) based

on i total Cu as the dose, ii free Cu2+ activity as the dose, and iii

fraction (f) of total biotic ligand bound by Cu2+ The lines represent the TMM, the FIAM and the TBLM, respectively for i,

ii and iii 154 Figure 8.1b The observed vs predicted tomato shoot yield (TSY); i the

TMM, ii the FIAM, and iii the TBLM The solid lines represent the linear regression for which the equations and the R2 values are given, and the dotted lines represent the 1:1 line 155

Figure 8.2a Dose response relationships for F candida juvenile production

(FJP) based on i total Cu as the dose, ii free Cu2+ activity as the

dose, and iii fraction (f) of total biotic ligand bound by Cu2+ The lines represent the TMM, the FIAM and the TBLM, respectively for i, ii and iii 156

Figure 8.2b The observed vs predicted F candida juvenile production (FJP);

i the TMM, ii the FIAM, and iii the TBLM The solid lines represent the linear regression for which the equations and the R2values are given, and the dotted lines represent the 1:1 line 157

Figure 8.3a Dose response relationships for E fetida cocoon production

(ECP) based on i total Cu as the dose, ii free Cu2+ activity as the

dose, and iii fraction (f) of total biotic ligand bound by Cu2+ The lines represent the TMM, the FIAM and the TBLM, respectively for i, ii and iii 158

Figure 8.3b The observed vs predicted E fetida cocoon production (ECP); i

the TMM, ii the FIAM, and iii the TBLM The solid lines represent the linear regression for which the equations and the R2values are given, and the dotted lines represent the 1:1 line 159 Figure 8.4a Dose response relationships for potential nitrification rate (PNR)

based on i total Cu as the dose, ii free Cu2+ activity as the dose,

and iii fraction (f) of total biotic ligand bound by Cu2+ The lines represent the TMM, the FIAM and the TBLM, respectively for i,

ii and iii 160 Figure 8.4b The observed vs predicted potential nitrification rate (PNR); i

the TMM, ii the FIAM, and iii the TBLM The solid lines represent the linear regression for which the equations and the R2values are given, and the dotted lines represent the 1:1 line 161 Figure 8.5a Dose response relationships for glucose induced respiration (GIR)

based on i total Cu as the dose, ii free Cu2+ activity as the dose,

and iii fraction (f) of total biotic ligand bound by Cu2+ The lines represent the TMM, the FIAM and the TBLM, respectively for i,

ii and iii 162 Figure 8.5b The observed vs predicted glucose induced respiration (GIR); i

the TMM, ii the FIAM, and iii the TBLM The solid lines represent the linear regression for which the equations and the R2

Figure 8.6a Dose response relationships for tomato shoot yield (TSY) based

on i total Ni as the dose, ii free Ni2+ activity as the dose, and iii

fraction (f) of total biotic ligand bound by Ni2+ The lines represent the TMM, the FIAM and the TBLM, respectively for i,

ii and iii 164 Figure 8.6b The observed vs predicted tomato shoot yield (TSY); i the

TMM, ii the FIAM, and iii the TBLM The solid lines represent the linear regression for which the equations and the R2 values are given, and the dotted lines represent the 1:1 line 165

Figure 8.7a Dose response relationships for F candida juvenile production

(FJP) based on i total Ni as the dose, ii free Ni2+ activity as the

dose, and iii fraction (f) of total biotic ligand bound by Ni2+ The lines represent the TMM, the FIAM and the TBLM, respectively for i, ii and iii 166

Figure 8.7b The observed vs predicted F candida juvenile production (FJP);

i the TMM, ii the FIAM, and iii the TBLM The solid lines represent the linear regression for which the equations and the R2values are given, and the dotted lines represent the 1:1 line 167

Figure 8.8a Dose response relationships for E fetida cocoon production

(ECP) based on i total Ni as the dose, ii free Ni2+ activity as the

dose, and iii fraction (f) of total biotic ligand bound by Ni2+ The lines represent the TMM, the FIAM and the TBLM, respectively for i, ii and iii 168

Figure 8.8b The observed vs predicted E fetida cocoon production (ECP); i

the TMM, ii the FIAM, and iii the TBLM The solid lines represent the linear regression for which the equations and the R2values are given, and the dotted lines represent the 1:1 line 169 Figure 8.9a Dose response relationships for glucose induced respiration (GIR)

based on i total Ni as the dose, ii free Ni2+ activity as the dose,

and iii fraction (f) of total biotic ligand bound by Ni2+ The lines represent the TMM, the FIAM and the TBLM, respectively for i,

ii and iii 170 Figure 8.9b The observed vs predicted glucose induced respiration (GIR); i

the TMM, ii the FIAM, and iii the TBLM The solid lines represent the linear regression for which the equations and the R2values are given, and the dotted lines represent the 1:1 line 171

Figure 8.10a Dose response relationships for potential nitrification rate (PNR)

based on i total Ni as the dose, ii free Ni2+ activity as the dose,

and iii fraction (f) of total biotic ligand bound by Ni2+ The lines represent the TMM, the FIAM and the TBLM, respectively for i,

ii and iii 172 Figure 8.10b The observed vs predicted potential nitrification rate (PNR); i

the TMM, ii the FIAM, and iii the TBLM The solid lines represent the linear regression for which the equations and the R2values are given, and the dotted lines represent the 1:1 line 173 Figure 8.11 A comparison of the predicted vs observed EC50 total metal

concentrations for Cu toxicity to barley root elongation based on a) the TMM, b) the FIAM, and c) the TBLM The solid lines represent the 1:1 ratio and the dotted lines represent a factor of 2 variation above and below the 1:1 lines The linear regression relationships between the log of predicted and observed EC50 values are also given 176 Figure 8.12 A comparison of the binding constants for (a) Cu toxicity and (b)

Ni toxicity 179 Figure 8.13 The protective effects of the competitive cations in (a) Cu toxicity

and in (b) Ni toxicity The legend “None” represents the theoretical EC50 when none of the competitive cations are present The reduction in Cu toxicity is computed at pH 6.5, 5.5, 4.5 and at 4.5 with Ca (= 2.87 mM) and Mg (= 0.73 mM)

indicated by the legend “pH 4.5*” in (a) The reduction in Ni toxicity (b) is computed at pH 5.5 (“Only H”), with only Ca (=

2.87 mM) (“Only Ca”), with only Mg (= 0.73 mM) (“Only Mg”) and with Ca (= 2.87 mM), Mg (= 0.73 mM) at pH 5.5 (“All”) 182 Figure 8.14 The relationship between the predicted and the observed EC50

values for (a) Cu toxicity and (b) Ni toxicity to various endpoints

The solid lines represent the 1:1 ratio and the dotted lines are a factor of 2 above and below the 1:1 line 184 Figure 9.1a The dose response relationship for Cu toxicity to barley root

elongation (BRE) in the calcareous soils The models represented are (i) the TMM, (ii) the FIAM and (iii) the TBLM The lines represent the best fit models developed for the non-calcareous soils in Chapter 7 197

Figure 9.1b The relationship between the predicted vs measured barley root

elongation (BRE) for Cu toxicity based on (i) the TMM, (ii) the FIAM and (iii) the TBLM The linear regression lines (solid lines), the associated equations, R2 values and the perfect-fit line (dotted lines) are shown 198 Figure 9.2a The dose response relationship for Ni toxicity to barley root

elongation (BRE) in the calcareous soils The models represented are (i) the TMM, (ii) the FIAM and (iii) the TBLM The lines represent the best fit models developed for the non-calcareous soils in Chapter 7 199 Figure 9.2b The relationship between the predicted vs measured barley root

elongation (BRE) for Ni toxicity based on (i) the TMM, (ii) the FIAM and (iii) the TBLM The linear regression lines (solid lines), the associated equations, R2 values and the perfect-fit line (dotted lines) are shown 200 Figure 9.3 Predicted vs observed EC50 cation activities of (a) Cu2+ and (b)

Ni2+ for the various endpoints in the solutions of the of calcareous soils The solid line represents the 1:1 ratio and the dotted lines are a factor of 5 above and below the 1:1 line 203

ABSTRACT

Contamination of soils with metals is a world-wide problem that could threaten the sustainability of essential soil functions Therefore, risk assessments utilizing appropriate and prudent soil quality criteria are needed Although it is widely recognized that the total metal concentration in soils does not represent its

bioavailability, soil quality criteria and risk assessment of metals continue to be

predominantly based on the total metal concentrations This study provides a

theoretical model, the Terrestrial Biotic Ligand Model (TBLM), based on which ecotoxicities of metals in soil systems can be assessed The TBLM achieves a

mathematical integration of the soil-solution-organism interaction by way of chemical equilibrium models The active sites on the organism are treated as a ligand, the Biotic Ligand (BL), to which metal ions bind and may cause a toxic response The level of toxicity is correlated only to the fraction of the total BL sites bound by the metal Additionally, other cations, principally protons, Ca2+, and Mg2+, also compete to bind

on the BL sites and, consequently, may alleviate metal toxicities

For the development of the TBLM, a speciation study was conducted in

up to nineteen soils spiked with different levels of Cu, and Ni in the laboratory under field moisture conditions The soils cover nine countries in the European Union and represent seven major soil groupings A whole soil approach with Windermere Humic Aqueous Model (WHAM VI) using the total metal concentration, the organic carbon (OC) content, the solution pH, and dissolved concentrations of major cations, as inputs

calcareous soils (pH < 7 and CaCO3 content ≅ 0%) Dissolved Cu concentrations (log transformed) were predicted with a root mean squared error (RMSE) of 0.39 and the free copper ion (Cu2+) activities (log transformed) were predicted with a higher RMSE

of 0.77 most likely due to errors introduced during sample preparation, handling, and measurements Similarly, the dissolved Ni concentrations in the non-calcareous soils were predicted with a RMSE of 0.26 for non-calcareous soils with OC content > 1% For the remaining two non-calcareous soils with OC content < 1%, the importance of clay fraction in the partitioning was indicated

The whole soil approach using WHAM VI was then applied to speciate

Cu, and Ni in the soil solutions associated with the bioassays The bioassays

represented Cu, and Ni toxicities to plants (barley (Hordeum vulgare cv Regina) root elongation and tomato (Lycopersicon esculentum cv Moneymaker) shoot yield), invertebrates (reproduction of redworms (Eisenia fetida) and springtails (Folsomia candida)) and microbes (potential nitrification rate and glucose induced respiration) in

the same soils that were used for the corresponding speciation studies, and under similar experimental conditions

The analyses of the bioassay results using the TBLM framework showed that protons compete with Cu2+ ions in their interaction with the BL sites associated with the six endpoints In addition to the protons, the major cations Ca2+ and/or Mg2+also compete with Ni2+ ions in these endpoints These competing cations (protons,

Ca2+, and Mg2+) at their average ambient soil solution concentrations reduce the EC50 free metal ion activities of both Cu and Ni by more than two orders of magnitude from the values in the absence of their competition The significance of this reduction in

metal toxicities by these major cations is currently not incorporated in models for ecotoxicities of metals in soils

A comparison of the TBLM predictions to the predictions based on models employing as the dose the total metal concentration or solution free metal ion activity in the solution shows that the TBLM is consistently able to achieve a better normalization of the wide variation in toxicological results in the non-calcareous soils The TBLM predictions of the EC50 total metal concentrations were generally within a factor of two of the observed values with RMSE of the predicted values (log

transformed) of 0.26 for Cu toxicity and 0.17 for Ni toxicity For Cu, 48 out of the 61 observed EC50 total metal concentrations associated with the six endpoints were predicted with within a factor of two For Ni, almost all of the predicted EC50 total metal concentrations (43 out of 45) were within a factor of two of the observed values Additionally, 90% of the predicted EC50 total Cu concentrations and 100% of the predicted EC50 total Ni concentrations were within a factor of three of the observed values The application of the TBLM to the calcareous soils based on solution

speciation using solution composition predicted the EC50 Ni2+ activities generally within a factor of five However, for Cu in these soils, the predictions of EC50 Cu2+activities were poor most likely due to the inaccurate speciation using solution

composition and the possible interaction of CuOH+ and CuHCO3+ with the associated biotic ligand sites

This study, which incorporates Cu, and Ni toxicities to six different

endpoints associated with higher plants, invertebrates, and microbes for up to eleven non-calcareous soils of disparate properties is, to my knowledge, the first one to use a single theoretical framework for modeling metals’ toxicities in terrestrial systems Its

encouraging performance in the majority of the soils considered in this study suggests that the TBLM provides a reasonable theoretical approach for insights into the

competitive interaction of the cations, and assessing metal toxicities in a complex and heterogeneous soil system Therefore, this study has achieved a significant

advancement in assessing the bioavailability and toxicities of metals in terrestrial systems, which is a significant part of environmental risk assessment

Chapter 1 INTRODUCTION

predominantly based on the total metal concentrations (Steenbergen et al., 2005)

The current measures of metal bioavailability invariably involve some form of chemical extraction (Haddad and Evans, 1993; Saeki et al., 2002; Cela and Sumner, 2002) In these methods, the fraction extractable of the total metal is

correlated to be the bioavailable fraction Another measure of metal bioavailability that is gaining currency is the “diffusive gradient in thin films (DGT)” measurements (Zhang et al., 2001 and 2004) In this method, an effective “bioavailable”

concentration in the soil pore water is determined based on the diffusion of metal cations across a film of resin However, the relationships between the bioavailable metals represented in these methods and their ecotoxicities are inconsistent (Alloway

et al., 1990; Haddad and Evans, 1993; Cela and Sumner, 2002; Saeki et al., 2002;

Alternatively, empirical models based on regression analyses have also been developed for predicting soil-specific critical limits for metals in soils These involve correlating a level of toxicity (e.g EC10, EC50) to various soil properties, principally the cation exchange capacity (CEC), soil pH, and soil organic matter (SOM) content (Lock et al., 2001; Oorts et al., 2006; Rooney et al., 2006) Albeit they are useful in relating the toxicities of metals to soil chemistry, they remain purely empirical and may thus lack wide scale applicability

Still another approach has been to use the free metal ion activities in soil pore water as the toxic dose The approach is based on the Free Ion Activity Model (FIAM) that was originally developed for aquatic systems (Morel, 1983; Campbell, 1995) Calculated free metal ion activities were suggested to be a better predictor of ecotoxicity than total metal in soils (Sauvé et al., 1998; Dumestre et al., 1999)

indicating that they better represent the most bioavailable species of metals in soil pore water However, evidence has been reported (Plette et al., 1999; Van Gestel and Koolhaas, 2004; Lofts et al., 2004; Smolders et al., 2004; Steenbergen et al., 2005; Oorts et al., 2006; Zhao et al., 2006) suggesting that interactions similar to those incorporated in the biotic ligand model (BLM) for aquatic systems (Di Toro et al., 2001) may be present in soil pore water

The BLM assumes that there is a site of action, the biotic ligand (BL), on the organism which when bound by a metal cation exerts a toxic effect Furthermore, the extent of the toxic effect depends on the fraction of the total BL sites occupied by the metal cation and when cations (such as protons, Ca and Mg), which are typically not toxic, compete with the metal cation to bind on the BL sites, they effectively reduce its toxicity The competitive interaction of these cations, particularly that of the

proton, has been indicated in metal toxicities to a whole range of the soil organisms such as soil microbes, invertebrates, and higher plants (Plette et al., 1999; Allen, 2002; Van Gestel and Koolhaas, 2004; Lofts et al., 2004; Smolders et al., 2004; Steenbergen

et al., 2005; Oorts et al., 2006; Zhao et al., 2006)

However, a BLM-type approach to soil systems, which treats the soil solution as the starting point, is essentially the same as the aquatic BLM Such an approach completely ignores the interaction among different phases in the soil

Additionally, this approach necessitates the determination of the soil solution

chemistry for its application But soil solution chemistry is not easily determined and the data are not readily available A Terrestrial Biotic Ligand Model (TBLM) must be able to simultaneously integrate all the interactions in a soil system

Naturally, there have been attempts to achieve such a holistic integration

of the interactions in soil systems Plette et al (1999) and Van Gestel and Koolhaas (2004) developed empirical models to predict accumulation of metals in the biota by treating the biota as one of the sorbing phases in soils They employed various

isotherms for metal accumulation both on soils and on the biota Saxe et al (2001) also considered the interaction of the organism as a sorbent phase in soil systems in modeling metal accumulation by earthworm However, all of these models suffer from two limitations First the empirical nature of the sorption models may limit their wide-scale applicability and second, metal accumulation in the organisms does not always correlate well with the observed toxic effects (Lanno et al., 2004)

Hence, an explicit model which directly links the interactions, first

between the soil phase and the soil solution and thence, between the soil solution and the organism is currently lacking Development of such a model would not only

further our understanding of metals toxicities in soils but also would greatly facilitate metals risk assessment in soils, which constitutes a significant part of environmental risk assessment

1.2 The overall objective

The overall objective of this project is to develop a TBLM for Cu, and Ni ecotoxicities in soils This TBLM will link the soil chemistry, the soil solution

chemistry, and the interactions with an organism, with respect to Cu and Ni To do so the TBLM will integrate two sub-models, a speciation model and a toxicity model The speciation model will consider the interaction between the soil phases with the soil solution The toxicity model will then link the Cu, and Ni speciation in the soil

solution with the level of their toxicities to plant growth (barley (Hordeum vulgare cv Regina) root elongation and tomato (Lycopersicon esculentum cv Moneymaker) shoot yield), invertebrate reproduction (redworm (Eisenia fetida) cocoon production and springtail (Folsomia candida) juvenile production) and microbial activity (glucose

induced respiration and potential nitrification rate) in up to eighteen different soils from Europe The objective of this study was to develop the TBLM and demonstrate its applicability to Cu, and Ni toxicities to six different endpoints in a wide range of soils In doing so, the study will provide a potentially valuable tool for metals risk assessment in soils

1.3 The structure of the dissertation

To address the issues and the goals identified here, the thesis is organized

as follows:

In Chapter 2 a review of the existing literature on metals ecotoxicities

and speciation in soils is given The evidence for a BLM-type interaction of the

cations with the soil biota is presented A discussion of various approaches to

modeling partitioning and speciation of metals in soils and on the available

computational models are also presented

In Chapter 3 the TBLM is described and formulated Its sub-model for

metal speciation is discussed and the toxicity model is mathematically formulated Finally the approach used in this study to estimate the model parameters is discussed

In Chapter 4 the properties of the soils, the speciation and the bioassay

studies, which form the source of the data for this study, are described

In Chapter 5 Cu speciation in non-calcareous soils is modeled using

WHAM VI A discussion of the model performance in terms of the precision in

predicting free Cu2+ ion activities and dissolved Cu concentrations in soil solutions is provided

In Chapter 6 Ni speciation in non-calcareous soils is modeled with

WHAM VI with the same approach used in Cu speciation in Chapter 5 and its results discussed

In Chapter 7 the TBLM for Cu, and Ni toxicities to barley root

elongation is developed and its features and significance are illustrated

In Chapter 8 the TBLM developed for Cu, and Ni toxicities to barley root

elongation in Chapter 7 is extended to the remaining five endpoints to demonstrate the general applicability of the approach

In Chapter 9 the possible incorporation of calcareous soils into the

TBLM framework is discussed Instead of the whole soil approach to speciation in

these soils, for reasons that will be discussed, WHAM VI will be applied for

speciation of metals based on the soil solution composition The computed speciation

is then coupled with the estimated TBLM constants to evaluate the applicability of the TBLM to these soils

Finally in Chapter 10, a summary of all the major findings is given along

with the scope and the limitation of the current TBLM The suggestions and the recommendations are also made for future refinement and extension of the TBLM

1.4 References

Allen, H E Terrestrial ecosystems: an overview In: Bioavailability of metals in

terrestrial ecosystems: importance of partitioning for bioavailability to

invertebrates, microbes and plants; Allen, H E., Ed.; SETAC Press:

Pensacola, FL, 2002; pp 1-5

Alloway, B J.; Jackson, A P.; Morgan, H The accumulation of cadmium by

vegetables grown on soils contaminated from a variety of sources Sci Total

Environ 1990, 91, 223-236

Campbell, P G C Interactions between trace metals and aquatic organisms: A

critique of the Free-Ion Activity Model In: Metal Speciation and

Bioavailability in Aquatic Systems; Tessier, A., Turner, D R., Eds.; John

Wiley & Sons: New York, NY, 1995; pp 45-102

Cela, S.; Sumner, M E Critical concentrations of copper, nickel, lead, and cadmium

in soils based on nitrification Commun Soil Sci Plan 2002, 33, 19-30

Di Toro, D M.; Allen, H E.; Bergman, H L.; Meyer, J S.; Paquin, P R.; Santore, R

C Biotic ligand model of the acute toxicity of metals 1 Technical basis

Environ Toxicol Chem 2001, 20, 2383-2396

Dumestre, A.; Sauvé, S.; McBride, M.; Baveye, P.; Bethelin, J Copper speciation and

microbial activity in long-term contaminated soils Arch Environ Contam

Toxicol 1999, 36, 124-131

Giller, K E.; Witter, E.; McGrath, S P Toxicity of heavy metals to micro organisms

and microbial processes in agricultural soils: A review Soil Biol Biochem

1998, 30, 1389-1414

Haddad, K S.; Evans, J C Assessment of chemical methods for extracting zinc,

manganese, copper, and iron from New-South-Wales soils Commun Soil Sci

Plant Anal.1993, 24, 29-44

Koster, M.; Reijnders, L.; Van Oost, N R.; Peijnenburg, W J G M Comparison of

the method of diffusive gels in thin films with conventional extraction

techniques for evaluating zinc accumulation in plants and isopods Environ

Pollut 2005, 133, 103-116

Lanno, R.; Wells, J.; Conder, J.; Bradham, K.; Basta, N The bioavailability of

chemicals in soils for earthworms Ecotoxicol Environ Safety, 2004, 57,

39-47

Lock, K.; Janssen, C R Ecotoxicity of zinc in spiked artificial soils versus

contaminated field soils Environ Sci Technol 2001, 35, 4295-4300

Lofts, S.; Spurgeon, D J.; Svendsen, C.; Tipping, E Deriving soil critical limits for

Cu, Zn, Cd, and Pb: A method based on free ion concentrations Environ Sci

Technol 2004, 38, 3623-3631

Morel, F M M Principles of Aquatic Chemistry John Wiley & Sons: New York,

NY, 1983

Nriagu, J O.; Pacyna, J M Quantitative assessment of worldwide contamination of

air, water and soils by trace-metals Nature, 1988, 333, 134-139

Oorts, K.; Ghesquiere, U.; Swinnen, K.; Smolders, E Soil properties affecting the

toxicity of CuCl2 and NiCl2 for soil microbial processes in freshly spiked soils

Environ Toxicol Chem 2006, 25, 836-844

Plette, A C C.; Nederlof, M M.; Temminghoff, E J M.; Van Riemsdijk, W H

Bioavailability of heavy metals in terrestrial and aquatic systems: A

quantitative approach Environ Toxicol Chem 1999, 18, 1882-1890

Rooney, C P.; Zhao, F J.; McGrath, S P Soil factors controlling the expression of

copper toxicity to plants in a wide range of European soils Environ Toxicol

Chem 2006, 25, 726-732

Saeki, K.; Kunito, T.; Oyaizu, H.; Matsumoto, S Relationships between bacterial

tolerance levels and forms of copper and zinc in soils J Environ Qual 2002,

31, 1570-1575

Sauvé, S.; Dumestre, A.; McBride, M.; Hendershot, W Derivation of soil quality

criteria using predicted chemical speciation of Pb2+ and Cu2+ Environ

Toxicol Chem 1998, 17, 1481-1489

Saxe, J K.; Impellitterri, C A.; Peijnenburg, W J G M.; Allen, H E A novel model

describing heavy metal concentrations in the earthworm, Eisenia andrei

Environ Sci Technol 2001, 35, 4522-4529

Smolders, E.; Buekers, J.; Oliver, I.; McLaughlin, M E Soil properties affecting

toxicity of zinc to soil microbial properties in laboratory-spiked and

field-contaminated soils Environ Toxicol Chem 2004, 23, 2633-2640

Steenbergen, N T T M.; Iaccino, F.; De Winkel, M.; Reijnders, L.; Peijnenburg, W

J G M Development of a biotic ligand model and a regression model

predicting acute copper toxicity to the earthworm Aporrectodea caliginosa

Environ Sci Technol 2005, 39, 5694-5702

Van Gestel, C A M.; Koolhaas, J E Water-extractability, free ion activity, and pH

explain cadmium sorption and toxicity to Folsomia candida (Collembola) in

seven soil-pH combinations Environ Toxicol Chem 2004, 23, 1822-1833

Zhang, H.; Zhao, F J.; Sun, B.; Davison, W.; McGrath, S P A new method to

measure effective soil solution concentration predicts copper availability to

plants Environ Sci Technol 2001, 35, 2602-2607

Zhang, H.; Lombi, E.; Smolders, E.; McGrath, S P Kinetics of Zn release in soils and

prediction of Zn concentration in plants using diffusive gradients in thin films

Environ Sci Technol 2004, 38, 3608-3613

Zhao, F J.; Rooney, C P.; Zhang, H.; McGrath, S P Comparison of soil solution

speciation and DGT measurement as indicators of copper bioavailability to

plants Environ Toxicol Chem 2006, 25, 733-742

Chapter 2 LITERATURE REVIEW

2.1 Bioavailability and ecotoxicity of metals in soils

It is widely recognized that the total metal concentration in soils does not represent its bioavailability to the soil organisms (Allen, 2002) Most recently,

Bradham et al (2006) reported that the mortality of earthworms (Eisenia Andrei)

varied between 0 to 100% in twenty one soils of varying properties even though each was amended with 2000 mg Pb kg-1 soil In a study of Cu toxicity to barley root

elongation and tomato shoot yield, the EC50 total metal concentration varied by fifteen and thirty nine fold, respectively, among eighteen soils (Rooney et al., 2006) Similarly the toxicity thresholds for Cu, and Ni toxicities to microbial processes varied nineteen to ninety fold in eighteen soils (Oorts et al., 2006) Clearly, the total metal concentration in soils is a poor indicator of its bioavailability to soil invertebrates, plants and microbes In the context of Pb toxicity to earthworms in soils, Lanno et al (2004) reports that the bioavailability of metals in soils can be modified dramatically

by soil properties and consequently, exposure assessments based on the total metal concentration in soils will likely lead to erroneous estimation of its risk Therefore, a reasonable approach for bioavailability estimation of metals in soils is necessary to correctly assess their ecological risks

Operationally defined extractions using weak salt solutions have been

assessment of metals in soils (Prüeβ, 1997) Because its concentration is similar to that

of average soil solution, a 0.01 M CaCl2 solution is often preferred for extraction of metals from soils (Koster et al., 2005) Houba et al (1996) even suggested using this solution as a universal extractant However, the extractions using this solution have not correlated consistently with metal uptake (Haddad and Evans, 1993; Brun et al 2001; Saeki et al., 2002; Cela and Sumner, 2002; Koster et al., 2005) Haddad and

Evans (1993) studied metal uptake by clover (Trifolium subterraneum cv

Woogenellup) in 60 different soils to assess eight chemical extractants as predictors of

Zn, Mn, Cu and Fe bioavailability A 0.05 M HCl extractable Zn produced the best correlation (R2 = 0.78) for acidic to near neutral soils, whereas the DTPA-extractable

Zn produced the best correlation (R2 = 0.80) for calcareous soils The 0.01 M extractable Mn and the 0.31 M HNO3-extractable Cu produced the best correlations for Mn (R2 = 0.64) and Cu (R2 = 0.72) uptake, respectively In a Cu uptake study in

CaCl2-vineyard soils, Brun et al (2001) reported that the Cu content in maize (Zea mays cv

Gaucho) roots correlated best with EDTA, DPTA, and ammonium acetate-extractable

Cu with R2 values of 0.90, 0.93, and 0.89, respectively, and only weakly with the extracted by 0.01 M CaCl2 (R2 =0.44) But in the same study, the Cu content in the maize shoot correlated the best with CaCl2-extractable Cu (R2 = 0.71) compared to R2

Cu-< 0.35 for EDTA, DPTA, and ammonium acetate-extractable Cu These studies show that there is not a universal method of extraction that consistently provides the best estimate of bioavailable metals in soils Therefore, it is difficult to choose an

appropriate method of extraction that is able to reflect the bioavailability of metals in soils

Recently, a new approach based on a chemical measurement of metal content in soils by use of a gel technique, Diffusion Gradients in Thin Films (DGT), has been used to assess metal bioavailability in soils (Zhang and Davison, 1995; Harper et al., 1998; Hooda et al., 1999) A DGT device consists of a plastic assembly with a layer of resin embedded in gel with another layer of gel and a protective filter over it A DGT device is placed into a soil for a given period of time during which metal ions accumulate in the resin layer An effective concentration, CE, is determined based on the diffusive flux of the metal ions in the thin film (Zhang et al., 2001) A remarkably high correlation (R2 = 0.95) between Cu uptake by pepperwort (Lepidium heterophyllum cv Benth) and the CE (log transformed) in 29 soils was reported by

Zhang et al (2001) compared to the correlation with EDTA-extractable Cu (R2 = 0.55), Cu2+ activities (R2 = 0.67), and dissolved Cu (R2 = 0.85) in the soil solutions

Nolan et al (2005) studied Cd, Cu, Pb, and Zn uptake by wheat (Triticum aestivum L.)

in 13 metal contaminated soils and reported the correlations between metal uptake by wheat and various estimations of bioavailability, including CE, dissolved metal

concentration, metal activity, metal extractable with 0.01 M CaCl2, and total metal content Metal uptake was correlated the best with CE for Zn (R2 = 0.96) and Cd (R2 = 0.90), with dissolved metal concentrations for Pb (R2 = 0.95), and with the total metal for Cu (R2 = 0.87) The performance of CE was better than dissolved metal

concentrations for Zn and Cd, but not for Pb and Cu; the R2 values for CE vs

dissolved metal concentrations were 0.96 vs 0.86 for Zn, 0.90 vs 0.83 for Cd, 0.94

vs 0.95 for Pb, and 0.67 vs 0.69 for Cu This study indicates that there is no clear evidence that the DGT-technique consistently provides the best estimation of metal bioavailability to plants in soils

Koster et al (2005) also compared the DGT-based CE and the

conventional extraction methods in assessing Zn bioavailability to plants (grass-

Lolium perenne L., lettuce- Lactuca sativa L., and lupine- Lupinus nanus) and an

isopod (Oniscus asellus) in soils The results for field soils are shown in Table 2.1

Clearly, the DGT-method does not show a consistent improvement in the correlations

compared to CaCl2 extraction Given the observed correlations and the inconvenience

of the procedures, Koster et al (2005) concluded that the DGT approach offers no

additional advantage over the conventional extraction methods in predicting Zn

bioavailability in terrestrial systems Therefore, appropriateness of the DGT-technique

in assessing metal bioavailability in soils remains debatable

Table 2.1 The Coefficient of Determination for the regression between Zn

uptake rate of plants, Zn uptake by an isopod, and various

bioavailability estimations in field soils (extracted from Koster et al

(2005))

Organisms Measurements Grass Lettuce Lupine Isopod

0.01 M CaCl 2 -extractable Zn 0.79 0.77 0.81 0.91

DGT-based C E 0.71 0.87 0.70 0.65

0.43 M HNO 3 -extractable Zn 0.65 0.62 0.40

Pore water Zn concentration 0.42 0.81 0.91

An alternative approach using the free metal ion activities in the soil pore

waters has also been used in metal bioavailability studies (Sauvé et al., 1996 and

1998; McGrath et al., 1999; Dutta and Young, 2005; Hough et al., 2005) This

for metals toxicities in aquatic environment The basic assumption of this model is

that the active sites on an organism interact principally with the free metal ions and

this interaction can be thought of as a surface complexation reaction The biological

effect elicited is then correlated to the concentration of the metal bound active sites

Under the assumption that the metal binding to the active sites does not appreciably

affect its speciation, the biological effect is directly correlated to the free ion activity

of the cation in the solution Checkai et al (1987) reported that plant Cu content

appeared to correlate with its activity in solution cultures of tomato (Lycopersicon

esculentum) Sauvé et al (1996) showed that the Cu2+ activities determined in

CaCl2-extracted soil solutions provided a much better correlation (R2 values ranging from

0.76 to 0.85) than the total Cu concentration in soils (R2 values ranging from 0.45 to

0.76) with Cu uptake by lettuce (Lactuca sativa cv Buttercrunch, radish (Raphanus

rativa cv Cherry Belle) leaves and hypocotyls, and ryegrass (Lolium perenne cv

Barmultra) in urban contaminated soils Sauvé et al., (1998) even derived soil quality

criteria based on predicted Cu2+ and Pb2+ activities Based on available bioassay data

on plant biomass yield and microbial activities, the following relationships were

determined for Pb (Equation 2.1) and Cu (Equation 2.1):

0010239,

0.41,R

(13.5)210.1

pPb(1.5)19.5

(21.8)Inhibition

0.42,R

(8.8)151.2

pCu(1.0)13.2

(22.0)Inhibition

activities Although, there is a significant correlation between the endpoints and the

explained by the regressions Besides, treating these endpoints together is a gross simplification and provides no mechanistic or theoretical justification for the

specificity of the different organisms and processes with regard to the metal toxicities

Theoretical treatments of the FIAM has been reviewed by Campbell (1995) and Brown and Markich (2000) for aquatic systems and by Parker et al (1998) for hydroponic studies of higher plants These studies have recognized that the

exceptions to the FIAM are found in at least two situations: i when other cations (principally protons, Ca2+ and Mg2+) compete with the free metal ions to bind on the site of action on the organism (Di Toro et al., 2001; Santore et al., 2001), and ii metal species other than the free metal ion in question binds on the site of action (De

Schamphelaere and Janssen, 2002)

It is likely that the soil solution is the predominant route of metal exposure

to the organisms and the free metal ions in the soil solution are the most bioavailable species (McGrath, 1994; Dumestre et al., 1999; Saxe et al., 2001; Peijnenburg, 2002) However, the competitive interactions of cations incorporated in the aquatic biotic ligand model (BLM) (Di Toro et al., 2001; Santore et al., 2001) have been suggested for the terrestrial systems in the case of plants (Cheng and Allen, 2001; Kinraide et al., 2004; Zhao et al., 2006), invertebrates (Van Gestel and Koolhaas, 2004; Steenbergen

et al., 2005), and microbial activities/processes (Lofts et al., 2004; Smolders et al., 2004; Oorts et al., 2006) The aquatic BLM assumes the following:

i) The free metal ions in the aquatic environment are the most

bioavailable species and bind to the biotic ligand (BL), the site of action, in the organism and can cause a toxic effect