Integration of bioavailability, ecology and ecotoxicology by three lines of evidence into ecological risk indexes for contaminated soil assessment

https://doi.org/10.1016/j.scitotenv.2007.08.032Get rights and content

Abstract

A Weight of Evidence approach was applied to define three integrated effect indexes estimating the impairment on terrestrial ecosystems caused by the stressor(s) of concern. According to a Triad approach, the integrated effect indexes combined the information provided by the measurement endpoints of each line of evidence (chemistry/bioavailability, ecology and ecotoxicology) and allowed to analyse the impairment degree highlighted by each measurement endpoint as difference from the reference condition. Multi-Criteria Decision Analysis (MCDA) was used for the aggregation of the complementary Triad information, including expert judgement and a weighted procedure based on the endpoint sensitivity and the sensitivity of the test for ecosystem effects. The developed methodology was implemented in the DSS-ERAMANIA, Module 2, and is presented in this paper as “Integrated Effect Indexes” (IEI) sub-module. The latter has been preliminary applied to the Acna di Cengio (Italy) contaminated site; the results of this application are presented and discussed.

Introduction

Contaminated sites and megasites have become an issue of major concern due to their worldwide occurrence and the related extensive environmental and socio-economic impacts (EEA (European Environment Agency), 2000, CLARINET Contaminated Land Rehabilitation Network for Environmental Technologies, 2002). In order to address the rehabilitation of contaminated ecosystems, Ecological Risk Assessment (ERA) is the appropriate process for identifying environmental quality objectives and the ecological aspects of major concern (US-EPA — U.S. Environmental Protection Agency, 1998, Suter et al., 2000). Within the risk characterization step of ERA procedure, the application of Weight of Evidence (WOE) methods (Burton et al., 2002a, Chapman et al., 2002), was recently proposed, to determine possible ecological impacts based on multiple Lines of Evidence (LOE) (US-EPA, 1998). A WOE incorporates judgements concerning the quality, extent, and congruence of the data contained in the different LOE (Menzie et al., 1996). It includes both observational (e.g. ecologically based) and investigative components (e.g. toxicological determinations of cause-and-effect relationships for chemical and/or physical stressors) (Burton et al., 2002b). As one of the WOE approaches, the Triad requires three major lines of evidence for an accurate assessment of the effects of contamination: one based on chemical contaminant characterization including literature toxicity data to estimate the effects; one based on laboratory toxicity to surrogate organisms; the third one concerning indigenous biota community characterization (Long and Chapman, 1985). The Triad approach was mainly applied as Sediment Quality Triad (SQT), a procedure developed by Long and Chapman (1985) to assess the quality of sediments. However, recently a Triad has been defined also for the assessment of soil quality (Rutgers and Den Besten, 2005). In this approach the information provided by the three LOE (i.e. chemistry, ecology and ecotoxicology), is unique and complementary: the combination of components (i.e. each LOE) is necessary because no individual component provides comprehensive information (Chapman, 1990). Each LOE in the Triad is aiming at providing information on the risk (or effect) by means of appropriate measurement endpoints. For instance, environmental chemistry provides information on total contamination and contaminant bioavailability; ecotoxicology provides direct evidence of soil toxicity related to contamination, but the included tests are often conducted under laboratory conditions and may not be directly applicable to in situ conditions; and finally ecology provides usually direct evidence of contaminant-related effects in the environment but only if non-pollution-related effects (e.g. competition, soil type, temperature) can be neglected (Chapman, 1990). Moreover, allowing to proceed through subsequent investigation levels (i.e. tiers), Triad enables risk assessment to be stopped as soon the provided answer is satisfactory being affected by a relatively small uncertainty, thus ensuring an adequate financial investment (Rutgers and Den Besten, 2005).

The Triad and other WOE approaches can be qualitative or quantitative (Buset, 2004). Qualitative approaches are defined as the “simple” combination of various LOE results in a non-quantitative manner. Agreement or disagreement among the different LOE are evaluated but no quantitative risk estimation is provided (Burton et al., 2002a, Burton et al., 2002b).

Quantitative approaches include the elaboration of the results from different LOE in order to address the evaluation of the ecosystem impairment. These approaches can be further divided into two main categories, approaches based on matrices and approaches based on indices, although approaches using both matrices and indices have been proposed (Buset, 2004). The main advantage of matrix-based approaches is the high degree of sensitivity, applicability and transparency (Burton et al., 2002a). As far as indices are concerned, a single index can be calculated for each LOE or an overall index can integrate information provided by all different LOE. Chapman (2000) criticized the use of a single index value, because it can lead to an oversimplification of available information and can hide the results derived from independent LOE. Use of indices has been criticized also by Burton et al. (2002a), because it results in information compression. On the other hand, this approach has the great advantage to facilitate the communication to non-professional stakeholders.

To support the experts in both quantitative and qualitative evaluation of the multiple information provided by a Triad-based ERA, the decision support tool “Integrated Ecological Risk Indexes” was developed and implemented by applying Multi-Criteria Decision Analysis (MCDA). It is the second module of the Decision Support System DSS-ERAMANIA, based on a framework structured in three subsequent tiers of investigation, as presented elsewhere by Semenzin et al. (2007a). In that paper the modular structure of the DSS-ERAMANIA is reported, where Module 1, “Comparative Tables”, includes the sub-modules: Comparative Tools Table for bioavailability (BAV), Comparative Observations Table for ecology (ECO) and Comparative Tests Table for ecotoxicology (ETX), and Module 2, “Integrated Ecological Risk Indexes”, includes the sub-modules: Integrated Effect Indexes (IEI), Ecosystem Impairment Matrix (EcoIM) and Global ecosystem impairment Evaluation Matrix (GEM).

In general, a Decision Support System (DSS) is a computer-based system including integration of different information, modelling capability and incorporation of multiple stakeholders perspectives to identify suitable and wide-shared solutions (Loucks, 1995, Shim et al., 2002). The MCDA methods are used within DSSs because of their capacity of drawing attention to similarities or potential areas of conflict among stakeholders or experts with different views, and of building specific values functions and aggregation functions as discussed in Giove et al. (2007).

The site-specific risk assessment framework developed for the DSS-ERAMANIA and its first module (i.e. Module 1), structured in three Comparative Tables (one for each Triad LOE) and concerning the tests (i.e. measurement endpoints) comparison and selection within each Triad LOE, were presented in Critto et al. (2007) and Semenzin et al. (2007b). The second Module (i.e. Module 2), composed of three sub-modules: IEI, EcoIM and GEM, integrates the results of the multiple measurement endpoints (belonging to the three LOE) selected by Module 1.

The overall objective of this paper is to present the first part of Module 2 of the DSS-ERAMANIA, composed of “Integrated Effect Indexes” (i.e. IEI) designed to quantitatively aggregate the complementary information obtained by the application of measurement endpoints. Moreover, the preliminary application of IEI sub-module to the Acna di Cengio contaminated site (located in Savona province, Italy) is presented and discussed.

The second part of the Module 2, concerning the qualitative assessment of impairment on ecosystem biodiversity and functional diversity (i.e. EcoIM and GEM) will be addressed by another paper (Semenzin et al., in preparation).

Section snippets

Methods

A MCDA-based methodology was developed to calculate Integrated Effect Indexes (IEIs). These indexes aim at quantifying the terrestrial ecosystem impairment by integrating the complementary information obtained by the measurement endpoints belonging to each Triad LOE. Specifically, the methodology includes two procedures: the Endpoint's Impairment Analysis and the procedure for calculating the Integrated Effect Indexes (IEIs).

Results and discussion

The normalization and aggregation procedure presented in the Methods section offers a structured solution to the process of integrating and evaluating the results of different tests in order to obtain a risk assessment estimation and characterization. By including expert judgement and Multi-Criteria Decision Analysis (MCDA) it allows to normalize each test result and to weight the highlighted impairment by taking into account the relevance of the response for the ecosystem. The process is

Conclusions

In order to estimate the impairment occurring on the terrestrial ecosystem of concern, due to soil contamination, a new quantitative methodology for integrating multiple information provided by the Triad LOE was developed. It allows to aggregate the results from measurements endpoints belonging to each LOE into Integrated Effect Indexes (IEIs) by means of Multi-Criteria Decisional Analysis (MCDA) methods. Particular attention was paid to the scaling phase of the process, where different experts

Acknowledgement

This work and the ERA-MANIA project have been funded by the Italian Government Commissary for the rehabilitation of the Bormida Valley and were developed by University of Venice, Ca' Foscari, Interdepartmental Centre for Sustainable Development (IDEAS) in collaboration with Italian National Environmental Protection Agency (APAT), Dutch National Institute for public Health and Environment (RIVM) and Consortium Venezia Ricerche (CVR).

We thank Prof. Silvio Giove for proposing and including MCDA

References (45)

  • ANPA Agenzia Nazionale per la Protezione dell'Ambiente

    Guida tecnica sui metodi di analisi per il suolo e i siti contaminati — Utilizzo di indicatori biologici ed ecotossicologici. RTI CNT_TES 1/2004

    (2004)
  • BertaG. et al.

    Morphogenetic modifications induced by the mycorrhizal fungus Glomus strani E3 on the root system of Allium porrum L.

    New Phytol

    (1990)
  • BurtonG.A. et al.

    Weight-of-evidence approaches for assessing ecosystem impairment

    Hum Ecol Risk Assess

    (2002)
  • BurtonG.A. et al.

    Weight-of-evidence framework or assessing sediment (or other) contamination: improving certainty in the decision-making process

    Hum Ecol Risk Assess

    (2002)
  • Buset M. Weight of Evidence e l'Analisi di Rischio Ecologico. Thesis for Environmental Sciences degree, Ca' Foscari...
  • CesaroP. et al.

    Phytotoxicity test to be used in ecological risk assessment of heavy metal polluted soils

  • ChapmanP.M.

    The sediment quality triad: then, now and tomorrow

    Int J Environ and Pollut

    (2000)
  • ChapmanP.M. et al.

    Weight of evidence frameworks for sediment quality (and other) assessment

    Hum Ecol Risk Assess

    (2002)
  • CLARINET Contaminated Land Rehabilitation Network for Environmental Technologies

    Sustainable management of contaminated land: an overview

    A report from the Contaminated Land Rehabilitation Network for Environmental Technologies

    (2002)
  • GastaldiL. et al.

    Evaluation of soil toxicity using a battery of stress biomarkers on the earthworm Eisenia andrei

  • GioveS. et al.

    Decision Support Systems for the management of contaminated sites: a multi-criteria approach

  • Cited by (51)

    • Soil Biological Quality index (QBS-ar): 15 years of application at global scale

      2018, Ecological Indicators
      Citation Excerpt :

      The major aim of those studies was to test the effects of forest cutting, grazing, trampling, industrial activities, emission, agriculture, heavy metals and other anthropogenic disturbance on soil quality. Semenzin et al. (2008) inserted QBS-ar in a study aimed to define three integrated effect indexes estimating the impairment on terrestrial ecosystems caused by stressors of concern. The increasing number of applications offers the opportunity to assess the potential of this index to evaluate soil quality.

    View all citing articles on Scopus
    View full text