The practicalities and pitfalls of establishing a policy-relevant and cost-effective soil biological monitoring scheme

Authors


Abstract

A large number of biological indicators have been proposed over the years for assessing soil quality. Although many of those have been applied in monitoring schemes across Europe, no consensus exists on the extent to which these indicators might perform best and how monitoring schemes can be further optimized in terms of scientific and policy relevance. Over the past decade, developments in environmental monitoring and risk assessment converged toward the use of indicators and endpoints that are related to soil functioning and ecosystem services. In view of the proposed European Union (EU) Soil Framework Directive, there is an urgent need to identify and evaluate indicators for soil biodiversity and ecosystem services. The recently started integrated project, Ecological Function and Biodiversity Indicators in European Soils (EcoFINDERS), aims to address this specific issue within the EU Framework Program FP7. Here, we 1) discuss how to use the concept of ecosystem services in soil monitoring, 2) review former and ongoing monitoring schemes, and 3) present an analysis of metadata on biological indicators in some EU member states. Finally, we discuss our experiences in establishing a logical sieve approach to devise a monitoring scheme for a standardized and harmonized application at European scale. Integr Environ Assess Manag 2013; 9: 276–284. © 2013 SETAC

INTRODUCTION

Soils are a crucial part of the Earth's system and play fundamental roles in its functioning, linking the atmosphere, the subsurface, and the aquatic compartments. As asserted by the Soil Framework Directive (SFD) proposal (EU 2006), soil is a nonrenewable resource that performs key environmental, social, and cultural functions that are vital to human life and the sustainability of global ecosystems. Essential environmental services delivered by soils include: nutrient cycling, C storage and turnover, water retention, regulation of soil structure, resistance to pests and diseases, and regulation of aboveground diversity. These services result from the belowground functioning of trophic and biochemical interactions between organisms (i.e., microbes, fauna, and plant roots). Progress made over the last decade confirmed that there is a wide range of soil microbial diversity and activity that remains mostly unexplored (Curtis et al. 2002), and there is growing evidence that the role of soil invertebrates and their interactions with soil microbes are essential to ecosystem functioning (Wardle 2002; Bardgett 2005; Mulder et al. 2011).

A major challenge is to link soil biodiversity with soil functioning and to assess how soil biodiversity contributes to the delivery of ecosystem services. Some major relationships among soil biodiversity, soil functioning, and ecosystem services are given in Figure 1, which include threats to soil biodiversity. Hence, there is a requirement within SFD to identify and evaluate indicators for soil biodiversity and ecosystem services. Thus, our objectives here are to: 1) discuss the use of the concept of ecosystem services in ecological monitoring for soils, 2) review former and ongoing monitoring schemes, 3) present an analysis of metadata on biological indicators, and 4) outline what is needed to establish a logical sieve approach for devising a future monitoring scheme.

Figure 1.

Relationships between soil biodiversity, ecosystem functioning, and ecosystem services (modified after Brussaard et al. 2007a).

This article is rather about the process toward indicators, with practicalities and pitfalls en route, than about the final results.

THE CONCEPT OF ECOSYSTEM SERVICES IN ECOLOGICAL MONITORING FOR SOILS

The majority of soil processes are mediated by the soil biota, and although a mechanistic understanding of the relationship between soil biodiversity and ecosystem functioning is still sought (Bardgett et al. 2005), biological indicators should be considered for policy evaluation (Francaviglia 2004) and are already in use, mainly from an ecotoxicological perspective (Becaert and Deschenes 2006; Feld et al. 2009). The application of biological indicators at a European scale was first mentioned in an Organization for Economic Co-Operation and Development report (OECD 2002) on agri-biodiversity measures to ascertain the impacts of agriculture on biodiversity. However, the concept of biological indicators has strengthened in its application, and biological measures are now used as indicators of impact and change both to biodiversity itself and associated ecosystem services (CBD 2010).

In this article, the concept of ecosystem services is explicitly used to derive the most suitable assessment parameters to tailor site investigations to land managers' and stakeholders' goals in land use (Faber et al. 2006; Faber and van Wensem 2011; Rutgers and Jensen 2011). The approach will work in 2 ways: the land-manager gains insight into which ecosystem services are beneficial and desirable, and thus the required soil management, whereas the risk assessor can remain focused on which indicators have to be applied (Breure et al. 2005; Thomsen et al. 2012).

The concept of ecosystem services can be used as a guiding principle in environmental quality assessment (Figure 2). A crux is the identification of relevant indicators of ecosystem services, through recognition of essential structures and processes that are key in the delivery of these services. For example, in 1997 the Biological Indicator for Soil Quality (BISQ) was designed in the Netherlands to assess Life Supporting Functions for soil biota in relation to soil texture type and land use (Schouten et al. 1997; Rutgers et al. 2009; Mulder et al. 2011). In a recent inventory, soil attributes at 4 farms were linked to ecosystem services via consultation with soil professionals and stakeholders (Rutgers et al. 2012). The performance of ecosystems services could plausibly be linked to the management characteristics of the farms, i.e., organic, conventional, or intensive. However, a further scientific underpinning of ecosystem service quantification schemes is required.

Figure 2.

Derivation of relevant indicators for evaluation of ecosystem service provision in environmental quality assessment, through ecological requirements that are conditional for sustainable land use of specific type. (A) Literature toxicity data for relevant indicators may be used for derivation of soil quality criteria for sustainable land use; (B) In site-specific ecological risk assessment the indicator status may be assessed through bioassays or field inventories, if vulnerable to the particular stressor in case; (C) indicators for monitoring may be selected with respect to relevant ecosystem services required in association to land use (modified from Faber and Van Wensem 2011).

PAST AND PRESENT SOIL MONITORING SCHEMES

Previous research at the European level has provided a selection of indicators for the characterization of soil, including studies into the decline in soil biodiversity (Bispo et al. 2009), but many other European projects and studies have also looked at biological parameters as indicators of soil quality (Römbke and Breure 2005).The selection of potential biological indicators is only a step in developing practical procedures (Doran and Zeiss 2000), as there are operational issues to be solved, such as ease of application, sensitivity, interlaboratory comparability, throughput, costs, descriptiveness, and communicability. Previous experiences may now be integrated and finalized. One of the outcomes of the EU FP7 project Ecological Function and Biodiversity Indicators in European Soils (EcoFINDERS) (http://ecofinders.dmu.dk/) will be to identify potential bioindicators of biodiversity and ecosystem function applicable at a European scale.

The use of soil organisms, both in terms of their community structure or trophic function, for an evaluation of soil quality is not a novel approach (historical overviews by Wardle [2002] and Breure et al. [2005]). Today, the aforementioned BISQ concept seems to be among the most advanced soil monitoring approaches. BISQ contains 200 repeatedly monitored locations on 10 combinations of soil texture type and land use, covering both agricultural and natural sites. Besides environmental parameters (land use, land history, soil parameters, physicochemical characteristics, climate factors, etc.), bacterial community diversity and functions, as well as free-living nematodes, collembolans, mites, enchytraeids, and earthworms have been monitored (Mulder et al. 2003; Rutgers et al. 2008). The aims of the BISQ project can be summarized as follows:

  • Baseline establishment of abundance and diversity of soil organisms under different land uses and soil textures in Dutch soils

  • Estimation of the impact of land management regimes (livestock, pesticides, liming, tillage) on soil biodiversity

  • Identification of reference state of soils with good quality regarding life support functions and ecosystem services

The latter aspect is given emphasis via the contribution of soil biota to support a set of ecosystem services (Mulder et al. 2011) but also via other (abiotic) system characteristics (Rutgers et al. 2012).

In parallel to national activities, the European Union (EU) supported the Framework Programme (FP6) project Environmental Assessment of Soil for Monitoring (ENVASSO) to design a single, integrated, and operational set of indicators for EU-wide application. A proposed set of suitable indicators for monitoring soil biodiversity (Bispo et al. 2007, 2009) was selected from a literature review and an inventory of national monitoring programs. Whereas the literature review allowed the identification of approximately 100 potential indicators, the inventory of existing monitoring networks showed that few indicators were actually used. Stringent criteria were applied to select appropriate indicators and to improve practicability and flexibility, and a 3-tiered system was developed, meaning that different sets of indicators were identified depending on purpose (aim and conditions of a specific monitoring program) (Table 1). The basic set of indicators (Tier I) would be applied in all cases and, depending on availability of resources and specific requirements, can be extended to include Tiers II and III. The potential adoption of a European-wide soil biodiversity monitoring scheme requires a minimum set of indicators, and for this reason Tier I was established. The 3 indicators were selected to act as surrogate measures for an overall decline in biodiversity and related functions. As the 3 indicators represent different taxonomical groups, functional levels and body sizes, they would differ considerably in terms of exposure to chemical or physical stressors. Finally, according to the literature review carried out, the amount of information available for these 3 indicators was by far the highest of all indicators studied. When using soil biodiversity indicators, methods have to be standardized at least for Tier I, preferably according to International Organization for Standardization (ISO) standards (Römbke et al. 2006). In addition, it was stressed that baselines (i.e., reference values) for the diversity and abundance of (at least) the 3 basic indicator groups must be defined in Europe (Gardi et al. 2009). As a starting point, ranges published for selected land use and/or soil texture categories in the Netherlands (Rutgers et al. 2009) or Germany (Römbke et al. 2012) may be used.

Table 1. Priority level of indicators to monitor the decline in soil biodiversity (ENVASSO)a
Key issueGroups of speciesTier I: (for all samples)Tier II (if relevant for specific issues and resources available)Tier III (optional)
  • PLFA = phospholipid-derived fatty acids.

  • a

    ENVASSO, ENVironmental ASsessment of Soil for mOnitoring. Project funded as Scientific Support to Policy (SSP) under the European Commission 6th Framework Programme (Contract 022713, 2006-8).

Species diversityMacrofaunaEarthwormsAll macrofauna 
 MesofaunaCollembolan species; Enchytraeids if no earthwormsMites suborders 
 Microfauna Nematode diversity (e.g., trophic groups)Protista
 Microflora Bacterial and fungal diversity (e.g., PLFA) 
 Vegetation  For grassland and pastures
Biological functionsMacrofauna  Macrofauna activity
 Mesofauna  Mesofauna activity (e.g., litter bags or bait lamina)
 MicrofloraSoil respirationBacterial and fungal activity 

Tier II indicators are recommended for more intensive studies looking at the effects of a stressor. However, these indicators include organism groups that require expert identification, such as soil micro-arthropods. Such increased effort is only required when specific sites or questions have to be addressed. At Tier III, indicators integrating the activity of the whole soil community are used—an approach though ecologically relevant is likely to be less sensitive in establishing effects.

Each indicator selection (cf. soil biodiversity: Schouten et al. 1997; Bispo et al. 2009; Ritz et al. 2009) is subject to context specificity in terms of time and place, professional involvement, stakeholder preferences, budgetary restrictions, and specific objectives and requirements of the monitoring network (Table 2). The EU-FP7-project EcoFINDERS addresses this context specificity in a structured approach by:

  • exploring and evaluating metadata of soil biodiversity indicators in Europe

  • combining high-throughput molecular biological and conventional methods to assess soil biodiversity across Europe

  • linking soil biodiversity to ecosystem services using state of the art ecological theory and modeling

  • focusing on standardization and applicability of soil biodiversity indicators

Table 2. List of criteria by which indicators would be assessed and the guidelines used during the assessment.
CriteriaMeasured byLow scoreHigh score
  1. SOP = standard operation procedure.

MeasurableLab equipmentVery few labs have the equipment neededAll labs would be able to carry out the work.
 Specialist skillsSpecialist skills are neededGeneral skills would suffice.
Cost efficiencyCapital start-upMore than €100 000Less than €2000
 Cost per sampleMore than €100Less than €2
 Labor needed in the laboratoryHigh labor demandLow labor demand
 Labor needed in the fieldHigh labor demandLow labor demand
Policy relevanceFor “Biodiversity”  
 For “Ecosystem Services”  
SensitivityTo soil typeNo response or idiosyncratic responseThe indicator responds characteristically to change
 To land useNo response or idiosyncratic responseThe indicator responds characteristically to change
 To disturbanceNo response or idiosyncratic responseThe indicator responds characteristically to change
Use as an indicator Not in use alreadyIn use already
Fit for use as an indicatorMeaningful  
 StandardizedMethods are not ready for general use (i.e., low experience, no SOPs available)Methods are ready for general use or are already in general use.
 Spatio-temporally relevant  
 UnderstandableWill the indicator be useful in a policy development situationWill the indicator be useful in a policy development situation

CURRENT USE OF INDICATORS IN EUROPE

Many initiatives for soil monitoring with biodiversity indicators have been proposed in the literature, and several reviews have been published (Winder 2003; Morvan et al. 2008; Feld et al. 2009; Gardi et al. 2009; Jeffrey et al. 2010; Ranjard et al. 2010; Griffiths et al. 2011), but a comprehensive synthesis is not available. Thus questionnaires were sent from the EcoFINDERS project to all partners and other contacts in EU (national contact points for soil data from the European Soil Bureau Network, soil ecologists, and ecotoxicologists; at least one person per member state) to collect data on soil biodiversity monitoring activities in Europe (Table 3). Few questionnaires were returned demonstrating the general lack of structured soil biodiversity monitoring in Europe. In a second stage, additional data were welcomed also from scientific studies on soil biodiversity when data were obtained from a range of natural and managed soil systems (not experimental systems). Additional data were obtained from literature searching for the keywords “soil, biodiversity, and monitoring.” All data were collated, providing an overview on the use of particular soil biodiversity indicators within specific programs, and cross-referenced information, such as number of soil samples analyzed, geographic coverage, contact data, and literature references. A total of 135 entries uncovered information of approximately 14 200 unique soil biodiversity measurements (Figure 3).

Figure 3.

Soil biodiversity indicators used in approximately 14.200 measurements in European soils. The indicators were grouped in four categories: soil fauna (blue), microbes (yellow), soil processes (red) and ‘omics’ including multi endpoint indicators like PLFA(green). amoA = amoA genes in archaea (coding for the alpha-subunit of the ammonia monooxygenase; ARISA = automated rRNA intergenic spacer analysis fingerprints; ITS = internally transcribed spacer sequences; PLFA = phospholipid-derived fatty acids; SIR = substrate-induced respiration; t-RFLP = terminal restriction fragment length polymorphism.

This overview is quite incomplete as no stringent criteria were used to exclude data, and the effectiveness to encourage persons to return questionnaires was unequally distributed over Europe and research institutes. Furthermore not all soil biodiversity survey initiatives are published in the scientific or “gray” literature. A summary of the scattered results shows that, when amalgamated into groups, soil biodiversity indicators have been applied on a significant number of replicated locations: fauna, 4860; microbes, 3360; soil processes, 4230; and approximately 1900 “omics” tools and other multi-endpoint indicators (regarding various soil processes through different functional groups). Multi-endpoint indicators and omics tools have a shorter history than the more classic indicators, which may explain their limited application so far. However, developments in these techniques are fast, so that potential for future use is large where functional relationships can be established.

The finding that all these soil biodiversity indicators have been used on many occasions (Figure 3) suggests that ideas about the usefulness of these indicators for soil biodiversity monitoring are not consolidated. There is no consensus about a basic set of soil biodiversity indicators and all soil biodiversity indicators have useful characteristics (e.g., sensitivity, ease of use, ecological meaning, etc.) that prevail under particular circumstances. Some are used relatively often. Nematodes have been used most frequently as a faunal indicator, soil microbial biomass (including substrate-induced respiration) is the most used microbial indicator, basal respiration (including potential C-mineralization) is most used for soil processes, and bacterial automated rRNA intergenic spacer analysis (B-ARISA) is most often used as a multi-endpoint tool depicting various soil processes through different functional groups (Figure 3). Such frequently used indicators may originate as a result of inclusion in large programs, e.g., nematodes in the Netherlands (3 monitoring cycles starting from 1993) (Rutgers et al. 2009), and B-ARISA in France (Dequiedt et al. 2009, 2011).

Although all soil biodiversity indicators have been used extensively, their use across Europe is rather heterogeneous (Figure 4). To achieve a European-wide approach, harmonization of soil monitoring initiatives is highly recommended. The selection of indicators within the FP7 EcoFINDERS project aims to contribute here by establishing a policy-relevant and cost-effective set of indicators.

Figure 4.

Use of soil biodiversity indicators across Europe. The surface area of a pie represents the total number of soil biodiversity measurements performed. The soil biodiversity indicator applied is depicted as percentage in four categories, i.e., soil fauna (blue), microbes (yellow), soil processes (green) and ‘omics’ and multi-endpoint techniques (red).

THE LOGICAL SIEVE APPROACH

The logical sieve assessment proposed by Ritz et al. (2009) was used as a rationale to derive a range of indicators for the purpose of monitoring soil biodiversity and ecosystem function. This approach allows a structured discrimination of biological indicator methods through a series of queries with regard to their potential within monitoring. To apply the concept laid out in Ritz et al. (2009) a number of steps must be taken:

  • Establishment of the purpose for which the monitoring will be applied; e.g., in EcoFINDERS the application of soil biological indicators will be used to monitor changes in soil biodiversity and ecosystem function across Europe

  • Listing of potential biological indicators—this has been derived from a wide range of sources including literature, past European and national scale studies, and assessment of indicators in national monitoring networks

  • Classification of indicators into a range of categories that include morphological, microbial assay, molecular, and functional techniques

The indicators should meet and be scored against 6 general criteria (OECD 2002; UNEP 2003; Ritz et al. 2009; Turbé et al. 2010), the specific definitions of which needed to be modified to make them relevant to EcoFINDERS (Table 2). Thus they should be:

  • 1)Measurable, which was related to the availability of the necessary laboratory equipment and technical skills
  • 2)Cost-effective, when looking at capital and consumable costs as well as the labor intensiveness in the field and the laboratory
  • 3)Policy relevant, to provide information on biodiversity and ecosystem function
  • 4)Sensitive, in that there was a characteristic but not idiosyncratic response to soil texture, land use and disturbance
  • 5)Fit for use, in that it was meaningful, standardized, spatio-temporally relevant, and understandable
  • 6)Currently in use in a soil monitoring scheme (and so has benchmarks available)

A complete list of potential indicators is prohibitively long, more than 180 described by Ritz et al. (2009). A more manageable short-list was selected from the analysis of metadata (described above) plus suggestions from a collaborative workshop of experts within the EcoFINDERS project. To meet the purpose of the monitoring scheme (i.e., the EcoFINDERS project) indicators were short-listed for relevance to ecosystem services (namely water retention, C sequestration, and nutrient provision). This included process measurements, organism-based assays, and biodiversity (that included organism-based assays as well as developing molecular biology tools). 3

Table 3. Excerpt of the questionnaires distributed to persons (one or more per country) in soil ecology, the soil ecotoxicology group of the Society of Environmental Toxicology and Chemistry (SETAC), and the FP7 EcoFINDERS project consortium.
We would like to define what data is currently available across Europe in terms of indicators of biodiversity and ecosystem functions/services of soils. Can you please indicate for your country/institution which data is available?
1What data is available from your country (e.g., for which organism groups)?
2Is the data a national database?
3At what scale is the data recorded/available?
4What is the current status of the data?
 Available ………. ________
 Metadata available………. ________
 Summary output of indicators available…………. ________
5What sampling and analysis methods were applied?
6Can you provide a summary description of the database?
7Can you provide a contact name for this database?
8Do you know of any databases from countries which that do not have representatives in EcoFINDERS?

STANDARDIZATION AND HARMONIZATION OF METHODS ACROSS EUROPE

One of the biggest constraints in European scale biological monitoring is the application of standard methods across Europe. Although many generic biological methods exist, many are adapted to local conditions or constraints of the laboratory performing the analysis. This nonuniformity can be differentiated as follows:

  • 1.Process-based measurements, usually performed in the laboratory (e.g., microbial biomass, respiration). Many of these methods have long been in use and ISO standards are available for most.
  • 2.Identification of microbial or invertebrate taxa, in particular with genetical methods. This is a relatively new, but rapidly developing field, where few methods have been standardized (and some may soon be in danger of being outdated).
  • 3.Sampling and extraction procedures (partly in the field, partly in the laboratory). These methods have been used for many years, and have been modified often. ISO standards exist for several years, but acceptance is a problem due to the fact that comparability with old data is compromised.
  • 4.Design of monitoring studies. This aspect is probably of the highest impact to the outcome of sampling programs, but no specific guideline is possible because campaign design depend on the aim of the monitoring. However, basic rules for the design of such studies have been established in 2 guidance articles (ISO 2004, 2010).

A problem with many biological methodologies is a lack of reference control materials, which would allow for direct comparison between and within laboratories, i.e., assessment of the repeatability and reproducibility of the method. Although such control materials are standard in soil chemistry methods (Carter and Gregorich 2007), in biological measurements this is rarely applied due to the nature of the measurement (most biological methods require “fresh” soil material). Instead, the efficiency and reliability of a method relative to others can be determined by application on the same plot at the same time or by using subsamples of the same soil batch. Such comparative information is required as part of the development of international standard methodologies and is being carried out as part of many collaborative projects at national (Creamer et al. 2009) and European (e.g., EcoFINDERS) level.

In policy evaluation schemes, there is increasing favor for the application of ISO methodologies for biological indicators. Methods are now available for a spectrum of soil biological measurements (Römbke et al. 2006; Philippot et al. 2012). These methods should be applied consistently across Europe and should be accompanied by specific standard operation procedures (SOPs) and training courses.

DISCUSSION

Although soil ecosystem services are mostly supporting services, to some stakeholder groups it might be expedient for risk assessment and soil quality monitoring schemes to focus on ecosystem services indirectly related to soil, i.e., the provisioning, regulating, and cultural services. The regulation of water and climate clearly are services that are crucially associated with soil quality. Here, key biota with clear and quantifiable relation to the mediation of soil processes underlying these services are best adopted as indicators.

A second factor to inform stakeholders and decision makers is the possibility for valuation of ecosystem services, in a monetary sense or otherwise. The indicators selected for testing in EcoFINDERS will be evaluated for cost-effectiveness; not only does an indicator need to work biologically, it also has to represent good value for money. The subsequent implicit valuation of relevant ecological indicators may enhance the support for ecological risk assessment and biodiversity monitoring results. Economic valuation will also facilitate the use of the outcome from site-specific risk assessment in cost–benefit analysis. This also applies for policy-relevant indicators.

The concept of ecosystem services allows for the simultaneous assessment of different stressors affecting their suitability for use. To this extent, we expect that indicators that are vulnerable to the various stressors involved will be useful. Thus, the concept and the aforementioned approach for selection of assessment endpoints will meet the requirements put forward by developments in soil policy, e.g., the EU Soil Strategy, and will be tested in EcoFINDERS.

Conclusions

  • 1.Ecological risk assessment and soil biological monitoring show similarities in the selection of endpoints and indicators that relate to soil functioning and the provision of ecosystem services. Harmonization in approaches will benefit communication about assessments using biological indicators.
  • 2.The use of indicators that are clearly linked to ecosystem services is more relevant to stakeholders in society, and scientific conclusions based on that use therefore will meet greater acceptance and make greater impact in decision making (i.e., are more policy-relevant).
  • 3.Present use of soil biological indicators in monitoring schemes is highly variable across Europe and would benefit from standardization of sampling and identification methods, preferably on the basis of ISO guidelines or guidance articles.
  • 4.Modifying the selection criteria, such as in the Logical Sieve approach, is necessary to come up with an objectively filtered short list.

EDITOR'S NOTE

This paper is one of 8 articles generated from the SETAC Special Symposium: Ecosystem Services, from Policy to Practice (15–16 February 2012, Brussels, Belgium). The symposium aimed to give a broad overview of the application of the ecosystem services concept in environmental assessment and management, against the background of the implementation of the European environmental policies such as the biodiversity agenda, agricultural policy, and the water framework directive.

Acknowledgements

This work was supported by the European Union within the projects EcoFINDERS (FP7-264465) and NOMIRACLE (FP6 contract 003956), and by the Dutch Secretary of Agriculture, Nature Conservation, and Food Quality within the strategic research program “Sustainable spatial development of ecosystems, landscapes, seas and regions” (KB-14-002-022). The metadatabase was constructed with the generous help of: Arnold Arnoldussen, Mark Bailey, Richard Bardgett, Csaba Csuzdi, Ciro Gardi, Mariangela Girlanda, Jakub Hofman, Sigbert Huber, Anna Hug, Philippe Lemanceau, Visa Nuutinen, Guénola Peres, Olaf Schmidt, Sari Timonen, Anne Winding, and Sophie Zechmeister-Boltenstern. Figure 1 was reproduced with permission of Lijbert Brussaard. Thanks are also due to 2 anonymous reviewers for their encouragement and constructive criticism.

Ancillary