1、raising standards worldwideNO COPYING WITHOUT BSI PERMISSION EXCEPT AS PERMITTED BY COPYRIGHT LAWBSI Standards PublicationPD CEN/TR 16151:2011W a t e r quality Guidance on the design of Multimetric IndicesPD CEN/TR 16151:2011 PUBLISHED DOCUMENTNational forewordThis Published Document is the UK imple
2、mentation of CEN/TR 16151:2011.The UK participation in its preparation was entrusted to T e c h n i c a l C o m m i t t e e E H / 3 / 5 , B i o l o g i c a l M e t h o d s .A list of organizations represented on this committee can be obtained on request to its secretary.This publication does not pur
3、port to include all the necessary provisions of a contract. Users are responsible for its correct application. BSI 2011 ISBN 978 0 580 71891 5 ICS 13.060.45 Compliance with a British Standard cannot confer immunity from legal obligations.This Published Document was published under the authority of t
4、he Standards Policy and Strategy Committee on 30 April 2011.Amendments issued since publicationDate T e x t a f f e c t e dPD CEN/TR 16151:2011TECHNICAL REPORT RAPPORT TECHNIQUE TECHNISCHER BERICHT CEN/TR 16151 April 2011 ICS 13.060.45 English Version Water quality - Guidance on the design of Multim
5、etric Indices Qualit de leau - Lignes directrices pour la conception des indices multimtriques Wasserbeschaffenheit - Anleitung zur Planung und Erstellung Multimetrischer Indices This Technical Report was approved by CEN on 27 December 2010. It has been drawn up by the Technical Committee CEN/TC 230
6、. CEN members are the national standards bodies of Austria, Belgium, Bulgaria, Croatia, Cyprus, Czech Republic, Denmark, Estonia, Finland, France, Germany, Greece, Hungary, Iceland, Ireland, Italy, Latvia, Lithuania, Luxembourg, Malta, Netherlands, Norway, Poland, Portugal, Romania, Slovakia, Sloven
7、ia, Spain, Sweden, Switzerland and United Kingdom. EUROPEAN COMMITTEE FOR STANDARDIZATION COMIT EUROPEN DE NORMALISATION EUROPISCHES KOMITEE FR NORMUNG Management Centre: Avenue Marnix 17, B-1000 Brussels 2011 CEN All rights of exploitation in any form and by any means reserved worldwide for CEN nat
8、ional Members. Ref. No. CEN/TR 16151:2011: EPD CEN/TR 16151:2011CEN/TR 16151:2011 (E) 2 Contents Page Foreword 3Introduction .41 Scope 52 Terms and definitions .53 Principle 64 Procedure .74.1 General 74.2 Selection of Candidate Metrics 84.3 Exclusion of redundant metrics .84.4 Definition of Upper a
9、nd Lower Anchors 84.5 Transformation into a 0 to 1 score .94.6 Selection of core metrics 94.7 Combination of Core Metrics to a Multimetric Index indicating a single stressor or indicating general degradation (general multimetric approach) 94.8 Combination of Core Metrics to a Multimetric Index separ
10、ating the impact of different stressors 10Annex A (informative) Examples for metrics used to assess individual Biological Quality Elements, assigned to metric types 11Bibliography . 12PD CEN/TR 16151:2011CEN/TR 16151:2011 (E) 3 Foreword This document (CEN/TR 16151:2011) has been prepared by Technica
11、l Committee CEN/TC 230 “Water analysis”, the secretariat of which is held by DIN. Attention is drawn to the possibility that some of the elements of this document may be the subject of patent rights. CEN and/or CENELEC shall not be held responsible for identifying any or all such patent rights. PD C
12、EN/TR 16151:2011CEN/TR 16151:2011 (E) 4 Introduction Multimetric Indices are among the commonly used tools for classification of the quality of fresh water and brackish water ecosystems (rivers, lakes, transitional waters, wetlands). A Multimetric Index combines several individual metrics, the resul
13、ts of which are finally combined into a Multimetric result. Thus Multimetric Indices integrate several attributes of a community (“metrics”) to describe and assess condition. Different categories of metrics (e.g. taxa richness, share of sensitive and tolerant species, trophic structure) reflecting d
14、ifferent environmental conditions are combined into one Multimetric Index. Multimetric Indices can be applied to different aquatic ecosystems (rivers, lakes, transitional waters, wetlands) and to different Biological Quality Elements (fish, benthic invertebrates, macrophytes, phytoplankton, phytoben
15、thos). They are flexible in terms of the composition of metrics, since different metrics are suited for the assessment of different ecosystems or different stressors. In recent years, a wide variety of Multimetric Indices has been developed and is now being applied, particularly for the purpose of i
16、mplementing the Water Framework Directive. It can be expected that many existing Multimetric Indices will be adapted and many new ones will be developed within the next years. To enhance comparability between Multimetric assessment systems the procedure of developing and applying a Multimetric Index
17、 is described. PD CEN/TR 16151:2011CEN/TR 16151:2011 (E) 5 1 Scope This document describes methods for developing and applying Multimetric Indices used for assessing rivers, lakes, transitional waters or wetlands. It is suitable for use with data on fish, benthic invertebrates, macrophytes, phytopla
18、nkton, and phytobenthos. 2 Terms and definitions For the purposes of this document, the following terms and definitions apply. 2.1 Anchors (Upper and Lower Anchors) values of a metric, which are empirically set and defined as “1” (Upper anchor) and “0” (Lower anchor), respectively, for transferring
19、a metrics result into a 0 to 1 score NOTE The Upper Anchor relates to the reference value (i.e. the metrics value under reference conditions). The Upper Anchor can be calculated from the median or mean of reference samples or by other appropriate statistical methods as described in 4.3. The Lower An
20、chor is related to the lower limit of the metrics value under the worst ecological quality conditions. 2.2 fresh water or brackish water type (river type, lake type, transitional water type) division into an ecologically meaningful entity of sites with limited biotic and abiotic variation and a reco
21、gnisable discontinuity with other types NOTE Fresh water or brackish water types serve as “units“, to which an assessment system can be applied. 2.3 metric measurable part or process of a biological system empirically shown to change in value along a gradient of human influence 2 NOTE It reflects sp
22、ecific and predictable responses of the community to human activities, either to a single factor or to the cumulative effects of all events and activities within a watershed. 2.4 metric type metrics addressing comparable aspects of a community, regardless of the stressor to which the metrics are res
23、ponding NOTE The following metric types can be distinguished (see Annex A): composition / abundance metrics: all metrics giving the share of a taxon or taxonomic group in relation to the total number of individuals counted; all metrics giving the abundance of a taxon or taxonomic group; metrics comp
24、aring reference and observed taxa (e.g. similarity indices); richness / diversity metrics: all metrics giving the number of taxa within a certain taxon (including the total number of taxa), all diversity indices; sensitivity / tolerance metrics: all metrics giving the ratio of taxa sensitive and ins
25、ensitive to stress in general or to a certain stress-type, either using presence/absence or abundance information; functional metrics: all metrics addressing the characteristics of taxa other than their taxonomic definition (biological or ecological traits, ecological guilds): feeding types, habitat
26、 preferences, ecosystem type preferences, current preferences, life-history parameters, body-size parameters; they can be based on taxa abundance or richness. PD CEN/TR 16151:2011CEN/TR 16151:2011 (E) 6 2.5 Multimetric Index combination of the results of three or more metrics 2.6 stressor category o
27、f direct or indirect human impact to a fresh water, which potentially influences the composition and / or abundance of stream biota NOTE The following stressors can be distinguished: organic pollution: organic matter input induced by human activities; eutrophication: nutrient input induced by human
28、activities; acid stress: permanently or temporarily decreased pH value due to human activities; temperature stress; toxic stress: effects of toxic contaminants released by human activities; degradation in stream morphology: bed and bank alteration, habitat degradation, riparian land use, straighteni
29、ng, migration barriers, siltation; hydrological stress: alteration of flow regime, e.g. residual flow, pulse releases; general degradation: simultaneous and inseparable impacts of more than one stressor. 2.7 stressor gradient set of sites of a fresh water ecosystem type with a varying intensity of a
30、 stressor 3 Principle Two ways of calculating Multimetric Indices can be distinguished: the “general approach” and the “stressor-specific approach”. In the “general approach”, various metrics are calculated from a taxa list. The metric results are individually compared to the respective metric value
31、s under reference conditions. From this comparison, a score for each metric is determined. These scores are finally combined into a Multimetric Index (Figure 1). The “stressor-specific” approach sorts a suite of metrics according to their ability to detect a certain stressor. Thus, the scores of the
32、 metrics addressing a single stressor are first combined into a value reflecting the intensity of this stressor. The assessment results for all stressors are finally combined into the Multimetric Index (Figure 2). Both ways of calculating Multimetric Indices may have advantages in certain situations
33、: The “general approach”, carefully applied, provides an overview of a water bodys status and is, thus, mainly suited if the specific effects of individual stressors on the targeted organism group are not known in detail. It can, for example, be applied for the general ecological quality assessment
34、and for intercalibration purposes. The “stressor specific approach” can only be applied if precise information on the effects of different stressors (e.g. acidification, organic pollution) on the targeted organism group are known and it is most suited for investigative monitoring purposes, tailored
35、towards the identification of alteration causes. The results of a Multimetric Index can be viewed at different levels: at the upper level there is the Ecological Quality Class, at the second level (in case of the stressor specific approach) are the results of the stressor specific modules (quality c
36、lasses “organic pollution” and “stream morphology degradation” in Figure 2) and at the third level the results of the individual metrics are produced. PD CEN/TR 16151:2011CEN/TR 16151:2011 (E) 7 Figure 1 The “general approach” of multimetric assessment Figure 2 “Stressor-specific approach” of multim
37、etric assessment 4 Procedure 4.1 General The procedure of developing a Multimetric Index is composed of the following steps: Selection of Candidate Metrics; Exclusion of Redundant Metrics; Definition of Upper and Lower Anchors; Transformation of Core Metrics into a 0 to 1 score; Selection of Core Me
38、trics; PD CEN/TR 16151:2011CEN/TR 16151:2011 (E) 8 Combination of Core Metrics to a Multimetric Index. 4.2 Selection of Candidate Metrics Select only those metrics showing a quantitative dose-response change across a stressor gradient that is reliable, interpretable and not dispersed or obscured by
39、natural variation. Selecting Candidate Metrics consists of the following steps: a) Compilation of a data set of the freshwater type the Multimetric Index will apply to. This dataset shall include data on the Biological Quality Element that the Multimetric Index will use, for different sites covering
40、 the widest range of alteration. b) If the required information is available, metrics of all metric types should be calculated (composition / abundance metrics, richness / diversity metrics; sensitivity / tolerance metrics; functional metrics). c) Definition of a stressor gradient within the dataset
41、 by: 1) using abiotic data on the individual sites, describing the impact of a single stressor (e.g. data on BOD5or on oxygen content for a Multimetric Index addressing organic pollution; 2) using abiotic data on the individual sites, describing the impact of general degradation (e.g. data on catchm
42、ent land use; hydromorphological modification and water pollution combined); 3) defining a gradient within an ordination space of the taxonomic composition of the Biological Quality Element reflecting “general degradation” or the major single stressor. The stressor gradient can either be defined cov
43、ering the whole range from “high” to “bad” quality (preferred option), or by just defining a part of it. It can be defined either in five classes (“high”, “good”, “moderate”, “poor”, “bad”), or as “unstressed sites” and “stressed sites” or as a continuous gradient. d) Correlating the results of a me
44、tric to the stressor gradient. If the stressor gradient has been defined in quality classes, t-Test, Mann-Whitney U test or rank correlation should be used. If the stressor gradient is defined as a continuous gradient, Pearsons R or Spearmans R should be used, or any appropriate statistical method.
45、Using the appropriate test a Candidate Metrics results must show a significant correlation to the stressor gradient. This correlation can be positive or negative, either across the whole stressor gradient measured or for a part of the stressor gradient measured (e.g. only moderate to high quality si
46、tes). Metrics fulfilling this criterion are in principle suitable for the assessment of degradation of the fresh water ecosystem type and can be selected as Candidate Metrics. 4.3 Exclusion of redundant metrics Two Candidate Metrics belonging to the same Metric Type with a comparable ecological sign
47、ificance and delivering similar results within a data set, should not be simultaneously included into a Multimetric Index. To avoid this, a correlation matrix of all metrics selected as Candidate Metrics needs to be produced. If two analogous metrics have a Spearmans R or Pearsons R of 0,8 one of th
48、ose metrics should be excluded from the Multimetric Index, preferably the one with the poorest correlation with the stressor gradient. 4.4 Definition of Upper and Lower Anchors The Upper Anchor is coherent to the reference value of the metric, i.e. a robust and predictable metrics value under refere
49、nce conditions. Reference conditions might be set site specifically (e.g. using a predictive system) or type specifically and relate to the site or types condition if they were undisturbed by stressors. If numerous data on reference sites are available, the Upper Anchor can be set as a percentile (25 % or 10 %) of all metric values of the reference sites. If few data are available for reference sites the most robust statistic, i.e. the median or the mean, should be used t
