1、BRITISH STANDARDBS ISO 5667-20:2008Water quality Sampling Part 20: Guidance on the use of sampling data for decision making Compliance with thresholds and classification systemsICS 13.060.45g49g50g3g38g50g51g60g44g49g42g3g58g44g55g43g50g56g55g3g37g54g44g3g51g40g53g48g44g54g54g44g50g49g3g40g59g38g40g
2、51g55g3g36g54g3g51g40g53g48g44g55g55g40g39g3g37g60g3g38g50g51g60g53g44g42g43g55g3g47g36g58BS ISO 5667-20:2008This British Standard was published under the authority of the Standards Policy and Strategy Committee on 30 April 2008 BSI 2008ISBN 978 0 580 55352 3National forewordThis British Standard is
3、 the UK implementation of ISO 5667-20:2008.The UK participation in its preparation was entrusted by Technical Committee EH/3, Water quality, to Subcommittee EH/3/6, Sampling (of technical committee EH/3 Water quality).A list of organizations represented on this committee can be obtained on request t
4、o its secretary.This publication does not purport to include all the necessary provisions of a contract. Users are responsible for its correct application.Compliance with a British Standard cannot confer immunity from legal obligations.Amendments/corrigenda issued since publicationDate CommentsRefer
5、ence numberISO 5667-20:2008(E)INTERNATIONAL STANDARD ISO5667-20First edition2008-03-15Water quality Sampling Part 20: Guidance on the use of sampling data for decision making Compliance with thresholds and classification systems Qualit de leau chantillonnage Partie 20: Lignes directrices relatives l
6、utilisation des donnes dchantillonnage pour la prise de dcision Conformit avec les limites et systmes de classification BS ISO 5667-20:2008ii iiiContents Page Foreword iv Introduction . vi 1 Scope . 1 2 Summary of key points 1 3 Types of error and variation 2 3.1 General. 2 3.2 Analytical error 3 3.
7、3 Overall uncertainty . 3 4 Activities 4 4.1 Estimation of summary statistics . 4 4.2 Thresholds for water quality and compliance . 6 4.3 Confidence of failure 7 4.4 Methods for thresholds expressed as percentiles 7 4.5 Non-parametric methods . 10 4.6 Look-up tables 13 5 Definition of thresholds 14
8、5.1 General. 14 5.2 Ideal thresholds 14 5.3 Absolute limits 15 5.4 Percentage of failed samples 18 5.5 Calculating limits for effluent discharges 18 6 Declaring that a substance has been detected . 19 7 Detecting change 20 8 Classification. 23 8.1 General. 23 8.2 Confidence that class has changed 25
9、 Annex A (informative) Calculation of confidence limits. 27 Annex B (informative) Calculation for the binomial distribution. 29 Annex C (informative) Sample results with high error or reported as less than a limit of detection 32 Bibliography . 34 BS ISO 5667-20:2008iv Foreword ISO (the Internationa
10、l Organization for Standardization) is a worldwide federation of national standards bodies (ISO member bodies). The work of preparing International Standards is normally carried out through ISO technical committees. Each member body interested in a subject for which a technical committee has been es
11、tablished has the right to be represented on that committee. International organizations, governmental and non-governmental, in liaison with ISO, also take part in the work. ISO collaborates closely with the International Electrotechnical Commission (IEC) on all matters of electrotechnical standardi
12、zation. International Standards are drafted in accordance with the rules given in the ISO/IEC Directives, Part 2. The main task of technical committees is to prepare International Standards. Draft International Standards adopted by the technical committees are circulated to the member bodies for vot
13、ing. Publication as an International Standard requires approval by at least 75 % of the member bodies casting a vote. Attention is drawn to the possibility that some of the elements of this document may be the subject of patent rights. ISO shall not be held responsible for identifying any or all suc
14、h patent rights. ISO 5667-20 was prepared by Technical Committee ISO/TC 147, Water quality, Subcommittee SC 6, Sampling (general methods). ISO 5667 consists of the following parts, under the general title Water quality Sampling: Part 1: Guidance on the design of sampling programmes and sampling tech
15、niques Part 3: Guidance on the preservation and handling of water samples Part 4: Guidance on sampling from lakes, natural and man-made Part 5: Guidance on sampling of drinking water from treatment works and piped distribution systems Part 6: Guidance on sampling of rivers and streams Part 7: Guidan
16、ce on sampling of water and steam in boiler plants Part 8: Guidance on the sampling of wet deposition Part 9: Guidance on sampling from marine waters Part 10: Guidance on sampling of waste waters Part 11: Guidance on sampling of groundwaters Part 12: Guidance on sampling of bottom sediments Part 13:
17、 Guidance on sampling of sludges from sewage and water treatment works Part 14: Guidance on quality assurance of environmental water sampling and handling Part 15: Guidance on preservation and handling of sludge and sediment samples Part 16: Guidance on biotesting of samples BS ISO 5667-20:2008v Par
18、t 17: Guidance on sampling of bulk suspended solids Part 18: Guidance on sampling of groundwater at contaminated sites Part 19: Guidance on sampling of marine sediments Part 20: Guidance on the use of sampling data for decision making Compliance with thresholds and classification systems The followi
19、ng parts are under preparation: Part 21: Guidance on sampling of drinking water distributed by non-continuous, non-conventional means Part 22: Guidance on design and installation of groundwater sample points Part 23: Determination of significant pollutants in surface waters using passive sampling BS
20、 ISO 5667-20:2008vi Introduction This part of ISO 5667 concerns the use of information on water quality obtained by taking samples in taking decisions in measuring success, failure or change, in the context of the inevitable uncertainties associated with sampling. This part of ISO 5667 provides guid
21、ance on controlling the risk of such uncertainties leading to non-optimal decisions. Non-optimal decisions can also stem from the way in which thresholds for discharges and targets for environmental waters are formulated or set out in regulations and permits. This part of ISO 5667 also examines the
22、problems caused when compliance with these thresholds is assessed using data obtained by sampling. This part of ISO 5667 aims to ensure that future laws, regulations, and guidance assert the requirement to assess and report statistical significance. NOTE 1 Decisions might result in the commendation
23、or criticism of people, sites, companies, sectors or nations. Decisions can give rise to legal action and/or expensive and time-consuming remedial actions to improve water quality. Figure 1 shows the links between the following topics: a) the setting up of thresholds for taking decisions on the need
24、 to improve water quality, possibly including criteria to minimize water quality deterioration; b) the establishment of sampling programmes to satisfy the requirements of these thresholds and the need to assess performance against them; c) making use of the outcome of sampling programmes to take dec
25、isions. Figure 1 Links between topics associated with sampling and taking decisions BS ISO 5667-20:2008viiThis part of ISO 5667 deals with topic c). Topics a) and b) are huge and wide ranging in their own right, and their detailed treatment lies outside the scope of this part of ISO 5667. Neverthele
26、ss, this part of ISO 5667 does make recommendations for the expression of targets and thresholds for water quality topic a), which are important when using sample data to take decisions. This part of ISO 5667 also gives advice on what is required for sampling programmes topic b) in order that they b
27、e compatible with the way thresholds are defined, and so as to place no unnecessary difficulties and errors in the process of taking decisions. Other areas which lie outside the scope of this part of ISO 5667 are: the detailed mechanics of taking and handling samples; assurance that samples are repr
28、esentative over time of the body of water being sampled; and performance of chemical analyses on samples. These are all covered in other documents. Nonetheless, if poorly obtained results from these areas can add substantially to overall sampling uncertainties and impose extra difficulties in taking
29、 decisions. This part of ISO 5667 describes some of these extra difficulties. This part of ISO 5667 does not cover the full range of statistical techniques that may be applied and the circumstances in which they should be used. The main purpose is to establish the principle that uncertainty from sam
30、pling and analysis (and errors generally) should always be assessed and taken into account as part of the process of taking decisions. If this is not done, incorrect decisions can result, for example, on where action is needed, and the scale of that action. NOTE 2 Some statistical techniques are use
31、d as illustrative examples. These are techniques that have seen routine use in some regulatory regimes that take proper account of statistical uncertainties. They are suitable for use in situations that resemble the worked examples discussed. It is not the purpose of this part of ISO 5667 to direct
32、the development of regulatory conditions. This part of ISO 5667 provides principles and tools to support management, including regulation. It is recognised that regulatory thresholds are developed using a range of strategies that incorporate technical, social and legal considerations. It is also rec
33、ognised that tools other than statistical data analysis are likely to be used in interpreting and applying thresholds. BS ISO 5667-20:2008blank1Water quality Sampling Part 20: Guidance on the use of sampling data for decision making Compliance with thresholds and classification systems 1 Scope This
34、part of ISO 5667 establishes principles, basic requirements, and illustrative methods for dealing with the use of sample data for decision making based on the assessment of the confidence that water quality: a) meets targets and complies with thresholds; b) has changed; and/or c) lies in a particula
35、r grade in a classification system. This part of ISO 5667 also specifies methods for preliminary examination of the sensitivity of decisions to error and uncertainty, although it does not cover the full range of statistical techniques. This part of ISO 5667 provides general advice on decision making
36、 related to constraint formulation for expression of thresholds and targets and the form and scale of sampling programmes. NOTE 1 In the water industry, “standard” is commonly used to indicate the value or limit of a parameter of interest. However, in this part of ISO 5667, the term “threshold” is u
37、sed to avoid confusion with published national, regional, and International Standards. NOTE 2 This document is framed in terms of sampling and measurement of chemical concentrations, in particular those subject to strong day-to-day temporal variations. The principles apply, however, to any item esti
38、mated by sampling which is subject to random error, including microbiological and biological data, and data subject to strong spatial variations. 2 Summary of key points Water quality is often assessed by the results of chemical analysis of a number of samples taken over a period of time. Uncertaint
39、y is introduced by the action of random chance in taking samples. It can be present in any set of measurements of water quality taken over a period of time. The values for chemical analysis of these samples depend on the quality of the particular small volumes of water that are extracted or measured
40、. If water quality varies in space or time, a second set of samples taken over the same period will have different values because these samples are made up of different small volumes of water taken at different times. Each set of samples allows an estimate of the true water quality. These estimates
41、will differ: they will have a different mean and span a different range. They have the potential, if taken at face value, to suggest different conclusions about compliance with thresholds and targets. Sampling uncertainty (or sampling error) is the term often given to this effect. Sampling uncertain
42、ty includes uncertainties and errors associated with chemical analysis, and occurs even in the case of trivial errors in chemical analysis and if there are no mistakes in the methods by which samples are taken and handled. BS ISO 5667-20:20082 Sampling uncertainty is reduced if more samples are take
43、n, but the scale of the uncertainty is often unappreciated. In this part of ISO 5667, “overall uncertainty” includes these chance sampling effects and all the other sources of variation in a set of samples. This variability reflects the underlying signals generated by natural or perhaps unnatural pr
44、ocesses; it includes the effects of errors in chemical analysis and the handling of samples. It might contain systematic variations from trends and diurnal, weekly, and seasonal cycles. In this context, the more appropriate term is “overall uncertainty”, “overall error” or “total assay error” (ISO/I
45、EC Guide 99:19935). Overall uncertainty should be quantified, at least approximately, and taken into account in all cases where water quality varies and sampling is used to estimate information used in decision making. This includes assessing compliance with thresholds (see Clause 5), deciding wheth
46、er water quality has changed (see Clause 7), and putting waters into grades in classification systems (see Clause 8). This part of ISO 5667 recommends that: a) thresholds for which compliance is assessed by sampling should be defined or used so that the overall uncertainty can be estimated and dealt
47、 with appropriately (see 5.2); b) thresholds defined as absolute limits should be treated as percentiles when assessing compliance using sampling (see 5.3); c) thresholds defined as limits to be met by a percentage of samples should be defined or used as the corresponding percentiles (see 5.4); d) t
48、he degree of confidence should be estimated when assessing compliance with thresholds (see Clause 4); and, e) the degree of confidence in changes or differences should be estimated when aiming to demonstrate change or no change (see 8.2). 3 Types of error and variation 3.1 General In many procedures
49、 by which sample data are used to take decisions, there is a set of results taken over a period of time (e.g. a year). This information might be used to make such judgements as whether: a) water quality in a river failed to meet required thresholds; b) a treatment works performed better this year than last; c) water quality in a lake needs improvement; d) one company has better effluent discharge compliance than another; or e) most of the risk of environmental impact is from a particular type of effl
