1、Designation: D 7048 04Standard Guide forApplying Statistical Methods for Assessment and CorrectiveAction Environmental Monitoring Programs1This standard is issued under the fixed designation D 7048; the number immediately following the designation indicates the year oforiginal adoption or, in the ca
2、se of revision, the year of last revision. A number in parentheses indicates the year of last reapproval. Asuperscript epsilon (e) indicates an editorial change since the last revision or reapproval.1. Scope1.1 The scope and purpose of this guidance is to present avariety of statistical approaches f
3、or assessment, complianceand corrective action environmental monitoring programs.Although the methods provided here are appropriate and oftenoptimal for many environmental monitoring problems, they donot preclude use of other statistical approaches that may beequally or even more useful for certain
4、site-specific applica-tions.1.2 In the following sections, complete details of selectstatistical procedures used in assessment and corrective actionprograms for environmental monitoring (soil, ground water,air, surface water, and waste streams) are presented.1.3 The statistical methodology described
5、 in the followingsections should be used as guidance. Other methods may alsobe appropriate based on site-specific conditions or for moni-toring situations or media that are not presented in thisdocument.1.4 This practice offers an organized collection of informa-tion or a series of options and does
6、not recommend a specificcourse of action. This document cannot replace education,experience and professional judgements. Not all aspects of thispractice may be applicable in all circumstances. This ASTMstandard is not intended to represent or replace the standard ofcare by which the adequacy of a gi
7、ven professional servicemust be judged without consideration of a projects manyunique aspects. The word Standard in the title of this documentonly means that the document has been approved through theASTM consensus process.1.5 This standard does not purport to address all of thesafety concerns, if a
8、ny, associated with its use. It is theresponsibility of the user of this standard to establish appro-priate safety and health practices and determine the applica-bility of regulatory requirements prior to use.2. Referenced Documents2.1 ASTM Standards:2D 5092 Practice for Designing and Installation o
9、f GroundWater Monitoring Wells in AquifersD 5792 Practice for Generation of Environmental DataRelated to Waste Management Activities: Development ofData Quality ObjectivesD 6250 Practice for Deveration of Decision Points andConfidence Limits for Statistical Testing of Mean Concen-tration in Waste Ma
10、nagement DecisionsD 6312 Practice for Developing Appropriate Statistical Ap-proaches for Ground Water Detection Monitoring Pro-grams3. Terminology3.1 Definitions:3.1.1 assessment monitoringinvestigative monitoring thatis initiated after the presence of a contaminant has beendetected in ground water
11、above a relevant criterion at one ormore locations. The objective of the program is to determine ifthere is a statistical exceedance of a standard or criteria at aPotential Area of Concern (PAOC) or at the ground waterdischarging to surface water interface, and/or to quantify therate and extent of m
12、igration of constituents detected in groundwater above applicable criteria.3.1.2 compliance monitoringas specified under 40 CFR264.99, compliance monitoring is instituted when hazardousconstituents have been detected above a relevant criterion atthe compliance point during RCRA detection monitoring.
13、Ground-water samples are collected at the compliance point,facility property boundary, and upgradient monitoring wells foranalysis of hazardous constituents to determine if they areleaving the regulated unit at statistically significant concentra-tions above background.1This guide is under the juris
14、diction of ASTM Committee D18 on Soil and Rockand is the direct responsibility of Subcommittee D18.21 on Ground Water andVadose Zone Investigations.Current edition approved May 1, 2004. Published June 2004.2For referenced ASTM standards, visit the ASTM website, www.astm.org, orcontact ASTM Customer
15、Service at serviceastm.org. For Annual Book of ASTMStandards volume information, refer to the standards Document Summary page onthe ASTM website.1Copyright ASTM International, 100 Barr Harbor Drive, PO Box C700, West Conshohocken, PA 19428-2959, United States.3.1.3 corrective action monitoringunder
16、RCRA, correc-tive action monitoring is instituted when hazardous constitu-ents from a RCRA regulated unit have been detected atstatistically significant concentrations between the compliancepoint and the downgradient facility property boundary asspecified under 40 CFR 264.100. Corrective action moni
17、toringis conducted throughout a corrective action program that isimplemented to address ground-water contamination. At non-RCRA sites, corrective action monitoring is conductedthroughout the active period of corrective action to determinethe progress of remediation and to identify statistically sign
18、ifi-cant trends in ground-water contaminant concentrations.3.1.4 detection limit, DLthe true concentration at whichthere is a specified level of confidence (for example, 99 %confidence) that the true concentration is greater than zero.3.1.5 detection monitoringa program of monitoring forthe express
19、purpose of determining whether or not there hasbeen a release of a contaminant to ground water. Under RCRA,Detection Monitoring involves collection of ground-watersamples from compliance point and upgradient monitoringwells on a semi-annual basis for analysis of hazardous con-stituents of concern, a
20、s specified under 40 CFR 264.98. Resultsare evaluated to determine if there is a statistically significantexceedance of the ground-water protection criterion and/orbackground. At non-RCRA sites, monitoring is conducted in asimilar manner and results are compared to criteria to deter-mine if there is
21、 a statistically significant exceedance.3.1.6 direct push samplingground-water sampling con-ducted with a device that is temporarily pushed into the groundwith a hydraulic system or with a hammer. After ground-watersample collection, the device is removed from the ground.Examples include Geoprobet,
22、Hydropuncht direct push, andenvironmental soil probe.3.1.7 false negative ratethe rate at which the statisticalprocedure does not indicate contamination when contamina-tion is present.3.1.8 false positive ratethe rate at which the statisticalprocedure indates contamination when contamination is notp
23、resent.3.1.9 lognormal distributiona frequency distributionwhose logarithm follows a normal distribution.3.1.10 lower confidence limit, LCLa lower limit that has aspecified probability (for example, 95 %) of including the trueconcentration (or other parameter). Taken together with theupper confidenc
24、e limit, forms a confidence interval that willinclude the true concentration with confidence level thataccounts for both tail areas (for example, 90 %).3.1.11 lower prediction limit, LPLa statistical estimate ofthe minimum concentration that will provide a lower bound forthe next series of k measure
25、ments from that distribution, or themean of m new measurements for each of k sampling locations,with specified level of confidence (for example, 95 %).3.1.12 nonparametrica term referring to a statistical tech-nique in which the distribution of the constituent in thepopulation is unknown and is not
26、restricted to be of a specifiedform.3.1.13 nonparametric prediction limitthe largest (or sec-ond largest) of n background samples. The confidence levelassociated with the nonparametric prediction limit is a functionof n, m and k.3.1.14 normal distributiona frequency distribution whoseplot is a conti
27、nuous, infinite, bell-shaped curve that is sym-metrical about its arithmetic mean, mode and median (whichare numerically equivalent). The normal distribution has twoparameters, the mean and variance.3.1.15 outliera measurement that is statistically inconsis-tent with the distribution of other measur
28、ements from which itwas drawn.3.1.16 parametrica term referring to a statistical tech-nique in which the distribution of the constituent in thepopulation is assumed to be known.3.1.17 quantification limit, QLa lower limit on the con-centration at which quantitative determinations of an analytesconce
29、ntration in the sample can be reliably made duringroutine laboratory operating conditions. The QL is typicallydescribed quantitatively as the true concentration at which thesignal to noise ratio of measured concentration or instrumentresponse is 10:1. The signal to noise ratio is often determinedby
30、a percent relative standard deviation of 10 %.3.1.18 potential area of concernareas with a documentedrelease or likely presence of a hazardous substance that couldpose an unacceptable risk to human health or the environment.3.1.19 phase I environmental site assessmentnon-intrusive investigation that
31、 identifies PAOCs which may re-quire further investigation in subsequent phases of work.3.1.20 phase II environmental site assessment, ESIintrusive survey to confirm or deny existence of a release intothe environment at a PAOC at levels which may adverselyimpact public health or the environment.3.1.
32、21 upper confidence limit, UCLan upper limit that hasa specified probability (for example, 95 %) of including thetrue concentration (or other parameter). Taken together withthe lower confidence limit, the UCL forms a confidenceinterval that will include the true concentration with confidencelevel th
33、at accounts for both tail areas.3.1.22 upper prediction limit, UPLa statistical estimate ofthe maximum concentration that will not be exceeded by thenext series of k measurements from that distribution, or themean of m new measurements for each of k sampling locations,with specified level of confide
34、nce (for example, 95 %) basedon a sample of n background measurements.3.2 Symbols: = the true population mean of a constituentx= the sample-based mean or average concentration of aconstituent computed from n background measurements whichdiffers from because of sampling variability, and other errors2
35、= the true population variance of a constituents2= the sample-based variance of a constituent computedfrom n background measurementss = the sample-based standard deviation of a constituentcomputed from n background measurementsy= the mean of the natural log transformed data (also thenatural log of t
36、he geometric mean)D7048042sy= the standard deviation of the natural log transformeddatan = the number of background (offsite or upgradient) mea-surementsk = the number of future comparisons for a single monitor-ing event (for example, the number of downgradient monitor-ing wells multiplied by the nu
37、mber of constituents to bemonitored) for which statistics are to be computeda = the false positive rate for an individual comparison (thatis, one sampling location and constituent)m = the number of onsite or downgradient measurementsused in computing the onsite mean concentrationa* = the site-wide f
38、alse positive rate covering all samplinglocations and constituentst = the 100(1 a) percentage point of Studentst-distribution on n 1 degrees of freedomHL= the factor developed by Land (1971) (1)3to obtain thelower 100(a) % confidence limit for the mean of a lognormaldistributionHU= the factor develo
39、ped by Land (1971) (1) to obtain theupper 100(a) % confidence limit for the mean of a lognormaldistribution4. Summary of Guide4.1 The guide is summarized as Figs. 1-7. These figuresprovides a flow-chart illustrating the steps used in computingthe comparisons to regulatory or health based ground-wate
40、rprotection standard (GWPS) in assessment and correctiveaction environmental monitoring programs.5. Significance and Use5.1 The principal use of this standard is in assessment,compliance and corrective action environmental monitoringprograms (for example, for any facility that could potentiallyconta
41、minate ground water). The significance of the guidance isthat it presents a statistical method that allows comparison ofground-water data to regulatory and/or health based limits.5.2 Of course, there is considerable USEPA support forstatistical methods applied to detection, assessment and cor-rectiv
42、e action monitoring programs that can be applied toenvironmental investigations. For example, the 90 % upperconfidence limit (UCL) of the mean is used in SW846 (Chapter9) for determining if a waste is hazardous. If the UCL is lessthan the criterion for a particular hazardous waste code, thenthe wast
43、e is not a hazardous waste even if certain individualmeasurements exceed the criterion. Similarly, in the USEPAStatistical Analysis of Groundwater Monitoring Data at RCRAFacilities Addendum to the Interim Final Guidance (1992) (2),confidence intervals for the mean and various upper percentilesof the
44、 distribution are advocated for assessment and correctiveaction. Interestingly, both the 1989 and 1992 USEPA guidancedocuments (2, 3) suggest use of the lower 95 % confidencelimit (LCL) as a tool for determining whether a criterion hasbeen exceeded in assessment monitoring. The latest USEPAguidance
45、in this area (that is, the draft USEPA UnifiedStatistical Guidance) calls for use of the LCL in assessmentmonitoring and the UCL in corrective action. In this way,corrective action is only triggered if there is a high degree ofconfidence that the true concentration has exceeded the crite-rion or sta
46、ndard, whereas corrective action continues untilthere is a high degree of confidence that the true concentrationis below the criterion or standard. This is the general approachadopted in this guide, as well.5.3 There are several reasons why statistical methods areessential in assessment and correcti
47、ve action monitoring pro-grams. First, a single measurement indicates very little aboutthe true concentration in the sampling location of interest, andwith only one sample there is no way of knowing if themeasured concentration is a typical or an extreme value. Theobjective is to compare the true co
48、ncentration (or some intervalthat contains it) to the relevant criterion or standard. Second, inmany cases the constituents of interest are naturally occurring(for example, metals) and the naturally existing concentrationsmay exceed the relevant criteria. In this case, the relevantcomparison is to b
49、ackground (for example, off-site soil orupgradient ground water) and not to a fixed criterion. As such,background data must be statistically characterized to obtain astatistical estimate of an upper bound for the naturally occur-ring concentrations so that it can be confidently determined ifonsite concentrations are above background levels. Third, thereis often a need to compare numerous potential constituents ofconcern to criteria or background, at numerous samplinglocations. By chance alone there will be exceedances as thenumber of comparisons becomes
