Sampling and analytical quality Control (SAX) for improved error estimation in the measurement of Pb in the environment using robust analysis of variance

doi:10.1016/S0883-2927(09)80027-9

Applied Geochemistry

Volume 8, Supplement 2, January 1993, Pages 149-153

https://doi.org/10.1016/S0883-2927(09)80027-9 Get rights and content

Abstract

Two geochemical surveys of Pb in the environment were assessed for random errors originating from the sampling and analytical procedures. The levels of measurement error that were considered acceptable were set by comparison with the geochemical variation in the individual survey, rather than at a predetermined absolute threshold. Robust analysis of variance was used to separate the analytical and sampling errors from the “true” geochemical variation in the survey areas. Robust statistics were shown to be much more reliable than classical statistics for such environmental surveys where there are usually a small number of outlying values. A limit of 20% was set as the maximum acceptable proportion of the total variance that could be contributed by the measurement processes. A stream sediment survey for Pb showed a 4.3% contribution from measurement variance, whereas an urban soil survey gave a corresponding value of 43.3%. The distortion evident in the geochemical information for the soil traverse confirms the suitability of this approach to Sampling and Analytical Quality Control (SAX).

References (12)

Robust statistics—how not to reject outliers. Part 1 Basic Concept
Analyst
(1989)
Robust statistics—how not to reject outliers. Part 2 Inter-laboratory Trials
Analyst
(1989)
M.J. Barcelona
Overview of the sampling process
C. Eisenhart
The assumptions underlying the analysis of variance
Biometrics
(1947)
R.G. Garrett
The determination of sampling and analytical errors in exploration geochemistry
Econ. Geol.
(1969)
R.G. Garrett et al.
The evaluation of sampling and analytical variation in regional geochemical surveys
Assoc. Explor. Geochem., Spec.
(1979)

There are more references available in the full text version of this article.

Cited by (44)

Spatial and geochemical aspects of heavy metal distribution in lacustrine sediments, using the example of Lake Wigry (Poland)
2020, Chemosphere
Citation Excerpt :
The obtained geochemical data was quality tested by way of an analysis of variance with Robust statistics method. It assumes that technical variance (combined sampling and analytical variances) for the geochemical data should not exceed 20% of the total variance (Ramsey et al., 1992; Ramsey, 1993). Technical variance for the Wigry Lake data did not exceed the limit value and amounted to 20% for Fe, 11% for Mn and 16% for Zn.
Heavy metals which pollute aquatic environments typically bond with bottom sediments and the analysis of the spatial distribution of metals allows to assess the geochemical purity of deposits and to identify the potential pollution sources. Research carried out on the Wigry Lake involved the collection of almost 500 samples of sediments, and the specification of the depth of their residence (0.2–71.4 m) as well as the level of concentration of three metals: Fe (80.3–32 857 mg kg⁻¹), Mn (17.8–1698 mg kg⁻¹) and Zn (3.14–632 mg kg⁻¹). The geochemical and bathymetric data was interpolated using geostatistical methods and mapped with the consideration of 5 types of sediments: lacustrine chalk, carbonate gyttja, fluvial-lacustrine sediment, organic gyttja and clastic sediment. As a result, a significant increase in the concentration of metals was revealed in deeper zones, at a considerable distance from the lake shore, wherein the respective values of correlation coefficients were as follows: depth–Mn 0.77; depth–Fe 0.60; depth–Zn 0.58. A strong dependency between the concentration of analysed metals and the type of sediment, attributed to the granular and chemical composition of sediments, was also revealed. Correlations between individual metallic pairs (Fe–Mn 0.77; Fe–Zn 0.80; Mn–Zn 0.75) indicated that similar factors influence spatial distribution of metals in sediments. The implementation of 3 different geochemical backgrounds allowed to conclude that the Wigry Lake is slightly polluted with the analysed metals, and that the origin of Mn is mainly natural, while in the case of Fe and Zn anthropogenic influence can also be identified.
Estimating the uncertainty from sampling in pollution crime investigation: The importance of metrology in the forensic interpretation of environmental data
2018, Forensic Science International
Citation Excerpt :
This is particularly troublesome when dealing with environmental samples where systematic and standardized procedures are more difficult to be carried on and, for this, require more research efforts. Before that, there were some articles on the subject of sampling uncertainty, mostly in the field of applied geochemistry (e.g. [6–10] until 2007, when two major publications were released, the NORDTEST Handbook “Uncertainty from sampling — A Nordtest handbook for sampling planners on sampling quality assurance and uncertainty estimation” [11] and the guidance upon which this is based, the EURACHEM/CITAC Guide “Measurement uncertainty arising from sampling: a guide to methods and approaches” [12]. After that, there were few publications addressing the topic of uncertainty of sampling in environmental investigations (e.g. [1,13–18]).
The forensic interpretation of environmental analytical data is usually challenging due to the high geospatial variability of these data. The measurements’ uncertainty includes contributions from the sampling and from the sample handling and preparation processes. These contributions are often disregarded in analytical techniques results’ quality assurance. A pollution crime investigation case was used to carry out a methodology able to address these uncertainties in two different environmental compartments, freshwater sediments and landfill leachate. The methodology used to estimate the uncertainty was the duplicate method (that replicates predefined steps of the measurement procedure in order to assess its precision) and the parameters used to investigate the pollution were metals (Cr, Cu, Ni, and Zn) in the leachate, the suspect source, and in the sediment, the possible sink. The metal analysis results were compared to statutory limits and it was demonstrated that Cr and Ni concentrations in sediment samples exceeded the threshold levels at all sites downstream the pollution sources, considering the expanded uncertainty U of the measurements and a probability of contamination >0.975, at most sites. Cu and Zn concentrations were above the statutory limits at two sites, but the classification was inconclusive considering the uncertainties of the measurements. Metal analyses in leachate revealed that Cr concentrations were above the statutory limits with a probability of contamination >0.975 in all leachate ponds while the Cu, Ni and Zn probability of contamination was below 0.025. The results demonstrated that the estimation of the sampling uncertainty, which was the dominant component of the combined uncertainty, is required for a comprehensive interpretation of the environmental analyses results, particularly in forensic cases.
Use of measurement uncertainty in a probabilistic scheme to assess compliance of bottled water with drinking water standards
2010, Journal of Geochemical Exploration
Citation Excerpt :
Therefore, the sampling procedure to be followed is to take the first primary water sample at a specified hour, and subsequently, after a suitable interval, but within the same hour, to collect the second primary water sample. Both water samples should be analysed in replicate in order to produce results according to a hierarchical nested balance design (AMC, 2009; Argyraki and Petrakaki, 2010; Johnson, 2011; Lee and Ramsey, 2001; Ramsey, 1997, 1993, 1998, 2009; Ramsey and Argyraki, 1997; Ramsey and Ellison, 2007; Ramsey et al., 1992). In this paper, the quadruple replicate analyses, performed at different periods on the same bottled water, were used in order to show the procedure, and in this case the main source of heterogeneity, and resulting uncertainty, was caused by the temporal changes of the geochemistry of water from the same brand of bottled water, and any effects due to the bottling material.
Ground water bodies are important resources for drinking water, including bottled water, and national regulatory bodies should assess their quality continuously. For this purpose, an effective assessment system of bottled water at source should be installed. A hierarchical nested balance design for the collection of random primary duplicate water samples, and their replicate analyses, is described, and the use of robust analysis of variance to estimate measurement uncertainty. The latter is subsequently used for the development of four probabilistic categories for the classification of element concentrations in bottled water with respect to legislative standard values, i.e., (a) compliant (below Lower Threshold Limit), (b) possibly non-compliant (possibly above Standard Value), (c) probably non-compliant (probably above Standard Value), and (d) non-compliant (above Upper Threshold Limit), for the reliable assessment of compliance to European Union and national drinking water standards. Overall, the quality of European bottled water is considered good, with the exception of a few that have concentrations in Mn, B, Ba, As, Fe, Ni, Se, and Al, which are definitely above the estimated respective Upper Threshold Limit and, thus, exceed the corresponding legislative standard value defined by European Union directives. National regulatory bodies should verify these results, and install an efficient assessment system of compliance to regulatory limits using the methodology described in this paper.
Measurement uncertainty in the determination of total petroleum hydrocarbons (TPH) in soil by GC-FID
2008, Chemometrics and Intelligent Laboratory Systems
In this investigation two methods were used for estimating the measurement uncertainty due to sampling and analysis of petroleum hydrocarbon contaminated soil. Analysis of variance (ANOVA) was used for type A evaluation of the measurement uncertainty. The results showed that the statistical evaluation of measurement uncertainty can be complicated by the log-normality and heteroscedasticity of the data. Although mathematical transformation of raw data is widely suggested for overcoming the discrepancy between data and ANOVA assumptions, its use results in problems with the interpretation of the ANOVA results at the original scale.
The measurement uncertainty was also estimated from the calculated precision equations for sampling and analysis. Comparison of measurement uncertainty values with the equivalent values obtained with ANOVA revealed that ANOVA overestimates the expanded uncertainty at both low and high TPH concentrations. Consequently, correct selection of the statistical analysis method needs comprehensive knowledge of the assumptions and limitations of statistical methods and careful consideration of the special characteristics (distribution, constancy of measurement variance) of the raw data as these may affect the validity of the estimated uncertainty. The expanded uncertainty obtained in this study for the results of TPH determinations with linear measurement precision modelling was moderate, ranging from 21% at a TPH concentration of 895 mg/kg to 9% at a TPH concentration of 10 019 mg/kg. If a single sample taken in a survey is analyzed only once, then the analytical variance contributes the most to the measurement variance, ranging from 68– 80% at a TPH concentration of 100–10 000 mg/kg.
Distribution of trace metals in the Odra River system: Water-suspended matter-sediments
2005, Limnologica
Extensive investigations of trace metals concentrations in water, suspended particulate matter (SPM) and bottom sediments of the whole Odra River system were carried out over the years 1997–2000. The vertical distribution of selected metals and their mobility were also studied in the sediment cores from upper and middle river sections. Significant levels of metal contamination were found. Median concentrations (Cd, Pb, Cu, Zn and As) in the SPM and sediments were (mg kg⁻¹) 7.1 and 8.9 Cd, 128 and 146 Pb, 81 and 119 Cu, 1198 and 1204 Zn, 48 and 54 As, respectively. The highest metal pollution of the Odra River solids was found with cadmium, zinc, lead and arsenic, showing high similarity in their frequency distributions in both SPM and sediments. Cd, Zn and As appear to be of particular concern because of the high levels, that appear to be bioavailable, and their high mobility. The exchangeable and carbonate chemical forms of Cd and Zn reached up to 50% of their total amount. Besides the determination of total metal concentration, the metal chemical forms in river solids were investigated. The results of very wide studies of the Odra River system through 4 years suggest that metal pollution decreased, especially for Zn, Pb and Cu. Among all metals studied in the Odra River sediments, substantial reductions of Cd contamination were observed neither in the period after ’97 flood, nor if compared with the earlier results obtained before ’97. No essential differences of the metal contents were observed among the samples for the same river compartment, from the same locality, taken within the five sampling campaigns. The pattern of spatial and vertical metal distributions in the river solids indicates that a variety of sources might be responsible for the contamination; very intensive, historical and current mining and smelting activities probably are the most important ones.
III.1 Assessment of pollution potential from solid waste
2004, Waste Management Series

View all citing articles on Scopus

View full text

Sampling and analytical quality Control (SAX) for improved error estimation in the measurement of Pb in the environment using robust analysis of variance

Abstract

Robust statistics—how not to reject outliers. Part 1 Basic Concept

Analyst

Robust statistics—how not to reject outliers. Part 2 Inter-laboratory Trials

Analyst

Overview of the sampling process

The assumptions underlying the analysis of variance

Biometrics

The determination of sampling and analytical errors in exploration geochemistry

Econ. Geol.

The evaluation of sampling and analytical variation in regional geochemical surveys

Assoc. Explor. Geochem., Spec.