External Quality-Assurance Results for the National Atmospheric Deposition Program/ National Trends Network and Mercury Deposition Network, 2004

View or download the PDF
(1.8 mb)

Abstract

The U.S. Geological Survey (USGS) used five programs to provide external quality-assurance monitoring for the National Atmospheric Deposition Program/National Trends Network (NADP/NTN) and two programs to provide external quality-assurance monitoring for the NADP/Mercury Deposition Network (NADP/MDN) during 2004. An intersite-comparison program was used to estimate accuracy and precision of field-measured pH and specific-conductance. The variability and bias of NADP/NTN data attributed to field exposure, sample handling and shipping, and laboratory chemical analysis were estimated using the sample-handling evaluation (SHE), field-audit, and interlaboratory-comparison programs. Overall variability of NADP/NTN data was estimated using a collocated-sampler program. Variability and bias of NADP/MDN data attributed to field exposure, sample handling and shipping, and laboratory chemical analysis were estimated using a system-blank program and an interlaboratory-comparison program.

In two intersite-comparison studies, approximately 89 percent of NADP/NTN site operators met the pH measurement accuracy goals, and 94.7 to 97.1 percent of NADP/NTN site operators met the accuracy goals for specific conductance. Field chemistry measurements were discontinued by NADP at the end of 2004. As a result, the USGS intersite-comparison program also was discontinued at the end of 2004.

Variability and bias in NADP/NTN data due to sample handling and shipping were estimated from paired-sample concentration differences and specific conductance differences obtained for the SHE program. Median absolute errors (MAEs) equal to less than 3 percent were indicated for all measured analytes except potassium and hydrogen ion. Positive bias was indicated for most of the measured analytes except for calcium, hydrogen ion and specific conductance. Negative bias for hydrogen ion and specific conductance indicated loss of hydrogen ion and decreased specific conductance from contact of the sample with the collector bucket.

Field-audit results for 2004 indicate dissolved analyte loss in more than one-half of NADP/NTN wet-deposition samples for all analytes except chloride. Concentrations of contaminants also were estimated from field-audit data. On the basis of 2004 field-audit results, at least 25 percent of the 2004 NADP/NTN concentrations for sodium, potassium, and chloride were lower than the maximum sodium, potassium, and chloride contamination likely to be found in 90 percent of the samples with 90-percent confidence.

Variability and bias in NADP/NTN data attributed to chemical analysis by the NADP Central Analytical Laboratory (CAL) were comparable to the variability and bias estimated for other laboratories participating in the interlaboratory-comparison program for all analytes. Variability in NADP/NTN ammonium data evident in 2002-03 was reduced substantially during 2004. Sulfate, hydrogen-ion, and specific conductance data reported by CAL during 2004 were positively biased. A significant (a = 0.05) bias was identified for CAL sodium, potassium, ammonium, and nitrate data, but the absolute values of the median differences for these analytes were less than the method detection limits. No detections were reported for CAL analyses of deionized-water samples, indicating that contamination was not a problem for CAL.

Control charts show that CAL data were within statistical control during at least 90 percent of 2004. Most 2004 CAL interlaboratory-comparison results for synthetic wet-deposition solutions were within ±10 percent of the most probable values (MPVs) for solution concentrations except for chloride, nitrate, sulfate, and specific conductance results from one sample in November and one specific conductance result in December.

Overall variability of NADP/NTN wet-deposition measurements was estimated during water year 2004 by the median absolute errors for weekly wet-deposition sample concentrations and precipitation measurements for two collocated NADP/NTN sites. One pair of samplers was collocated in New Mexico, and a second pair was collocated in Texas. MAEs were less than 10 percent for nitrate and sulfate concentrations, specific conductance, and collector catch for both collocated sites. MAEs were between 10 and 28 percent for calcium, magnesium, sodium, potassium, ammonium, chloride, and hydrogen-ion concentrations. MAEs for precipitation depth were between 5 and 11 percent. Upon converting oncentrations to deposition amounts, MAEs increased for both collocated sites for all analytes.

For the 2004 NADP/MDN system-blank program, the median system-sample minus bottle-sample difference was 0.018 nanograms per liter (ng/L), which is nearly an order of magnitude less than the 0.15 ng/L Mercury Analytical Laboratory (HAL) minimum reporting limit (MRL). The 2004 system-blank data indicate that maximum contamination in 95 percent of NADP/MDN samples was less than the MRL with 95-percent confidence. In the interlaboratory-comparison program, HAL data were in statistical control throughout most of 2004 except for four samples in March. The median difference between the HAL-reported concentrations and the MPVs was zero. No bias was detected in the interlaboratory-comparison program data produced by the HAL.

Contents

Abstract

Introduction

Statistical Approach

National Trends Network Quality Assurance Programs

Intersite-Comparison Program

Results for Intersite-Comparison Studies 52 and 53

Intersite-Comparison Followup Study

Sample-Handling Evaluation Program

Variability of Sample-Handling Evaluation Program Results

Bias of Sample-Handling Evaluation Program Results

Field-Audit Program

Assessment of Field-Audit Data

Variability and Bias in Field-Audit Data

Effects of Field Exposure on Sample Concentrations

NTN Interlaboratory-Comparison Program

Interlaboratory-Comparison Program Variability and Bias

Interlaboratory-Comparison Program Control Charts

Collocated-Sampler Program

Collocated-Sampler Data Analysis

Assessment of Absolute Error in Collocated-Sampler Data

Mercury Deposition Network Quality Assurance Programs

System-Blank Program

MDN Interlaboratory-Comparison Program

MDN Interlaboratory-Comparison Program Control Charts

Evaluation of Interlaboratory Variability and Bias

Results for NADP/MDN Interlaboratory-Comparison Program Blanks

Summary

References Cited

Figures

1. Flowchart showing intersite-comparison program of the U.S. Geological Survey external quality-assurance project.

2. Graph showing distribution of pH and specific-conductance values for intersite-comparison studies 52 and 53.

3. Flowchart showing field-audit program of the U.S. Geological Survey.

4–6. Boxplots showing:

4. Comparison of 2003 and 2004 field-audit bucket-minus-bottle concentration differences.

5. Comparison of 2003 and 2004 field-audit bucket-minus-bottle differences for pH and specific conductance.

6. Relation of paired bucket-minus-bottle differences and sample volume for 2004 field-audit data.

7–8. Graphs showing:

7. Maximum contamination in 2004 field-audit samples represented by the 90-, 95-, and 99-percent upper confidence limits for bucket-minus-bottle paired differences.

8. Maximum contamination in National Atmospheric Deposition Program/National Trends Network samples estimated by 3-year moving 90-percent upper confidence limits for 90th percentiles of field-audit paired differences.

9. Flowchart showing interlaboratory-comparison program of the U.S. GeologicalSurvey for the National Trends Network.

10–24. Graphs showing:

10. Difference between the measured calcium concentration values and the median calcium concentration value calculated by solution for all participating laboratories in the interlaboratory-comparison program during 2004.

11. Difference between the measured magnesium concentration values and the median magnesium concentration value calculated by solution for all participating laboratories in the interlaboratory-comparison program during 2004.

12. Difference between the measured sodium concentration values and the median sodium concentration value calculated by solution for all participating laboratories in the interlaboratory-comparison program during 2004.

13. Difference between the measured potassium concentration values and the median potassium concentration value calculated by solution for all participating laboratories in the interlaboratory-comparison program during 2004.

14. Difference between the measured ammonium concentration values and the median ammonium concentration value calculated by solution for allparticipating laboratories in the interlaboratory-comparison program during 2004.

15. Difference between the measured chloride concentration values and the median chloride concentration value calculated by solution for all participating laboratories in the interlaboratory-comparison program during 2004.

16. Difference between the measured nitrate concentration values and the median nitrate concentration value calculated by solution for all participating laboratories in the interlaboratory-comparison program during 2004.

17. Difference between the measured sulfate concentration values and the median sulfate concentration value calculated by solution for all participating laboratories in the interlaboratory-comparison program during 2004.

18. Difference between the measured pH values and the median pH value calculated by solution for all participating laboratories in the interlaboratory-comparison program during 2004.

19. Difference between the measured specific conductance values and the median specific conductance value calculated by solution for all participating laboratories in the interlaboratory-comparison program during 2004.

20. Percentage difference between measured values and median values calculated by solution for Central Analytical Laboratory, Champaign, Illinois, for 2004 interlaboratory-comparison program.

21. Comparison of median absolute percent differences determined for collocated-sampler sites NM07/07NM and TX22/22TX during water year 2004 and 41 collocated-sampler sites during 1989-2001.

22. Mercury Deposition Network system-blank program of the U.S. Geological Survey.

23. 2004 Mercury Deposition Network system-blank data in order of increasing system-sample minus bottle-sample difference.

24. Maximum contamination in 2004 Mercury Deposition Network system-blank samples represented by the 90-, 95-, and 99-percent upper confidence limits for system-sample minus bottle-sample paired differences.

25. Flowchart showing interlaboratory-comparison program of the U.S. Geological Survey for the Mercury Deposition Network.

26–27. Graphs showing:

26. Control charts for laboratories participating in 2004 Mercury Deposition Network interlaboratory-comparison program.

27. Comparison of total mercury concentration results from the Mercury Analytical Laboratory (HAL) to all other laboratories participating in the 2004 Mercury Deposition Network interlaboratory-comparison program for blank samples.

Tables

1. Site-operator responses and summary statistics for 2004 intersite-comparison program studies 52 and 53.

2. Results of followup studies for intersite-comparison study number 52.

3. Solutions used in 2004 sample-handling evaluation, field-audit, and interlaboratory-comparison programs.

4. Target values for solutions used in 2004 U.S. Geological Survey sample-handling evaluation, field-audit, and interlaboratory-comparison programs.

5. Selected statistics for 2004 sample-handling evaluation program paired bucket-sample minus bottle-sample concentration differences.

6. Relative and absolute bucket-minus-bottle differences calculated as a percentage of the target concentration or value for each analyte for 2004 sample-handling evaluation program.

7. Summary of paired bucket-sample minus bottle-sample concentration differences for 2004 field-audit program.

8. Comparison of the maximum likely analyte contamination levels in 90-percent of 2004 field-audit samples with 2004 concentration quartiles for the National Atmospheric Deposition Program/National Trends Network.

9. Most probable values for solutions used in 2004 U.S. Geological Survey interlaboratory-comparison program.

10. Comparison of the differences between reported concentrations and most probable values for synthetic wet-deposition samples in the 2004 interlaboratory-comparison program.

11. Number of analyte determinations greater than the method detection limits for each participating laboratory and each ion for deionized-water samples during 2004.

12. Comparison of median absolute differences determined for collocated-sampler sites NM07/07NM and TX22/22TX during water year 2004 to median values obtained during 2004 for all National Atmospheric Deposition Program / National Trends Network sites and to median absolute differences for 41 collocated-sampler sites during 1989-2001.

13. Most probable values for solutions used in 2004 U.S. Geological Survey Mercury Deposition Network interlaboratory-comparison program.

14. Comparison of the differences between reported mercury concentrations and most probable values for 2004 Mercury Deposition Network interlaboratory-comparison program samples.

For additional information contact:
Gregory A. Wetherbee
U.S. Geological Survey
Office of Water Quality
Denver Federal Center
P.O. Box 25046, MS 401
Lakewood, CO 80225
Telephone: 1-303-236-1837