External quality assurance project report for the National Atmospheric Deposition Program’s National Trends Network and Mercury Deposition Network, 2015–16

The U.S. Geological Survey Precipitation Chemistry Quality Assurance project operated five distinct programs to provide external quality assurance monitoring for the National Atmospheric Deposition Program’s (NADP) National Trends Network and Mercury Deposition Network during 2015–16. The National Trends Network programs include (1) a field audit program to evaluate sample contamination and stability, (2) an interlaboratory comparison program to evaluate analytical laboratory performance, and (3) a colocated sampler program to evaluate bias and variability attributed to automated precipitation samplers. The Mercury Deposition Network programs include the (4) system blank program and (5) an interlaboratory comparison program. The results indicate that NADP data continue to be of sufficient quality for the analysis of spatial distributions and time trends for chemical constituents in wet deposition.

The field audit program results indicate increased sample contamination for calcium, magnesium, and potassium relative to 2010 levels, and slight fluctuation in sodium contamination. Nitrate contamination levels dropped slightly during 2014–16, and chloride contamination leveled off between 2007 and 2016. Sulfate contamination is similar to the 2000 level. Hydrogen ion contamination has steadily decreased since 2012. Losses of ammonium and nitrate resulting from potential sample instability were negligible.

The NADP Central Analytical Laboratory produced interlaboratory comparison results with low bias and variability compared to other domestic and international laboratories that support atmospheric deposition monitoring. Significant absolute bias above the magnitudes of the detection limits was observed for nitrate and sulfate concentrations, but no analyte determinations exceeded the detection limits for blanks.

Colocated sampler program results from dissimilar colocated collectors indicate that the retrofit of the National Trends Network with N-CON Systems Company, Inc. precipitation collectors could cause substantial shifts in NADP annual deposition (concentration multiplied by depth) values. Median weekly relative percent differences for analyte concentrations ranged from -4 to +76 percent for cations, from 5 to 6 percent for ammonium, from +14 to +25 percent for anions, and from -21 to +8 percent for hydrogen ion contamination. By comparison, weekly absolute concentration differences for paired identical N-CON Systems Company, Inc., collectors ranged from 4–22 percent for cations; 2–9 percent for anions; 4–5 percent for ammonium; and 13–14 percent for hydrogen ion contamination. The N-CON Systems Company, Inc. collector caught more precipitation than the Aerochem Metrics Model 301 collector (ACM) at the WA99/99WA sites, but it typically caught slightly less precipitation than the ACM at ND11/11ND, sites which receive more wind and snow than WA99/99WA.

Paired, identical OTT Pluvio-2 and ETI Noah IV precipitation gages were operated at the same sites. Median absolute percent differences for daily measured precipitation depths ranged from 0 to 7 percent. Annual absolute differences ranged from 0.08 percent (ETI Noah IV precipitation gages) to 11 percent (OTT Pluvio-2 precipitation gages).

The Mercury Deposition Network programs include the system blank program and an interlaboratory comparison program. System blank results indicate that maximum total mercury contamination concentrations in samples were less than the third percentile of all Mercury Deposition Network sample concentrations (1.098 nanograms per liter; ng/L). The Mercury Analytical Laboratory produced chemical concentration results with low bias and variability compared with other domestic and international laboratories that support atmospheric-deposition monitoring. The laboratory’s performance results indicate a +1-ng/L shift in bias between 2015 (-0.4 ng/L) and 2016 (+0.5 ng/L).