NADP Measurement Confidence Intervals for Critical Load Contingency Planning

Gregory A. Wetherbee
U.S. Geological Survey

Accounting for the variability inherent in National Atmospheric Deposition Program (NADP) measurements is important for prudent implementation of critical load contingency plans. The overall variability of NADP measurements is used to compute confidence limits for annual wet-deposition estimates, which provide confirmation of whether annual wet-deposition loads are increasing or decreasing from year to year.

The U.S. Geological Survey, Water Resources Discipline, Branch of Quality Systems (USGS) operates a Co-located Sampler Program to estimate the overall variability of NADP measurements. For more than 17 years this program has been an integral part of the external quality assurance project for NADP. Co-located samplers are defined by the USGS program as two sets of identical wet/dry precipitation collectors and precipitation depth gages located at the same site within 5 to 30 meters of each other. Replicate weekly precipitation samples and depth measurements were obtained from 47 co-located NADP National Trends Network sites during 1989-2006. These data were compiled to estimate annual dissolved wet-deposition of selected constituents for each of the paired sites.

Average relative standard deviations of annual wet-deposition measurements for nitrate plus ammonium as nitrogen (N), sulfate, and hydrogen ion were calculated from the co-located program replicate data. These data were used to compute statistical confidence intervals. The confidence intervals broaden as concentration increases. For example, the 90% confidence interval for annual N wet-deposition is 1.7 to 2.4 kilograms per hectare (kg/ha) for a measurement of 2 kg/ha. For a measurement of 4 kg/ha, the confidence interval is 3.4 to 4.8 kg/ha. In other words, one can discern between 2 kg/ha and 1.7 kg/ha with 90% confidence.

The confidence intervals can be used by natural resources managers to determine whether observed changes in annual wet-deposition should trigger contingency plans to control atmospheric emissions sources. In addition, they may be used to decide whether the atmospheric wet-deposition is proceeding along a path toward the critical load.