Estimating Bias within the National Atmospheric Deposition Program and the National Trends Network

Approach:

As there is no `standard’ network against which the NADP/NTN monitoring program can be measured, the network has attempted to estimate bias by comparing NADP/NTN data to those of other similar networks. This comparison is being accomplished through the collection and analysis of data from other networks that have located there sampling equipment within 30 meters of NADP/NTN sampling equipment. This type of collocation however, only identifies systematic differences between the collocated networks. Collocated equipment precision (random error) must also be known so that overall network measurement error can be estimated as the sum of the systematic and random error. This approach follows in part, the general framework of collaborative testing outlined by Youden (1967).

Chemical Analysis and Rain Gage Volumes:

Pooled estimates of bias are derived using the median differences found between networks at each collocation site. The efficiency of the estimates are proportional to the number of networks and sites participating in collocation studies.

Bucket Volumes and Rain Gage Amounts:

Distributions of the differences between bucket volumes collected by the network as compared to the network’s standard gage are used to estimate regional and seasonal biases.

A comparison of NADP/NTN standard gage depths to those of nearby National Weather Service gage depths estimates differences between rain fall estimates made by our national measurement standard and the network.

The classification of daily precipitation amounts by type show the distribution of rain, snow and mixed events by region and season across the network.

Special Studies:

Special studies of collection devices both for rainfall amounts and chemical composition have been conducted by the network to explore the nature and magnitude of these biases.

  • Nipher Rain Gage StudyEnvironmental Protection Agency. 1990. Progress Report: Comparison of Precipitation Measurements by Nipher-Shielded and Standard Belfort Recording Rain Gages at NADP/NTN Sites. EPA 600-3-90-064.Bigelow, D.S., A.S. Denning, and J. Baron. 1990. Differences Between Nipher and Alter Shielded Universal Rain Gages at Two Colorado Deposition Sites. Environmental Science and Technology, 24, (5), pp 758-760.
  • Leonard Mold and Die Collector Study
  • Geotech Collector StudyGraham, R.C.; Obal, J. 1989. The Non-parametric Statistical Evaluation of Precipitation Chemistry Sampler Intercomparison Data; Atmos. Environ., 23, 1117-1130.
  • Graham, R.C.; Robertson, J.K. 1990. An Assessment of Performance of Wet Atmospheric Deposition Samples. U.S. Geological Survey; Reston, VA. Water-Resources Investigations Report 90-4042.
  • Graham, R.C.; Robertson, J.K.; Obal, J. 1987. An Assessment of the Variability in Performance of Wet Atmospheric Deposition Samples. U.S. Geological Survey; Denver, CO. Water-Resources Investigations Report 87-4125.
  • Graham, R.C.; Robertson, J.K.; Schroder, L.J.; Lafemina, J. 1988. Atmospheric Deposition Sampler Intercomparison; Water, Air, Soil Pollut. 37, 139-147.