A statistical comparison of active and passive ammonia measurements collected at Clean Air Status and Trends Network (CASTNET) sites.

Atmospheric concentrations of ammonia (NH3) are not well characterized in the United States due to the sparse number of monitors, the relatively short lifetime of NH3 in the atmosphere, and the difficulty in measuring non-point source emissions such as fertilized agricultural land. In this study, we compare measured weekly concentrations of NH3 collected by two denuder systems with a bi-weekly passive NH3 sampler used by the National Atmospheric Deposition Program's (NADP) Ammonia Monitoring Network (AMoN). The purpose of the study was to verify the passive samplers used by AMoN and characterize any uncertainties introduced when using a bi-weekly versus weekly sampling time period. The study was conducted for 1 year at five remote Clean Air Status and Trends Network (CASTNET) sites. Measured ambient NH3 concentrations ranged from 0.03 μg NH3 m(-3) to 4.64 μg NH3 m(-3) in upstate New York and northwest Texas, respectively, while dry deposition estimates ranged from 0.003 kg N ha(-1) wk(-1) to 0.47 kg N ha(-1) wk(-1). Results showed that the bi-weekly passive samplers performed well compared to annular denuder systems (ADS) deployed at each of the five CASTNET sites, while the MetOne Super SASS Mini-Parallel Plate Denuder System (MPPD) was biased low when compared to the ADS. The mean relative percent difference (MRPD) between the ADS and MPPD and the ADS and AMoN sampler was -38% and -9%, respectively. Precision of the ADS and MPPD was 5% and 13%, respectively, while the precision of the passive samplers was 5%. The results of this study demonstrate that the NH3 concentrations measured by AMoN are comparable to the ADS and may be used to supplement the high-time resolution measurements to gain information on spatial gradients of NH3, long-term trends and seasonal variations in NH3 concentrations.