Item talk:Q71773

Evaluation of candidate rain gages for upgrading precipitation measurement tools for the National Atmospheric Deposition Program

The National Atmospheric Deposition Program (NADP) was established in 1977 to investigate atmospheric deposition and its effects on the environment. Since its establishment, precipitation records have been obtained at all NADP sites using a gage developed approximately 50 years ago-the Belfort 5-780 mechanical rain gage. In 1998 and 1999, a study was done by the U.S. Geological Survey to evaluate four recently developed, technologically advanced rain gages as possible replacement candidates for the mechanical gage currently (2002) in use by the NADP. The gage types evaluated were the Belfort 3200, Geonor T-200, ETI Noah II, and the OTT PLUVIO. The Belfort 5-780 was included in the study to compare the performance of the rain gage currently (2002) used by NADP to the performance of the more recently developed gages. As a reference gage, the NovaLynx Model 260-2510 National Weather Service type stick gage also was included in the study. Two individual gages of each type were included in the study to evaluate precision between gages of the same type. A two-phase evaluation was completed. Phase I consisted of indoor bench tests with known amounts of simulated rainfall applied in 20 individual tests. Phase II consisted of outdoor testing by collecting precipitation during a 26-week period near Bay St. Louis, Mississippi. The ETI Noah II, OTT PLUVIO, and NovaLynx stick gages consistently recorded depths more commensurate with the amounts of applied simulated rainfall in Phase I testing than the Geonor T-200, Belfort 5-780, and Belfort 3200 gages. Gages where both the median difference between the measured and applied simulated rainfall and the interquartile range of all of their measured minus applied simulated rainfall differences were small (less than or equal to 0.01 inch) were judged to have performed very well in Phase I testing. The median and interquartile-range values were 0.01 inch or less for each of the ETI Noah II gages, OTT PLUVIO gages, and NovaLynx stick gages. The performance of the Geonor T-200 and Belfort 3200 gages was affected by technical problems during Phase I testing. As part of the evaluation of Phase II results, the average weekly precipitation totals obtained from the Belfort 5-780 gages and from each of the gages under consideration as possible replacements for the Belfort 5-780 gage were all compared with the average precipitation weekly totals obtained from two NovaLynx stick gages. The median absolute differences between a particular gage model and the NovaLynx stick reference gage for the 26 weeks of outdoor testing ranged from 0.04 inch for the ETI Noah II and OTT PLUVIO gages to 0.06 inch for the Geonor T-200. The total absolute difference between a particular gage type and the reference gage ranged from 1.23 inches for the Belfort 5-780 to 1.83 inches for the Geonor T-200 gages. Because the Belfort 3200 gages were inoperable for most of the Phase II testing, it is not meaningful to include the results from that gage type in a calculation of median or total absolute differences. The OTT PLUVIO proved to be the most reliable gage in Phase I and II testing, operating trouble free over the duration of the study.