Tipping bucket mechanical errors and their influence on rainfall statistics and extremes P. La Barbera, L.G. Lanza and L. Stagi Department of Environmental Engineering, University of Genova, 1 Montallegro, 16145 Genova, Italy Abstract Based on the error figures obtained after laboratory tests over a wide set of operational rain gauges from the network of the Liguria region, the bias introduced by systematic mechanical errors of tipping bucket rain gauges in the estimation of return periods and other statistics of rainfall extremes is quantified. An equivalent sample size is defined as a simple index that can be easily employed by practitioner engineers to measure the influence of systematic mechanical errors on common hydrological practice and the derived hydraulic engineering design. A few consequences of the presented results are discussed, with reference to data set reconstruction issues and the risk of introducing artificial climate trends in the observed rain records. Keywords Rainfall measurement; extremes, rain gauge; mechanical errors; calibration; statistics Introduction The measurement of liquid precipitation at the ground is crucial to most hydrological appli- cations. Cumulated water depth over a given sampling interval in time is the usual output of traditional rain gauges, while inference of the rain intensity is possible by assuming uni- form distribution of the rain process over the sampling interval. Such a measure of precipi- tation at the ground is usually assumed as the “ground truth” and newly developed techniques for rainfall monitoring based on remote sensing (weather radar, radiometers borne on aircrafts or satellite platforms, etc.) are commonly compared with the rain gauge figures for calibration and/or validation purposes. As quoted by Marsalek (1981) the tipping-bucket rain gauge dates back to the 18th cen- tury and – though further refined over the years – it has become probably the most popular recording rain gauge, used by most national weather service agencies. The reason for such a widespread popularity comes from the very simple mechanics exploited for direct meas- urement of rainfall, and the reliability of the instrument. Also, it can be easily updated in its data acquisition and storage components as long as new electronic devices become opera- tionally available. Finally, maintenance loads are reasonable and the cost is affordable even in the case of rather extended networks. However many different types of water collectors have been implemented to convey rainfall from a standardised orifice into the measuring bucket and a large variability still exists world-wide in presently operational gauges. Simple gauges based on the tipping- bucket technology are also used for the measurement of other hydrological variables such as runoff and sediment transport. The measurement of liquid precipitation at the ground is affected by different sources of both systematic and random errors, mainly due to wind, wetting and evaporation induced losses (e.g. Sevruk, 1982) which make the measurement of light to moderate rainfall scarcely reliable in the absence of an accurate calibration. Wind induced errors still have an influence at rainfall rates in the order of 20–50 mm/h with an incidence around 5% being reported at daily scale after comparison between paired ground level and elevated gauges in a few intercomparison stations in central Europe (Sevruk and Hamon, 1984). Solid precipitation measurements (snow) are even more difficult as snow is more sensitive than Water Science and Technology Vol 45 No 2 pp 1–9 © IWA Publishing 2002 1 Downloaded from http://iwaponline.com/wst/article-pdf/45/2/1/425098/1.pdf by guest