Precipitation, sampled once per hour from an accumulating gauge and converted into an hourly rate. Any phase (liquid or solid) is captured to the extent possible and given as liquid water equivalent. Antifreeze and oil is added to the gauges to limit freezing or evaporation of the bucket contents. Note that the data series has not been corrected for the efficiency of the precipitation gauge in catching precipitation, so called undercatch. This is known to be particulay problematic during high winds, and snow fall. Sensor model changes: 1995-08-17: Belfort 5915, unshielded; 2014-06-10 11:00: Pluvio2, shielded. Data Comments: 1995-08-17 The belfort sensor suffered from freezing of the spring mechanism due to water invasion. This resulted in bucket weight increases due to individual events being registered as a single event once the device thawed. This is clear to see in the data. No attempt has been made to redistribute these single large events back to the correct time stamps. It is recommended to study the time series together with snow depth and temperature to understand when these events occurred. In addition, it is recommended to consider the hydrological year for annual precipitation totals to ensure winter precipitation is attributed to the correct year. The Belfort sensor did not have a wind shield and therefore is likely to have a higher undercatch (less efficient at catching the precipitation) than the Pluvio sensors used from 2014.; 2014-06-10 The hourly rainfall rates are derived from the raw bucket weight of the OTT Pluvio2 gauge by applying the 'neutral aggregating filter' (Smith et al. 2019). Reprocessing of the data in 04/2025 led to changes in the data in the period 1995-08-17 - 2014-06-10.
