Investigation and quality assessment of the Past Weather Code from the Integrated Surface Database

Quantitative SYNOP Code weather variables such as rainfall amount, although of high societal and environmental importance, are frequently subject to recording errors and inhomogeneities resulting in uncertain conclusions. Here we assess the viability of the more qualitative Past Weather Code (PWC) for its use in robust climate analysis in the belief that it is less prone to both random and systematic
errors. The Past Weather Code data, from a selection of the National Oceanographic and Atmospheric Administration’s Integrated Surface Database (ISD) (4731 sufficiently long stations), is quality assessed by searching for inhomogeneities in station PWC time series, removing the offending stations and averaging the remaining stations into a global gridded dataset. PWCs 6 (Rainfall), 7 (Snowfall) and 9 (Thunderstorms) are found to robustly exhibit seasonal features, e.g. the Indian monsoon and peak Northern Hemispheric winter snowfall. Precipitation responses to the North Atlantic Oscillation are also detected in winter PWC 6 data over Europe.