Quality‐control of an hourly rainfall dataset and climatology of extremes for the UK

Abstract

Sub‐daily rainfall extremes may be associated with flash flooding, particularly in urban areas but, compared with extremes on daily timescales, have been relatively little studied in many regions. This paper describes a new, hourly rainfall dataset for the UK based on ∼1600 rain gauges from three different data sources. This includes tipping bucket rain gauge data from the UK Environment Agency (EA), which has been collected for operational purposes, principally flood forecasting. Significant problems in the use of such data for the analysis of extreme events include the recording of accumulated totals, high frequency bucket tips, rain gauge recording errors and the non‐operation of gauges. Given the prospect of an intensification of short‐duration rainfall in a warming climate, the identification of such errors is essential if sub‐daily datasets are to be used to better understand extreme events. We therefore first describe a series of procedures developed to quality control this new dataset. We then analyse ∼380 gauges with near‐complete hourly records for 1992–2011 and map the seasonal climatology of intense rainfall based on UK hourly extremes using annual maxima, n‐largest events and fixed threshold approaches. We find that the highest frequencies and intensities of hourly extreme rainfall occur during summer when the usual orographically defined pattern of extreme rainfall is replaced by a weaker, north–south pattern. A strong diurnal cycle in hourly extremes, peaking in late afternoon to early evening, is also identified in summer and, for some areas, in spring. This likely reflects the different mechanisms that generate sub‐daily rainfall, with convection dominating during summer. The resulting quality‐controlled hourly rainfall dataset will provide considerable value in several contexts, including the development of standard, globally applicable quality‐control procedures for sub‐daily data, the validation of the new generation of very high‐resolution climate models and improved understanding of the drivers of extreme rainfall.

Towards identification of critical rainfall thresholds for urban pluvial flooding prediction based on crowdsourced flood observations

“…Strongest correlations were found with the variables d 0.25 and d 1 . This confirms the relevance of sub-daily temporal resolutions when investigating the impact of rainfall on urban hydrology (Blenkinsop et al, 2017;Ochoa-Rodriguez et al, 2015;Romero et al, 2011;Simões et al, 2015). It should be noted that relations found are strongly dependent on the spatial aggregation scale.…”

“…Heavy rainfall events, which are most relevant for flooding, are rare, limiting the type of observation-based analyses that can be performed. Numerous scientific investigations emphasize the importance of variable temporal resolutions when the urban hydrologic response is relevant (Blenkinsop et al, 2017;Ochoa-Rodriguez et al, 2015;Romero et al, 2011;Simões et al, 2015). In an urban environment, the critical response time is in the range of minutes to hours (Romero et al, 2011).…”

“…In an urban environment, the critical response time is in the range of minutes to hours (Romero et al, 2011). Therefore, variable sub daily temporal resolutions are important to enhance our understanding of climate extremes (Blenkinsop et al, 2017). Research has shown that these variable resolutions 5 influence hydrodynamic model outputs (Ochoa-Rodriguez et al, 2015).…”

“…For the d 0.25 and d 1 variables, the maximum depth during an event was selected as a 10 characteristic rainfall value. It follows from other research that sub-daily durations are the most relevant for urban hydrologic response (Blenkinsop et al, 2017;Ochoa-Rodriguez et al, 2015;Romero et al, 2011;Simões et al, 2015). Report and OP data were plotted versus rainfall depth and Spearman's rank correlation coefficients were computed.…”

Projected changes in extreme precipitation over Scotland and Northern England using a high-resolution regional climate model

“…Total mean precipitation can be expressed as the sum of the mean intensity above the "wet" threshold times the frequency of "wet hours", plus the remainder that is truncated below the threshold. Many observations have higher minimum thresholds than 0.1 mm/h ; for instance, many UK tipping bucket gauges have 0.2 mm ticks (Blenkinsop et al 2017). Return levels and periods are standard metrics for extreme risk assessments, and are estimated using extreme value theory (Coles 2001).…”