Trends and related causes in extreme hourly precipitation

Abstract

The intensity and frequency of (extreme) precipitation are likely aff ected by climate change. While many studies have focused on trends in extreme daily sums, extreme precipitation often occurs on shorter timescales. The present work aims to assess and understand trends in intensity and frequency of hourly extreme precipitation for 5 stations in the Netherlands (De Bilt, De Kooy, Vlissingen, Maastricht and Eelde) for the period 1958 until 2015.

Robust signals are found in hourly precipitation data with spatial and seasonal variation in magnitude and number of signi cant trends. Multiple signi cant trends (p > 95%) for independent 2-day periods show a rapid increase of the (25-5%) highest intensities, a decrease in the frequency of wet hours (of the mean and 50-0.1% wettest 2-days), and a constant to increasing signal in the mean and (10-1% highest) precipitation sums. Our study confi rms previous fi ndings (IPCC, 2013) that the "high" extremes increase in a disproportional way, compared to the yearly mean and "low" extremes.

To explain the observed trends, we studied from theory and by statistical analysis four important factors for extreme precipitation, (i) temperature, (ii) dewpoint temperature, (iii) CAPE and (iv) wind speed. The observed trends in precipitation intensity are likely caused by the increase in temperature, via Clausius-Clapeyron related increase of the atmospheric "moisture-holding capacity". This fi nding matches well with our observed negative trends in the frequency of wet hours, as warmer air becomes less easily saturated.

Promising for future research is the integration with high-resolution (e.g. satellite and radar) observational data, testing of models' capability to capture extreme precipitation and scaling up the research to larger regions such as Europe.