Storyboard: Identification of historic "Heavy Precipitation Events" using a tracking algorithm

Introduction

Thunderstorms are among the most fascinating weather phenomena. They are characterized by lightning and thunder, cause strong wind gusts and typically they bring intense rain. In this way the most severe thunderstorms can be perceived as heavy precipitation events (HPEs). We here will present a methodology that, upon objective criteria, identifies HPEs in convection-permitting observations and numerical models.

Flash floods rank among the most severe hazards caused by weather extremes. Typically the precipitation events have a strong convective character, are relatively small in spatial scale and short-lived. Thus it is challenging to accurately forecast them and to represent them in climate models with coarse grid spacing.

Motivation

Regional Climate Models integrated on convection permitting grids (dx<=4km; cpRCMs) improve the representation of precipitation extremes, as shown by Pichelli et al. (2021), e.g. in their Figure 3. However, information about the properties of the events, causing these extremes, is not yielded.

Tracking algorithms

Transform the problem from the Eulerian into the Lagrangian frame of reference and provide valuable information about the heavy precipitation events themselves. The schematic shows the Intensity in precipitation on the y-axis and the scale of events on the x-axis. In this way we identify here convective phenomena sorted by their typical scales and intensity.

Trackers in principle work the following: first a threshold is imposed to precipition rate (prTH), second grid cells are clustered in space and time to form objects, and third all objects that are smaller than a minimal volume (minvol) are dismissed.

We here apply a well-established tracker (MET MODE-TD, see https://dtcenter.org/met-online-tutorial-metv8-0/mode-time-domain) on a composite of high-resolution observations and two cpRCMs. With this we both demonstrate the capability of convection-permitting resolutions and the usefulness of the tracker.

Interactive map of heavy precipitation events in observations

We identify from a composite of convection-permitting observational datasets, covering France, Italy, Switzerland and Germany and the years 2001 to 2009, heavy precipitation events using the MODE-MTD tracker.

The setup is given by a precipitation threshold of 5 mm/hour, a minimum volume of 100 grid cells and a smoothing in space by 3*3 grid cells.

In this interactive map one can zoom and find historic events, and look into their properties, which show up when hovering over a circle, which represents a HPE.

Properties given are: mean area, duration, distance traveled, propagation speed, total precipitation, mean and maximum precipitation rate, and several percentiles of precipitation rate. The diameter of each circle is proportional to the mean area and the color represents the intensity, by the 90th percentile of precipitation rate.

Carrara, Italy, in September 2003

This HP and flash flood event is summarized on Italian Wikipedia: https://it.wikipedia.org/wiki/Alluvione_di_Carrara_del_2003. The region was affected by torrential rain: "in about two and a half hours 200 mm of rain was falling". It may as well be described as a landfalling event, hot and humid air from the Mediterranean sea, the tense south-western flow and the orography of upper Tuscany, were among the triggers that played a major role.

The tracker identifies the event in the observations and attributes fitting properties: It is short-lived, of small scale and intense.

Also in a cpRCM such events may be well-represented, as indicated.

Gard, France, in September 2002

This event is documented in detail by Delrieu et al.,2005. A landfalling mesoscale convective system remained stationary for almost two days, trapped by the orography. It brough horrendous amounts of rainfall and caused wide-spread flooding.

It is well-represented in the observations on the left. Our tracker identifies it and describes it correctly with a duration of 32 hours, a high intensity and a large scale and total precipitation amount. Also in a cpRCM such events are presented and can be identified.

Western Germany, 2016/05/26-2016/06/09

A stationary synoptic weather pattern, characterized by a ridge stretching from Great Britain to Scandinavia, caused a blocking situation. The pressure gradient and the resulting wind speed in the lower troposphere were very weak. Consequently, thunderstorms were almost stationary, resulting in large precipitation accumulations in local areas.

The tracker identifies a multitude of precipitation events in the observations.

Conclusions

Through the application of trackers valuable information about the events causing hazards like flash floods can be obtained. Here statistics obtained from 9 year-periods of an observational dataset, and two cpRCM ensembles, one driven by reanalysis ("evaluation") and one driven by CMIP5-GCMs are presented. On left we find properties describing the scale of the HPEs and on the right properties that describe their intensity. In this way models can be evluated against observations.

Eventually the same can be done for climate projections, promising valuable information about how HPEs are going to change in a future warmer climate.

Models and Observations provided through the project CORDEX-FPSCONV: https://www.hymex.org/cordexfps-convection/wiki/doku.php?id=Home

References:

Pichelli, E., Coppola, E., Sobolowski, S., Ban, N., Giorgi, F., Stocchi, P., ... & Vergara-Temprado, J. (2021). The first multi-model ensemble of regional climate simulations at kilometer-scale resolution part 2: historical and future simulations of precipitation. Climate Dynamics, 56(11), 3581-3602.

Delrieu, G., Nicol, J., Yates, E., Kirstetter, P. E., Creutin, J. D., Anquetin, S., ... & Wobrock, W. (2005). The catastrophic flash-flood event of 8–9 September 2002 in the Gard Region, France: a first case study for the Cévennes–Vivarais Mediterranean Hydrometeorological Observatory. Journal of hydrometeorology, 6(1), 34-52.

Piper, D., Kunz, M., Ehmele, F., Mohr, S., Mühr, B., Kron, A., & Daniell, J. (2016). Exceptional sequence of severe thunderstorms and related flash floods in May and June 2016 in Germany–Part 1: Meteorological background. Natural Hazards and Earth System Sciences, 16(12), 2835-2850.

Photos of flash floods:

• Gard
• Carrara