In order to better assess flood hazards, hazard maps should include historical data. This is what researchers at the CEDIM — Center for Disaster Management and Risk Reduction Technology at the Karlsruhe Institute of Technology (KIT) are advocating. CEDIM has presented a first report on the flood disaster in Rhineland-Palatinate and North Rhine-Westphalia. Regarding the role of climate change, the combination of more available water in the atmosphere and increasing persistence of large-scale weather patterns holds an increasing potential for extreme precipitation events.
Last week’s flood disaster claimed more than 170 lives in Germany (as of July 21, 2021). People are still missing. The damage to buildings and infrastructure can only be roughly determined and is in the double-digit billions — of which at least two billion euros alone for transport infrastructure. In the meantime, the German Insurance Association (GDV) has calculated the insured damage to buildings and infrastructure. (GDV) has estimated the insured damage at four to five billion euros in Rhineland-Palatinate and North Rhine-Westphalia alone. How did the floods, which mainly affected Rhineland-Palatinate and North Rhine-Westphalia, come about? How can flood hazards — especially rare, extreme events — be better estimated in advance? These are the questions that CEDIM’s Forensic Disaster Analysis (FDA) Group has been addressing, and it has produced its first report.
As the researchers explain, enormous amounts of precipitation led, for example, to the water level at the Ahr (Altenahr) significantly exceeding its previous record of 2016 (3.71 meters, discharge: 236 m³/s). However, due to flooding, the gauging station failed at a value of 5.05 meters (discharge: 332 m³/s). The Rhineland-Palatinate State Office for the Environment used model calculations to calculate a level of up to seven meters for the night of the catastrophe; based on this, the experts estimated a discharge of between 400 and 700 m³/s.
Several factors led to the extremely high precipitation totals
From a meteorological perspective, several factors led to the extremely high precipitation totals. In addition, the highly indented terrain of the affected regions, especially in the district of Ahrweiler, with partly deeply incised river valleys, increased the surface runoff. The already nearly saturated soil due to partly heavy precipitation in the preceding days further aggravated the situation.
“Within 48 hours, more rain fell in parts of North Rhine-Westphalia and Rhineland-Palatinate than usually falls there in the entire month of July; the majority even fell within only about ten hours.”
- Professor Michael Kunz, CEDIM Spokesperson
To estimate the flooded areas in the hardest-hit areas of Kreis Ahrweiler and Rhein-Erft-Kreis, the research team combined satellite data with aerial photos from (amateur) drones and helicopters, as well as photos from social media. According to these estimated flooded areas, there are just over 19,000 buildings in the affected areas with a value of around nine billion euros. Combined with empirical data from past flood disasters (infrastructure damage, natural hazards, and other damage), the researchers estimated total damage between eleven and 24 billion euros (first CEDIM estimate: July 21, 2021). It should be noted, however, that flooded areas represent only a portion of the total affected area.
More available water in the atmosphere and increasing persistence of large-scale weather patterns increase risk
According to the Karlsruhe disaster researchers, whether a single extreme event or the sequence of several extremes can already be attributed to climate change can neither be precisely proven nor completely denied, especially when it comes to events on short time and spatial scales that are strongly influenced by local factors. However, for the large-scale processes in the atmosphere that lead to the development of extreme events, the following is true: The combination of more available water in the atmosphere as a result of temperature increase and an increasing persistence of large-scale weather patterns with a tending northward shift of the jet stream, the strong wind band in the upper troposphere, has a high hazard potential. “As a positive trend is expected for these three factors, the potential for extreme precipitation events will also increase in the future,” Kunz explains.
Significant flood events in the Ahr Valley as early as 1804 and 1910
“There have already been two particularly significant flood events in the Ahr Valley in the past, in 1804 and 1910, but a comparison with historical records suggests that this year’s values should be classified lower than those of 1804,” says CEDIM deputy spokesperson Dr. James Daniell. For the 1804 flood event, the discharge was already estimated at about 1,100 m³/s by the University of Bonn. This year’s event may have been hydrologically similar in magnitude to the 1910 event with a discharge of 500 m³/s. “The current flood maps for the Ahr valley are currently based on discharge statistics with data since 1947, as homogeneous measurement series have been available since that time. However, this means that the two historical events have not yet been taken into account in the hazard assessment,” says Dr. Andreas Schäfer, lead author of the report. For example, the current estimate of a hundred-year flood as a design basis for flood protection for the Ahr River is 241 m³/s.
CEDIM’s FDA Group urges the inclusion of historical data in flood hazard maps, including data from before continuous measurement records, to better assess flood hazards. “Admittedly, when analyzing and interpreting the data, we must generally keep in mind that both infrastructures and flood protection measures have changed in recent years. As a result, it’s harder to compare readings directly, and we should focus less on water levels,” Daniell explains. “We can use gauge levels from 1804 and 1910 as indirect indicators to identify flood years. However, measured data on discharge, on trends over time, and on precipitation totals are more important for interpretation. Ultimately, however, both historical variables — gauges and discharge — should be included when developing hazard maps.”
