Quantitative historical hydrology in Europe

Abstract

Abstract. In the last decades, the quantification of flood hydrological characteristics (peak discharge, hydrograph shape, and runoff volume) from documentary evidence has gained scientific recognition as a method to lengthen flood records of rare and extreme events. This paper describes the methodological evolution of the quantitative historical hydrology under the influence of developments in hydraulics and statistics. In the 19th century, discharge calculations based on flood marks was the only source of hydrological data for engineering design, but later was left aside on favour of systematic gauge records and conventional hydrological procedures. In the last two decades, there is growing scientific and public interest to understand long-term patterns of rare floods, maintain the flood heritage and memory of extremes, and to develop methods for deterministic and statistical application to different scientific and engineering problems. A compilation of 45 case studies across Europe with reconstructed discharges demonstrates that (1) in most cases present flood magnitudes are not unusual within the context of the last millennium, although recent floods may exceed past floods in some temperate European rivers (e.g. the Vltava and Po rivers), (2) frequency of extreme floods have decreased since the 1950s, although some rivers (e.g. the Gardon and Ouse rivers) show a reactivation of rare events over the last two decades. There is a great potential of gaining understanding of individual extreme events based on a combined multiproxy approach (palaeoflood and documentary records) providing high-resolution time flood series and their environmental and climatic changes; and to develop non-systematic and non-stationary statistical models based on relations of past floods with external and internal covariates under natural low-frequency climate variability.

Trends in flood losses in Europe over the past 150 years

“…In the Mediterranean region, there is a smaller increase, or even decrease in more recent decades, of extreme precipitation than in other parts of Europe which is also consistent with trends in number of events in the two sub-domains. The overall upward trend and the contrast between northern and southern Europe is consistent with other studies, both for extreme precipitation 13 , 58 , 59 and large flood occurrences 15 , 16 . However, the number of events and flooded area must have had less pronounced trends for the continent as a whole, since the records of past floods have grown more complete over time, as shown in Fig.…”

“…Qualitative and quantitative hydrological studies for Europe have indicated no general continental-wide trend in river flood occurrences, extreme precipitation, or annual maxima of runoff 12 – 14 . However, substantial variations between different catchments have been observed, ranging from an increase in north-western Europe to no trend or a decrease in other parts of the continent 15 , 16 . Similar findings were reported for storminess along the European coasts 17 , 18 .…”

Reducing uncertainty with flood frequency analysis: The contribution of paleoflood and historical flood information

“…Historical flood information and paleoflood records can supplement systematic gauge records to improve at‐site FFA, but unlike gauge records, they are noncontinuous and are typically described in terms of above, or below, a flood threshold [ Swain et al ., ]. Historical flood information, which can be derived from flood marks on old buildings, newspaper reports, and oral descriptions [ Herget and Meurs , ], is well documented in Europe with records dating back to A.D. 1500 [ Brázdil et al ., ] and is now being incorporated into FFA [e.g., Benito et al ., ; Machado et al ., ]. However, in localities where European settlement occurred relatively recently, such as in Australia, historical flood information has limited capacity to extend the analysis period for FFA.…”

The challenges of including historical events using Bayesian methods to improve flood flow estimates in the United Kingdom: A practitioner's point of view

“…Depending on the documents, the information available may include the origins, scale, and impacts of a flood (Brázdil et al, ). Often accounts in different documents will contradict each other making cross‐referencing of various sources a necessity (Benito, Brázdil, Herget, & Machado, ; Brázdil, Kundzewicz, & Benito, ).…”