Space-Time scAling of the Rainfall to FLOOD transformation

Informazioni relative al progetto

STARFLOOD

ID dell’accordo di sovvenzione: 793558

DOI

10.3030/793558

Progetto chiuso

Data della firma CE 21 Marzo 2018

Data di avvio 1 Agosto 2018

Data di completamento 31 Luglio 2020

Finanziato da

• EXCELLENT SCIENCE - Marie Skłodowska-Curie Actions

Costo totale

€ 166 156,80

Contributo UE

€ 166 156,80

Coordinato da

TECHNISCHE UNIVERSITAET WIEN
Austria

Periodic Reporting for period 1 - STARFLOOD (Space-Time scAling of the Rainfall to FLOOD transformation)

Periodo di rendicontazione: 2018-08-01 al 2020-07-31

Sintesi del contesto e degli obiettivi generali del progetto

What is the problem/issue being addressed in STARFLOOD?

Reliable approaches for estimating flood magnitudes of a given probability in space and time are indispensable to optimize flood risk management and minimize future damage and loss of life. A common approach in engineering practice to estimate a flood without available river records is the design storm method. In the design storm method, probabilities of the rainfall are first estimated and the rainstorm is then used as input data for a rainfall-runoff model, which translates the rainfall event into a flood. The flood event is then assumed to have the same probability as the rainfall event. However, up until today, it is not fully clear under which climatic and geological conditions this assumption really holds, which is why it is widely debated. The project Space-Time scAling of the Rainfall to FLOOD transformation (STARFLOOD) addresses this lack of understanding of the relationship between rainfall and flood probabilities. This is conducted across Austria (i) based on analysis of observed rainfall and river flow time series and (ii) using coupled framework of rainfall and rainfall-runoff models, i.e. extrapolation of time series.

Why is the topic of STARFLOOD important for society?

Floods are among the costliest of all natural hazards. The recent wide-spread flood in Central Europe in June 2013, for example, incurred more than EUR 12 billion economic losses, and flood risks are expected to increase significantly in the future due to increasing exposure and due to climatic changes. One must be aware that the basic assumption of equal probabilities in common engineering methods described above has far-reaching consequences for the society: In cases where the assumption does not hold (i) flood hazard assessment may lead to either over-design of a structure i.e. potentially unnecessary use of public tax payer money, or (ii) under-design i.e. a too low protection level for a given location with potentially dire consequences for the infrastructure, economy and for life.

What are the overall objectives of STARFLOOD?

The project STARFLOOD responds to the research gaps addressed above by investigating how the probabilities of rainfall transform into probabilities of floods from a space-time perspective. This is achieved in two ways: (i) analyses based on pure data analysis, i.e. analyses based on observed time series of rainfall and floods to gain improved understanding; (ii) analyses based on the use of stochastic weather models and rainfall-runoff model. The analyses are conducted in the country of Austria with long available observations of rainfall of runoff at good spatial density.

Objective I: STARFLOOD explores the performance of the full cascade of rainfall to flood probabilities
Objective II: STARFLOOD explores the physical causes of flood probabilities
Objective III. STARFLOOD improves the understanding of the scaling behavior of hydrological processes

Lavoro eseguito dall’inizio del progetto fino alla fine del periodo coperto dalla relazione e principali risultati finora ottenuti

Statistics of numerous rainfall time series have been analyzed and set into relation with dominating rainfall mechanisms (Objective I, scientific article published). For Objective II, statistics of floods have been analyzed at over 400 gauged river catchments of Austria and set into relation to the intensity-duration-frequency statistics from rainfall by means of a novel approach (scientific article in preparation), which allows to better understand the transformation from rainfall into flood probabilities depending on topography, soils, geology and climate. To understand the scaling behavior of hydrological processes and their link to flood generation (Objective III), a rainfall-runoff modelling framework is currently being set up. In that context, a new non-parametric rainfall model has been developed (scientific article published). In addition, synergies could be built between STARFLOOD and the ongoing project WETRAX+ that examines how synoptic weather situations lead to flash flooding in the Upper Danube basin (Jacobeit, J., Blöschl, G., Komma, J., Stahl, N., Hofstätter, M., Haslinger, K., & Pistotnik, G. (2018). WETRAX+: welche Wetterlagen führen zu Sturzfluten? Forum für Hydrologie und Wasserbewirtschaftung 40 (18)).

During the STARFLOOD project, considerable effort has been devoted to dissemination.
- Seven conferences were attended, including the (i) European Geosciences Union (EGU) General Assembly 2019 and (ii) 2020 (Vienna, Austria), (iii) the International Union of Geodesy and Geophysics (IUGG) General Assembly 2019 (Montreal, Canada), (iv) International Disaster Response Expo 2019 (London, UK), (v) European Forum on Disaster Risk Reduction (EFDRR) hosted by United Nations Office for Disaster Risk Reduction 2018 (Rome, Italy), (vi) Natural Hazards and Disaster Science Forum 2018 hosted by the Centre of Natural Hazards and Disaster Science CNDS (Uppsala, Sweden) and (vii) Symposium on Statistical Hydrology 2019 (Uppsala, Sweden). Contributions included research talks (EGU, CNDS), poster presentations (EGU, IUGG), keynotes (International Disaster Response Expo), and development and moderation of a working session (EFDRR), with a large audience of researchers from various domains, practitioners, stakeholders, policy makers and general public.
- Two workshops have been attended, namely the Vienna Catchment Science Symposia (VCSS) 2019 and 2020 (Vienna, Austria).
- On social media, namely LinkedIn and Twitter, Horizon 2020 and ongoing research progress has been reported in posts/tweets at regular intervals.
- Two school visits have been initiated for spring 2020 with support of the Young Science Center Austria (youngscience.at). Due to the Covid19-pandemic, these school visits had to be postponed and are currently planned for 2021.

Progressi oltre lo stato dell’arte e potenziale impatto previsto (incluso l’impatto socioeconomico e le implicazioni sociali più ampie del progetto fino ad ora)

The STARFLOOD project is innovative in various aspects:

1. The project comprehensively explores the physical causes of flood probabilities: it provides crucial insights into flood probabilities in time and space over a large range of scales under various climatic and geological conditions.
2. The project explores the full cascade of rainfall to flood probabilities: the innovation consists of evaluating the sources of random errors and biases (specific components of the stochastic rainfall and the runoff model). Attributing the error sources facilitates the development of more accurate and more robust models of flood probabilities.
3. The project supports the improvement of current and future rainfall algorithms for flood hazard estimation: The field of stochastic space-time rainfall models is fragmented and it remains unclear which types of rainfall models are most suitable for reliable flood hazard assessment.

We believe that the research line that was initiated by the Marie Curie fellowship, and that is being continued by funding of TU Wien during the next years, will create socio-economic impact by:
- Providing insights that allow for more robust flood hazard assessment in any region across the globe.
- Improving flood zonation in Austria from the knowledge gained, a topic TU Wien has been involved in for many years.
- Providing other methods developed during STARFLOOD that may be useful for the detection of long-term variability in rainfall and runoff time series, which may support research on climate change.

Overview of the project