@misc{SamprognaMohorThiekenKorup2021, author = {Samprogna Mohor, Guilherme and Thieken, Annegret and Korup, Oliver}, title = {Residential flood loss estimated from Bayesian multilevel models}, series = {Postprints der Universit{\"a}t Potsdam : Mathematisch-Naturwissenschaftliche Reihe}, journal = {Postprints der Universit{\"a}t Potsdam : Mathematisch-Naturwissenschaftliche Reihe}, issn = {1866-8372}, doi = {10.25932/publishup-51774}, url = {http://nbn-resolving.de/urn:nbn:de:kobv:517-opus4-517743}, pages = {1599 -- 1614}, year = {2021}, abstract = {Models for the predictions of monetary losses from floods mainly blend data deemed to represent a single flood type and region. Moreover, these approaches largely ignore indicators of preparedness and how predictors may vary between regions and events, challenging the transferability of flood loss models. We use a flood loss database of 1812 German flood-affected households to explore how Bayesian multilevel models can estimate normalised flood damage stratified by event, region, or flood process type. Multilevel models acknowledge natural groups in the data and allow each group to learn from others. We obtain posterior estimates that differ between flood types, with credibly varying influences of water depth, contamination, duration, implementation of property-level precautionary measures, insurance, and previous flood experience; these influences overlap across most events or regions, however. We infer that the underlying damaging processes of distinct flood types deserve further attention. Each reported flood loss and affected region involved mixed flood types, likely explaining the uncertainty in the coefficients. Our results emphasise the need to consider flood types as an important step towards applying flood loss models elsewhere. We argue that failing to do so may unduly generalise the model and systematically bias loss estimations from empirical data.}, language = {en} } @article{SamprognaMohorThiekenKorup2021, author = {Samprogna Mohor, Guilherme and Thieken, Annegret and Korup, Oliver}, title = {Residential flood loss estimated from Bayesian multilevel models}, series = {Natural Hazards and Earth System Sciences}, volume = {21}, journal = {Natural Hazards and Earth System Sciences}, publisher = {European Geophysical Society}, address = {Katlenburg-Lindau}, issn = {2195-9269}, doi = {10.5194/nhess-21-1599-2021}, pages = {1599 -- 1614}, year = {2021}, abstract = {Models for the predictions of monetary losses from floods mainly blend data deemed to represent a single flood type and region. Moreover, these approaches largely ignore indicators of preparedness and how predictors may vary between regions and events, challenging the transferability of flood loss models. We use a flood loss database of 1812 German flood-affected households to explore how Bayesian multilevel models can estimate normalised flood damage stratified by event, region, or flood process type. Multilevel models acknowledge natural groups in the data and allow each group to learn from others. We obtain posterior estimates that differ between flood types, with credibly varying influences of water depth, contamination, duration, implementation of property-level precautionary measures, insurance, and previous flood experience; these influences overlap across most events or regions, however. We infer that the underlying damaging processes of distinct flood types deserve further attention. Each reported flood loss and affected region involved mixed flood types, likely explaining the uncertainty in the coefficients. Our results emphasise the need to consider flood types as an important step towards applying flood loss models elsewhere. We argue that failing to do so may unduly generalise the model and systematically bias loss estimations from empirical data.}, language = {en} } @article{RoezerMuellerBubecketal.2016, author = {R{\"o}zer, Viktor and M{\"u}ller, Meike and Bubeck, Philip and Kienzler, Sarah and Thieken, Annegret and Pech, Ina and Schr{\"o}ter, Kai and Buchholz, Oliver and Kreibich, Heidi}, title = {Coping with Pluvial Floods by Private Households}, series = {Water}, volume = {8}, journal = {Water}, publisher = {MDPI}, address = {Basel}, issn = {2073-4441}, doi = {10.3390/w8070304}, pages = {24}, year = {2016}, abstract = {Pluvial floods have caused severe damage to urban areas in recent years. With a projected increase in extreme precipitation as well as an ongoing urbanization, pluvial flood damage is expected to increase in the future. Therefore, further insights, especially on the adverse consequences of pluvial floods and their mitigation, are needed. To gain more knowledge, empirical damage data from three different pluvial flood events in Germany were collected through computer-aided telephone interviews. Pluvial flood awareness as well as flood experience were found to be low before the respective flood events. The level of private precaution increased considerably after all events, but is mainly focused on measures that are easy to implement. Lower inundation depths, smaller potential losses as compared with fluvial floods, as well as the fact that pluvial flooding may occur everywhere, are expected to cause a shift in damage mitigation from precaution to emergency response. However, an effective implementation of emergency measures was constrained by a low dissemination of early warnings in the study areas. Further improvements of early warning systems including dissemination as well as a rise in pluvial flood preparedness are important to reduce future pluvial flood damage.}, language = {en} } @misc{RoezerMuellerBubecketal.2017, author = {R{\"o}zer, Viktor and M{\"u}ller, Meike and Bubeck, Philip and Kienzler, Sarah and Thieken, Annegret and Pech, Ina and Schr{\"o}ter, Kai and Buchholz, Oliver and Kreibich, Heidi}, title = {Coping with pluvial floods by private households}, url = {http://nbn-resolving.de/urn:nbn:de:kobv:517-opus4-400465}, pages = {24}, year = {2017}, abstract = {Pluvial floods have caused severe damage to urban areas in recent years. With a projected increase in extreme precipitation as well as an ongoing urbanization, pluvial flood damage is expected to increase in the future. Therefore, further insights, especially on the adverse consequences of pluvial floods and their mitigation, are needed. To gain more knowledge, empirical damage data from three different pluvial flood events in Germany were collected through computer-aided telephone interviews. Pluvial flood awareness as well as flood experience were found to be low before the respective flood events. The level of private precaution increased considerably after all events, but is mainly focused on measures that are easy to implement. Lower inundation depths, smaller potential losses as compared with fluvial floods, as well as the fact that pluvial flooding may occur everywhere, are expected to cause a shift in damage mitigation from precaution to emergency response. However, an effective implementation of emergency measures was constrained by a low dissemination of early warnings in the study areas. Further improvements of early warning systems including dissemination as well as a rise in pluvial flood preparedness are important to reduce future pluvial flood damage.}, language = {en} } @phdthesis{Wolf2017, author = {Wolf, Julia}, title = {Schadenserkennung in Beton durch {\"U}berwachung mit eingebetteten Ultraschallpr{\"u}fk{\"o}pfen}, url = {http://nbn-resolving.de/urn:nbn:de:kobv:517-opus4-397363}, school = {Universit{\"a}t Potsdam}, pages = {ix, 142}, year = {2017}, abstract = {Die zerst{\"o}rungsfreien Pr{\"u}fungen von Bauwerken mit Hilfe von Ultraschallmessverfahren haben in den letzten Jahren an Bedeutung gewonnen. Durch Ultraschallmessungen k{\"o}nnen die Geometrien von Bauteilen bestimmt sowie von außen nicht sichtbare Fehler wie Delaminationen und Kiesnester erkannt werden. Mit neuartigen, in das Betonbauteil eingebetteten Ultraschallpr{\"u}fk{\"o}pfen sollen nun Bauwerke dauerhaft auf Ver{\"a}nderungen {\"u}berpr{\"u}ft werden. Dazu werden Ultraschallsignale direkt im Inneren eines Bauteils erzeugt, was die M{\"o}glichkeiten der herk{\"o}mmlichen Methoden der Bauwerks{\"u}berwachung wesentlich erweitert. Ein Ultraschallverfahren k{\"o}nnte mit eingebetteten Pr{\"u}fk{\"o}pfen ein Betonbauteil kontinuierlich integral {\"u}berwachen und damit auch stetig fortschreitende Gef{\"u}ge{\"a}nderungen, wie beispielsweise Mikrorisse, registrieren. Sicherheitsrelevante Bauteile, die nach dem Einbau f{\"u}r Messungen unzug{\"a}nglich oder mittels Ultraschall, beispielsweise durch zus{\"a}tzliche Beschichtungen der Oberfl{\"a}che, nicht pr{\"u}fbar sind, lassen sich mit eingebetteten Pr{\"u}fk{\"o}pfen {\"u}berwachen. An bereits vorhandenen Bauwerken k{\"o}nnen die Ultraschallpr{\"u}fk{\"o}pfe mithilfe von Bohrl{\"o}chern und speziellem Verpressm{\"o}rtel auch nachtr{\"a}glich in das Bauteil integriert werden. F{\"u}r Fertigbauteile bieten sich eingebettete Pr{\"u}fk{\"o}pfe zur Herstellungskontrolle sowie zur {\"U}berwachung der Baudurchf{\"u}hrung als Werkzeug der Qualit{\"a}tssicherung an. Auch die schnelle Schadensanalyse eines Bauwerks nach Naturkatastrophen, wie beispielsweise einem Erdbeben oder einer Flut, ist denkbar. Durch die gute Ankopplung erm{\"o}glichen diese neuartigen Pr{\"u}fk{\"o}pfe den Einsatz von empfindlichen Auswertungsmethoden, wie die Kreuzkorrelation, die Coda-Wellen-Interferometrie oder die Amplitudenauswertung, f{\"u}r die Signalanalyse. Bei regelm{\"a}ßigen Messungen k{\"o}nnen somit sich anbahnende Sch{\"a}den eines Bauwerks fr{\"u}hzeitig erkannt werden. Da die Sch{\"a}digung eines Bauwerks keine direkt messbare Gr{\"o}ße darstellt, erfordert eine eindeutige Schadenserkennung in der Regel die Messung mehrerer physikalischer Gr{\"o}ßen die geeignet verkn{\"u}pft werden. Physikalische Gr{\"o}ßen k{\"o}nnen sein: Ultraschalllaufzeit, Amplitude des Ultraschallsignals und Umgebungstemperatur. Dazu m{\"u}ssen Korrelationen zwischen dem Zustand des Bauwerks, den Umgebungsbedingungen und den Parametern des gemessenen Ultraschallsignals untersucht werden. In dieser Arbeit werden die neuartigen Pr{\"u}fk{\"o}pfe vorgestellt. Es wird beschrieben, dass sie sich, sowohl in bereits errichtete Betonbauwerke als auch in der Konstruktion befindliche, einbauen lassen. Experimentell wird gezeigt, dass die Pr{\"u}fk{\"o}pfe in mehreren Ebenen eingebettet sein k{\"o}nnen da ihre Abstrahlcharakteristik im Beton nahezu ungerichtet ist. Die Mittenfrequenz von rund 62 kHz erm{\"o}glicht Abst{\"a}nde, je nach Betonart und SRV, von mindestens 3 m zwischen Pr{\"u}fk{\"o}pfen die als Sender und Empf{\"a}nger arbeiten. Die Empfindlichkeit der eingebetteten Pr{\"u}fk{\"o}pfe gegen{\"u}ber Ver{\"a}nderungen im Beton wird an Hand von zwei Laborexperimenten gezeigt, einem Drei-Punkt-Biegeversuch und einem Versuch zur Erzeugung von Frost-Tau-Wechsel Sch{\"a}den. Die Ergebnisse werden mit anderen zerst{\"o}rungsfreien Pr{\"u}fverfahren verglichen. Es zeigt sich, dass die Pr{\"u}fk{\"o}pfe durch die Anwendung empfindlicher Auswertemethoden, auftretende Risse im Beton detektieren, bevor diese eine Gefahr f{\"u}r das Bauwerk darstellen. Abschließend werden Beispiele von Installation der neuartigen Ultraschallpr{\"u}fk{\"o}pfe in realen Bauteilen, zwei Br{\"u}cken und einem Fundament, gezeigt und basierend auf dort gewonnenen ersten Erfahrungen ein Konzept f{\"u}r die Umsetzung einer Langzeit{\"u}berwachung aufgestellt.}, language = {de} } @article{SchroeterKreibichVogeletal.2014, author = {Schroeter, Kai and Kreibich, Heidi and Vogel, Kristin and Riggelsen, Carsten and Scherbaum, Frank and Merz, Bruno}, title = {How useful are complex flood damage models?}, series = {Water resources research}, volume = {50}, journal = {Water resources research}, number = {4}, publisher = {American Geophysical Union}, address = {Washington}, issn = {0043-1397}, doi = {10.1002/2013WR014396}, pages = {3378 -- 3395}, year = {2014}, abstract = {We investigate the usefulness of complex flood damage models for predicting relative damage to residential buildings in a spatial and temporal transfer context. We apply eight different flood damage models to predict relative building damage for five historic flood events in two different regions of Germany. Model complexity is measured in terms of the number of explanatory variables which varies from 1 variable up to 10 variables which are singled out from 28 candidate variables. Model validation is based on empirical damage data, whereas observation uncertainty is taken into consideration. The comparison of model predictive performance shows that additional explanatory variables besides the water depth improve the predictive capability in a spatial and temporal transfer context, i.e., when the models are transferred to different regions and different flood events. Concerning the trade-off between predictive capability and reliability the model structure seem more important than the number of explanatory variables. Among the models considered, the reliability of Bayesian network-based predictions in space-time transfer is larger than for the remaining models, and the uncertainties associated with damage predictions are reflected more completely.}, language = {en} }