Historic Floods Database

1. Historic Floods Database Ad de Roo, Jose Barredo

2. Background • ENV (water unit)* wants: • to establish a database of historic floods • As part of the Flood Directive deliverables • With consultation/approval by MS • existing databases not always verified data • to have it included in WISE • development to be done by JRC+EEA+ENV • Building on existing databases (EMDAT, DFO, MunichRe etc) • To be included in the database (proposed): • Location, extent (shapefile) • Start date, peak date, end date, duration • Type of flood (river flood, coastal, flashflood, pluvial flood) • Magnitude, return period • People affected, victims, evacuated/displaced • Economic damage (insured losses, uninsured losses) • Costs of measures * pers. communication with Maria Brättemark 18 May 2010

3. Aims of the database • Provide an easy overview mapping of floods in Europe, in a WISE application (note: thus publically available) • Comparing historic floods sites against MS reported Flood Directive (preliminary) flood risk maps (PFRM) and flood risk management plans (FRMP) • Calculate total damages and victims for floods • Assessment of trends of losses, casualties and number of damaging events • Provide a unique and large set of weblinks to information on historic floods

4. Work done sofar • JRC made – on request of ENV - a draft working paper on the design of an historic floods database • Checking existing databases • Proposed design

5. EM-DAT (CRED) What is EM-DAT? The EM-DAT database presents core data on the occurrence and effects of over 14,500 disasters from 1900 to present, including: • Natural disasters • Technological disasters Disasters in EM-DAT are defined as: “A situation or event which overwhelms local capacity, necessitating a request to the national or international level for external assistance, or is recognized as such by a multilateral agency or by at least two sources, such as national, regional or international assistance groups and the media” Criteria For a disaster to be entered into the database at least one or a combination of the following criteria must be fulfilled: • 10 or more people reported killed • 100 people or more reported affected • A declaration of a state of emergency • A call for international assistance

7. Source: International Strategy for Disaster Reduction (ISDR) and CRED (2010). Disaster statistics OCCURRENCE: trends-century. Effect of improvements in disaster data collection or anthropogenic forcings [?]

8. “One of the main contributors to this apparent increase of natural disasters is the launch of active data collection by the Office of US Foreign Disaster Assistance (OFDA) in 1960 and CRED in 1973. The punctual increases are indicated in the Figure bellow.” Source: Guha-Sapir, Hargitt and Hoyois (2004) Thirty years of natural disasters 1974-2003: the numbers (CRED).

9. Dartmouth Flood Observatory (DFO) • Global Active Archive of Large Flood Events (from 1985) • The information presented in DFO is derived from a wide variety of news, governmental, instrumental, and remote sensing source. • It is presented in order to facilitate research into the causes of extreme flood events, provide international warning of such floods, and improve widespread access to satellite-based measurements and mapping.

10. Dartmouth Flood Observatory (DFO)

12. Summary from existing databases • Main problems: • Location information not sufficiently detailed for the aimed purpose • Verification / validity of the data (feedback from several MS hydro authorities, e.g. on DFO) • General problems: • Available information from global databases of natural disasters is limited and suffers from a number of drawbacks. • One of the most significant limitations is the effect of the increase in the reporting of events in recent decades as result of improvements in data collection and flows of information. • Additionally, records are usually sourced from different institutions and hence collected using a wide range of different assessment methods and rationales. • All the above might further increase uncertainty, and so caution is needed in assessing datasets from global databases of natural disasters.

13. Proposal for a database • A European Historical Floods database could look similar to the Dartmouth and EM-DAT database. In addition, it would be useful to add: • Additional/more detailed information on the location (latitude/longitude centroid, or even a inundation area vector) • One aim of the database is to make sure that historic floods sites are included in the flood risk maps (provided sufficient measures are taken such that there is no significant flood risk anymore) • a more detailed estimate of the return period of the hydrologic flood event, and other hydrometeorological conditions causing the flood (amount of precipitation, rain on snow, “Gross Wetter Lage, e.g. Vb” • a weblink to more detailed information on the flood, images, flood inundation maps etc • A collaboration could/should be envisaged with CRED and DFO to keep the databases streamlined.

14. Applications of the database • A database such as suggested above would allow: • an easy overview mapping of floods in Europe (e.g. in a WISE application) • comparing historic floods sites with Flood Directive flood risk maps • calculate total damages and victims for floods • assessment of trends of losses, casualties and number of damaging events • restrict the analysis based on date of the flood, magnitude, or a specific country c.q. region of Europe • provide a unique and large set of weblinks to information on historic floods

15. Risks and difficulties • an incomplete database can lead to erroneous conclusions, e.g. on trends in the number of flood disasters, losses and casualties and the links with climate change • many floods are transnational: should they be mapped for each country individually? Or one database entry for the entire flood • The setup of this database requires a one-time only large effort from each MS (since MS verification is needed) • A limitation for the implementation of a EU-wide database could be the comparability of the datasets provided by MS. Available databases (e.g. Italy, Switzerland, Ireland, EM-DAT, DFO) use different criteria for the inclusion of disasters. Indeed in some cases the criteria have changed over time • A database such as this requires permanent maintenance and regular (annual?) input from MS • In some cases the “owner” of the database can be reluctant to share the baseline datasets

16. Next steps • JRC draft working paper is currently with ENV for comments • During last WG-F, COM asked MS to inform COM about existing databases • At next WG-F, COM will make a proposal to MS how to proceed

17. More information: http://floods.jrc.ec.europa.eu http://efas.jrc.ec.europa.eu Contacts: ad.de-roo@jrc.ec.europa.eu WMO RAVI meeting, Toulouse, 25-27 March 2009

18. Proposal for a historic floods database Practical way forward: • Merge existing CRED and DFO information into single database • Send list per MS for consultation / modification / additions • Merge updated lists into single database

19. Content: DFO includes the following fields: DFO# - An archive number is assigned to any flood that appears to be "large", with, for example, significant damage to structures or agriculture, long (decades) reported intervals since the last similar event, and/or fatalities. GLIDE# - GLobal IDEntifier Number. A globally common Unique ID code for disasters. Country - Primary country of flooding. Other affected countries are listed in three separate fields to the right of the main Country column. Locations - Includes names of the states, provinces, counties, towns, and cities. Rivers - Names of rivers. Begin - Ended - Ocassionally there is no specific beginning date mentioned in news reports, only a month; in that case the DFO date will be the middle of that month. Ending dates are often harder to determine - sometimes the news will note when the floods start to recede. We make an estimate based on a qualitative judgement concerning the flood event. Duration - Derived from start and end dates. Known Dead - News reports are usually specific about this, but occasionally there is only mention of 'hundreds' or 'scores' killed; in this case we estimate as follows: "hundreds"=300; "scores"=30; "more than a hundred =110 (number given plus 10%). If there is information on the number of people 'missing', the DFO does not include them in the total of deaths. We require an exact number for analytical purposes, but caution that our numbers are never more than estimates. Number Displaced - This number is sometimes the total number of people left homeless after the incident, and sometimes it is the number evacuated during the flood. News reports will often mention a number of people that are 'affected', but we do not use this. If the only information is the number of houses destroyed or damaged, then DFO assumes that 4 people live in each house. If the news report only mentions that "thousands were evacuated", the number is estimated at 3000. If the news reports mention that "more than 10,000" were displaced then the DFO number is 11,000 (number plus 10%). If the only information is the number of families left homeless, then DFO assumes that there are 4 people in each family. Damage (US $) - This number is never more than an estimate and we use no independent criteria for determining such. Instead we accept the latest and apparently most accurate number available in all the relevant sources. Main Cause - One of eleven main causes is selected: Heavy rain, Tropical cyclone, Extra-tropical cyclone, Monsoonal rain, Snowmelt, Rain and snowmelt, Ice jam/break-up, Dam/Levy, break or release, Brief torrential rain, Tidal surge, Avalanche related. Information about secondary causes is in the Notes and Comments section of the table. Severity Class - Assessment is on 1-2 scale. These floods are then divided into three classes. Class 1: large flood events: significant damage to structures or agriculture; fatalities; and/or 1-2 decades-long reported interval since the last similar event. Class 1.5: very large events: with a greater than 2 decades but less than 100 year estimated recurrence interval, and/or a local recurrence interval of at 1-2 decades and affecting a large geographic region (> 5000 sq. km). Class 2: Extreme events: with an estimated recurrence interval greater than 100 years. Geographic Flood Extents (sq km) - This is derived from our global map of news detected floods. Polygons representing the areas affected by flooding are drawn in a GIS program based upon information acquired from news sources. Note: These are not actual flooded areas but rather the extent of geographic regions affected by flooding. Magnitude (M) - Flood Magnitude =LOG(Duration x Severity x Affected Area) Notes and Comments - Name of associated typhoon, hurricane or storm. Notes on the weather associated with the flood event, if floods were accompanied by landslides, or if floods were caused or exacerbated by dam or levy failures. Notes of recurrance intervals, record flooding or rainfall. Notes on infrastructure and agricultural damages, including hectares of crops or arable land flooded, or total amount of land flooded. Breakdowns of dead and displaced by country or region. Direct quotes of interesting information from news and UN sources. If there is no information for a particular category available in our sources, then the cell is empty. GIS Files - Available in Mapinfo interchange format. Each flood event in the Register of Major Flood Events table has an associated GIS polygon representing the area affected by the flooding in that event. The same news and governmental sources that are used to complete the entries in the tables are used to determine an approximate geographic area that is affected by the flood event; this should not be confused with actual areas of inundation. When DFO has obtained satellite data and produced a Rapid Response Inundation Map for an event a link to that map is provided in the Country column of the Register of Major Flood Events.

20. EM-DAT (CRED) Content: EM-DAT includes the following fields: DISNO: A unique disaster number for each disaster event (8 digits: 4 digits for the year and 4 digits for the disaster number – for example, 19950324). Country: Country(ies) in which the disaster occurred. Disaster group: Two groups of disasters are distinguished in EM-DAT – natural disasters and technological disasters. Disaster type and subset: Description of the disaster according to a pre-defined classification. For example, type: Windstorm and subset: Cyclone or type: Transport; and subset: Rail. Date (start and end): The date when the disaster occurred and ended. (Month/Day/Year.) Killed: Persons confirmed dead and persons missing and presumed dead. Injured: People suffering from physical injuries, trauma or an illness requiring immediate medical treatment as a direct result of a disaster. Homeless: People needing immediate assistance for shelter. Affected: People requiring immediate assistance during a period of emergency; it can also include displaced or evacuated people. Total affected: Sum of injured, homeless and affected. Estimated damage: Several institutions have developed methodologies to quantify these losses in their specific domain. However, there is no standard procedure to determine a global figure for economic impact. Estimated damage is given in US dollars and/or euro. Additional fields: Other geographical information (such as location, latitude and longitude), the value and scale of the events (such as the Richter scale value for an earthquake), the international status (OFDA/EU response, request for international assistance, disaster/emergency declaration), the aid contribution (in US dollars) as well the sectors affected. EM-DAT is validated and updated daily. It is compiled from various sources, including UN agencies, governmental institutions, insurance companies, research institutes and the media according to a priority list set up by CRED.