Background to the PFRA European Overview - UC9810.5b

Background to the PFRA European Overview - UC9810.5b The individual Member State Reports reflect the situation as reported by the Member States to the European Commission in 2012 The situation in the MSs may have altered since then Assessment of data and information reported by Member States on their Preliminary Flood Risk Assessments and identification of Areas of Potentially Significant Flood Risk under the Floods Directive Member State Report: [UK] - [United Kingdom] The main outcomes of the assessment were: 1. There are significant differences in how the requirements of the Floods Directive have been implemented in the UK. In England and Wales Article 4 has been applied in terms of flooding from ordinary (minor) watercourses and all other sources of local flooding, groundwater, and pluvial, and Article 13.1.b to the main rivers and large raised reservoirs and sea water. In Scotland Article 4 has been applied to all relevant types of flood, and in Northern Ireland, Article 4 has been applied to fluvial and pluvial flooding. Finally in Gibraltar, Article 4 has been applied specifically to sea water floods. There are also significant differences in the methodological approaches between the 4 regions of the UK. 2. Article 4 has been applied in Scotland, Northern Ireland and Gibraltar for all types of flooding considered as relevant. In England and Wales, this Article has only been applied by Lead Local Flood Authorities (LLFA) who are responsible for pluvial and groundwater flooding, and from minor watercourses which include ditches and streams not included as main rivers. 3. Article 13.1(a) has not been applied to any specific area or specific type of flood for any of the RBDs in the UK. 4. Article 13.1(b) has only been applied to the UoMs in England and Wales and is applied to floods from sea water, main rivers and large raised reservoirs. The only information reported to WISE on methodologies was the overall approach, with the same text as reported for Article 4. 5. Only Scotland reported details (type and consequence) of significant historical flood events to WISE. England (361) and Wales (113) (E&W) provided summaries of flood events instead of the specific details requested. Northern Ireland also reported summaries of 27 flood events rather than in the detail requested. In Gibraltar, a summary of 1 flood event was reported to WISE. 6. In Scotland, the location of future floods is primarily predicted and information on historic floods is used to validate the predictions. In England and Wales, LLFAs used computer models to generate information on future floods. In Northern Ireland, the future flood risk is assessed using a GIS based Source Pathway Receptor model that combines the output from predictive flood models with a digital terrain model and a host of readily available receptor datasets. In Gibraltar, no modelling for future flooding was undertaken and the assessment was based on largely 1 of 25

anecdotal evidence of past flooding. Only flooding from rainfall and storm surges from the sea were considered to be significant for future risk assessments. 7. Adverse consequences were defined and considered, with regional differences in approach and methodology. 8. All regions have given consideration to the impact of climate change on future flood risk. Scotland, England, Wales and Northern Ireland have also considered the effect of building and planning. Question 0: Contextual information regarding the Member State. There are 3 main regions in the main part of the UK: Northern Ireland (3 UoMs); Scotland (1 main UoM (UK01) plus part of the UoM shared with England (UK02); and, England and Wales (10 UoMs, one shared with Scotland). In addition Gibraltar (UKGI17) reported a PFRA and APSFR. Information from 17 XML (two for UK02) files was included in the WISE aggregation reports and was therefore used in the assessment. In addition, a number of relevant documents were downloaded from the relevant national web sites to obtain more detailed and specific information where necessary. The documents downloaded and examined were: Title The National Flood Risk Assessment, December 2011(of Scotland), Flood Risk Management (Scotland) Act 2009 Appraisal Method for Flood Risk Management Strategies (not dated) Flood Risk Management (Scotland) Act 2009 (FRM Act) National Flood Risk Assessment Methodology, (not dated) Preliminary Flood Risk Assessment (PFRA), Annexes to the final guidance Report GEHO1210BTHF-E-E, 2 March 2011. Flood Risk Management Plans: Guidance for Risk Management Authorities in England and Wales, A Living Draft, August 2013 Selecting and reviewing Flood Risk Areas for local sources of flooding, Guidance to Lead Local Flood Authorities, Flood Risk Regulations 2009, (2010) Flooding in England: A National Assessment of Flood Risk (2009) Source downloaded from the Scottish Environment Protection Agency web site: http://www.sepa.org.uk/flooding.aspx downloaded from the Scottish Environment Protection Agency web site http://www.sepa.org.uk/flooding.aspx downloaded from the Scottish Environment Protection Agency web site. http://www.sepa.org.uk/flooding.aspx downloaded from Environment Agency (England) web site. http://a0768b4a8a31e106d8b0-50dc802554eb38a24458b98ff72d550b.r19.cf3.r ackcdn.com/geho1210bthf-e-e.pdf downloaded from Environment Agency (England) web site. http://a0768b4a8a31e106d8b0-50dc802554eb38a24458b98ff72d550b.r19.cf3.r ackcdn.com/lit_8649_4e4b09.pdf downloadable from www.defra.gov.uk or www.cymru.gov.uk downloaded from Environment Agency (England) web site. http://www.environmentagency.gov.uk/research/library/publications/10 2 of 25

Preliminary Flood Risk Assessment and Methodology for the Identification of Significant Flood risk Areas December 2011, Rivers Agency of Northern Ireland, Government of Gibraltar, Preliminary Flood Risk Assessment, Final Report, January 2011 8660.aspx downloaded from web http://www.dardni.gov.uk/final-pfra-report.pdf Downloaded from the web. https://www.gibraltar.gov.gi/new/sites/default/f iles/docs/preliminary%20flood%20risk%20as sessment%20report.pdf Flooding is a devolved responsibility for Wales, Scotland and Northern Ireland and so there are significant differences in how the requirements of the Floods Directive have been implemented in the UK. In England and Wales Article 4 has been applied in terms of flooding from ordinary (minor) watercourses and all other sources of local flooding, groundwater, and pluvial, and Article 13.1.b to the main rivers and large raised reservoirs and sea water. In Scotland Article 4 has been applied to all relevant types of flood, and in Northern Ireland, Article 4 has been applied to fluvial and pluvial flooding. Finally in Gibraltar, Article 4 has been applied specifically to sea water floods. There are also significant differences in the methodological approaches between the 4 regions of the UK. Table 1 The application of Articles, 4, 13.1.a and 13.1.b in the Units of Management of the United Kingdom Source: WISE Flood aggregation report FD 1.1 Specific Areas to which each Article has been applied UoM Article 4 Article 13.1.a Article 13.1.b UK01 1 0 0 UK02_England 1 0 1 UK02_Scotland 1 0 0 UK03 1 0 1 UK04 1 0 1 UK05 1 0 1 UK06 1 0 1 UK07 1 0 1 UK08 1 0 1 UK09 1 0 1 UK10 1 0 1 UK11 1 0 1 UK12 1 0 1 UKGBNIIENB 1 0 0 UKGBNIIENW 1 0 0 UKGBNINE 1 0 0 UKGI17 1 0 0 Note: if the Articles have not been applied to or reported for any specific area, it is assumed that they have been applied to the entire UoM. In which case the values in the table above will equate to 1. Values of zero for any Article or UoM indicate that that the Article has not been applied to that UoM. 3 of 25

Table 2 Specific types of floods to which Article 4, 13.1.a and 13.1.b have been applied Source: WISE Flood aggregation report: FD 1.2 Types of flood to which each Article has been applied Article UoM Source * Mechanism * Characteristic * All types ** Article 4 UK02_Englan d, UK03, UK04, UK05, UK06, UK07, UK08, UK09, UK10, UK11, UK12 "Ordinary (minor) watercourses and all other sources of local flooding", Groundwater, Article 4 UKGBNIIEN B, UKGBNIIEN W, UKGBNINE Pluvial Fluvial, Pluvial Article 4 UKGI17 Sea water Defence exceedance Article 4 UK01, UK02_Scotlan d Article 13.1.b UK02_Englan d, UK03, UK04, UK05, UK06, UK07, UK08, UK09, UK10, UK11, UK12 "Main Rivers and large raised reservoirs", Sea water "Natural Flood" * Source, mechanism and characteristics in quotation marks is source, mechanism and characteristics specified by member state ** No specific flood types were reported and it is assumed that Article 4 is applied to all flood types yes Question 1: Are all the types of flood that might be reasonably expected in the Member State included in the assessment of the risk of flooding under Article 4, Article 13.1(a) or Article 13.1(b)? Source Included Not included but Yes/No might be expected Fluvial Pluvial Yes Yes Groundwater Yes Sea water Yes Artificial water-bearing infrastructure Yes 4 of 25

Other (provide details in the summary below) No Yes There are 4 distinct regions in the UK relevant to this assessment: Scotland (UK01 and part of UK02); England and Wales (UK03 to UK12); Northern Ireland (one report covering the 3 UoMs); and Gibraltar. There are some differences in the reported information. In Scotland the National Flood Risk Assessment concentrated on rivers, coastal flooding and heavy rainfall with some (though limited) consideration of the influence of groundwater. No explicit mention is made of flooding from natural lakes (this is the other options ticked above) or artificial water bearing infrastructure including reservoirs or flooding from sewerage systems. Scotland has subsequently indicated that natural lakes were part of the fluvial network for which flood extents were developed. Additionally, flood risk information on a sub-set of Scotland s sewerage systems was developed after the NFRA but has been embedded into Scotland s surface water flooding maps. It is not considered a nationally significant source of flooding. Finally, groundwater was included but is not considered a primary source of flooding; rather it is a contributing factor to other types of flooding. Therefore, there is limited data on groundwater flooding that would support further investigation. In England and Wales, the Environment Agency has applied Article 13.1.b and is adapting the current hazard and risk maps to the requirement of the Floods Directive (WISE aggregation report). They are responsible for the main rivers, the sea and large raised reservoirs. England and Wales have subsequently indicated that PFRAs were not produced for river and coastal flooding as flood hazard and flood risk maps have been produced for the whole of England and Wales instead of identifying APSFRs. Future work will produce Flood Risk Management Plans for the whole of England and Wales. Lead Local Flood Authorities have applied Article 4 to surface runoff, groundwater and ordinary watercourse and any interactions these have with drainage systems and other sources of flooding including sewers (except in terms of blockages or failures) and also excluding burst (potable) water mains. Flooding from canals that are not main rivers should also be included in the PFRA. The EA considers snowmelt as precipitation that would lead to surface run-off and tsunamis as a form of sea flooding: it is not clear how the risk from these latter two sources would have been assessed. No explicit mention is made of flooding from natural lakes (this is the other options ticked above). In Northern Ireland the main flooding sources considered were rivers, the sea, surface water run-off and impounded water bodies (dams and reservoirs) though the latter source was not conclusively assessed because of the lack of readily available information to make this assessment: this shortfall will be dealt with in the future when new regulations will 5 of 25

enable the collection for the relevant data on reservoirs and dams. Groundwater flooding was excluded because it was considered insignificant in Northern Ireland compared to the other sources. There is also no explicit mention of natural lakes as a source of flooding. In Gibraltar, 5 sources of flooding were considered: rivers, sea, surface run-off, groundwater and infrastructure (flood defence) failure. There are no watercourses (rivers) in Gibraltar and the risk from groundwater was considered as minimal (because of the hydrogeology) and these sources were considered as representing no risk. Table 3 Time period covered by different types of historic flood events Source: WISE Flood aggregation report: FD 2.0 Time period covered by historic flood events Source Total flood events Range of years 2000 onwards 1950 to 1999 1900 to 1949 1800s Fluvial 92 1900-2009 19 27 46 0 0 Pluvial 15 1907-2010 5 2 8 0 0 Sea water 16 1949-2005 1 14 1 0 0 Artificial water-bearing 10 1990-2009 2 8 0 0 0 infrastructure Other: Peat 2 1979-2004 1 1 0 0 0 Slide/Debris Flow/Bog Burst No data 13 1914-2008 5 7 1 0 0 Before 1800 6 of 25

Table 4 Types of historical significant flood events and types of reported consequences Source: WISE Flood aggregation report: FD 2.1 Types of historical significant flood events Member State Article Source Mechanism Characteristics Number of Historical floods Number of flood events with no data Human Health Number with reported consequences Environment Cultural Heritage UK Article 4 no data 13 0 14 13 13 13 Economic Activity UK Article 4 "Peat Slide/Debris Flow/Bog Burst" 2 0 2 2 2 2 UK Article 4 Artificial water-bearing infrastructure 10 0 10 10 10 12 UK Article 4 Fluvial 92 0 95 92 92 124 UK Article 4 Pluvial 15 0 16 15 15 19 UK Article 4 Sea water 16 0 16 16 16 20 Table 5 Types of potential future significant flood events and types of consequences Source: WISE Flood aggregation report: FD 3.1 Types of potential future significant flood events Member State Article Source Mechanism Characteristics UK Article 4 "Main rivers" Natural exceedance Medium onset flood UK UK Article 4 "Ordinary (minor) watercourses" Article 4 "Ordinary (minor) watercourses" Number of Potential Future floods 7 of 25 Number of flood events with no data Human Health Number with reported consequences Environment Cultural Heritage Economic Activity 24 0 24 24 24 24 Natural exceedance Flash flood 2 0 2 2 2 2 Natural exceedance Medium onset flood 11 0 11 11 11 11

UK Article 4 Artificial waterbearing infrastructure Defence or infrastructural failure UK Article 4 Groundwater Natural exceedance Medium onset flood UK Article 4 Pluvial Defence exceedance Medium onset flood Flash flood 3 0 3 3 3 3 14 0 14 14 14 14 1 0 1 1 1 1 UK Article 4 Pluvial Natural exceedance 2 0 2 2 2 2 UK Article 4 Pluvial Natural exceedance Flash flood 1 0 1 1 1 1 UK Article 4 Pluvial Natural exceedance Medium onset flood 134 0 134 134 134 134 UK Article 4 Sea water Defence exceedance "Natural Flood" 1 0 1 1 1 1 8 of 25

Question 2a: What aspects required by Article 4 were not considered in the application of Article 4? Has Article 4 been applied? Yes, information has been reported on this aspect This Article has been applied in Scotland, Northern Ireland and Gibraltar for all types of flooding considered as relevant. In England and Wales, this Article has only been applied by Lead Local Flood Authorities (LLFA) who are responsible for pluvial and groundwater flooding, and from minor watercourses which include ditches and streams not included as main rivers (which are the responsibility of the Environment Agency of England and Wales who are applying Article 13.1.b for this type and some other types of flood). Based on the information reported to WISE, in Scotland all the expected aspects have been included in the PFRA. In England and Wales the Environment Agency provided the LLFAs guidance on what was required in a PFRA. Based on the guidance and the WISE report (and not an example of a PFRA produced by a LLFA) it seems that most aspects would have been included. However, there may be some limitations in the assessment of the risk from groundwater flooding as the dataset used (Areas Susceptible to Groundwater Flooding) is quoted not to be interpreted as identifying areas where groundwater is actually likely to flow or pond: rather it should be used to identify where further studies would be useful. England and Wales has subsequently indicated that further assessment of groundwater flooding is being considered as a result of flooding events that occurred in 2014. In terms of surface water flooding there was also no readily available or derivable information about the effectiveness of existing man made infrastructure (drainage). LLFAs are also response for assessing the hazard and risk from minor water courses. Some of these may have been covered by existing datasets but there may also have been gaps in the aspects such their geomorphological characteristics that were not included. For the PFRA in Northern Ireland most aspects have been considered. There was an initial consideration of the geomorphological characteristics of watercourses. Northern Ireland has subsequently indicated that it was not possible to develop a practicable broad-scale approach that would be sufficiently robust to support a meaningful assessment of the potential for geomorphological activity to change the future flood risk from the many hundreds of rivers under consideration. The effectiveness of flood defences was also ignored in the indicative (strategic) flood models used to assess the potential adverse consequences of future floods because of the uncertainty associated with the actual levels of protection offered by the existing defences (river walls, flood banks, culverts etc.). This approach was adopted to ensure that communities located behind the existing major flood defence systems were identified as areas of potential significant flood risk and thereby create the opportunity to develop detailed predictive flood models and produce hazard and risk maps that illustrate the estimated level of protection provided to the areas by the defences. The PFRA for Gibraltar also covers all aspects that are relevant to the characteristics of water bodies in Gibraltar. 9 of 25

Question 2b: What aspects required by Article 4 were not considered when producing an assessment of the risk of flooding under Article 13.1(a)? Has Article 13.1(a) been applied? No, it is explicitly stated that this Article has not been applied This Article has not been applied to any specific area or specific type of flood for any of the RBDs in the UK. Question 3: What aspects required by Article 4 were not considered when producing Flood Hazard Maps and flood risk maps, and Flood Risk Management Plans under Article 13.1(b)? Has Article 13.1(b) been applied? Yes, information has been reported on this aspect In the UK, this Article has only been applied to the UoMs in England and Wales (part of UK02 and UK03 to UK12), and is applied to floods from sea water, main rivers and large raised reservoirs. Large reservoirs are defined by a size criterion of 25,000 m3 capacity though this is to be decreased to 10,000 m3 capacity on implementation of new legislation. The Environment Agency of England and Wales will produce flood hazard and risk maps and flood risk management plans for flood risk from main rivers, large raised reservoirs and the sea by adapting the existing maps and plans to meet the requirements of the Floods Directive. The only information reported to WISE on Article 13.1.b methodologies was on the overall approach: this was the same text as reported for Article 4 and only describes the respective role of the EA and Lead Local Flood Authorities. A search of the Environment Agency of England and Wales web site did not locate any detailed methodological reports on the basis of existing flood hazard and flood risk maps. The statement that existing maps will be adapted to meet the requirements of the Floods Directive implies that the current maps do not cover all aspects outlined in Article 4. In addition no specific information was found on how the risk from large raised reservoirs would be assessed. 10 of 25

Question 4: What are the types of floods considered/not considered within the auspices of the Floods Directive? Considering the UK as a whole, fluvial, pluvial, groundwater, sea water and floods from artificial water bearing infrastructure have been considered but there are significant regional differences. Peat slide/debris flow/ bog burst floods were reported (in WISE) for Scotland (UK01) as historic but not potential future flood event. Scotland has subsequently indicated that the lack of available, detailed and reliable information on these types of flood prevented them from being included in the future flood assessment. Historic flooding in Scotland has also resulted from snow melt and ice blockages but again these were not included in the future flood assessment due to a lack of information and because they are infrequent and not considered significant. In Scotland the National Flood Risk Assessment (2011) concentrated on rivers, coastal waters and pluvial flooding with a further consideration of the influence of groundwater. However in the Scotland report (Appraisal method for Flood Risk Management Strategies First edition) it is stated that Groundwater flooding will not be considered within the appraisal process. Accurate data relating to groundwater risk in Scotland is not currently available, and there is uncertainty related to the potential for groundwater to cause flooding. It is understood that groundwater issues can exacerbate flooding from other sources however, the levels of flooding attributed to groundwater are uncertain. Raised reservoirs over 10,000 m3 capacity (artificial water bearing structures) are included in England Wales but not in Northern Ireland (because of a lack of information even though they are considered as being potentially significant) or in Scotland (no reason found). Scotland has subsequently indicated that this was due to a lack of available and reliable information prior to submission. It also noted that the Reservoirs (Scotland) Act 2011 includes a responsibility for SEPA to assign a risk designation to all reservoirs over 10,000 m3 capacity. This will be based on the potential adverse consequences of an uncontrolled release of water and the probability of such a release. The outputs from these assessments will be considered for inclusion in future reporting cycles. Canals are also considered in England and Wales. In England and Wales a differentiation is made between main rivers and ordinary water courses (e.g. stream, ditches and sluice) in terms of fluvial flooding for which, respectively, the Environment Agency of England and Wales and the Local Lead Flood Authorities (municipal authorities) are responsible (WISE report). Scotland has subsequently indicated that canals were not included in the assessment for Scotland as flood risk from canals is not considered significant and little information was available, Flooding from sewers does not have to be considered in England and Wales unless wholly or partly caused by rainwater entering or otherwise affecting the system (WISE). 11 of 25

Groundwater flooding is also considered to be insignificant in Northern Ireland (PFRA report) compared to other sources, as confirmed by a desktop review of hydrogeology and flooding history that demonstrated no areas at significant risk from groundwater flooding. In Gibraltar (UK17) (PFRA document) only two sources of potential flooding (pluvial and sea water) were considered to represent a risk (i.e. significant): there are no surface watercourses and the hydrological characteristics of the groundwater aquifers led to the conclusion that they represented no risk. A residual risk was considered to arise from sea flood defence failure during extreme tidal events. Question 5: What were the criteria used to define the historical significant floods and what were the reasons for not including some types of flood that occurred in the past? Only Scotland reported details (type and consequence) of significant historical flood events to WISE. Historic flood events were collated from a number of sources including SEPA data and newspaper records. 15,000 individual flood events were found covering all of Scotland. An impact score was assigned to each event ranging from 0 (very low impact) to 4 a very high impact. Impact was based on the effects on receptors including Human Health, Economic Activity, Environment and Cultural Heritage. Threshold criteria (not reported or found) were associated with the impacts on each receptor, the exceedance of which indicated significance. Any event which has a very high impact (4) was deemed to be significant. England (361) and Wales (113) (E&W) provided summaries of flood events rather than details, as permitted in this reporting cycle. Defra has subsequently indicated that this was because information on past flood events was not readily derivable. The reported information for E&W is for the PFRAs undertaken by the Lead Local Flood Authorities (e.g. county council and unitary administrative authorities) which have responsibility for flooding from defined sources such as pluvial and groundwater. The Environment Agency (EA) of E&W has responsibly for other sources of flooding such from main rivers and sea water and they have applied Article 13.1.b and hence are not producing PFRAs but are going straight to the development/revision of existing Flood Hazard (FH) and Flood Risk (FR) Maps. The Environment Agency (EA) provided guidance to the LLFAs on the production of PFRA which listed a number of sources of historic flood information. It is not clear how (or if) the EA used the historic information in the development of the FH and FR maps as their method seems to be solely based on modelling. The relevant information was sought for in a number of EA publications downloaded from their web site including National Assessment of Flood Risk. In terms of the LLFAs, the factors used to determine the significance of any harmful consequences of historic flood events included those that were significant on a national scale; how memorable the event was; scale of the flooding and consequences and the level of response; severity of impacts; and 12 of 25

on whether the quality of the information was sufficient to determine if there were significant harmful consequences. Northern Ireland also reported summaries of 27 flood events rather than in detail, as permitted within this reporting cycle. In Northern Ireland a library and media review of flood events was undertaken to establish the extent of major floods events dating back to early 1900 s. Since 1980 floods have been recorded by aerial photography. The Flood risk associated with post 1980 flood events has been assessed using a GIS application which calculated key Flood Risk Indicators associated with their adverse consequences to Human Health, economic activity, Environment and cultural heritage. No information was found on what criteria equated to a historic event being significant in Northern Ireland. In Gibraltar, information on past floods was found from a wide range of sources including Government departments, museums and utility companies. A summary of 1 flood event was reported to WISE. There was no information on what equated to a historic significant floods though groundwater floods were not considered in the PFRA because there were no recorded events from this source. No information was found from the other parts of the UK on the basis of exclusion of specific flood types from the consideration of historic flood events. Question 6: What methods and criteria were used to identify potentially significant future floods and what were the reasons for not including some types of potential future floods? In Scotland the location of future floods is primarily predicted. Information on historic floods is used to validate the predictions. The 1 in 200 year (annual exceedance probability of 0.5%) indicative river and coastal flood extents were used and these were integrated with national (Scottish) pluvial and groundwater datasets. A systematic 1km 2 grid approach was used in assessing flood hazard to defined receptors from fluvial, coastal and pluvial flood sources. A groundwater factor was also applied for areas which were classified as having a high or very high susceptibility to flooding. In general however the focus of the National Flood Risk Assessment (NFRA) was on fluvial, pluvial and coastal flooding rather than that from groundwater. The significance of the records of historic flood events was assessed using a scoring method incorporating the impact, source and reliability of the event. This data was then used to validate the NFRA grid output ensuring that significant historic events were used in the determination of Potentially Vulnerable Areas. Because of some uncertainty in the methods for assessing flood risks, a manual review was undertaken to take account of any uncertainties within the grid outputs. In terms of the sources of flooding (pluvial, groundwater and ordinary watercourse) for which the LLFAs are responsible in England and Wales, computer models were used to generate information on future floods. In terms of surface run-off the derived national datasets included areas susceptible to surface water flooding and the flood map for surface 13 of 25

water. The assessment was based on flooding to a depth of 0.3 metres by a rainfall event with a chance of 1 in 200 of occurring in any given year (equivalent to 'in the order of' 1 in 100 chance of flooding). These datasets were supplemented by locally agreed surface water information which best represented local conditions. For groundwater a national 1 km 2 grid datasets was used where no local information was available. In terms of ordinary water course there were no datasets that dealt solely with this source and so national surface run-off datasets were used when possible supplemented by any local information. No specific information relating to the Environment Agency s methods for assessing the risk of flooding from coastal waters, main rivers and from raised reservoirs was found in spite of extensive search on the relevant web sites. In Northern Ireland the future flood risk is assessed using a GIS based Source Pathway Receptor model that combines the output from predictive flood models with a digital terrain model and a host of readily available receptor datasets. The assessment of the potential adverse consequences of flooding is based on three different probability events (low, medium and high) for each of the flood sources. By producing flood outlines with different return periods it was possible to annualise Flood Risk Indicator values to produce estimates of the adverse effects that best represent the true long term annual average. The AEP of the events used in the assessment for each source is fluvial: 10%, 1% and 0.1%, coastal; 10%, 0.5% and 0.1% and pluvial 10%, 0.5% and 0.1%. No information was reported on the methods for groundwater and artificial water bearing infrastructure. In Gibraltar no modelling for future flooding was undertaken and the assessment was based on largely anecdotal evidence of past flooding. Only flooding from rainfall and storm surges from the sea were considered to be significant for future risk assessments. Question 7: What types of flood were considered but not assessed as being significant, and what were the reasons given? In Scotland the significance of historic floods was assessed by assigning an impact score to each event ranging from 0-4, where 0 indicates a Very Low impact event and 4 indicates a Very High Impact event. This allowed areas historically prone to flooding to be highlighted. An impact threshold matrix was developed to ensure consistency when assigning impact scores. The matrix contained the most common examples of each impact for each receptor. Event descriptions must include the threshold criteria in order to be designated as a significant event. A score multiplier was added after the impact scoring. This took into account the age of the flood event and ensured that events were more fairly scored given the changes in, for example, building standards. Any event which had a Very High Impact was deemed to be Significant. Where significant flood events occurred within an APSFR, it was considered that significant adverse consequences of similar future flood events may be envisaged. The PFRA in Scotland concentrated on flooding from rivers, the coast and heavy rainfall with a further consideration of the influence of groundwater. 14 of 25

There is no mention of potential floods from impounded water bodies (e.g. reservoirs or canals) or natural lakes or specifically of flooding from sewers. No reasons were found why they were not considered. Scotland has subsequently indicated that natural lakes were part of the fluvial network for which flood extents were developed. In England and Wales reservoirs below 10,000m³ in capacity were considered as unlikely to present significant flood risks in the context of implementing the requirements of the Floods Directive. In addition the Lead Local Flood Authorities (LLFAs) do not need to assess flooding from sewers, unless wholly or partly caused by rainwater or other precipitation entering or otherwise affecting the system. Floods of raw sewage caused solely, for example, by a sewer blockage do not fall under the national Regulations implementing the Floods Directive. The Regulations also do not apply to floods from water supply systems, e.g. burst water mains, Snow melt and tsunamis. Snowmelt would count as precipitation and so could lead to surface runoff. Tsunamis are considered as a form of flooding from the sea. LLFAs also did not include past floods of a kind that are not likely to occur now due to improvements in drainage or flood risk management measures. In Northern Ireland an initial review of historical flooding was undertaken to scope the potential risk from all flood sources. From this it was concluded that the main flood sources likely to give rise to significant flooding were rivers, the sea, surface water runoff and impounded water bodies (such as dams and reservoirs). However, the flood risk from impounded water bodies was not conclusively assessed in the PFRA as there was at the time insufficient available or readily available information to conduct a robust assessment of the risk from this source. The reason for this lack of information is that NI does not have legislation for the management of reservoir safety and as a consequence the owners of impoundments have not been required to collate such information as would be necessary to assess the potential risk of their failure. NI will address this risk through the introduction of new reservoir safety legislation. It is stated that Groundwater flooding is insignificant in Northern Ireland compared with fluvial, coastal and pluvial flooding. As a consequence it was decided not to explore the development of predictive flood models for this source. The only sources of flooding considered to be of potential significance in Gibraltar were from the sea and from surface water run-off. Neither of these was considered to represent a significant risk. Flooding from the two groundwater bodies in Gibraltar was also considered but because of their hydrological characteristics they were not considered as being relevant. A strategic assessment of the drainage and sewerage network within the Gibraltar RBD was additionally used to assess the level of flood risk. Question 8: What types of flood were not considered at all, and why? 15 of 25

In Scotland it appears that flood from impounded water bodies (e.g. reservoirs or canals), natural lakes and from sewers were not considered. No reasons for this were found. In England and Wales all types of flood except from natural lakes seem to have been considered. England and Wales has subsequently indicated that it is included in the flood maps but is not considered a separate source of flooding. From the reported information for Northern Ireland it is not clear whether canals are included in and considered as part of impounded water bodies. Natural lakes such as Lough Neagh were also not mentioned. Flooding from sewers is also not described as a potential source of flooding. Rivers do not occur on Gibraltar, hence this source was not considered. It is not known whether or not impounded water bodies occur in Gibraltar; no mention was made of them in the WISE reports. Question 9: What criteria were used to define an adverse consequence? In Scotland the following receptors of flooding were considered: Human Health (A) People: No. of People/social flood vulnerability; Human Health (B) Community: Community facilities that could cause community disruption if affected by flooding Economic Activity (A) Businesses: No. of business properties and the estimated damages; Economic Activity (B) Transport: Roads, railways and airports; Economic Activity (C) Agriculture: Agricultural land and forestry areas; Environment: Designated Areas and the susceptibility and resilience of the species/habitats to the impacts of flooding; Cultural Heritage: the importance of cultural sites exposed to flooding. The criteria that defined an adverse consequence on each receptor were not reported to WISE. Scotland has subsequently indicated that flood risk was calculated as a product of likelihood, hazard, vulnerability, exposure and value. The other aspects required to be considered under Article 4.2.d (hydrological and geomorphological characteristics; effectiveness of existing manmade flood defence infrastructures; climate change; future land planning) were included in the PFRA. In England and Wales Lead Local Flood Authorities (responsible for pluvial, groundwater and minor watercourse flooding) used the following flood risk indicators to assess the potential adverse consequences of future flooding. Human health; number of people affected, number of critical services; Economic activity; number of non-residential properties, infrastructure network (length of roads and rail), and area of agricultural land; Environment and heritage, the consequences of pollution, the impact on internationally and nationally designated environmental sites, and the impact on internationally and nationally 16 of 25

designated heritage assets. It is assumed that consequences of pollution might include any effect on water body status (i.e. WFD status). Elements representing the topography, the position of watercourses and their general hydrological and geomorphological characteristics, including floodplains as natural retention areas were also included in the modelling for future potential floods. The position of populated areas, areas of economic activity were also described. The criteria that defined an adverse consequence on each receptor were not reported to WISE. No equivalent information was reported to WISE for the other sources of floods for which England and Wales is applying Article 13.1.b. However, as the authority responsible for these other sources of flooding (Environment Agency of England and Wales) provided guidance to the LLFAs in producing PFRAs, it seems likely that the same approach and criteria would have been applied. In Northern Ireland a GIS application was developed to combine the flood extent outlines with a wide range of receptor datasets. Embedded within the application is a Flood Risk Query Tool which applies algorithms to the data to quantify the flood risk in terms of flood risk indicators. A broad range of flood risk indicators have been generated to measure the adverse impact of potential flooding on groups of receptors on an average annual basis. For example there are Flood Risk Indicators that quantify the annual flood risk in terms of the number of people and their vulnerability, the number of properties (and their various types), the number of key services, the length of key transportation infrastructure flooded, the economic damages, the areas of environmental sites inundated, the number of Integrated Pollution Prevention and Control sites and the number of nationally important cultural heritage sites (i.e. Listed Buildings, Gardens, Sites and Monuments Records and Sites of Archaeological Interest). An example of a key flood risk indicator is the Aggregated Average Annual Damage (AAAD). This is the estimated average economic damages arising from all sources of flooding which, taken over the very long-term, is likely to occur on an annual basis. The Flood Risk Indicators are supported as necessary by information on the vulnerability of particular receptors. For example, UK census data was used to identify more vulnerable populations, such as conglomerations of elderly population. The spatial distribution of the Economic Deprivation Index computed by the Northern Ireland Statistical Service was computed for flooded properties and used in the assessment of the vulnerability of people to flooding. The criticality of key infrastructure and designated environmental sites susceptible to flooding was determined through a broad ranging consultation exercise that included the owner/operators of the asset owners and the NI Environment Agency. The criteria associated with the Flood Risk Indicators that defined an adverse consequence on each receptor were not reported to WISE. In Gibraltar the criteria used to identify and assess significant flood risks included, but were not limited to public health impacts, damage to property and infrastructure, number of people affected along with the impacts of pollution. These were considered to be the main indicators that are applicable to Gibraltar. Gibraltar did not identify any areas that 17 of 25

they considered to be at a significant flood risk but areas of land were identified that might become inundated during flooding events from storm (sea) surges or heavy rainfall, but where significant damage is unlikely to occur. Question 10: What adverse consequences were excluded or not considered, and what were the reasons for their exclusion? None seem to have been excluded in Scotland s PFRA (WISE and downloaded documents). In England and Wales the LLFAs may not have considered some of the issues identified under Article 4.2(d) in some of the future flood information where it was considered not appropriate, or the information was not available, for assessment of local flood risk. For example, the broad scale modelling of local flood risk has limited or no readily available or derivable information about the effectiveness of existing manmade infrastructure. In the case of surface water infrastructure it is usually a matter of very small local interventions, including elements of the drainage and pumping network, or even variations in kerb heights or the camber of a road that can influence the effectiveness of drainage. For future floods such as the Areas Susceptible to Groundwater Flooding, and some local future flood information, no automatic analysis was available about the position of populated areas, and areas of economic activity as contained within the Environment Agency's National Receptor Dataset. For groundwater LLFAs made their own assessment of the impacts from local knowledge or overlaying with other maps of spatial information. The influence of geomorphology was considered in the PFRA (PFRA document) for Northern Ireland and an activity index was generated and based on land cover, drift geology and stream-power with validation based on inspecting digital terrain / photography images. Reaches with the greatest potential for geomorphological activity that might affect flood risk were highlighted. However, the analysis was a broad-scale approach and more site specific detailed studies of deposition and flood risk would be required. Given the uncertainty associated with the approach used it was concluded that geomorphology should not be considered further when assessing the potential consequences of future floods. However, geomorphology data will be taken into account when options are being considered for flood risk management plans. Flood defence infrastructure has been ignored in the predictive modelling of future flooding in Northern Ireland. This is a precautionary approach because there are uncertainties regarding the levels of protection that flood defences would provide in practice given the detailed information and analysis required for proper assessments. 18 of 25

There is little detailed information in the PFRA downloaded document for Gibraltar. The text implies that adverse impacts on human health, economy, environment and cultural heritage have been considered. Question 11: What methods were used to identify and quantify potential future adverse consequences and impacts? In Scotland a 1 in 200 year indicative river and coastal flood extents were used for the PFRA. These represent the floodplains as natural retention areas. National pluvial and groundwater datasets were also developed and integrated into the PFRA. An annual exceedance probability factor of 0.5 was applied. Pluvial flooding extents were generated using a 0.1 m and 0.3 m depth contours for use within the assessment. Each receptor was assessed in terms of exposure to flood hazards, and also the characteristics of receptors (i.e. the value/vulnerability to impacts of flooding). A 1km² grid approach was adopted where grid cells were attributed with a risk score depending on the characteristics of the receptors within the cell. A grid output was created for fluvial, coastal and pluvial flood sources, for each receptor category, these were combined to provide a total score per cell. A groundwater factor was then included for areas which had been classified to have a High or Very High susceptibility to groundwater flooding. A quantitative method was then used for translating the grid output to a higher geographic level called the sub catchment unit (SCU). This summed the scores of all the cells within each SCU and categorised the SCUs from Very Low to Very High. Additional information only relevant or applicable at the catchment scale was then included in the analysis. Flood defences were considered in a 2 phase process: the 1st phase identifies the protection offered by defences; the 2nd phase is a manual, qualitative assessment of the area behind a defence and the potential benefit offered- both are carried out at the SCU level. The results of the assessment of historic floods was used to validate the grid output, ensuring that significant past flood events have been identified and analysed to inform the categorisation of SCUs and the definition of Potential Vulnerable Areas (PVA). All SCUs categorised as Very High, High or Medium categories were included as PVAs. The receptor groups used in the PFRA for the identification of flood risk areas in Scotland were: 1. Human Health (A) People (No. of Residential Properties and the social vulnerability of the area); 2. Human Health (B) - Community (Important facilities that could cause community disruption if affected e.g. schools, hospitals); 3. Economic Activity (A) - Businesses (No. of business properties and the estimated weighted annual average damage related to the property); 4. Economic Activity (B) - Transport (Roads, railways and airports); 5. Economic Activity (C) - Agriculture (Agricultural land and forestry areas); 6. The Environment (Areas designated for natural heritage purposes and their vulnerability to flooding); and, 7. Cultural Heritage (Cultural sites such as UNESCO 19 of 25

World Heritage Sites). The receptors were assessed using risk categories from Very Low to Very High. A large range of thresholds and methods have been tested to ensure that that the resulting categorisation most accurately represents flood risk in Scotland. These are described in detail in Flood Risk Management (Scotland) Act 2009 (FRM Act), National Flood Risk Assessment Methodology, Scottish Environment Protection Agency, downloaded from internet. The main source of information about future floods in England and Wales is in the form of maps produced by computer models. National maps of surface water flooding called the Areas Susceptible to Surface Water Flooding (AStSWf) and the Flood Map for Surface Water (FMfSW) include elements representing the topography, the position of watercourses and their general hydrological and geomorphological characteristics, including floodplains as natural retention areas. The position of populated areas, areas of economic activity are described by the Environment Agency's for England and Wales National Receptor Dataset, and analysis of FMfSW and AStSWF were informed by this. The FMfSW contains two rainfall events, divided into two depth bandings: 1 in 200 rainfall and 1 in 200 rainfall deep, as well as 1 in 30 rainfall and 1 in 30 rainfall deep. The AStSWF dataset contains one rainfall event, with three susceptibility bandings: less, intermediate and more. Some Local Lead Flood Authorities (responsible for pluvial, groundwater and flooding from minor water course etc) were also able to use other future flood scenarios produced to assess the impact of local sources of flooding from past studies of specific areas. National maps of river and sea flooding were available in addition to the surface water maps to assess potential areas where different sources of flooding could interact. Information about potential groundwater flooding was provided from Areas Susceptible to Groundwater Flooding map to help assess areas where this type of flooding could pose an additional risk. Using the available maps and data sets, LLFAs were able to assess the potential adverse consequences of future flooding. The Environment Agency of England and Wales provided a very broad scale map showing groundwater flood areas on a 1km² grid, called "Areas Susceptible to Groundwater Flooding". This was deemed sufficient for the PFRA where no local information exists. There are no Environment Agency products that deal solely with flooding from ordinary watercourses, however the national surface water maps, and the Flood Map for rivers and sea (for which the Environment Agency of England and Wales is responsible) contain some relevant information, and so were included in PFRAs. Where additional information about future flooding from ordinary watercourses exists locally, it was recorded in the PFRA. In England and Wales Defra and the Welsh Assembly Government provided guidance (Selecting and reviewing Flood Risk Areas for local sources of flooding: Guidance to Lead Local Flood Authorities Flood Risk Regulations 2009, Department for Environment, 20 of 25