Publication - Research and analysis

Mapping flood disadvantage in Scotland 2015: report

Published: 23 Dec 2015
Directorate:
Environment and Forestry Directorate
Part of:
Environment and climate change
ISBN:
9781785448478

This research identifies and maps the neighbourhoods in Scotland that would be most disadvantaged by flooding.

Mapping flood disadvantage in Scotland 2015: report
4. Results

4. Results

This section reports the results of the assessment of flood disadvantage in Scotland. Section 4.1 presents the results of investigating the exposure of residential properties to different types of flooding in Scotland, aggregated to data zones and local authorities. Section 4.2 provides the outcomes of the assessment of social vulnerability to flooding. In section 4.3, the flood disadvantage in relation to different types and return periods of flooding is analysed. In section 4.4, the geography of social vulnerability to flooding and flood disadvantage is explored with regard to urban-rural classification of the data zones and their proximity to the coast. Section 4.5 discusses the results against the PVAs identified in NFRA. Finally, section 4.6 presents the three local authority case studies.

4.1. Exposure of residential properties to flooding in Scotland

In total, over 4% of residential properties in Scotland (just over 108,000) are estimated to be exposed to one or more sources of flooding of low probability (Table 5). River flooding affects the greatest number of residential properties, followed by coastal and surface water flooding.

Table 5. Exposure of data zones and residential addresses to flooding

Flood type and return period Number of data zones exposed 1 % data zones exposed % residential addresses exposed
C25 286 4.4 0.4
C200 375 5.8 0.5
C200+cc 552 8.6 1.2
R30 1327 21.2 1.0
R200 1619 25.0 1.9
R200+cc 1821 27.8 2.9
S30 1431 22.1 0.3
S200 1924 29.7 0.6
S200+cc 2043 31.5 0.7
Any 200+cc 2 3166 48.7 4.4

1 Considering the 6500 data zones identified to contain population on the day of the 2011 census
2 These figures refer to the percentage of neighbourhoods which are exposed to any flood type, whether it is coastal, river or surface water (or any combination) at 1:200 years including climate change return period.

Nearly half of all data zones in Scotland contain residential properties which may be exposed to any type of flooding of 1 in 200 years return period, considering the impacts of climate change (Table 5). Surface water flooding at low probability affects the greatest number of data zones, indicating the widespread character of surface water flooding, compared to the lower number of data zones exposed to river and coastal flooding.

Figure 2 presents the spatial distribution of flood hazard-exposure (all types of flooding combined; the largest extents considered). The highest concentration of residential properties at risk of any type of flooding is present in Falkirk [5] and Stirling, followed by Scottish Borders, Orkney Islands and West Dunbartonshire (Figure 3).

Figure 4, Figure 5 and Figure 6 present the percentage of residential properties exposed to river, coastal and surface water flooding at different probability levels. Falkirk, Orkney Islands and West Dunbartonshire have the highest proportion of residential properties exposed to coastal flooding. Stirling, Scottish Borders, and Perth and Kinross and Moray have the highest proportion of residential properties exposed to river flooding, whilst surface water flooding may affect the highest proportion of residential properties in Aberdeen City, Highland and Moray, followed by Renfrewshire and Falkirk.

Figure 2. Percentage of residential properties exposed to any type of flooding (1:200+cc). Base map is Ordnance Survey data © Crown Copyright and database right 2015. Derived from OS AddressBase and SEPA data.

Figure 2. Percentage of residential properties exposed to any type of flooding (1:200+cc).

Figure 3. Percentage of residential properties in local authorities exposed to flooding. [6]

Figure 3. Percentage of residential properties in local authorities exposed to flooding.

Figure 4. Percentage of residential properties in local authorities exposed to river flooding.

Figure 4. Percentage of residential properties in local authorities exposed to river flooding.

Figure 5. Percentage of residential properties in local authorities exposed to coastal flooding. [7]

Figure 5. Percentage of residential properties in local authorities exposed to coastal flooding.

Figure 6. Percentage of residential properties in local authorities exposed to surface water flooding.

Figure 6. Percentage of residential properties in local authorities exposed to surface water flooding.

4.1.1. Recently constructed properties exposed to flooding

The Statement of Principles, the agreement currently in place between the Association of British Insurers ( ABI) and the Scottish Government (2008), ensures that insurance is available to many previously flooded or at-risk customers. The ABI proposed a new scheme to safeguard the availability and affordability of flood insurance for those at high risk, called Flood Re. While this scheme is being developed, ABI members voluntarily continue to meet their commitments to their existing customers under the old Statement of Principles agreement. However, this commitment does not apply to any new property built after 1 January 2009 in order to discourage the development of properties on flood plains. Therefore, some of these properties may not be insured.

The extent of residential properties built or re-developed after 1 st January 2009 that are exposed to flooding was therefore of interest, and was estimated from available data. Since no alternative data was available, residential addresses with a 'start date' on or after 1 st January 2009 were identified in the OS AddressBase dataset.

It should be noted that this assessment is exploratory in character and therefore has a number of caveats including:

  • Significant uncertainty on the property dataset used;
  • The paucity of data on detailed flood risk assessments and management of coastal and surface water flooding around the property which may enable development that is appropriate within the context of Scottish planning policy; and,
  • An absence of information on the property lowest floor level and any PLP present.

As such, it might only be concluded that, of around 100,000 properties noted in the database with a start date on or after 1 January 2009, a very small proportion (single-digit percentage) could potentially be located in areas affected by flooding. Further investigation is required to better understand such development, including the nature of the property dataset (to validate the records relating to new development) and into the supporting information that may enable appropriate development in line with Scottish planning policy. This is being further investigated by a number of ongoing projects (see section 5.2).

4.2. Social vulnerability to flooding in Scotland

A third of the neighbourhoods in Scotland have below average social vulnerability to flooding (table 6). Just below 8% of the data zones are classified as having an extremely high or acute vulnerability to flooding. These are mainly located within large Scottish cities, with Glasgow containing 191 such data zones, Edinburgh - 82; Dundee - 44 and Aberdeen - 27. Figure 7 presents the number of neighbourhoods with above average social vulnerability to flooding per local authority. Section 4.6 provides more detailed information for case study local authorities (Dumfries and Galloway, Dundee City and Scottish Borders). Figure 8 presents the spatial distribution of social vulnerability to flooding in Scotland.

Table 6. Data zones of different levels of social vulnerability to flooding

Vulnerability Number of data zones Percentage of data zones
Extremely low 262 4.0
Relatively low 1881 28.9
Average 2620 40.3
Relatively high 1226 18.9
Extremely high 399 6.1
Acute 112 1.7

However, not all data zones classed as extremely or acutely vulnerable are likely to be exposed to flooding (see section 4.1). In fact, just under half of them are exposed to flooding. This means that disadvantage to flooding is only likely to occur in a small number of neighbourhoods. This is explored in the next section.

Figure 7. Number of data zones classified as having above average social vulnerability to flooding in local authorities.

Figure 7. Number of data zones classified as having above average social vulnerability to flooding in local authorities.

Figure 8. Social vulnerability to flooding in Scotland. Base map is Ordnance Survey data © Crown Copyright and database right 2015. Data sources used in developing the social vulnerability to flooding index are listed in Table 2.

Figure 8. Social vulnerability to flooding in Scotland.

4.3. Flood disadvantage in Scotland

Flood disadvantage occurs where high levels of social vulnerability to flooding coincide spatially with high level of hazard-exposure, i.e. areas where a high percentage of residential properties are exposed to flooding.

Flood disadvantage has been categorised into six classes (Table 3), where negative values indicate lower than average flood disadvantage. The higher the positive values, the higher the flood disadvantage; the highest values of the index are in the 'acute' category. The values close to zero are near the national average.

Flood disadvantage has only been calculated for the data zones exposed to a given type of flooding (see Table 4). Table 7 summarises the levels of flood disadvantage in Scotland with regard to different types of flooding. Figure 9 presents the spatial distribution of flood disadvantage in Scotland with regard to all types of flooding combined; the data zones not exposed to any type of flooding are shaded out in grey.

The next sections focus on the extreme and acute flood disadvantage (see Table 7) and its distribution among local authorities. The data zones where high social vulnerability to flooding coincides with high exposure to flooding should be prioritised for action to protect the most vulnerable communities from the impacts of flooding.

Table 7. Levels of flood disadvantage in Scotland

Flood type and return period Number of data zones of varying level of flood disadvantage % of extremely or acutely disadvantaged data zones
Acute Extremely high Relatively high Average Relatively low Extremely low Not exposed Of all data zones of data zones exposed to flooding
C25 11 13 33 134 95 0 6214 0.4 5.3
C200 12 18 54 162 129 0 6125 0.5 8.0
C200+cc 16 35 79 228 194 0 5948 0.8 9.2
R30 38 55 181 634 460 4 5128 1.4 6.8
R200 53 70 213 720 561 2 4881 1.9 6.8
R200+cc 55 78 264 780 638 6 4679 2.0 7.2
S30 27 57 192 718 433 4 5069 1.3 5.9
S200 42 72 278 899 623 10 4576 1.8 5.9
S200+cc 47 84 296 941 663 12 4457 2.0 6.4
Any 200+cc 98 138 444 1411 1055 20 3334 3.6 7.4

Figure 9. Flood disadvantage in Scotland (any flood source 1:200+cc). Base map is Ordnance Survey data © Crown Copyright and database right 2015. Derived from OS AddressBase, SEPA data and data sources listed in Table 2.

Figure 9. Flood disadvantage in Scotland (any flood source 1:200+cc).

4.3.1. Extreme flood disadvantage - any type of flooding

With regard to any type of flooding, with the widest spatial extent considered, 236 neighbourhoods (3.6% of all data zones or 7.4% of those exposed to flooding) can be classified as extremely (138) or acutely (98) disadvantaged [8] .

Figure 10 presents the number of extremely and acutely flood disadvantaged neighbourhoods with regard to any type of flooding per local authority. Four local authorities (Na h-Eileanan Siar, Fife, Midlothian and Shetland Islands) contain no data zones that would be extremely or acutely disadvantaged with regard to flooding and hence are not presented in figure 10. However, whilst the assessment at the data zone level shows no extreme disadvantage, individual households and people may still be vulnerable and exposed to flooding. Therefore, these local authorities should not be exempt from local assessment of flood risk.

The highest number of neighbourhoods with acute flood disadvantage are present in Falkirk, Glasgow, North Ayrshire and Stirling. Extremely high flood disadvantage is present mainly in Glasgow and Edinburgh, followed by Dumfries and Galloway.

The next sections look in more detail at levels of flood disadvantage in individual local authorities with regard to coastal, river and surface water flooding at different return periods. As highlighted in section 3.4, the disadvantage index has been calculated for each of the flood types and return periods separately, therefore whilst some of the local authorities do not contain acutely/extremely disadvantaged data zones in relation to 'any type of flooding', they may contain high levels of disadvantage with regard to individual flood types and return periods (for example, Fife).

Firstly, the number of extremely and acutely flood-disadvantaged neighbourhoods in all local authorities is presented. Secondly, the number of data zones is shown as a percentage of extremely and acutely flood-disadvantaged data zones in Scotland, therefore showing the relative contribution of the local authority to the overall flood disadvantage in Scotland.

Figure 10. Number of data zones classified as acutely or extremely disadvantaged (any type of flooding, 1:200+cc) per local authority.

Figure 10. Number of data zones classified as acutely or extremely disadvantaged (any type of flooding, 1:200+cc) per local authority.

4.3.2. Extreme flood disadvantage - coastal flooding

With regard to coastal flooding, fifteen local authorities contain extremely and acutely disadvantaged data zones [9] . Falkirk, West Dunbartonshire and Orkney Islands have the highest percentage of data zones classified as extremely disadvantaged (Figure 11). Falkirk, West Dunbartonshire, Highland and Dumfries and Galloway contribute the highest proportion of extremely/acutely flood disadvantaged data zones in relation to Scotland as a whole (Figure 12).

4.3.3. Extreme flood disadvantage - river flooding

Considering river flooding, 26 local authorities contain extremely or acutely disadvantaged neighbourhoods. Aberdeen and Scottish Borders have the highest percentage of extremely/acutely disadvantaged neighbourhoods with regard to high probability (1:30) river flooding. Stirling, Moray, Scottish Borders and Aberdeen have the highest percentage of extremely/acutely disadvantaged neighbourhoods with regard to medium probability (1:200) river flooding. When the low probability (1:200+cc) river flood events are considered, Stirling, Scottish Borders and East Ayrshire have the highest percentage of extremely/acutely disadvantaged neighbourhoods (Figure 13).

Aberdeen City contains 16% of the extremely/acutely disadvantaged neighbourhoods in Scotland for the high probability (1:30) river flooding. It is followed by Glasgow, Dumfries and Galloway and North Ayrshire. When the medium probability of river flood events (1:200) is considered, Aberdeen City, Glasgow and Edinburgh contain over a quarter of extremely/acutely disadvantaged neighbourhoods in Scotland. When the low probability (1:200+cc) flooding is considered, the acute and extreme flood disadvantage concentrates in Edinburgh, Stirling and Highland, followed by Falkirk and Aberdeen (Figure 14).

4.3.4. Extreme flood disadvantage - surface water flooding

Six local authorities do not contain any neighbourhoods that are extremely/acutely disadvantaged with regard to surface water flooding (Dundee City, East Lothian, Midlothian, Moray, Na h-Eileanan Siar and Shetland Islands).

Glasgow, Aberdeen, Falkirk and Orkney Islands (Figure 15) have the highest percentage of extremely/acutely flood disadvantaged data zones at all return periods [10] . When the contribution of individual local authorities to the overall number of extremely/acutely disadvantaged neighbourhoods in Scotland is considered, Glasgow presents the highest concentration of flood disadvantage with a third of the extremely disadvantaged neighbourhoods being located there (Figure 16). This is followed by City of Edinburgh and Aberdeen City.

Figure 11. Percentage of data zones classed as extremely or acutely flood disadvantaged with regard to coastal flooding in local authorities.

Figure 11. Percentage of data zones classed as extremely or acutely flood disadvantaged with regard to coastal flooding in local authorities.

Figure 12. Relative contributions to Scotland's total number of extremely or acutely flood disadvantaged neighbourhoods from the named local authority with respect to coastal flooding (%).

Figure 12. Relative contributions to Scotland's total number of extremely or acutely flood disadvantaged neighbourhoods from the named local authority with respect to coastal flooding (%).

Figure 13. Percentage of extremely and acutely flood disadvantaged neighbourhoods with respect to river flooding in local authorities.

Figure 13. Percentage of extremely and acutely flood disadvantaged neighbourhoods with respect to river flooding in local authorities.

Figure 14. Relative contributions to Scotland's total number of extremely or acutely flood disadvantaged neighbourhoods from the named local authority with respect to river flooding (%).

Figure 14. Relative contributions to Scotland's total number of extremely or acutely flood disadvantaged neighbourhoods from the named local authority with respect to river flooding (%).

Figure 15. Percentage of extremely and acutely flood disadvantaged neighbourhoods with respect to surface water flooding in local authorities.

Figure 15. Percentage of extremely and acutely flood disadvantaged neighbourhoods with respect to surface water flooding in local authorities.

Figure 16. Relative contributions to Scotland's total number of extremely or acutely flood disadvantaged neighbourhoods from the named local authority with respect to surface water flooding (%).

Figure 16. Relative contributions to Scotland's total number of extremely or acutely flood disadvantaged neighbourhoods from the named local authority with respect to surface water flooding (%).

4.3.5. Estimating flood-disadvantaged population

The flood disadvantage assessment has been carried out for the number of vulnerable households that may be exposed to flooding, rather than the number of people that may be affected. Some very cautious estimates of the number of people who may be exposed to flooding, or flood disadvantaged, can be provided based on the average household size in data zones. The uncertainty in estimating these figures is associated with the flood hazard data used and the use of a mean household size, which may vary considerably between individual households.

The average household size was estimated for each data zone based on the census 2011 data (population divided by number of households). This was multiplied by the number of households exposed to flooding using SEPA flood maps.

The total number of people that may be exposed to any type of flooding at low probability (1:200+cc) in Scotland (not differentiating between vulnerable and not vulnerable groups), is around 228,000. The highest number of people are likely to be exposed to river flooding (Figure 17). Whilst the number of people who may be exposed to medium probability (1:200) surface water flooding exceeds the number of people exposed to medium and high probability coastal flooding, the population numbers exposed to coastal flooding exceed those exposed to surface water flooding when low probability flood events are considered.

Figure 17. Estimated population that may be exposed to different types of flooding.

Figure 17. Estimated population that may be exposed to different types of flooding.

Flood disadvantage partially reflects the percentage of residential properties exposed to flooding within a data zone (see section 3.3). Therefore, generally the higher the level of disadvantage, the higher the proportion of residential properties exposed. Figure 18 shows that even though the number of extremely high and acutely disadvantaged data zones is relatively low (138 and 98 respectively), they contain a disproportionate number of people that may be negatively affected by flooding due to their combined vulnerability and exposure (around 40,000 and 60,000 respectively). Flood risk managers may wish to focus on these areas when implementing management plans in order to reduce the impact of flooding on the well-being of a high number of vulnerable people.

An estimated 100,000 people in Scotland are acutely or extremely flood disadvantaged.

Figure 18. Number of people and data zones of different flood disadvantage levels (any type of flooding at low probability - 1:200+cc)

Figure 18. Number of people and data zones of different flood disadvantage levels (any type of flooding at low probability - 1:200+cc)

With regard high probability (1:30) coastal flooding, around 10,000 people are extremely or acutely flood disadvantaged; however, if the low probability (1:200+ CC) flood risk is considered, over 28,000 people may be flood-disadvantaged.

In the case of river flooding, around 20,000 people are extremely or acutely flood-disadvantaged with regard to high probability (1:30) flood events. This number doubles for medium probability (1:200) and triples for low probability (1:200+cc).

Over 6,000 people are extremely or acutely flood-disadvantaged with regard to high probability (1:30) surface water flooding; over 10,000 in relation to the medium probability (1:200) and nearly 14,000 when low probability (1:200+cc) flooding is considered.

4.4. Geographical distribution of social vulnerability to flooding and flood disadvantage

The levels of social vulnerability to flooding and flood disadvantage were compared amongst data zones in different types of settlements, and for coastal versus inland areas, in order to identify any geographical patterns present and if any types of locations should be investigated further through analysis of fine-grained vulnerability and prioritised for consideration of flood management actions.

The six-fold Scottish Urban Rural Classification 2013-2014 (Table 8; Scottish Government, 2014c) was used, which differentiates between different sizes of settlements and different accessibility levels. Data zones were classified based on the location of the population-weighted centroids [11] . With regard to coastal areas, 1353 data zones were located within 1km of the coast and 2218 data zones were within 2km of the coast.

Table 8. Urban-rural classification of the data zones (Scottish Government, 2014c).

Class Class name Description Number of data zones
1 Large urban areas Settlements of 125,000 people and over. 2163
2 Other urban areas Settlements of 10,000 to 124,999 people. 2327
3 Accessible small towns Settlements of 3,000 to 9,999 people, and within a 30 minute drive time of a Settlement of 10,000 or more 614
4 Remote small towns Settlements of 3,000 to 9,999 people, and with a drive time of over 30 minutes to a Settlement of 10,000 or more. 231
5 Accessible Rural Areas Areas with a population of less than 3,000 people, and within a 30 minute drive time of a Settlement of 10,000 or more. 751
6 Remote Rural Areas Areas with a population of less than 3,000 people, and with a drive time of over 30 minutes to a Settlement of 10,000 or more. 419

4.4.1. Distribution of social vulnerability to flooding among urban and rural areas

Of the 511 extremely high or acutely vulnerable data zones, 373 were located in large urban areas and 116 in other urban areas. However, extremely low vulnerability also tended to focus in urban areas: of 262 data zones classed as having extremely low vulnerability, 76 were present in large urban areas and 130 were located in 'other urban' areas (Figure 19). Therefore, urban areas tend to contain the extremes of vulnerability. Local authorities in urban areas need to recognize the presence of contrasting areas, often in close proximity, and plan for the management of social vulnerability accordingly.

Social vulnerability to flooding has a strong urban component.

Figure 19. Social vulnerability to flooding: number of data zones by six-fold urban-rural classification.

Figure 19. Social vulnerability to flooding: number of data zones by six-fold urban-rural classification.

Accessible small towns and accessible rural areas have the highest proportion of neighbourhoods of below-average social vulnerability to flooding. Remote small towns and remote rural areas tend to have social vulnerability around the average value for Scotland. However, this assessment has been carried out for the average values at the level of a neighbourhood which may mask highs and lows in vulnerability of individual households or people.

As for remote small towns, when the individual vulnerability indicators are explored, they emerge as having potential issues with social and physical isolation and mobility of people, which may raise issues with regard to responding to flood events. On average, they have the highest proportion of people living in care and medical establishments and tend to have higher number of households without cars compared to accessible small towns, and both accessible and remote rural areas. They also have the highest proportion of single pensioner households. This raises issues with the provision of resources during the flood emergency. Also, remote small towns have the second highest proportion of people working far away from home (after remote rural areas). Therefore, whilst the communities living in small remote towns are usually regarded as close-knit and having strong levels of self-help, the high proportion of people with limited physical capabilities during the event of flooding, with a high proportion of the working-age population away from home, may require additional resources.

Remote small towns and remote rural areas tend to be vulnerable due to social and physical isolation combined with older populations

Also, the vulnerable populations in remote rural areas may be negatively affected by flooding. A high proportion of single pensioner households (second after remote small towns), combined with low road density, long distances to the nearest GP surgery, and a high proportion of people working far away from home again raises issues of the ability of the community to respond to and recover after flooding, and presents a challenge for local authorities to spread their resources over large, sparsely populated areas. However, remote rural areas also have the highest number of location- and community-specific charities; the literature suggests that non-governmental organizations can successfully target social isolation in remote rural areas. Thus, locally-based charities should be considered important stakeholders in actions aiming at reducing social vulnerability to flooding.

4.4.2. Distribution of flood disadvantage among urban and rural areas

Flood disadvantage in Scotland, when all types of flooding are considered, tends to be concentrated in urban areas; in particular the smaller urban areas (10,000 to 124,999 people) contain a high proportion of extremely and acutely disadvantaged neighbourhoods (Figure 20).

Flood disadvantage is concentrated in urban areas.

This pattern is also present for coastal flooding, with the highly disadvantaged data zones being predominantly in 'other urban areas'; in contrast, rural areas do not contain any extremely or acutely disadvantaged data zones with regard to coastal flooding. River flood disadvantage also has a strong urban component, with acute levels of disadvantage concentrated in smaller urban settlements followed by large urban areas. Surface water flooding-related disadvantage is mainly present in large urban areas, as this is where surface water flooding tends to occur due to the high proportion of sealed surfaces and the pressure on the drainage systems. Table 9 summarises the number of extremely and acutely disadvantaged data zones with regard to different types of flooding and different return periods.

Figure 20: Levels of flood disadvantage for any type of flooding at low probability (1:200+cc) by six-fold urban-rural classification.

Figure 20: Levels of flood disadvantage for any type of flooding at low probability (1:200+cc) by six-fold urban-rural classification.

Table 9. Number of acutely and extremely disadvantaged data zones by six-fold urban-rural classification.

Flood type and return period Number of extremely and acutely disadvantaged data zones
Large urban Other urban Accessible small towns Remote small towns Accessible rural areas Remote rural areas
C25 5 18 0 1 0 0
C200 6 19 1 3 1 0
C200+cc 9 39 1 2 0 0
R30 41 29 7 4 2 10
R200 47 50 11 3 5 7
R200+cc 43 67 11 3 2 7
S30 53 27 3 0 0 1
S200 73 37 2 1 0 1
S200+cc 87 38 4 1 0 1
Any 200+cc 89 110 15 6 6 10

4.4.3. Comparison of coastal and inland areas

Coastal areas (defined as 2km distance from the coast) have a higher proportion of extremely and acutely vulnerable and disadvantaged data zones than areas located further inland (Figure 21). Therefore, coastal areas should be considered as a priority for flood risk management actions in order to reduce the impacts on vulnerable communities.

Social vulnerability to flooding and flood disadvantage have a strong coastal dimension.

Figure 21. Percentage of data zones classified as extremely and acutely vulnerable or disadvantaged within and outside coastal areas

Figure 21. Percentage of data zones classified as extremely and acutely vulnerable or disadvantaged within and outside coastal areas

4.5. Social vulnerability to flooding and flood disadvantage in the context of NFRA

In Scotland, 243 Potentially Vulnerable Areas ( PVAs) have been identified in the NFRA (Figure 22). They contain 92% of the total number of properties at risk within Scotland ( SEPA, 2011a).

There are some substantial differences in the vulnerability assessment employed in NFRA and in this study. Whilst the assessment of social vulnerability to flooding and flood disadvantage in this project was focused largely on the characteristics of the population, the NFRA took into account the density of residential properties and the Social Flood Vulnerability Index ( SFVI) [12] (Tapsell et al., 2002), which considers some of the vulnerability factors included in the assessment of social vulnerability to flooding reported here alongside a variety of other factors in delineating PVAs, including Economic Activity, Cultural Heritage and Environment ( SEPA, 2011b).

In addition, the underlying social and environmental data differ between the two assessments. The flood maps produced by SEPA for the strategic flood risk assessment are being constantly updated and NFRA was developed based on a different version of flood maps, and the socio-economic data used is earlier than 2011.

Further, the spatial scale of the underlying data differs: this assessment is based on the averages for data zones, whilst SEPA used 1km 2 grid, adjusted to accommodate for Sub Catchment Unit boundaries ( SEPA, 2011). Both approaches have their advantages: whilst the 1km 2 grid offers the equal-size unit approach to the assessment, most of the socio-economic data is reported for census or administrative units. In addition, in densely populated urban areas, where data zones are quite small, the 1km 2 grid may be too coarse to allow identification of fine-scale variability in social vulnerability to flooding or flood disadvantage.

In order to assess to what extent vulnerable and disadvantaged areas identified in this assessment reflect the results of NFRA, data zones were spatially overlaid with the PVAs [13] .

The analysis indicates that over 82% (5349) of the data zones coincide with PVAs. The data zones identified as overlapping with PVAs were found to have higher levels of social vulnerability to flooding and flood disadvantage (in relation to any type of flooding) than the data zones located outside the PVAs (see Figures 23 and 24).

All 112 acutely vulnerable data zones were located within PVAs and 97.5% of the extremely vulnerable data zones were also located within PVAs. The remaining 10 extremely vulnerable data zones were in Dumfries and Galloway, Aberdeen City, Fife, Highland and Inverclyde (see Figure 21). When flood disadvantage in relation to any type of flooding (1:200+cc) is considered, only one of 98 acutely flood disadvantaged neighbourhoods and five of 138 extremely disadvantaged neighbourhoods fell outside PVAs. These were in Dumfries and Galloway, Highland and East Ayrshire (see Figure 22). These locations could be considered by SEPA in the next NFRA cycle for consideration as PVAs.

Figure 22. Flood disadvantaged and vulnerable areas located outside the Potentially Vulnerable Areas in Scotland (as identified in NFRA (( SEPA, 2011a)). Base map is Ordnance Survey data © Crown Copyright and database right 2015.

Figure 22. Flood disadvantaged and vulnerable areas located outside the Potentially Vulnerable Areas in Scotland (as identified in NFRA ((SEPA, 2011a)).

Figure 23. Social vulnerability to flooding within and outside PVAs

Figure 23. Social vulnerability to flooding within and outside PVAs

Figure 24. Flood disadvantage (any 1:200+cc) within and outside PVAs (for data zones exposed to flooding only).

Figure 24. Flood disadvantage (any 1:200+cc) within and outside PVAs (for data zones exposed to flooding only).

4.6. Case studies of local authorities

4.6.1. Introduction

For the flood disadvantage assessment to be meaningfully translated into strategies which focus on preparing for, responding to, and recovering from flooding, local authorities or other agencies must recognise that they need to address vulnerability beyond emergency response.

Local authorities play a central role in leading and supporting local places to become more resilient to a range of future risks, and effective solutions on how to support vulnerable groups are recommended to be found and led by the local community or local council (The Scottish Government, 2012b). Therefore, local authority perceptions of the flood disadvantage assessment method and outputs, as well as the potential uses of the datasets, were gathered through engaging with three different local authorities: Dumfries and Galloway, Dundee City Council and the Scottish Borders.

The three case study local authorities have different characteristics. For instance, Dundee is an urban local authority. Dumfries and Galloway and the Scottish Borders have no large urban areas, but they both contain smaller urban settlements; Dumfries and Galloway contains a substantial proportion of remote rural areas (Figure 25).

Figure 25. Number of data zones in case study areas by six-fold urban-rural classification (based on Scottish Government 2014c).

Figure 25. Number of data zones in case study areas by six-fold urban-rural classification (based on Scottish Government 2014c).

All three local authorities have access to the coast. In Dundee, two-thirds of the data zones are located within 2km from the coast. In Dumfries and Galloway, one-third of the data zones are located within 2km from the coast, whilst only 4% of the Scottish Borders' data zones are located within 2km from the coast.

All three local authorities are involved in Local Plan Districts where the draft Flood Risk Management strategies have been put out to consultation. These will be published in December 2015.

4.6.2. Flood disadvantage in case study local authorities

Table 10 presents the number of data zones that are exposed to different types of flooding and return periods in the three case study authorities. The exposure to flooding varies. In Dundee, just over 17% of the data zones contain residential properties exposed to any type of flooding. By contrast, the exposure to flooding in Dumfries and Galloway is much more widespread with nearly 77% of the data zones containing residential properties located in flood risk areas.

Table 10. Number of data zones exposed to flooding in the case study authorities

Flood type and return period Dundee City Scottish Borders Dumfries and Galloway
C25 2 1 31
C200 3 1 31
C200+cc 7 2 31
R30 8 70 110
R200 12 83 119
R200+cc 13 87 125
S30 9 42 53
S200 12 59 70
S200+cc 14 59 72
Any 200+cc 31 95 148
Total number of data zones 179 130 193

Dundee City contains the highest proportion of data zones characterised by extremely high or acute social vulnerability to flooding. By contrast, a very small proportion of the Scottish Borders' data zones are characterised as extremely vulnerable and this local authority has the highest proportion of data zones of below national average vulnerability (Figure 26). Dumfries and Galloway is located between these two local authorities in terms of the proportion of neighbourhoods of extreme social vulnerability to flooding.

Figure 26. Levels of vulnerability in the three case study areas (percentage of data zones in different classes of social vulnerability to flooding).

Figure 26. Levels of vulnerability in the three case study areas (percentage of data zones in different classes of social vulnerability to flooding).

Considering flood disadvantage gives more nuanced results. In Dundee City only four data zones have extremely high or acute flood disadvantage. Whilst high levels of vulnerability in Dundee City may be of concern to various council departments (high values of various vulnerability indicators suggesting high levels of material deprivation, social isolation, considerable proportion of sealed surfaces and so on), the majority of data zones classed as vulnerable are not exposed to any type of flooding (; see also Table 11). The levels of flood disadvantage are thus higher in the Scottish Borders and Dumfries and Galloway (8 and 12 data zones with extremely high or acute disadvantage respectively).

Figure 27. Levels of flood disadvantage (Any 200+cc) in the three case study local authorities.

Figure 27. Levels of flood disadvantage (Any 200+cc) in the three case study local authorities.

Figures 28-30 present the disadvantage maps (in relation to any flooding of 1 in 200 years return period) for the three local authorities, including the disaggregation to different aspects of vulnerability for selected disadvantaged areas.

The coloured blocks in the bar chart represent the values of the vulnerability dimensions relative to the average Scottish neighbourhood (represented by the horizontal axis). Bars above the horizontal axis show positive vulnerability dimension values (greater than the Scottish average for each of the five dimensions shown in the legend). Bars below the horizontal axis show negative vulnerability dimension values (lower than Scottish average for each of the five dimensions shown in the legend). Therefore, the bars pointing upwards indicate higher than average sensitivity (blue bar) and exposure (green bar), and high inability to prepare (yellow bar), respond (red bar) and recover (purple bar). Bars pointing downwards indicate lower than average sensitivity and exposure and higher than national average ability to prepare, respond and recover.

The dimensions of vulnerability and the underlying indicators were analysed for three data zones identified in the case study local authorities as having acute flood disadvantage (Figures 28-30). Table 11 presents the values of vulnerability indicators for the selected data zone for each local authority case study in relation to the national average. This is now discussed in depth to demonstrate how the dimensions of vulnerability, in conjunction with the individual indicators, can be analysed in order to learn more about the underlying reasons for vulnerability in a given location.

The Scottish Borders (Hawick)

In the selected data zone in Hawick (S01005374) the sensitivity levels are close to the national average (Figure 28). Therefore, when considering Table 11, Hawick does not have any particular issues with aspects relating to a high proportion of older people or those in ill-health. Furthermore, enhanced exposure levels are close to the national average primarily because the proportion of houses with the lowest level at ground level is much lower than the national average.

The inability to recover in Hawick is much higher than the national average, and higher than for any of the other neighbourhoods surrounding this particular data zone. Also the inability to prepare and respond to flooding are substantially higher than the national average. By looking at the values of individual indicators (Table 11), it can be seen that Hawick contains:

  • A higher number of pension credit claimants. Whilst the older population is close to the national average, those in Hawick claim more pension credits, which suggests that they may be materially deprived. Additionally, the proportion of older people living alone is higher than the Scottish average. These groups may require extra assistance before, during and after flood events.
  • A high proportion of new addresses in a flood risk area. Hawick is in Edinburgh's commuter belt which leads to pressures for new housing. Whilst the indicator should be treated with caution (see section 3.3.), it may indicate that a proportion of households will struggle to get insurance under the new Flood Re regime. They may also struggle to obtain affordable insurance presently because of the history of flooding in this area, judging from the number of historic flood events (Table 11).
  • A high proportion of houses rented from private landlords. Tenanted properties are less likely to have contents insurance. In addition, tenants may not be able to install PLP. This may indicate a need to collaborate with landlords on making properties better prepared for flooding and encouraging landlords to help their tenants obtain contents insurance
  • In terms of its strengths, there is a high proportion of properties within SEPA's flood warning areas, which suggests that there may be a higher awareness of flooding. In addition, there are a higher number of locally-based charities than the national average. Such charities may support the local authority in flood preparation, response and recovery - the local authorities have also confirmed that Hawick has an active flood group. Hawick has a lower than average crime rate which may make it easier to work with the community during a flood as people will be less fearful of leaving their homes unattended. Residents of Hawick also have good access to GPs which can aid the recovery of sensitive groups.

Dumfries and Galloway (Newton Stewart)

Data zone S01000960 in Newton Stewart, Dumfries and Galloway, has very high levels of sensitivity whilst the ability to prepare is close to the national average. These combine with higher than national average levels of the inability to respond and recover (Figure 29). The indicators show that there are:

  • High proportions of older people and those in poor health. These groups are particularly sensitive to flood events. Compared to the national average, there is also a higher proportion of people living in medical and care establishments who may have more difficulty evacuating during a flood event.
  • There is a history of flood events but the area is not within SEPA's flood warning areas. Households which are not in a flood warning area may have less awareness about flooding. SEPA is currently developing flood warnings for this area.
  • The neighbourhood has a settled population with few recent arrivals. This may indicate that the community is close-knit and with good potential for self-help. However, there are a high proportion of people working away from home which means that the people remaining at home (for example older people, those with young children or in poor health) may be unable to help themselves. In this case, the high number of locally-based charities could provide assistance to those who cannot help themselves.
  • The selected area has a low density of roads. This means that the area may get cut off in the event of flooding and it may be difficult to reach inhabitants during a flood event.

The analysis suggests that the local authority could focus on addressing the needs of the sensitive (older, poorer health) population during and after flood events.

City of Dundee, Waterfront area

The selected data zone for Dundee (S01001108) has different characteristics yet again. The levels of sensitivity are below the national average. However, the levels of inability to prepare for, respond to, and recover after flooding, are higher than the national average. Enhanced exposure is also higher than the national average (Figure 30). Analysis of the indicators demonstrates that:

  • There are high levels of enhanced exposure, which are linked to the low presence of green space. The local authority may wish to investigate the surface sealing in that area and the type of housing present. Strategies to vary the land cover may be assessed, e.g. increasing green spaces and providing sustainable drainage systems ( SUDS) that may help to mitigate the floods in the area.
  • The population are generally younger and healthier. These groups are less sensitive with respect to floods.
  • The number of Job Seekers Allowance and Income Support claimants is higher than national average. Those on low incomes may be less likely to have contents insurance and may be less likely to afford PLP to make their homes more resilient.
  • There are higher proportions of new arrivals from outside the UK. In addition there is a higher proportion of people not speaking English well. This may mean that the communication of flood risk, flood warnings and preparation for flood events may not reach these groups. In addition, there are high proportions of new residents from within the UK and a very high proportion of private lets: both groups may not be familiar with the area. This could suggest the need for a tailored communication strategy that may utilise third sector organisations and/or landlords.
  • There has been a very high number of previous flood events. Insurers may be reluctant to provide affordable insurance in these areas and therefore inhabitants may not have their buildings and contents insured.
  • Domestic break-ins are relatively high. This has implications for the response and recovery phases of a flood event because people may be reluctant to leave their belongings behind. A higher police presence may be needed during flood events.
  • There is a relatively limited presence of voluntary organisations. This was confirmed by workshop participants who noted the low provision of any community resources in this area (schools, nurseries and community centres). This may mean that social networks are poor and that there is little support to be garnered from the third sector should the area be flooded.

Table 11. Values of social vulnerability to flooding indicators for the selected data zones (identified by codes) in three case-study authorities

Indicator Dundee,
Waterfront
S01001108
Dumfries & Galloway,
Newton Stewart
S01000960
Scottish Borders,
Hawick
S01005374
National average
% people under 5 years old 5.42 3.18 5.17 5.42
% people over 75 years old 1.07 20.85 12.53 7.92
% people whose day-to-day activities are limited 10.72 35.87 20.69 20.08
% households with at least one person with long term limiting illness 15.52 50.18 30.27 35.03
% people in routine or semi-routine occupations 14.78 34.17 40.75 29.37
% of people who are long term unemployed or who have never worked 4.16 4.27 3.58 5.10
% households with dependent children and no adults in employment 3.93 3.25 2.34 3.94
Number of Income Support claimants 40 14 15.00 20
Number of Job seeker allowance claimants 50 13 31.25 22
Total pension credit claimants 31 53 76.25 39
Total number of families receiving tax credits 65 25 70.00 63
% people with <1 year residency in the UK 9.11 0.00 0.46 0.93
% people who do not speak English/no not speak English well 2.35 1.81 1.07 1.41
% new addresses (01.01.2009) in flood risk areas 2.84 2.80 32.92 0.20
Number of historic flood events 34 6 7 1
% addresses in Flood Warning Target Areas 54.44 0.00 62.14 1.66
% new residents (< 1 year) arriving from outside the local area 23.61 4.80 8.78 8.04
% social rented households 20.33 19.13 11.72 23.75
% private rented households 57.27 14.08 27.93 11.43
% of Incapacity Benefit/Severe Disablement allowance claimants 2.23 19 2.87 23
% people living in medical and care establishments 0.83 6.89 0.0 0.75
% households with no car or van 43.39 24.91 37.50 29.12
% children of primary school age 4.35 2.65 3.79 8.20
Number of voluntary organisations focused on local community 5 12 12 8
% single pensioner households 2.30 21.66 18.95 13.13
% people working further than 30km from home 4.66 13.44 7.82 6.05
Road density 32.86 9.81 22.03 13.86
Number of domestic break-ins 78 0 0 30
Travel time to GP surgery (private transport) 1.90 2.2 3.10 4.19
Travel time to GP surgery (public transport) 5.40 9.3 7.70 11.34
% households with the lowest floor level: ground floor 22.07 85.99 49.05 78.32
% households with the lowest floor level: basement or semi-basement 1.29 2.77 1.96 1.26
% caravan or other mobile or temporary structures in all households 0.00 0.00 0.0 0.17
% urban land cover 86.68 19.52 30.13 13.85

Figure 28. Flood disadvantage in Scottish Borders. Inset: Hawick. Base map is Ordnance Survey data © Crown Copyright and database right 2015. Derived from OS AddressBase, SEPA data and data sources listed in Table 2.

Figure 28. Flood disadvantage in Scottish Borders

Figure 29. Flood disadvantage in Dumfries and Galloway. Inset: Newton Stewart. Base map is Ordnance Survey data © Crown Copyright and database right 2015. Derived from OS AddressBase, SEPA data and data sources listed in Table 2.

Figure 29. Flood disadvantage in Dumfries and Galloway. Inset: Newton Stewart.

Figure 30. Flood disadvantage in Dundee City. Inset: Waterfront area. Base map is Ordnance Survey data © Crown Copyright and database right 2015. Derived from OS AddressBase, SEPA data and data sources listed in Table 2.

Figure 30. Flood disadvantage in Dundee City.

4.6.3. Local authorities' feedback on the flood disadvantage assessment

The project team met with representatives from each of the three local authorities. Attendance was varied across departments. In total there were 16 participants drawn from various departments (Table 12). The project team presented the maps and data (in a similar format to the information above) before asking a series of questions relating to:

  • The local authorities' current flood risk management work and the extent to which vulnerability had been covered
  • Ease of understanding the terminology
  • Presentation of the maps and data
  • How the data might be used
  • Any other resources that they might find helpful.

The remainder of this section presents the participants' views.

Table 12. Case study local authority workshop participants

Dumfries and Galloway

Dundee City Council

The Scottish Borders

Number of participants

6

6

4

Departments represented

Planning and Infrastructure;

Flood Risk Management;

Economic Development;

Community Resilience;

Social Work;

Chief Executive's office

City Engineer;

Housing;

Planning;

Environment;

Social Work

Community Resilience; Economic Development;

Flood Risk Management

The assessment framework

The local authorities examined were supportive of the framework used and of the explicit links made between the vulnerability of communities and the hazard of flooding, as these issues tend to be considered in separation in local authorities' work. Flooding, and climate change adaptation more broadly, is typically the remit of environmental departments. Particular sectors, such as social care, remain detached from this issue because they do not have a strong futures dimension. As one of the participants observed: ' it's a fundamental issue about capacity…this [agenda] has been very much been left to the flooding team to do it themselves…that's probably the same in most local authorities' (The Scottish Borders Council).

At the same time, it was felt that there is a scope for closer links between the environmental hazards and social care: flooding, for one attendee, could be ' a new angle' in terms of the council's work on reducing inequalities (Dumfries and Galloway).

Attendees were broadly supportive of the terminology due to existing policy: social justice, for example, has been used by the Scottish Government over the past twenty years. The notion of vulnerability was well-understood even though it was acknowledged that there was not as much focus on social vulnerability with regard to the local authorities' work on flood risk management. Vulnerability was mainly considered as the number of properties at risk. However, SEPA's methodology for identifying PVAs was well-recognised as all three local authorities are involved in Local Plan Districts where the draft Flood Risk Management strategies have been put to consultation.

None of the workshop attendees had previously encountered the term 'flood disadvantage' in the terms presented: ' Vulnerability has been one that people respond to…disadvantage I'm not sure, there is a question mark over that' (The Scottish Borders Council).

It was noted that the language could therefore help to make direct links with work on inequality, particularly health inequality. However, while this is fairly well-developed in terms of Scottish policy: ' the use of the terminology is not new and we are still trying to work out how to translate it into practice' (The Scottish Borders Council). The flood disadvantage assessment could therefore be usefully presented to public health representatives to feed into their work.

In the two local authorities where material deprivation is generally low (The Scottish Borders Council and Dumfries and Galloway), the framework allowed the identification of those who may not be in a difficult financial situation but have other issues that make them more vulnerable to flooding. Dumfries and Galloway raised the issue of fairly affluent retirees who move into attractive rural areas and within a few years of retirement become reliant on social care services, due to health issues and their social isolation from family. This emphasises that considering multiple factors contributing to vulnerability alongside material deprivation allows for a more comprehensive understanding of vulnerability.

Also, the participants broadly agreed that disaggregating the social vulnerability to flooding into sensitivity, ability to prepare, respond and recover, and exposure (enhanced) was a useful way of understanding the nuances of flood disadvantage.

Indicators and indices of vulnerability and disadvantage

Participants largely accepted the vulnerability indicators used in the assessment. There were no issues over the number of indicators or their selection. Some of the indicators required more explanation, for example the participants asked about the use of the number of domestic break-ins as an indicator for ability to respond to flooding. One participant in Dundee suggested looking at single occupancy households (of any age) as those that may be more vulnerable to flooding due to having the sole responsibility for preparation and response to flood events. In Dumfries and Galloway, it was noted that the area has one of the highest levels of homelessness in Scotland, which is a real cause of concern from the point of view of social vulnerability; however, this issue is not picked up by the dataset. The paucity of information on the homeless poses a problem for mapping generally.

Further flood disadvantage assessments could usefully consider the presence of social infrastructure: for example, in Dundee, it was pointed out that the extremely high and acutely disadvantaged areas also had a low amount of community resources ( e.g. schools, community centres, churches), which may further reduce the adaptive capacity of these areas. A representative from Social Work highlighted that it was important to identify the location of nursing and residential care homes. Whilst this information is to some extent captured in the current dataset through the indicator ' proportion of people living in medical and care establishments', future assessments could explicitly include the location of such institutions.

It was observed that analysing the flood impacts on commercial properties (particularly small businesses) would be useful, although this is outside of the scope of the current project. In addition, the indirect impact of flooding on the employment provided by affected businesses, in particular for casual workers or low-income groups, was thought to be an important angle in analysing flood disadvantage.

The insurance availability indicator relating to the number of properties built after 1 st January 2009, which is a particular innovation in the current project, was received well. Whilst respecting the caveats, it was acknowledged that it was an intuitive (if crude) way of drawing attention to the areas containing properties that may be more difficult to insure: 'it certainly gives us something to work on' (The Scottish Borders Council). However, in the other two authorities it was indicated that these buildings were made to be more resilient through urban design measures such as raising properties on stilts with car parking at ground floor level or inclusion of sustainable urban drainage systems (Dundee City Council); and that the Local Development Plan required 600mm minimum freeboard [14].

A number of reasons for properties built on flood plains after 1 st January 2009 were suggested, including:

  • Planning processes for these types of properties had been initiated long before they were constructed. This may mean that any flood risk assessment would have used earlier data that did not, for example, include surface water flooding.
  • There was a sense that there were many competing priorities besides flooding and planners often have to make pragmatic decisions.
  • Elected members can go against the advice of their planning officers and approve planning applications.

Also, it was noted in one of the local authorities that the new coastal and riverside properties are relatively expensive and therefore are largely occupied by affluent, non-vulnerable people, thus not contributing to the overall flood disadvantage. Therefore, this indicator can be useful to highlight areas for attention but needs to be supplemented by local knowledge on the characteristics of the properties that have been built and the reasons why the development was given planning permission.

In Dundee and Scottish Borders the maps of hazard-exposure and social vulnerability to flooding broadly conformed to the local authorities' knowledge of the local area. Where local knowledge might differ, this was not necessarily a problem but a way of opening up discussions: 'In my experience of presenting such maps to stakeholders they immediately think how does this pertain to their experience on the ground (…) that may not match with what they know, but that's when you start to have the interesting conversations' (The Scottish Borders Council).

The disaggregation of information from the high-level social vulnerability to flooding and flood disadvantage to individual indicators underpinning the assessment was thought to be particularly helpful: the dataset ' helps to understand what it means for a person living in Hawick' (The Scottish Borders Council).

Contrary to this, the attendees in Dumfries and Galloway questioned the identification of areas as flood disadvantaged and indicated that it did not cohere with their knowledge. It was felt that the data sources led to a focus on the more urban areas of that local authority; the national data sources often fail to pick up the more fine-grained disadvantages in rural areas. For example, small pockets of extreme deprivation are often not highlighted and, similarly, the social isolation of some living in sparsely populated rural areas can be overlooked: 'statistics are very problematic for us…for data to be useful for us we need to know vulnerability at a smaller scale' (Dumfries and Galloway). Therefore, whilst the methodology was not questioned, it became clear that the ability to supplement the data developed in this project with local information would be more useful.

Presentation of the maps

The visual immediacy of the maps was recognised to be a powerful tool and maps generated a high level of interest and discussion among the workshop participants. However, there were suggestions for improvement. For example, attention was drawn to the 'red, amber, green' colour coding for identifying areas as disadvantaged. It was felt that an area marked red may raise negative associations and the colour scheme may need to be revised [15] . Also, participants considered the presentation of the bar charts to be quite difficult to read and interpret although thought they were 'helpful' once explained.

The majority of issues associated with maps were linked to the static character of maps. Participants were presented with hard copies that were limited in terms of the amount of information that could be presented. Participants expressed a wish to be able to zoom in to view the smaller data zones more clearly.

The ability to see different indices behind the maps was regarded as important in order to understand an area in detail. It was suggested that the different indicators used in the social vulnerability to flooding could be presented on maps as that medium is 'powerful'. This is difficult to achieve on a standard, 2D map, but could be assisted if the map was presented on a spatial portal where users could see a separate call out box when, for example, they hovered over a particular area.

Therefore, a strong recommendation emerging from the meetings with local authorities is for the development of a spatial portal which would allow displaying selected layers of information and would bring together the underlying spreadsheets containing the data with the maps [16] .

Potential use of the data

There was a strong recognition that the project outputs could support cross-departmental working - one local authority had already been making steps in that direction through linking local development and flood risk management together to a greater extent. Indeed, in the 24 hours following one workshop, confirmation was received that an attendee had contacted someone in another department regarding their learning from the presentation. At another workshop, the participants stated that this was the first time they had come together in order to discuss flood risk management issues.

One workshop indicated that the presence of a 'champion' within the local authority was crucial for taking the work forward and making connections between different departments. Therefore, identifying such individuals in local authorities and providing an opportunity for them to become familiar with the data produced could help to progress the consideration of flood disadvantage in local authorities.

Workshop attendees indicated that the dataset would be useful to emergency services to highlight areas for greater attention. For example, the emergency services need to know where people whose health and well-being may be affected by electricity shortages are (Dundee City Council). This dataset also helps to identify where people with limited support networks are.

However, Dumfries and Galloway indicated that, in terms of community resilience, their social work department held a 'persons at risk' database and, in the event of a flooding, this could be used to prioritise who might need assistance before, during and after the flood. Thus, the maps and dataset ' would be of interest, but I am not sure [how] high up the agenda it would be' (Dumfries and Galloway).

The use of maps as a way into community planning in order to open up discussions was positively regarded in one workshop. For example, landlords were identified as an important group who are: ' very much focussed about what goes on inside their properties, this provides a starting point to begin to get them to think about the wider picture…about how their properties relate to others' (The Scottish Borders Council).

A number of participants highlighted potential challenges for them in terms of responding to questions regarding the identification of areas as flood disadvantaged when the maps are published. There was concern that elected members may also simply demand greater resources to be spent on particular areas. Thus, it will be important to ensure that elected members fully understand the data and its limitations. There needs to be a clear disclaimer attached to the mapping and datasets that highlights the broad nature of the work. The caveats need to be presented clearly and explicitly to the public. If used for wider public communications, the method of presentation would have to be changed to take into account the discrepancies between the data zone level statistics versus, for example, local knowledge of small-scale flooding.

Further support

In terms of further support, workshop attendees highlighted that a list of examples of how others have used the data would be incredibly useful. This could be also supported by the information of the use of data generated within the ClimateJust project for England. A set of easy to understand and simple recommendations and basic advice on 'what to do next' would also benefit the end users of the dataset.

Closer connections should be made between the maps of flood disadvantage generated in this project and SEPA's flood maps and the PVAs identified in NFRA, in particular considering the focus on catchment wide flood plans beyond the local authority. This is to some extent addressed in section 4.5., where the results of mapping social vulnerability to flooding and flood disadvantage are spatially analysed against the location of PVAs and areas identified as vulnerable with regard to human health.


Contact