A picture of future flood risk in Europe
Climate change is projected to increase the frequency and intensity of flooding globally, but this can vary greatly at the regional and local scale. One region where the pattern of change varies significantly is Europe. In this blog, JBA Climate Change Analyst, Anya Hawkins uses JBA’s Climate Change Flood Data to explore how projected changes in inland flood hazard and risk vary across highly populated areas in Europe.
Climate change will affect the likelihood and severity of flooding all over the world, posing a major risk to communities. Globally, surface water (pluvial) and river (fluvial) flooding are projected to intensify as the climate warms (Arnell NW, Gosling SN, 2016: Seneviratne SI, Zhang X, Adnan M, et al. 2021). Yet, this is not necessarily the case when we drill down to individual locations. Here, elements like the intersection of climate change-driven shifts in regional weather patterns and topography play an important role. Understanding how possible spatial patterns of change play out for flood risk is vital for adequate planning and risk management.
Using JBA’s Climate Change Flood Data, we can assess how flood hazard and risk is projected to change in the future for a range of Shared Socioeconomic Pathway (SSP) climate scenarios. (See our recent blog for more on these).
How is flood risk in Europe projected to change by 2050?
In Europe, flood is already the costliest natural hazard, with river flooding alone affecting more than 170,000 people in EU countries and costing €7.8 billion each year (Dottori F, Mentaschi L, Bianchi A, et al. 2020). This cost is only projected to increase as economies continue grow in high flood risk areas and climate change leads to greater and more intense precipitation (Kundzewicz ZW, Pińskwar I, Brakenridge GR, 2017: Myhre G, Alterskjær K, Stjern CW, et al. 2019).
Here, we analyse the projected change in flood hazard and the impact this has on the annual cost of flood damage for the most densely populated areas of Europe by 2050, for the climate scenarios SSP2-4.5 and SSP5-8.5. This is achieved using JBA’s Climate Change Flood Data for a portfolio of locations based on population density.
Changing flood depths
JBA’s Flood Depth Data let us quantify the change in flood hazard and assess whether locations are getting wetter or drier under different warming scenarios. By comparing present day flood depths with those projected for 2050, we can see how changes vary spatially across Europe. Here, change is assessed for a return period of 100 years (RP100), which means that there is a 1% chance of an event of this magnitude happening in any year.
The change in maximum depths for RP100 between present day and 2050 is shown for the 10 most populated countries in Europe in Figure 1 for both surface water and river flood. This focuses on locations that experience flooding in either the present, in 2050, or in both. Although the magnitude of the flood depth changes is consistently greater for the higher (SSP5-8.5) versus lower (SSP2-4.5) emissions scenario, the direction and variability of the changes are considerably different between both flood types. Whereas surface water flood depths are expected to increase for all countries, except for some locations in Spain, the change in river flood depths is much more variable.
Considering the median changes across all locations, while most of the countries see increasing river flood depths, the southern European countries of Spain, Italy, Romania, and (when considering the median) Turkey show a reversal in this trend. For instance, in Italy, median RP100 surface flood depths are projected to increase by ~10 mm for surface water but decrease by ~10 mm for river flood.
Despite median river flood depths decreasing for some countries , the upper range of change shows greater increases than that of surface water in several instances. This is the case in Turkey where the upper 5% of locations experience an increase in RP100 river flood depths of at least 210 mm despite the decrease in the median depth. For comparison, the upper 5% of locations for surface water flood experience a depth change of 130 mm.
Changes of this magnitude have the potential to translate into large changes in risk. For instance, even with no flooding in the present-day, the 210 mm change in RP100 river flood would be sufficient to overtop the average doorstep height of 200 mm and cause much greater damages.
Figure 1: Boxplots of the change in maximum (a) surface water and (b) river flood depth for RP100 between our baseline and 2050 views of hazard for both SSP2-4.5 (purple) and SSP5-8.5 (blue). Plots are shown for the 10 most populated European countries with locations that experience no flooding excluded. The boxplots show the average and spread of the change in flood depths for location at risk of flooding across the locations disaggregated by population. The coloured box shows the lower to upper quartile range (from the 25th and 75th percentiles) with the horizontal line within these showing the median (50th percentile). The whiskers extending above and below the box end at the 95th and 5th percentiles respectively. The white dot shows the value of the mean.
Changing flood risk
JBA’s Pricing Data are used to assess changing financial risk posed by flood by examining the change in the annualised cost of damage as a proportion of the sum insured, otherwise known as the Annual Damage Ratio (ADR). Unlike the RP100 focus of the above analysis, this calculation considers the depths across all return periods. These are used to produce 10,000 return period damage ratios which are aggregated into the final ADR value for each location.
Using the same European portfolio, we see that, as with flood depths, the change in ADR varies more for river flood than for surface water, with the average change across these countries being an order of magnitude higher for river.
In contrast to damage resulting from river flood, the ADR change due to surface water flooding is a broadly consistent increase and with all countries having at least 50% of locations showing an increase in the ADR. The trend in river flood is consistent with the change in depths presented in Figure 1 with the direction of median change being the same for all countries.
Figure 2: Boxplots of the change in the (a) surface water and (b) river Annual Damage Ratio (ADR) for RP100 between our baseline and 2050 views of hazard for both SSP2-4.5 (purple) and SSP5-8.5 (blue). Plots are shown for the 10 most populated European countries.
The ADR calculation combines the change in flood hazard with a representation of damage (or vulnerability), which is a function of absolute flood depth (Figure 3). This is key to quantifying a change in overall risk: while the change in depths may be similar between locations, different absolute depths can result in very different ADR values.
Due to this extra part of the risk calculation, we see some differences in the pattern of changes in ADR versus changes in depth between countries. For example, while the UK shows a similar increase in surface water flood depths to France and the Netherlands (Figure 1), the change in ADR is much lower. This is because a greater proportion of locations in the UK have flood depths remaining below 20 cm (the average doorstep height) under both baseline and climate change scenarios than France and the Netherlands. Under the simplified property type assumptions here (residential without basement), no damage is expected for these UK locations and the ADR remains at 0.
Figure 3: Vulnerability curve showing the relationship between flood depth and the damage caused to a property, represented as a proportion of total sum insured (the damage ratio).
What’s driving these changes?
The changes in flood hazard and risk are a result of changing precipitation and temperature projected by the global climate model used in the calculation of the Climate Change Flood Data. This is a model from CMIP6, the sixth and most recent phase of the Coupled Model Intercomparison Project.
For surface water flooding, changes in extreme precipitation from the model inform our climate change assessment. For projections of river flooding, we consider both temperature and precipitation data to feed into a rainfall runoff model to assess change in annual maximum river flow. The spatial pattern in the response of these metrics to global warming is shown in Figure 4.
Figure 4: Maps of the response of precipitation and temperature flow to global warming for Europe. These show the change in annual maximum (a) precipitation and (b) river flow for each degree of global warming for the global climate model MRI-ESM2-0. Darker green colours indicate a stronger increase with global warming whereas darker brown colours indicate a stronger decrease.
Precipitation increases with warming for most of Europe (Figure 4a). This is largely a result of a warmer atmosphere being able to hold more water vapour, increasingly the intensity of precipitation on average. This signal is reversed, however, over southern Spain and the Canary Islands. Here, climate change-driven shifts in storm tracks result in overall drying, a pattern projected by a number of models, and something widely accepted as a very likely response to climate change in the region (Ranasinghe R, Ruane AC, Vautard R, et al. 2021). This explains the difference between Spain and the other countries for the projections of flood depths and ADRs.
In terms of river flow, the pattern of change is more variable than that of precipitation, with a trend of decreasing flow when moving towards southern Europe (Figure 4b). When moving further south, increasing temperatures drive a decrease in river flow as the effects of increased evapotranspiration begin to outweigh the smaller increases in precipitation, therefore reducing flow rates (Ranasinghe R, Ruane AC, Vautard R, et al. 2021). This results in the much more variable picture of changing river flood hazard and risk between the countries presented above.
It's not just in the south of Europe where this trend of decreasing river flows is observed. It's also present in northern Scandinavia and Finland. This is a result of complexities arising as warming causes changing snowpack and snowmelt characteristics. As temperatures rise, winter snowpacks are projected to get smaller, resulting in lower river flows during the melt season and decreasing flood risk (Thober S, Kumar R, Wanders N, et al. 2018; Di Sante F, Coppola E, Giorgi F, 2021). More details on this effect are given in another JBA blog on Future Climate, Flooding and Snowmelt.
Summary
Projections of future surface and river flood risk in Europe are highly variable across the continent. Although location-level projections are uncertain, this variability means that spatial data are needed to tell a consistent story about changing flood risk for a particular climate scenario.
This need is addressed by JBA’s Climate Change Flood Data, which can be used to quantify projections of the changing risk posed by flooding at any location worldwide, under multiple future scenarios and time horizons.
For more information about our global Climate Change Flood Data, or our other climate change offerings, get in touch today.
REFERENCES
Arnell NW, Gosling SN (2016) The impacts of climate change on river flood risk at the global scale. Climatic Change 134: 387–401
Seneviratne SI, Zhang X, Adnan M, et al. (2021) Weather and Climate Extreme Events in a Changing Climate. Climate Change 2021: The Physical Science Basis Contribution of Working Group I to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change.
Dottori F, Mentaschi L, Bianchi A, et al. (2020) Adapting to rising river flood risk in the EU under climate change.
Kundzewicz ZW, Pińskwar I, Brakenridge GR (2017) Changes in river flood hazard in Europe: a review. Hydrology Research 49: 294–302.
Myhre G, Alterskjær K, Stjern CW, et al. (2019) Frequency of extreme precipitation increases extensively with event rareness under global warming. Sci Rep 9: 16063.
Ranasinghe R, Ruane AC, Vautard R, et al. (2021) Climate Change Information for Regional Impact and for Risk Assessment. Climate Change 2021: The Physical Science Basis Contribution of Working Group I to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change.
Thober S, Kumar R, Wanders N, et al. (2018) Multi-model ensemble projections of European river floods and high flows at 1.5, 2, and 3 degrees global warming. Environ Res Lett 13: 014003.
Di Sante F, Coppola E, Giorgi F (2021) Projections of river floods in Europe using EURO-CORDEX, CMIP5 and CMIP6 simulations. International Journal of Climatology 41: 3203–3221.