Climate change impacts on housing and water in the Baltic Sea Region (BSR)

There are not many modelling studies of climate change impacts on housing for northern Europe. However, one study which evaluated the future property damage from wind storms (van der Linden and Mitchell, 2009) was found. Climate change’s effects on housing is, nevertheless, linked with water related hazards. Climate change is expected to have an impact on flood and drought risks around the world (IPCC, 2007a). Many climate change water related studies have evaluated the future river discharge (Andréasson et al., 2004; BACC, 2006; EEA, 2008), but there are also studies where flood and drought risk have been evaluated (Lenher et al., 2006). The climate change impacts on freshwater resources are generally negative. Decreased runoff will likely cause a reduction in the value of services provided by water resources (IPCC, 2007a), however, areas with increased runoff does not necessarily have a positive impact on water resources. It could, for example, have a negative effect on water quality and cause increased risk flooding (ibid). 

The summary of the impacts on housing sector (including water) are presented in Table 1. Fur for further details about the each subsection and specific studies, click on the links below the table. For tips on how to interpret the information in the table, see the Swedish example on the right.

Table 1. Climate change impacts on housing and water in the BalticClimate countries – a summary of general outlooks for the found impact scenarios interpreted from different scientific studies
(↑↑ Considerable increase; ↑ Slight increase; ↓↓ Considerable decrease; ↓ Slight decrease; ○ No or insignificant change; ~ Outcome very uncertain; ~↑ Outcome uncertain, increase tendency; ~↓ Outcome uncertain, decrease tendency; ─ Not included in the analysis)

Climate change impacts on:

SWE

FIN

EST

LAT

LIT

RU

GER

Property storm loss ~↑
River discharge ↑ and ↓ ↑↑ ↑ and ↓ ↓↓ ↑↑ ↑ and ↓
Runoff In North ↑↑, in South ~↓
Flood events ↑↑ ↑↑ ~↑ ~↑ ~↑ ~↑
Drought events ~↑ ~↑ ~↑ ~↑ ~↑ ↑↑

For examples of impact scenarios reviewed from different scientific papers/reports, see the following subsections:

Property damage due to wind storms (Central Europe)
River discharge (Europe)
Runoff (Sweden)
Flood and drought risk (Europe)

 

Property damage due to wind storms (Central Europe)

The climate scenarios produced form nine general circulation models (GCM) and eight regional circulation models (RCM) were analyzed to evaluate future storm loss potentials related to changes in wind-/storm risk (van der Linden and Mitchell, 2009). The climate simulations were performed for the end of the 21st century (2071-2100) compared to the reference period 1961-1990, assuming the A1b emission scenario. A storm loss regression model was used in the calculations of storm loss potentials in some European countries; no adaptation to the changed wind speed were assumed to take place. 

The results of regional downscaling simulation showed that Germany is projected to have about a 10% increased storm loss potential (Figure 1).
The standard deviation illustrated in Figure 1 was aimed to show the uncertainty, however, as an uncertainty measure, the standard deviation is strongly influenced by outliers (ibid).
A general projection of property damage due to wind storms in Germany is illustrated in Table 2, interpreted from the results in van der Linden and Mitchell (2009).

 

 
Figure 1. Changes (%) of mean annual storm loss 2071-2100 relative to 1961-2000  based on nine GCMs (upper row) and eight RCMs (bottom row) assuming the SRES A1B emissions scenario. Values in parentheses are inter-model standard deviations. (Fig. 9.12 in van der Linden and Mitchell (2009)) (click to enlarge)

Table 2. General outlook for storm loss potential
(~↑ Outcome uncertain, increase tendency; ─ Not included in the analysis) 

  SWE FIN EST LAT LIT RU GER
Change ~↑ 

Back to the list of subsections

 

River discharge (Europe)

The report of EEA (2008) contains projections of future flood hazards in Europe based on the Dankers and Feyen (2008) study. Dankers and Feyen (2008) assessed the effects of climate change on future flood hazards in Europe using the hydrological model LISFLOOD driven by climate simulations from the regional HIRHAM climate model, based on the A2 emission scenario for 2071-2100 compared to 1961-1990.  

The result, Figure 2, indicates that Finland and Russia will have about 40 to 50% more flood events in the future. Estonia and Latvia are projected to have about 5 to 20% more flood events. Sweden, Lithuania and Germany are in general projected to have less flood events, about 5 to 40% decrease. There is nevertheless a large spatial variation; some regions of those countries are even projected to have an increased number of flood events. A general projection of future river discharge in the BSR is illustrated in Table 3, interpreted from the results in EEA (2008).

Figure 2. 100-year return level of river discharge, projected relative change between 2071-2100 and 1961-1990 (Map 5.25 in EEA (2008)) (click to enlarge)

Table 3. General outlook for river discharge
(↑↑ Considerable increase; ↑ Slight increase; ↓↓ Considerable decrease; ↓ Slight decrease) 

  SWE FIN EST LAT LIT RU GER
Change ↑ and ↓ ↑↑ ↑ and ↓ ↓↓ ↑↑ ↑ and ↓ 

Back to the list of subsections

 

Runoff (Sweden)

Andréasson et al. (2004) assessed hydrological impacts of climate change for a wide range of Swedish basins. They used different methods of transferring the climate change signal from climate models to hydrological. The four climate scenarios, shown in Figure 3, are RCAO-Had/AM3H A2, RCAO-ECHAM4/OPYC3 A2, RCAO-Had/AM3H B2 and RCAO- ECHAM4/OPYC3 B2. The HBV model was used for the hydrological modelling of the study.

Andréasson et al. (2004) concluded that climate change’s subregional impacts on the annual mean runoff vary greatly depending on the basins’ location - whether it is in the northern or the southern Sweden. The results vary much with climate models and scenarios. The northern half of Sweden is projected to have increased runoff for all climate scenarios, ranging from 2 to > 35% varying with location and scenario. The results for the southern part is more varied, although, the tendency is that the southeastern will have decreased runoff, ranging from < 35% decrease to 5% increase varying with location and scenario. A general projection of future runoff in BSR is illustrated in Table 4, interpreted from the results in Andréasson et al. (2004).

Figure 3. Difference (%) in annual mean runoff for four climate scenarios for 2071-2100 compared to 1961-1990 (Fig. 5 in Andréasson et al. (2004)) (click to enlarge)

Table 4. General outlook for runoff
(↑↑ Considerable increase; ~↓ Outcome uncertain, decrease tendency; ─ Not included in the analysis)

  SWE FIN EST LAT LIT RU GER
Change In North ↑↑, in South ~↓

Back to the list of subsections

 

Flood and drought risk (Europe)

Lehner et al. (2006) a studied global change impacts on flood and drought risk. Many studies have been performed to study the impacts of climate change on regional water resources but this study somewhat unique since it takes into account the effects of altered human water use. Hence, this is an integrated analysis of possible impacts of climate- and water use change on future flood and drought frequency. The integrated global water model, WaterGAP was used to calculate the discharge to further be able to analyze the flood and drought risk. WaterGAP consist of a Global Hydrology Model and a Global Water Use Model.

Macro-scale characteristics of the terrestrial water cycle behavior and water availability are estimated by the Global Hydrology Model. The Global Water Use Model consists of four different models, one for each of the sectors household, industry, irrigation and livestock. The model calculates the water use for each of the sectors and differentiates between water withdrawals and consumptive water use. When using this method, it is possible to combine the effects of climate change with the effects of demographic, socioeconomic and technological trends in the analysis of large-scale discharge regimes. The WaterGAP model was driven by two GCMs (the HadCM3 and the ECHAM4/OPYC3 model) and two scenarios. The scenarios are consistent with the IPCC-IS92a scenario and the National Institute of Public Health and Environment Baseline-A water use scenario. They represent business-as-usual for a set of assumptions regarding population, economic growth, carbon dioxide emissions etc.

The result of the projected 100-year flood recurrence (Figure 4) for the BSR indicates a slight difference between the two climate scenarios (the two GCMs) and an increase in frequency with time. Sweden and Finland are the countries to be most affected, both in 2020s and 2070s. The return period of a 100 year flood for many areas in Finland and Sweden in 2070s is projected to be in the range of 10 to 70 years, varying with location and climate scenario. The other countries in the BSR are not projected to have any considerable changes in frequency of floods. Though, Lithuania is projected to have 40 to 70-years return period of 100-year flood in 2020s for HadCM3. A general projection of future flood and drought events in the BSR is illustrated in Table 5 and Table 6, interpreted from the results in Lehner et al. (2006).

 
Figure 4. Projected changes in frequency of 100-year floods for 2020s and 2070s compared to 1961-1990’s climate and water use (Fig. 8 in Lehner et al. (2006)) (click to enlarge)

Table 5. General outlook for flood events
(↑↑ Considerable increase; ↑ Slight increase; ~↑ Outcome uncertain, increase tendency)

  SWE FIN EST LAT LIT RU GER
Change ↑↑ ↑↑ ~↑ ~↑ ~↑ ~↑


The result of the projected 100-year drought recurrence for the BSR demonstrates a rather drastic difference between the two climate scenarios for the 2020s (Figure 5). The ECHAM4 climate scenario projects increased frequency of 100-year drought events in Germany with a return period of about <10 to 40 years.

Southern Sweden and Lithuania are, with this climate scenario, projected to have increased frequency for a few regions with a return period of 40 to 70 years for those regions. The largest increase in frequency of 100 year droughts in the BSR with the HadCM3 is projected for Sweden, varying from <10 to 40-years return period. The other Baltic Sea countries are also projected to have increased frequency for some regions, however, not to the same extent.
For the 2070s time period, the two climate scenarios show rather similar projections. Only Germany and Finland in the BSR are projected to have increased frequency for this time period. Only some regions in Germany and Finland out of the BSR are in the 2070s projected to have increased 100 year drought return period, the return period is anticipated to be 50 years for those regions.

 
Figure 5. Projected changes in frequency of 100-year droughts for 2020s and 2070s compared to 1961-1990’s climate and water use (Fig. 9 in Lehner et al. (2006)) (click to enlarge)

Table 6. General outlook for drought events
(↑↑ Considerable increase; ○ No or insignificant change; ~↑ Outcome uncertain, increase tendency)

  SWE FIN EST LAT LIT RU GER
Change ~↑ ~↑ ~↑ ~↑ ~↑ ↑↑

Back to the list of subsections

 

Look at the impacts on other sectors:

» Agriculture
» Energy
» Overall economy
» Forestry
» Health
» Natural environment