The Scottish Government

« Previous | Contents | Next »

Listen

4 PREDICTED TRENDS IN CLIMATE CHANGE

4.1 INTRODUCTION

This section of the report details the predicted climate change trends used as the basis of further consideration in relation to the design and operation of the road network. Presenting the predicted changes in a simple and succinct form is not always possible, as some of the changes are complex in nature, but where possible at the end of each section a summary of the key findings is included. A development of the tables included in section 3 of this study, to identify the available climate change data and specify what further information might be required to supplement this, is included at the end of this section. This includes a ranking of whether the issue concerned is considered to be of High, Medium or Low significance in terms of its level of impact on the road network.

To illustrate the predicted changes the results of the Medium-High emissions scenarios are generally used in this section of the report. Appendix B presents the comparative 50km and 25km RCM grids for the predictions under all of the emissions scenarios.

4.2 TEMPERATURE

4.2.1 General

Over the next few decades daily mean temperatures are predicted to rise by up to 1¡C over all of Scotland for all emissions scenarios. Daily minimum and maximum temperatures are likely to increase by the same order of magnitude, and summer maximums are shown in Figures 4.1 and 4.2 for all emission scenarios, with winter minimums shown in Appendix B.

Figure 4.1 - Change in daily maximum temperatures (¡C) by the 2020's during the summer (June to August). Low scenario to left, Medium-Low to right. Source: Met Office.

Figure 4.2 - Change in daily maximum temperatures (¡C) by the 2020's during the summer (June to August). Medium-High scenario to left and High to right. Source: Met Office.

Analysis of predicted changes to daily climate in the UKCIP02 report included an assessment of changes in extreme daily temperatures. These were presented as probabilities that a daily temperature might exceed maximum temperatures or fall below minimum temperatures, for given thresholds. This analysis was completed for summer and winter seasons on four grid-squares taken from across the UK. A similar analysis has been completed for this study, focusing on two Scottish grid-squares. The first 50km square selected is located in the region around Glasgow, representing a lowland urban area. The second covers an area around Aviemore, representing a highland rural area.

4.2.2 Freezing Temperatures

Figure 4.3 shows the probability distribution of a minimum temperature being achieved and the probability of daily minimum temperatures falling below a given threshold during the winter season (December to February) by the 2080's, for the Medium-High scenario. Although the analysis is based upon a future climate towards the end of the century, changes can be taken as indicative of the way in which daily extremes of temperature are likely to change over the next few decades, in nature if not the scale of the change.

The upper panels in Figure 4.3 show that as the climate warms, daily minimum temperatures in winter will also increase, as shown by comparing the red line with the black. However, the way temperatures are distributed throughout a season will remain similar to the present day, that is, although the mean value is increased the variation and standard deviation do not change significantly. It should be noted that the most probable wintertime daily minimum temperature for Aviemore in the present day climate simulation is just below 0 ^oC but that this will shift to be above 0 ^oC by the 2080's under the Medium-High scenario.

Figure 4.3 - Winter daily minimum temperatures (¡C) for the region a(left) and Glasgow (right), modelled present day climate (1961-1990, black) and the 2080's (Medium-High, red). Source: Met Office.

The lower panels of Figure 4.3, which show the probability that minimum temperatures fall below a threshold, can be used to determine the number of days in a season which are likely to be below freezing. In the modelled present day climate there is a 0.6 probability, or 60% chance, that a minimum daily temperature will fall below 0 ^oC in the Aviemore region during the winter season. This is equivalent to an average of 54 days out of a 90 day winter season. By the 2080's, under the Medium-High scenario, this likelihood decreases to 40%, which is equivalent to approximately 36 days in a season. Thus the number of days in a winter season when temperatures fall below freezing is reduced by a third. Similarly, there is a modelled current probability of temperatures in the Glasgow region falling below freezing for a total of 40 days in a winter season. By the 2080's, under the Medium-High scenario, this is projected to decrease to approximately 20 days per season, a reduction of 50%.

These calculations are based upon daily minimum air temperature. However, the ability of the model to replicate present day extremes of temperature for these locations has not been validated using observed temperatures. It should be noted that a ground frost can occur when air temperatures are above 0 ^oC, as ground temperatures can be lower than air temperature. The modelled reduction in the number of days when the temperatures fall below freezing can, however, be taken as indicative of the type of changes that are likely to occur.

The predicted trends in freezing temperatures are summarised below in Table 4.1.

Table 4.1 - Summary of Occurrence of Freezing Conditions

4.2.3 High Temperatures

Figure 4.4 shows a complementary analysis based on summertime daily maximum temperatures for the same two regions discussed above. The modelled present day climate is shown in black, with the 2080's Medium-High climate shown in red. The upper row shows the probable distribution of temperatures and the lower row represents the likelihood of daily maximum temperatures being exceeded. As with wintertime minimum temperatures it is apparent that summertime daily maximum temperatures increase. However, the distribution of temperatures changes in character, particularly in the warmer extreme of the range. Temperatures in excess of 30¡C are not seen in the Aviemore region in the modelled present day climate. By the 2080's, under the Medium-High scenario temperatures in excess of 30¡C, or even 40¡C, may occasionally occur in an average summer season. Temperatures may increase to similar levels in the Glasgow region.

The lower panel suggests that there is less than a 10% chance that a day during summer will be warmer than 20¡C in Aviemore for the modelled present day climate. By the 2080's, under the Medium-High scenario, this increases to a 40% chance. Similarly, the chance of this temperature being reached in the Glasgow region under the modelled present day climate is 15%. By the 2080's, under the Medium-High scenario, the chance of this occurring in the Glasgow region increases to 55%.

Figure 4.4 - Summer daily maximum temperatures (¡C) for the region aroundAviemore (left) and Glasgow (right), modelled present day climate (1961-1990, black) and the 2080's (Medium-High, red). Source: Met Office.

The predicted trends in high temperatures are summarised below in Table 4.2.

Table 4.2 - Summary of Occurrence of Extreme High Temperatures

4.2.4 Growing Season

Temperature also affects the growing season, the length of which is defined (Hulme et al, 2002) as the longest period within a year that satisfies the twin requirements of:

• beginning at the start of a period when daily-average temperature is greater than 5.5 ^oC for five consecutive days
• ending on the day prior to the first subsequent period when daily-average temperature is less than 5.5 ^oC for five consecutive days.

In general, the thermal growing season will increase across the whole of the UK, with slightly smaller increases in Scotland than for the rest of Britain. Southern Scotland will experience slightly larger increases than northern Scotland. Quantitatively the UKCIP02 scenarios predict that by the 2080's the thermal growing season in the Highlands will have increased from the modelled present day climate of 150 days to about 200 days +/- 20. In southern Scotland the increase is likely to be slightly higher. Assuming linear scaling, then by the 2020's the growing season is predicted to increase by about 22 +/- 10 days.

The predicted trends in growing season temperatures are summarised below in Table 4.3.

Table 4.3 - Summary of Growing Season Temperatures

4.3 RAIN

4.3.1 Precipitation

The UKCIP02 scenarios of annual change for Scotland show little predicted change in annual mean precipitation over the next few decades, with any change being within the range that can be attributed to natural variability. However, by the 2020's a distinct seasonal pattern of change over Scotland can be discerned, even in the Low emission scenario. Although there is likely to be little significant change in either autumn or spring precipitation amounts, changes are likely for the other seasons. Wintertime precipitation is predicted to increase in eastern regions in this period by 10% to 15%, varying between Low and High emission scenarios. However, the rest of Scotland is not predicted to experience changes greater than can be accounted for by natural variability. In the summer, decreases in precipitation are predicted to be widespread, with only the far northwest of the country seeing little change. By the 2020's, under the High emissions scenario, these decreases are likely to be greatest in the south east of the country and may be as great as 20% below modelled present day levels.

Year to year, or inter-annual, variability in precipitation is also predicted to change. Although UKCIP02 only presents variability results for the 2080's, this may be taken as indicative of the type of change which may be experienced by the 2020's, although to a lesser degree. It is likely that variability in precipitation will increase in eastern Scotland during winter. However, it is likely to decrease over much of the country, particularly the south, during summer. It should also be noted that over the next few decades natural variability in rainfall is likely to continue to dominate any climate change related trend in the long term average. This means that while winters will become increasingly wetter than at present there may also be periods of below average winter rainfall. Any recent fluctuations in rainfall cannot therefore be taken as an indication of possible trends for the next few years.

In assessing these predicted changes, the benefit of increased horizontal resolution can be illustrated as shown in Figure 4.5. This figure presents the percentage change in wintertime precipitation totals by the 2020's, for the Medium-High emissions scenario, showing results from both the 50km RCM used in UKCIP02 and the 25km RCM used in the British-Irish Council study. While the pattern of change is broadly similar, the higher resolution means the highly coastal nature of the increases can be seen more clearly, with little change predicted over central Scotland. It should be noted that this is a single integration of the model and is therefore just one possible evolution of our climate. The consistency of large scale patterns does, however, give some level of confidence in the prediction. Similar figures for other emissions scenarios and for the summer season are presented in Appendix B.

Figure 4.5 - Winter (December to February) precipitation change (%) by the 2020's with the Medium-High emissions scenario. The 50km HADRM3 model result is shown to the left and the 25km BIC result is to the right. Source: Met Office.

A number of maps of likely change in return period amounts ( RPAs) were also calculated for UKCIP02, based upon HadRM3 data. Given the relatively short period of data that is available it was not possible to calculate very long return periods. However, mapped changes were produced for two, five, ten and twenty year return periods. Figure 4.6 shows percentage change in the two year return period daily precipitation by the 2020's as an example, which demonstrates the main characteristics of patterns of change. Each of the scenarios is shown for each season, as well as the annual mean.

It can be seen that in the summer there is a general pattern of decrease in the RPAs whilst in winter there are increases in the RPAs across southern and eastern Scotland, with a less coherent pattern in other regions. However, it should be noted that these return periods are calculated from a relatively short dataset, based around the 2080's, which has been values for the 2020's.

While the large scale patterns are consistent with predicted changes in seasonal rainfall, there is less confidence in the finer detail shown in these maps. In particular, climate models cannot resolve localised convective storm activity, that is, individual thunderstorms. The intense storms which can be produced by convective activity occur at a scale of a few kilometres, well below the model's resolution. Convective activity and related rainfall is parameterised within the model. However, the rainfall produced is output as a grid-square average, spreading the potentially intense period of rainfall over the 50km or 25km square. Parameterisations are being continually improved though it is not likely that regional climate models will be developed to explicitly resolve convective storms in the near future.

Figure 4.6 - 2-year return period daily precipitation change (%) by the 2020's for each UKCIP02 emissions scenario and season. Source: UKCIP02.

Some validation of the extreme events simulated by the regional climate model was completed for UKCIP02. It was shown that the HadRM3 model has a tendency to overestimate the heaviest rainfall events when compared to current climate, particularly in winter. Despite this bias the model does successfully reproduce aspects of daily precipitation climatology. Analysis of daily precipitation was completed for the 2080's, but as described above it is reasonable to take the predicted changes as indicative of what may occur in coming decades, in type if not the size of change. As shown in Figure 4.7, by the 2020's it is likely that south western Scotland will see an increase in the number of days of intense rainfall in an average season, particularly in winter, although there is little indication of change for other regions or seasons. In this case 'intense' is defined to be the uppermost 10% of total seasonal rainfall.

Figure 4.7 - Changes in 'intense' rainfall days per season by the 2020's. All seasons for all UKCIP02 scenarios. Source: UKCIP02.

UK Water Industries Research ( UKWIR, 2004) funded a project to consider the impact of climate change on the hydraulic design of sewerage systems. The project was wide ranging, but had a principal focus on the performance of sewerage systems under 2080's rainfall conditions. It also considered what changes might be needed in the hydraulic design of sewerage systems to address any problems that climate change might pose. One component of this study was an investigation into climate change effects on short-period rainfall.

Work was carried out by the Met Office to produce predictions for future extreme rainfall based on the UKCIP98 Medium-High scenario. A comparison of rainfall from the Flood Studies Report ( NERC, 1975) and Flood Estimation Handbook (Institute of Hydrology, 1999) methods was also carried out. The final aspect of the study was a seasonality study on the differences between the rainfall output from UKCIP98, based on the HadRM2 model and UKCIP02, based on the HadRM3 model. The findings indicated that while the changes in the future predicted by HadRM3 may be less extreme in terms of sub-daily rainfall than the changes predicted by HadRM2, both models should be considered as realistic scenarios for the future climate in the UK.

The return period amounts ( RPA) for 6-hour duration events were calculated for the modelled present day climate and future (2080's) climate of the RCM. Figure 4.8 presents examples taken from the above report, and show the 'uplift ratio' between the return period amounts for the modelled present day climate and the future climate derived from HadRM2 and HadRM3 data. It must be noted that little validity can be assigned to the results for Orkney, Shetland and the Hebrides, which are not represented as land in HadRM3, or HadRM2, so their mapped uplift ratios are extrapolated from the mainland data.

Figure 4.8 - Present day to 2080's uplift ratio calculated from HadRM2 and HadRM3 data for extreme six hourly rainfall with a one year return period. Source: UKWIR, 2004.

The HadRM3 model predicts that most areas of Scotland will experience heavier extreme 6-hour events in the future than in the current climate. When averaging the present day to future ratios over the whole UK, future extreme 6-hour events are predicted to be heavier. Figure 4.8 provides just two examples taken from an extensive study based upon both HadRM3 and HadRM2. Uplift ratios derived from HadRM2 are significantly larger than those predicted by HadRM3. Both model projections are considered an equally likely evolution of our climate and hence uplift ratios from both models should always be used in any study rather than ratios from one model in isolation. Any application of these uplift ratios must be within the context of uncertainties described previously and the UKWIR publication.

The predicted trends in storm event rainfall for short return periods are summarised below in Table 4.4.

Table 4.4 - Summary of Extreme Storm Event Rainfall (Short Return Periods)

4.3.2 Watercourse Flooding

River and burn flooding is not covered by the Met Office and UKCIP climate change scenarios. However, several research studies (Werritty el al, 2002, Price and McKeena, 2003, Kay et al, 2003, Reynard et al, 2004) have been undertaken to transfer the predicted changes in storm event rainfall and antecedent catchment wetness into predictions of how peak fluvial flows will change.

In general, peak fluvial flows in Scotland are predicted to progressively increase during the 21 ^st century. Some regional differences are suggested by Price and McKeena (2003), with eastern Scotland experiencing larger increases than north western Scotland. There is some limited evidence that suggests the physical characteristics of the catchments may influence the scale of the response to climate change. Snowmelt contributions are likely to be reduced in the future and flood characteristics in the central and eastern Highlands may, as a result, be influenced.

In terms of quantitative estimates of change, by the 2080's extreme design floods, typically with an annual probability of between 0.5 and 0.01, that is, 2-year to 100-year events, are predicted to increase in terms of peak flow by 10% to 30% compared to the modelled present day climate. This suggest that the current 100-year event would be approximately twice as likely by the end of the century. Linear scaling to the 2020's suggests that design peak flows will increase by 4% to 13% compared to the modelled present day climate.

The uncertainties with these predictions are relatively high since they are largely based upon future storm event characteristics. All of the studies have concentrated on medium to large catchments (>100km ²) and investigations into the response of minor watercourses, such as those often culverted, have not as yet received much attention.

Reliable long-term sustained trends associated with climate change are not evident in the Scottish hydrological flood records. However, changes in many of the flood related factors in Scotland have been observed during the last few decades of the twentieth century, as outlined in Section 2 of this study.

It is interesting to note that these changes are all, to some extent, suggested by or inferred from the Met Office climate model HadRM3.

The predicted trends in storm event rainfall for long return periods are summarised below in Table 4.5.

Table 4.5 - Summary of Extreme Storm Event Rainfall (Long Return Periods)

4.3.3 Ground Water

Changes in ground water level are not provided by the available climate change models. No such studies for Scotland are known to have been undertaken. These comments can therefore only refer to the changes in the driving climatic variables, principally precipitation and to a lesser extent potential evaporation.

In general, the UKCIP02 scenarios suggest little change in annual mean precipitation over the next few decades, with any change being within the range of natural variability. However, wintertime precipitation is predicted to increase, particularly on the eastern side of the country. There are likely to be widespread summer decreases in precipitation, with only the far north west of the country seeing little change. The spring and autumn seasons are predicted to be little effected. Year-to-year variability is also predicted to change and this is likely to increase in eastern Scotland during the winter, and decrease over much of the country, particularly the south, during the summer. Therefore, eastern Scotland, and to a slightly lesser extent southern Scotland, are likely to experience episodes of greater saturation than they do at present. This may lead to higher recharge rates in the future, bringing about higher levels of groundwater in those areas with permeable geology and drift.

The presence of snow is predicted to reduce. Locally in the uplands this may alter the timed release of the water into the catchment and may in such locations promote more continual winter recharge than presently experienced.

Potential evaporation is likely to increase across Scotland, which, coupled with the reduced summer rainfall, will tend to result on average in greater and more persistent soil moisture deficits. This may reduce the summer and autumn recharge to the ground water.

It is not possible with the information presently available to provide quantitative guidance on either ground water levels or changes to the rates of recharge to the ground water body.

The predicted trends in ground water are summarised below in Table 4.6.

Table 4.6 - Summary of Ground Water

4.3.4 Soil Moisture

The UKCIP02 report presented likely changes in soil moisture by the end of the century, where soil moisture was defined as the amount of moisture available in the root zone, that is, moisture available for evapo-transpiration. Although likely changes over the next few decades were not discussed, it is reasonable to assume that the trend from present day to future conditions can be taken as indicative of likely change by the 2020's. Figure 4.9 shows predicted changes in soil moisture content by the 2020's for the winter and summer months under the Medium-High scenario from HadRM3 and the BIC model.

Figure 4.9 - Average soil moisture content change (%) by the 2020's, Medium-High emissions scenario. Winter is top and summer is bottom. 50km HadRM3 ensemble mean is left and BIC 25km model is right. Source: Met Office.

Average soil moisture is likely to increase during winter and decrease during summer months. Although winter precipitation is likely to increase, higher temperatures and reductions in relative humidity mean that evaporation will increase. This may be the reason why soil moisture levels across Scotland are likely to increase less than may have otherwise been expected, based upon precipitation increases. Changes in autumn soil moisture levels are predicted to be very similar to those in summer. This is due to the long time taken to recover soil moisture levels following hotter and drier summers. The reduction in soil moisture will be most pronounced in southern and eastern Scotland.

By the 2080's it is predicted that the average soil moisture content will be between 3% and 5% higher during the winter, and between 10% and 30% lower during the summer and autumn. Using linear scaling, by the 2020's it is predicted that the average soil moisture content will be between 0% and 2% higher during the winter, and between 3% and 8% lower during the summer and autumn.

Recent slope failures affecting Scottish roads might suggest that during the last 5 years soil and sub-soil water contents have risen to above average conditions. However, many of these failures have occurred during the summer, which does not agree with the climate change model predicted trend towards drier summers that are less prone to intense rainfall. The recent slope failures may simply be related to the natural variability in climatic conditions, and do not supply adequate evidence of a sustained long-term trend in ground conditions. Given the limited nature of changes predicted for soil moisture, and that some seasonally related increases and decreases are expected, it is anticipated that subsidence due to drying out of soils is unlikely to represent a significant problem.

The predicted trends in soil moisture are summarised below in Table 4.7.

Table 4.7 - Summary of Soil Moisture

4.4 SNOW

It is predicted that as the climate warms both snowfall totals and the length of the snow season in Scotland will decrease. The UKCIP02 report showed predicted decreases in snowfall of more than 30% over the Scottish Highlands by the 2080's, even under the Low emission scenario. This increases to a reduction of more than 90% across eastern Scotland in the High emissions scenario by the end of the century. Figure 4.10 shows the predicted average change in snowfall for winter (December to February) by the 2020's, under the Medium-High emissions scenario, with results for other emission scenarios shown in Appendix B. It is clear that snowfall totals throughout the season will decrease significantly in coming decades. It is also likely that as winter temperatures increase, the length of the snow season will decrease, with snow arriving later in the year and melting earlier than at present. These two in combination imply a reduction in the number of days of snowfall in an average season.

Figure 4.10 - Change in average winter snowfall (%) for the Medium-High emissions scenario for the 2020's. The 50km HADRM3 model result is left and the 25km BIC result is right. Source: Met Office.

By the 2080's it is predicted that the average winter snowfall across most of Scotland is likely to reduce by 50% to 90%. The higher reductions are for the eastern fringes and the south of Scotland and the lower reductions are for the central upland areas. Using linear scaling, by the 2020's it is predicted that average winter snowfall across most of Scotland is likely to reduce by 20% to 40%, with the same geographical distribution. As noted previously, natural variability is likely to dominate any climate change related trends in snowfall over coming decades. It is possible that years may occur with the first snowfalls occurring earlier in the year than usual or that snow fall may be heavier than average. Although years with reduced amounts of snow will become more and more likely, it is also possible that heavy snowfalls could occur in some years.

The predicted trends in snow are summarised below in Table 4.8.

Table 4.8 - Summary of Snow

4.5 WIND

Changes in wind strength and the frequency of storms have proved to be a topic of much uncertainty in climate model simulations. The slightest shift in patterns of atmospheric pressure can lead to large changes in local weather patterns and the prevailing wind directions for a particular location. Any changes predicted are very model dependent and therefore little confidence can be placed in them. For guidance, however, it should be noted that the UKCIP02 scenarios of change for the 2080's indicate that the two year daily mean wind speeds across Scotland are likely to change by no more than +/- 5%, with the winter predictions associated with the biggest increases. Of possibly more relevance is the predicted extension to the growing season, with trees retaining leaf cover until later in the year. Autumn storms may have more of an impact in future as trees in leaf may be more vulnerable to damage and potential wind throw.

There is some evidence that as climate changes the North Atlantic Oscillation ( NAO), a measure of the westerliness of winter weather, will move into a predominantly more positive phase. This means that during winter months, on average, the UK will be windier, wetter and milder. The UKCIP02 report noted that under the Medium-High emission scenario, the future trend is for an increase in NAO, although year to year variability is large. The increase in the decadal NAO index becomes significant, or greater than natural variability, by the 2050's. This suggests that winter weather in the UK will become more westerly and potentially stormier in nature.

The predicted trends in extreme winds are summarised below in Table 4.9.

Table 4.9 - Summary of Extreme Wind

4.6 FOG

Predicted changes in fog are not routinely output by current climate models. For UKCIP02 an estimate of the change in the number of days of fog was calculated from changes in relative humidity simulated by the regional climate model. The calculation was based upon a relationship derived from weather forecasting. The method indicated that the number of fog days in winter may decrease by 20% across all areas of the UK by the 2080's, based upon the Medium-High emissions scenario. Assuming linear scaling, by the 2020's this would be a decrease of 9%.

More recent work conducted by the Met Office has confirmed this finding with the number of fog days expected to decrease over Scotland in both autumn and winter as climate changes. This is an area of active research and the feasibility of including fog as a standard output of the climate model is under investigation.

The predicted trends in fog are summarised below in Table 4.10.

Table 4.10 - Summary of Fog

4.7 COASTAL FLOODING

Sea level rise is an important consequence of climate change and global warming, mainly arising through thermal expansion of ocean water and through the melting of mountain glaciers. It leads to both the relatively slow inundation of low lying coastal areas and to an increase in the frequency of short-lived extreme high water events. Regional variations in sea level change occur because the climate induced warming of sea water is not uniform and neither, therefore, is the thermal expansion of the ocean. Changes in both atmospheric pressure and oceanic circulation also contribute to regional sea level changes. Of the large number of models used in the IPCC assessments there is little agreement on the regional differences in sea level rise, which can be as much as +/-50% of the global average.

The extremes of sea level rise are the 'storm surges', high water levels generated by storm or cyclone activity, especially over relatively shallow water, or in estuaries where a 'funnelling' effect can act to increase surge heights even further. When storm surges coincide with high tide the most extreme high water levels can be achieved. Predicting the changes to storm surges which may be expected as climate changes is an area of active research.

Predicted global mean sea-level changes until the year 2100, relative to 1860, as simulated by the Met Office climate model under the four different emissions scenarios, are shown in Figure 4.11. All scenarios show a rise by the end of the century, ranging from 0.30m for the Low emissions scenario, to 0.49m for the High emissions scenario. It should be noted that it is not until after about 2050, at which point the rise from present day is about 0.11m, that the lines for different scenarios diverge. The Hadley centre model predicts global mean sea level rise near the middle of the range of the IPCC 2001 assessment.

Figure 4.11 - Global mean sea-level changes, relative to 1860, in metres, simulated by Met Office climate model under four different emissions scenarios. Source: Met Office.

Global mean sea-level rise will be dominated by thermal expansion of the ocean as it warms. Additional contributions will come from the melting of land ice contained in small glaciers and in the Greenland ice sheet. Arctic sea ice and floating ice shelves around Antarctica will not contribute to sea level rise because they already displace sea water when they are floating. The land ice sheets in Antarctica are not expected to melt significantly before the end of the 21 ^st century.

By the 2020's average sea level around the UK is expected to follow the global average predictions of between 0.04m and 0.14m rise, taking the smallest and largest rise predicted by the IPCC climate models using the range of emission scenarios. Uncertainty in regional differences is estimated to be +/-50% of these values, not including any changes due to land movement. It must also be noted that the continuing isostatic rebound, as land recovers from the weight of ice during the last ice age, will go some way towards countering sea level rise around much of Scotland. For example, by the 2050's, Edinburgh and surrounding areas are predicted to rise by 0.05m above present day levels. The rate of rise in some areas of western Scotland is estimated at twice the rate by which Edinburgh is rising.

An assessment has been made to determine what changes would be necessary by the 2050's to provide the same level of protection as existed in 1990, to allow for the predicted increases in coastal flooding. This assessment was based on the UKCIP02 sea-level change information and the simple analysis adopted in this report, together with allowing for the continuing isostatic rebound. From this work, Price and McKenna (2003) concluded that coastal flooding defences would need to be raised by between 5cm and 18cm. Using output from other IPCC accepted models suggests heightening would need to be between 0cm and 40cm. These should probably be seen as an indication of the minimum likely change, since changes to wave regimes and surges are as yet poorly understood and should also be taken into account. Some initial tentative indications suggest that changes related to surge height could be of the same order of magnitude, if not higher, than the changes in mean-sea level.

The possible change in coastal flooding depends on the interaction between the modest sea level rise predicted over the next few decades, the continuing isostatic rebound and rise of Scotland, and the relatively low risk of storm surge which exists. Therefore, by the 2020's, it is reasonable to assume that risk of roads being inundated through coastal flooding will not increase significantly. However, it should be noted that changes in wind patterns are uncertain, as are possible changes in wave climatology. Thus the risk of on-shore winds and increased wave heights, which in combination would cause increased coastal flooding cannot, at this time, be described.

The predicted trends in coastal flooding are summarised below in Table 4.11.

Table 4.11 - Summary of Coastal Flooding

4.8 ANALYSIS OF AVAILABLE AND FURTHER INFORMATION ON CLIMATE CHANGE REQUIRED BY THIS STUDY

Tables 4.12 to 4.17 present a summary of how the possible road impacts identified in Section 3 under each of the weather event categories have been assessed in terms of their significance. It also identifies the information on predicted trends in climate change required to evaluate appropriate responses for the future design and operation of the road network, which is discussed in Section 5. The tables are numbered as follows:

• Table 4.12 - Temperature
• Table 4.13 - Rain
• Table 4.14 - Snow
• Table 4.15 - Wind
• Table 4.16 - Fog
• Table 4.17 - Coastal Flooding

« Previous | Contents | Next »