Gardiner V., Matthews H. The changing geography of the United Kingdom
Подождите немного. Документ загружается.
MIKE HULME
416
unusually hot and dry seasons or years. Abstraction rates more than trebled during the
drought of 1989–92 and again during 1994–5 as farmers compensated for the lack of
rainfall and high evaporation rates by spray irrigating their crops, particularly in the Anglian
Water region. A similar pattern is evident for subsidence insurance claims. During the
1989–92 drought subsidence claims increased by a factor of six to about £600 million per
year, and the dry weather of 1995 led to at least a trebling of claims. In contrast, the
1975–6 drought saw only a doubling of insurance claims for subsidence. This latter
observation illustrates an important point about the sensitivity of different economic sectors
to climate variability—the relative sensitivities of such sectors to climate variability
themselves change over time. This may either occur as a function of changing economic
or cultural circumstance (there were more home owners in 1991 than in 1976), but also as
a result of specific adaptations to climate extremes (more homes were insured against
subsidence in 1991 than in 1976). This consideration is most important when it comes to
assessing the likely impacts of future climate change on either economic or environmental
systems—past sensitivity is not necessarily a good guide to future sensitivity and a large
array of adaptation strategies may lessen the impact of climate anomalies in the future
(Subak et al. 1999).
We have demonstrated the inherent variability of UK climate on different time-
scales and we have shown how different environmental assets and economic activities
are sensitive to such climate variations. But just how rapidly is UK climate warming
FIGURE 19.5 Total catch of the moth Hebrew character, orthosia gothica (L.) (Lep. Noctuidae) in
light traps at four UK locations in 1989 compared to the mean for all previous years of trapping
Source: Cannell and Pitcairn (1993).
CLIMATE CHANGE
417
and what are the associated changes in non-temperature variables such as
precipitation, gale frequencies and sea-level? The next section presents some
indicators of climate variability over past centuries, while the fifth section (pp.
423–31) presents scenarios of UK climate change for the next century and a summary
of their possible impacts.
Some key indicators of UK climate change and variability
The UK possesses some of the longest instrumental climate time-series in the world,
the longest being the 340-year series of Central England Temperature referred to above.
This presents a unique opportunity to examine climate variability in the UK on long
time-scales based on observational data. It would be advantageous if we could treat
these long time series as describing purely natural climate variability, thus enabling us
FIGURE 19.6 Top panel: water abstraction for agricultural use (spray irrigation in ml/day) for 1975–
94 for England and Wales and for the Anglian Water region; bottom panel: subsidence-related domestic
insurance claims at 1995 prices for 1975–95
Source: Subak et al. (1999).
Note: Many claims pertaining to the 1994–5 climate anomaly would appear in 1996 or 1997 data
which are not shown.
MIKE HULME
418
better to identify what level of human-induced climate change may truly be significant.
This would not be a correct interpretation, however, since—at least during the most
recent century—human forcing of the climate system has been occurring through
increasing atmospheric concentrations of greenhouse gases and sulphate aerosols. This
has allowed scientists to detect a ‘discernible human influence on the global climate
system’ (IPCC 1996:4). What we observe therefore, and this is as true for the UK as it
is for the world as a whole, is a mixture of natural climate variability and human-
induced climate change, with the contribution of the latter increasing over time. It is
nevertheless very instructive, before we progress to examine future climate change
scenarios for the UK, to appreciate the level of climate variability that the UK has been
subject to over recent generations.
Temperature indicators
The Central England Temperature series has already been referred to and an annual
version of the series plotted in Figure 19.3. Here, we show the four seasonal components
of this series smoothed with a 100-year filter to emphasise the century time-scale trends
(Figure 19.7). The greater warming in winter compared to summer is clear, although
most of this winter warming occurred before the present century commenced. Autumns
over recent decades have averaged nearly 1°C above the full period mean, while the
cluster of warm springs at the end of the series has meant that this season also has never
been as mild as it has over the last decade. The unusual warmth of the decade 1989–98
in the UK is supported by the sequence of very warm months and seasons shown in
Table 19.2.
The Central England Temperature series can also be used to examine changes
in daily temperature extremes, although in this case data are only available since
1772. Figure 19.8 shows the annual frequencies of ‘hot’ and ‘cold’ days in Central
England over these two centuries. There has been a marked reduction in the frequency
of cold days since the eighteenth century, these frequencies falling from between
fifteen and twenty per year to around ten per year over recent decades. There has not
been a commensurate rise in the frequency of hot days, although as with annual
temperature the last decade has seen the highest frequency of such days. For example,
1995 recorded twenty-six hot days compared to a long-term average of less than four,
and the decade 1988–97 averaged 7.4 hot days per year, nearly double the 1961–90
average.
Other climatic indicators
The annual precipitation series for England and Wales and for Scotland are shown
in Figure 19.3 and display little long-term trend. Here, we examine the seasonal
distribution of precipitation with Figure 19.9 showing the proportion of annual
England and Wales precipitation falling in winter (December-February). This index
is a simple measure of the continentality of UK climate—the larger proportion of
precipitation falling in winter the more continental the climate—and this seasonal
distribution of precipitation has important implications for how water resources are
managed. The proportion of precipitation over England and Wales falling in winter
CLIMATE CHANGE
419
FIGURE 19.7 Seasonal Central England Temperatures for the period 1659 to 1997. Horizontal lines
show 1961–90 means. Smoothed curves emphasise century time-scale variability
MIKE HULME
420
has increased over time, rising from about 23 per cent in the nineteenth century to
a 1961–90 average of about 27 per cent. Two of the three years with the highest
index values this century have occurred in the last decade—1990 and 1995 —
although there have also been recent years with quite low proportions of winter
precipitation—1992 and 1997. This shift in the seasonal distribution is a result both
of increased winter precipitation and decreased summer precipitation (not shown)
and is repeated in the Scotland series.
We also show (Figure 19.10) an index of gale activity over the UK, updated
following the analysis of Hulme and Jones (1991). This series is shorter than those for
temperature and precipitation and as with annual precipitation it shows no long-term
trend. Gale activity is highly variable from year to year, however, with a minimum of
two gales in 1985 and a maximum of twenty-nine gales in 1887. The middle decades of
this century were rather less prone to severe gales than the early and later decades,
whilst the most recent decade—1988 to 1997—has recorded the highest frequency of
severe gales since the series began, 15.4 per year compared to an average of 12.3 (see
also Chapter 17).
TABLE 19.2 Exceptionally warm months, seasons and years during the 1990s as indicated by the
Central England Temperature series
Note: All the anomalies shown fall in the ten warmest respective months, seasons or years in the
complete 340-year CET series. By chance, and assuming the monthly data are uncorrelated, one would
expect only four months, seasons or years to appear in this table. In fact, twenty-five exceptionally
warm anomalies have occurred during the 1990s. Over this period, only one month—June 1991—has
been exceptionally cold (using the inverse definition). Anomalies are shown in °C with respect to the
1961–90 mean. Data complete to July 1998.
CLIMATE CHANGE
421
There are many factors in the global climate system that have contributed to
these observed changes, but one of the more important for the UK has been the
behaviour of the North Atlantic Oscillation (or NAO). The NAO is a major disturbance
of the atmospheric circulation and climate of the North Atlantic-European region,
linked to a waxing and waning of the dominant middle-latitude westerly wind flow.
The NAO Index we show here is based on the monthly mean sea-level pressure
difference between various stations to the south (Azores and/or Iberian peninsula)
and north (Iceland) of the middle-latitude westerly flow (Jones et al. 1997). It is
therefore a measure of the strength of these zonal winds across the Atlantic. When
the NAO Index is positive, the westerly flow across the North Atlantic and Western
Europe is enhanced. During the winter half-year, the strengthened westerly winds
bring warmer, maritime air over north-west Europe causing a rise in temperature and,
FIGURE 19.8 Frequency of ‘hot’ (Tmean >20°C) and ‘cold’ (Tmean <0°C) days extracted from the
Central England Temperature record for the period 1772 to 1997. Horizontal lines are 1961–90 means
(3.4 and 11.8 days per year respectively) and smooth curves emphasise thirty-year time-scale variability
MIKE HULME
422
usually, precipitation. When the Index is low or negative, the opposite occurs with
temperatures falling over north-west Europe and rising over the north-west Atlantic.
The net result is a ‘see-saw’ or oscillation in temperatures across the North Atlantic-
European sector, as well as changes in other climate variables such as precipitation
and sea-ice extent.
The NAO exerts a strong influence on year-to-year climate variability in the UK. For
example, the correlation between NAO and winter temperature in the UK is about 0.67
(Jones and Hulme 1997). The winter (November to March) NAO Index is shown in Figure
19.11. The period from about 1970 recorded rising Index values, with the highest value
being recorded in 1994/5. The change in NAO condition between the winters of 1994/5 and
FIGURE 19.10 Annual frequency of ‘severe gales’ affecting the UK for the period 1881 to
1997. The horizontal line shows the 1961–90 mean frequency (12.3 per year) and the smooth
curve emphasises decadal time-scale variability
FIGURE 19.9 Percentage of precipitation over England and Wales falling in winter for the period
1766–1997. The horizontal line is the 1961–90 mean (27.4 per cent) and the smooth curve
emphasises thirty-year time-scale variability
CLIMATE CHANGE
423
1995/6 was quite remarkable—from the highest twentieth-century value to the lowest
twentieth-century value in successive years. The very low Index value in winter 1995/
6 was associated with a cold winter in the UK. The 1996/7 Index value was close to the
long-term average and the UK winter was correspondingly milder. The NAO displays
variations on a number of different time-scales, most of which may be unrelated to
global warming. However, given its importance in determining winter weather over the
UK, trends in UK climate cannot fully be understood without reference to the NAO
(Osborn et al. 1999).
A final indicator of UK climate variability shown in this section is that of sea-
level rise. Climate warming is anticipated to lead to a rise in global-mean sea-level,
primarily because of thermal expansion of ocean water and land glacier melt. Figure
19.12 shows long-term series of tide-gauge data for five locations around the UK
coastline. All of these series indicate a rise in unadjusted sea-level, ranging from 0.7
mm/yr at Aberdeen to 2.2 mm/yr at Sheerness (Woodworth et al. 1999). These raw
estimates of sea-level change need adjusting, however, to allow for natural rates of
coastline emergence and submergence resulting from long-term geological
readjustments to the last glaciation. The adjusted net rates of rise resulting from
changes in ocean volume range from 0.3 mm/yr at Newlyn to 1.8 mm/yr at North
Shields, evidence of an expanding ocean at least around the shores of the UK (see
Chapter 17).
Future changes in UK climate
Having established from proxy indicators and from the observed record that UK climate
varies naturally on a variety of time-scales, we now wish to consider what the future may
hold. To do so we have to consider the mechanisms which control UK climate, which
requires an understanding of two sets of factors. These are, first, the array of physical
processes and interactions within the climate system that shape the variations in UK climate
on these different time-scales and, second, the range of forcings that are external to the
FIGURE 19.11 NAO Index for the extended-winter period (November to March) from 1825/6 to
1997/8 (dated by January). The Index is the normalised sea-level pressure difference between Gibraltar
and south-west Iceland. Units are dimensionless. The smooth curve emphasises variations on time-
scales of thirty years
MIKE HULME
424
climate system but which have a profound effect on the way it functions. To achieve this
understanding in the most comprehensive way possible we need to rely upon climate
models. Climate models enable us, on a range of spatial and temporal scales, to investigate
the variability of climate that results from internal feedbacks within the climate system.
We can also use these models to examine the relative consequences for climate of imposing
FIGURE 19.12 Relative changes in sea-level over the last 100–150 years as recorded by tide gauges at
five UK locations. Last year of data is 1995 or 1996 and units are millimetres. Data are unadjusted for
crustal movements
Source: Woodworth et al. (1999).
CLIMATE CHANGE
425
different magnitudes of external forcing on the system, whether this forcing be changes
in solar activity, aerosols resulting from volcanic eruptions, or changes in the concentrations
of greenhouse gases in the global atmosphere. In this section, we use results from global
climate model experiments to present a range of future climate scenarios for the UK. In
this approach we follow the science reported by the IPCC in their Second Assessment
Report (IPCC 1996) and also used in the UK national climate change scenarios prepared
for the UKCIP in 1998 (Hulme and Jenkins 1998).
Future climate without climate change
We consider first the possible UK climate of the period around the 2050s decade in
the absence of any external forcing. That is, we assume that there are no further
increases in greenhouse gas concentrations in the atmosphere and that there is no
significant change in solar output nor any major volcanic eruptions between now and
then—these are, of course, artificial assumptions, but it helps us to identify the range
of possible climates that we would need to adapt to in the absence of climate change.
For this exercise we rely upon the results of a 1,400-year simulation of unforced
global climate by the Hadley Centre global climate model, HadCM2. By using this
model simulation as a description of natural climate variability we can define the
probability distribution of thirty-year mean climate states for the UK centred on the
decade of the 2050s (Figure 19.13). While all of these seasonal distributions of mean
temperature and precipitation are centred on zero (that is, there is an equally likely
chance of mean temperature or precipitation increasing or decreasing by this period),
it is the tails of the distributions that are most important for policy. Thus, even in the
absence of any climate change, over south-east England there is a finite chance of,
for example, (a) annual precipitation for the period 2040–70 averaging 8 per cent
higher or lower than the 1961–90 average, or (b) mean annual temperature for this
thirty-year period being 0.5°C warmer or colder than at present. Similarly, there is a
50 per cent chance that the annual climate of this part of the UK will be more than 2
per cent wetter or drier than now and a 50 per cent chance of it being more than
0.15°C warmer or colder. Furthermore, Figure 19.13 shows that the possible ranges
for seasonal change—in particular summer precipitation—are considerably larger than
these annual values, although over northern UK the ranges are slightly smaller than
over southern UK.
We start our description of possible future UK climate with this analysis of
unforced natural variability in climate on thirty-year time-scales because it is often
assumed that the only thing we have to plan for is human-induced climate change
(popularly referred to as ‘global warming’). This is not the case, as we have shown.
To take a more specific example, consider what level of summer rainfall variability
the water industry has to be able to manage successfully. Without climate change
affecting the UK and if, hypothetically, we were able to experience a large number
of climate states for the 2050s, then on average summer rainfall during 2040–69
will be similar to what it has been during 1961–90. In reality, however, we only
experience the 2050s once and Figure 19.13 shows that for south-east England
there is a 10 per cent risk that over this period summer rainfall will be averaging
7 per cent less than the 1961–90 mean and a 10 per cent chance that it will be