Effects of Climate Change in Sweden

S/JV— 45'8-3

naturv A rdsverket forlag Orderingaddress Swedish Environmental Protection Agency Customer Service S-106 48 Stockholm Phone+46 8 698 10 00 E-mail: [email protected] Internet: http//www.environ.se

ISBN 91-620-4583-0 ISSN 0282-7298

f z

DISCLAIMER

Portions of this document may be illegible in electronic image products. Images are produced from the best available original document. Preface

What will happen to ecological and technical systems in Sweden in a milder climate? The following report attempts to shed some light on this question. The report constitute a background to the Swedish Environmental Protection Agency’s future study “Sweden 2021“. Within the research programme “Effects of climate change and UV-B radi ation", the Swedish Environmental Protection Agency is supporting research on the effects of a changed climate on ecosystems. Preliminary results have previously been presented in “Ds 1994:121 “ This report contains a summa ry of Ds 1994:121 plus an analysis of the impact of a changed climate on tech nical systems. The analysis of the response of the technical systems to a cli mate change are not based on research, but are rather to be regarded as a sen sitivity analysis of the vulnerability of the systems to changes in the climate. The report examines a few selected systems, with no attempt at being com prehensive. The importance of the problems in relation to other global envi ronmental problems has not been weighed in. The authors accept full responsibility for the contents of the technical sensitivity report. Marianne Lillieskold has been responsible for compiling the report and the research results on the effects of a changed climate on Sweden ’s ecosystems. The following persons have participated in the techni cal analysis:

Sten Bergstrom Swedish Meteorological and Hydrological Institute, SMHI Goran Brofeldt Vattenfall Stefan Hagg Korsnas AB Jan Kjellegard Port of Stockholm AB Tom Lindqvist Stockholm Energy AB Lennart Nilsson Swedish National Grid Hartmuth Pauldrach National Board of Housing, Building and Planning Christer Sjostrom Royal Institute of Technology in Gavle, KTH Bo Westergren Stockholm Water Limited Hans Wirstam National Road Administration

Stockholm, June 1996 Marianne Lillieskold, Project Manager

3 Location map and names referred to in the text. Contents 1. Introduction 6

2. Climatic variations 7 Greenhouse gases 9 Today’s climate 10 UN Intergovernmental Panel on Climate Change 12 Climate models - GCMs 13 Ozone/UV-B radiation 14

3. Impact of climate on ecological systems 16 Terrestrial ecosystems 16 Limnic-telmatic ecosystems 17 Aquatic ecosystems 18 Ecosystem-based economic activities 19 Effects of UV-B radiation 21 Summary 22

4. Impact of climate on technical systems 23 Corrosion 24 Hydrological systems 27 Geotechnical systems 35 Energy systems 37 Risk management 41 Summary 45 References 46

5 1. Introduction

It has been estimated that a doubling of the carbon dioxide concentration in the atmosphere, or an equivalent increase in the total concentrations of the various greenhouse gases, would lead to an increase in the global mean tem perature by 1.5-4.5°C (EPCC 1990). However, the emissions of aerosols will mask the warming over industrial areas, which may result in a somewhat lower degree of warming, 1.5-3°C (EPCC 1995). Fossil fuels account for most of the emissions of anthropogenic greenhouse gases. At the current rate of emission, such a temperature increase could take place within 50 years. Warming is expected to take place more rapidly over the big continents than at sea, and more at higher latitudes than in tropical and subtropical regions. A higher mean temperature is expected to lead to great local and regional variations, particularly in precipitation, but also in terms of the frequency and intensity of storms. Examples of the effects of extreme weather types on society can be seen in reports from both Sweden and the rest of the world, where flooding, hurricanes and heat waves have caused infrastructural problems. Great uncertainties are involved in discussions of climate change. The cover illustration may seem provocative, but is intended to express the paradox that the effect of a rising global mean temperature at high Nordic latitudes along the Atlantic coast may just as well be a cooling. This scenario is not discus sed in this report. The extreme weather situations seen so far probably fall within the range of natural climatic variations. Owing to the long time required for climate change to take place, it is difficult to see when a change has really occurred. Given the large variations in climatic conditions that are already occurring, it is therefore of the utmost importance that we take all indications of weather and climatic impact seriously and adapt socio-economic structures accor dingly. Greater variation in weather types requires more flexible systems, but also greater restrictiveness with emissions of polluting and climate-affecting gases. Such control measures also mitigate other environmental problems (acid rain and other air pollution) that have a more direct effect on human health.

6 2. Climatic variations Klaesson

Per

Photo:

Ever since the atmosphere was formed, its chemical composition and global circulation have been constantly changing. Climate changes can be studied millions of years back in time in geological drill cores, and hundreds of thou sands of years back in time in sediment and ice cores. Models in which large quantities of data are processed are used in attempts to predict the future cli mate. Global warming is the topic of discussion today, but 20 years ago the dis cussion was focused on the cooling that could be read from the cyclicity of the curves of ice- and deep sea sediment cores. This reflects both the diffi culty of making assessments of climate change based on short series of data, and the importance of choosing a representative time span as a base period. The debate concerning which direction climate change may take remains and has been intensified in some respects. The reason for this is the cooling effect aerosols have on temperature but also the recent data which suggest that the North Atlantic circulation pattern has been weakened. The latter phenomen means that less heat is being transported northward, and the consequence is a cooling. In comparisons of climatic variations over time, today’s climate is compa red with the climate after the ice age; during the Middle Ages; in the early 20th century, etc. But in a “geological perspective", comparison could also

7 be made with the preceding interglacial, the warm period that preceded the most recent glacial period that started about 100,000 years ago (Figure 1). The reason for such a comparison is that the climatic conditions and the distribu tion of ice, land and sea masses give more similar premises to todays climate. Drill cores from Greenland (GRIP, GISP) and the Antarctic (Jouzel et al., 1993) contain information showing that climatic fluctuations are generally much more extreme in the northern hemisphere. The preceding interglacial, which lasted for about 20,000 years, probably included two cold periods with an intervening warm period with a considerably higher average temperature than now. The current interglacial has been characterized by a stable climate, which distinguishes it from the preceding one, but brief periods with higher temperatures than now have occurred repeatedly (e.g. Karlen, 1982). More recent data indicate that in order to find historical astronomical conditions, such as the angle of the earth’s axis and the angle and intensity of solar irra diation, that can be compared with today’s conditions, we must go as far back in time as 400,000 to 200,000 years before present. Considering the past 2-3 million years (the Pleistocene), it is found that the mean temperature has usually been lower than today. Rapid fluctuations have occurred i.e. the global signal after the last ice age faded out in 20-50 years,

"SUPER-INTERGLACIAL" CAUSED BY CO,

THOUSANDS OF YEARS AGO

INTERGLACIAL = PERIOD BETWEEN ICE AGES

Fig. 1. We are currently in an interglacial that haslasted about 10,000 years so far. It is estimated that a new glaciation will have reached its culmination in about23,000 years, but an intensified warm period may set in within the next few hundred years due to accumulation of anthropogenic emissions on top of the natural carbon dioxide cycle (Imbrieandlmbrie, 1979)

8 and microfossils show that the level of the sea off the Swedish west coast (the Kattegat and the Skagerrak) rose 25 m in 580 years (5 cm/y). We also know that it was 10 warmer 7,000 years ago, ’’the climatic optimum”, and that it was 0.5° colder during the “Little Ice Age", which reached its peak about 300 years ago.

•V Greenhouse gases

The so called greenhouse gases are important for the earth’s climate. Without water vapour and other greenhouse gases such as carbon dioxide, methane, nitrous oxide and ozone, the earth’s climate would be about 15-30° colder.

Temperature

Figure 2. Variations in tem perature (deviation from current temperature), car bon dioxide and methane during the past 160,000 years determined by analy sis of an ice core from cen tral Antarctica (after Jouzel, 140 120 100 80 60 40 1987). Thousands of years before present

9 This is due to the fact that some of the long-wave radiation leaving the sur face of the earth is absorbed by the greenhouse gases and transformed to heat in the lower atmosphere. Some of this heat is re-radiated back to the surface of the earth, which is thereby warmed. This is a natural process, but it is enhanc ed if the concentrations of greenhouse gases increase (radiative forcing). The covariation of temperature, carbon dioxide and methane is shown by Figure 2.

Anthropogenic greenhouse gases

Carbon dioxide (C02) is mainly formed by burning of fossil fuels. Defore station also contributes to the carbon dioxide input to the atmosphere, since the carbon accumulated by the forest during its growth phase is then released. A release of carbon dioxide has a residence time of 100-200 years in the atmosphere. Methane (CH4 ) mainly comes from landfills, cattle and certain forms of agriculture, such as rice growing. Despite the fact that methane emissions are increasing, methane is of less importance in the long run, since it disappears from the atmosphere in about 10 years. Nitrous oxide (N20) is another greenhouse gas that is also increasing in the atmosphere. Principal sources are rice paddies, the use of manure in agri culture and various combustion processes. Nitrous oxide has an atmospheric residence time of about 100 years. Chlorofiuorocarbons (CFCs, known as Freons) absorb radiation in a por tion of the infrared spectrum where the atmosphere did not previously have any ability to absorb heat radiation. For this reason, CFCs have tens of thou sands of times more effect per molecule as greenhouse gases than carbon diox ide. Their residence time in the atmosphere is between 50 and 1500 years. Ozone (03) in the troposphere and the lower stratosphere, the lowest (10-) 15 kilometres of the atmosphere, absorbs infrared radiation from the surface of the earth, thereby warming the lower atmosphere.

Today’s climate

The earth’s climate is controlled by the global circulation of the atmosphere and the oceans, and their interaction with the biosphere. The oceans store heat better than the continents, but it also takes longer for the oceans to warm up. 10 This inertia means that it takes a long time for a global climate change to stabilize. The global mean temperature has increased during the past 100 years or so by 0.45°±0.15°C (Figure 3). The six warmest years fell during the period 1980-1990. The climate in the Nordic region is affected by meteorological activity along the polar front, the region ’s geographic situation at the edge of a great conti nent, and the influence of the North Atlantic and its currents. This gives the climate a highly varying character with great local variations. The temperature in the Nordic region during the past 130 years was domi nated by a rising temperature from the mid-19th century to the end of the 1930s. Then began a period with declining temperatures that lasted until the mid-1960s, after which the temperature once again began to rise. Severe low- pressure activity prevailed during the years 1988-1993 with frequent wes terly winds, leading to mild winters and an increase in winter precipitation in the mountains. During the past 150 years, emissions from human activities have increased the atmospheric levels of carbon dioxide (+28%), methane (+130%) and nitrous oxide (+20%), thereby reinforcing the natural greenhouse effect. Evidence that the increased atmospheric carbon dioxide levels derive from human activities is provided by the fact that they closely follow the increase in anthropogenic emissions of carbon dioxide. Despite the fact that carbon dioxide is well-mixed in the atmosphere, its concentration is slightly higher in the northern hemisphere. The reason is that emissions here are higher. Mea-

0.4 --

a 0.2 --

S 0.2 - -

Krakofoa

Year 1860

Figure 3. Variations in global mean temperature during the period 1860-1990 rela tive to the mean value for the thirty-year period 1951-1980.

11 surements of the concentrations of the carbon isotopes 13C and 14 C in the atmos pheric carbon dioxide support the supposition of anthropogenic influence; fossil fuels and biospheric carbon lack 14 C and have a lower 13C/I2C ratio than carbon dioxide from biological processes. A decrease in the I3CPC ratio and l4 C since pre-industrial times is fully consistent with the addition of fossil fuels and biospheric carbon as a result of human activities (EPCC 1994).

UN Intergovernmental Panel on Climate Change

In 1988, to evaluate the state of knowledge regarding an imminent climate change, the WMO (World Meteorological Organization) and the UNEP (Uni ted Nations Environment Programme) set up the EPCC (Intergovernmental Panel on Climate Change), made up of experts from all over the world. An initial assessment in 1990 concluded that the greenhouse effect caused by human activities (mainly combustion of fossil fuels) will result in a global warming. In 1992, ozone depletion and aerosols, which have a negative radiative for cing effect, were also included in the EPCC work. The volcanic eruption of Mount Pinatubo in 1991 had led to a cooling (negative forcing) due to relea se of aerosols into the stratosphere. Here, modelling showed good agreement with the actually observed temperature decrease (0.4°). The IPCC report 1994 shows that the rate of increase in the atmospheric carbon dioxide concentration, as well as methane which has shown a decli ning rate of increase during the past decade, declined more markedly during 1991 and 1992. When the effect of the Pinatubo eruption started to fade away, global warming started to increase again in 1993. The impact of aerosols from industrial emissions and combustion processes is more difficult to assess and is considered to have greater regional variation. The previous imbalance in the carbon cycle was explained by the fact that the growing forests in the nor thern hemisphere comprise a large sink for carbon dioxide. The most recent EPCC report (1995) comes to the conclusion that in spite of all uncertainties ’’the balance of evidence suggests that there is a dicemi- ble human influence on global climate”

12 Climate models - GCMs

The future climate is simulated in general circulation models (GCMs) for dif ferent scenarios of increased concentrations of greenhouse gases, based on data collected by the IPCC. These climate models give results with a certain range of variation, but show relatively good agreement in predicting climate change in the Nordic region (Figure 4). The range of variation in these results is greater at a regional level. Simulation of a future milder climate confirms that climatic fluctuations will be greater in the northern hemisphere, probably due to changed salinity in the North Atlantic, which can give rise to an altered circulation pattern for that area (e.g. Pearce 1994). A higher temperature means greater evaporation. This may cause drought, but also increase of cloud formation. How cloudiness and precipitation affect temperature has not yet been fully evaluated in the climate models. This also applies to the influence of vegetation. Precipitation is very difficult to forecast and therefore also to handle in modelling work. Nevertheless, the models indicate that certain regions can expect higher precipitation, but with large regional differences. Type of pre cipitation cannot be determined in models, but thunderstorms, which are now most common in more southern regions, may also become more common far ther north in a warmer climate. The rain that falls during thunderstorms is also more intense. Generally, it is probable that effects of a global warming lead to greater variability in weather types. A rise in sea level is another expected result of global warming, mainly due to expansion of the water in the global oceans.

What will climate change mean for Sweden?

Because Sweden extends over several degrees of latitude, from the subbore- al zone in the south to the boreal and arctic-alpine vegetation zone in the north, Sweden have a wide variation of species and biotopes. Both ecosystems and technical systems have adapted to prevailing local conditions. A temperatu re rise may lead to a shift in vegetation zones and the extinction of certain species (Ds 1994:121). Viewed in a geological-historical perspective, this has occurred with repeated cyclicity. 13 Areas that have a long ground frost period today may have thaw-frost-thaw cycles during a greater part of the year than now, with long periods with heavy wet snow alternating with ice formation. The projected changes in vegetation, precipitation pattern and evaporation, due to a milder climate, runoff may be affected, with large regional variations as a result. Sea level rise is not a primary problem in all of Sweden, since the northern part of the east coast is still slowly rising as a respond to the last deglaciation. However, the west coast and parts of the southern coast may be affected, par ticularly as regards erosion and undermining of coastal structures. Technical systems with a short useful life can be replaced and adjusted as the need arises. As far as more long-range investments are concerned, which also require large financial resources, certain problems may arise as a result of a changing climate. Extreme events are of very great interest for the dis cussion of the consequences of possible climate change, since much of the social infrastructure is designed on the basis of such values. Examples are wind and snow loads on buildings, water supply, electric power supply, dam safety and safety against flooding. It is therefore important not to consider average values only when assessing different climate scenarios.

Climate change in the Nordic region (except Iceland) in 2030 Temperature: +2° in the summer +3° in the winter Precipitation: +10 mm/month during the summer +15 mm/month during the winter Sea level: +20 cm

Figure 4. Scenario for the year 2030 (Rodhe, 1990), used as a basis for vulnerabi lity analyses of ecosystems and technical systems.

Ozone/UV-B radiation

The changes in atmospheric chemistry and emissions of ozone-depleting gases lead not only to a temperature rise, but also to increased ultraviolet radiation, which can have a synergistic effect on certain organic processes and an impact on engineering materials. The ozone layer protects organisms on earth from excessively strong UV-B radiation, which is the most harmful type of radia 14 tion for biological organisms. Nowadays, thinning of the ozone layer is sig nificant at Sweden’s latitudes as well, where the ozone layer has thinned by 5-8% the last decade (WMO 1994). Earlier, the increase in UV-B radiation has only been measurable in the southern hemisphere and at high altitudes in the Alps. This is probably due to the fact that ozone depletion is offset by atmospheric pollution in such a way that UV-B radiation is blocked by the hazier atmosphere caused by aerosols in the northern hemisphere. As a con sequence of industrial emissions, it may be that today’s levels of ultraviolet radiation are below natural levels. Since industrial emissions will decrease in the future, UV-B radiation will then rise, paradoxically enough. Ultraviolet radiation varies in both time and space. The strongest radiation is, naturally, found around equatorial latitudes, with a diminishing gradient towards the poles. Ultraviolet radiation is also higher at greater altitudes. Ozone-depleting gases (CFCs) have a long atmospheric residence times, i.e. it takes a long time for them to be broken down and disappear from the atmosphere. The result is that they will remain in the atmosphere for a long time to come even if emissions are reduced.

15 3. Impact of climate on ecological systems Lillieskold

Marianne

Photo:

The vulnerability of ecological systems was evaluated in Sweden’s national report on climate change (Ds 1994:121). There it was concluded that the forest in particular, due to its long rotation time, may be affected drastically in the event of a climate change. The species composition in both terrestrial and aqu atic ecosystems may change, and the present-day climate and vegetation zones will be pushed northward. A brief summary of the current state of research is presented below.

Terrestrial ecosystems

The forest changes slowly (200-year time scale) as a function of climate and soil conditions. Models for a temperature rise show that the boreo-nemoral forest region (the coniferous forest belt that stretches from the northern Uni ted States and Canada over northern Scandinavia to the Russian taiga) may be shifted northward (Prentice et al., 1991). This will affect the tundra zone, which will have repercussions on the hydrological balance. In the south, the re will be a shift in dominance from spruce to a mix of beech, oak and pine. 16 Subarctic species are sensitive due to slow growth, long generation times and irregular reproduction, resulting in less capacity for rapid adaptation (Callaghan et al., 1993). Since the temperature increase is expected to be greater at high latitudes (Mitchell et al., 1990), the alpine ecosystems will be quick to show effects of a climate change. In the short term, an increase in temperature and carbon dioxide concentration will lead to increased forest growth. Effects of the temperature variations up to the 1930s have in the north led to a movement of the tree line 50 vertical metres up the mountain slopes (Sonesson & Hoogesteger, 1983). In the middle and southern parts of the mountains the tree line has moved downward, probably due to falling tem peratures since the 1930s (Kullman, 1989,1993). A rising tree line could lead to the extermination of certain arctic-alpine species which cannot migrate upward and also result in complete afforestation of the subarctic mountain slopes.

Limnic-telmatic ecosystems

The wetlands ’ response to rising temperatures and increased precipitation is difficult to predict. An increase in the amount of water may lead to a growth of mires (bogs), and thereby an increase in methane formation. An increase in temperature could also lead to increased evaporation, which would cause the wetlands to shrink in area. A smaller amount of water would result in oxy genation of wetlands and thereby decomposition, so that stored carbon is retur ned to the atmosphere as carbon dioxide. A higher temperature accelerates the process, and mineralization releases nutrients, leading to afforestation (Hanell, 1988; Silvola, 1986).

Soil water and groundwater may be affected negatively by a change in cli mate. Like the impact on wetlands, however, the exact effects are difficult to predict. Increased precipitation promotes greater groundwater recharge, whi le higher temperatures promote evaporation. The increased vegetation that is expected to result from rising temperature and precipitation may counteract increased groundwater recharge (Cardenas & Jansson, 1994). An increase in the carbon dioxide concentration makes the plants more water-efficient, which means that uptake of carbon per unit of evaporated water increases. 17 Photo: Inge Lennmark water-loving cies has periodically to rivers. diminishing The tem leading Aquatic form 18 sequence greater water agricultural different trol 1993).

the The

not

the (Forsberg, may brackish

Gulf regime

northward to

percolation, been

rain annual

3°C groundwater be

is increased

land salinity

changed are today,

higher fully temperature species

ecosystems

Bothnia (Lundstrom water

1992). regime

expected

shift established.

but

which the

in salinities

eutrophication,

will ecosystem, due

the

regimes

(Carlsson

what of of north,

to rise

migrate

north

groundwater the

in and

prolong

an turn will

is permafrost which

In will

increase Ohman, estimated caused

& leads the the happen the northward.

Sanner, lead

the which

previously Baltic long groundwater

to by

level

vegetation

in to

1990).

more in

line to inflow

term, the

may greater

Sea,

1994). the cause

fluctuations.

and

oxygen rapid load vegetation increase, rise. Lakes dissolved ter This term not ased and

event Higher cause belonged will the

winter leaching ice

period. Oxygen-rich

fresh

(Sverdrup

Mild 20% a will runoff on

composition

weathering, by

moderate

and

of changes

concentration. season. affected

the temperatures affected

higher

also

change water a

winters

to precipitation since

cover.

watercourses A

These

change in sea substances

another

shift result the

nutrients

& from

will

Transport precipitation

in in temperature water there The

above

and

mean factors wintertime. of Warfvinge, the

in the

the in

increase,

fish

the

the nutrient ground- the

ecosys climate

Warm-

lead a inflow

enters incre is

in all

short large shor

from

con con spe four will

will

the

by no of to

Marine ecosystems will have time to adapt to a slow change. In the long term, however, changes are expected in the shore zone. Changes in the size of bays (estuaries) and increasing salinity will affect spawning and fry areas. A chang ed species composition and geographic distribution will change the entire food chain. Mixing with deeper water layers will change the transport of nutrients and toxic organic compounds to the seas. This will favour blooms of toxic algae. Changes of ocean currents (the North Atlantic in the northern hemis phere and El Nino in the southern), as is suggested by climate models, will have consequences for both the marine fauna and the climate. Phytoplankton assimilate carbon dioxide from the air, so that the sea acts as a sink for greenhouse gases. Hypothetically, an increased phytoplankton pro duction due to eutrophication could increase the uptake of carbon dioxide.

Ecosystem-based economic activities

Agriculture. The growing season is limited by factors such as temperature and solar irradiance. With a rise in temperature, the spring growing season will become longer. Even with greater cloud formation, a limited effect will be obtained for the whole year. If the cultivation limit is shifted northward, annual production will increa se. However, increased crop production requires increased fertilizer applica tion, and the aggregate effect with increased mineralization as a result of incre ased precipitation and ground temperature will be increased nutrient leaching. To reduce the risk of this happening, the proportion of autumn- and winter- planted soil should be increased. Pesticide use will probably increase when the occurrence of insect pests and fungi increases with a warmer climate. Climatic and weather fluctuations are common in Sweden, which means that agricultural crops possess great tolerance to variations, providing good potential for adaptation to a changed climate. Furthermore, Swedish farmers cultivate a small number of plant spe cies with a short rotation period, which also favours adaptation to different conditions. However, farming practices will have to be adapted to the exten ded vegetation season.

Forestry. Higher temperatures and increased precipitation should result in improved forest growth. However, negative effects are likely since the natu

19 Foto: Bengt Olof Olsson 20 bottom peratures probably and southern nate, common expected. gat) Fishing. nish, by ed. carbon energy period. of sity.

The The tree

increasing In

whitefish It

will particularly even

the

Swedish

change species, dioxide to The conditions,

difficult Baltic,

Open-sea

flourish. be long in Many replace

the

forestry

coastal affected

shallow

will

the term, north.

concentration provenance

forest are southern

for if

the

fossil decline Cod amount

fishing particularly

which the

areas, industry the

forests At

marginally. salinity

water. helps

and

use

present, carbon energy.

species

which are

in of

herring of

and

to adapt, reduce

importance Eel, can

the

of and carbon biomass affected

reservoir

important

Factors natural

cold-water

the adapt

will of western

our Both pollution

Sweden fish,

Baltic

since only affect stored

further by

fuels

by positive regeneration.

such

while such in

changing for seas

warm-water the

many

the fish Sea ’ affect

coastal s increases.

cod. as

net as

out

perch exchange (the forest

such biomass

is the sea and

species

emission

to It the difficult

forests poorly which lead mid in ral of changes. Cultivated of species expected es

and

their Skagerrak

fishing.

perch, climate, is

sea as

and negative

The

amount a broad-leaved requires likely generation

cod

climate. not to continental milder spruce fish,

at methods,

have and roach of

have

are higher will

adapted of most and to expected higher composition

salt that to the

and

producing

carbon

predict. of

and effects

spruce

grow

and trees flatfish become low will cold

long A

forests

water atmospheric

cod, carbon

lobster

of vulnerable water

e.g. proportion shift the

biodiver more

increase.

to less pine

species. winters, to are

rotation

dioxide salmon

origin, can forests

choice

Katte Deep- domi in

in dimi rapid more

tem

stor well bio

will

will also

tre

hu the are the be

Coastal fishing in the Baltic Sea is relatively flexible, due to the occurrence of many species (of limnic-marine origin and both warm- and cold-water spe cies). A climate change is expected to lead to a greater number of storm days and therefore fewer fishing days, but a higher water temperature will lead to less ice formation, which means more fishing days.

Effects of UV-B radiation

Research is under way concerning the effects of a depleted ozone layer, resul ting in higher UV-B radiation, on plants and animals. The research, which is being funded by the Environmental Protection Agency, has focused on plants at high latitudes (e.g. Ambio 1995) and marine algae. Results obtained so far indicate that terrestrial plants form toxins, flavonoids (the invisible protecti ve pigments in plants), which absorb UV-B radiation and thereby give plants some protection. In general, plants have difficulty adapting, but those species that originate from regions with higher ultraviolet radiation can be bred more readily. Forest trees are probably most sensitive, but long-term experiments are lacking. Aquatic life such as fish roe, fry and small crustaceans are sensitive to ultra violet radiation, which penetrates down to a depth of about 0.5 metre. Sing le-celled algae (phytoplankton lack a protective mechanism) are most vulne rable to high UV-B radiation. These algae are responsible for half of the seas’ uptake of carbon dioxide from the atmosphere. If this balance is disturbed, this could contribute to an increase of the greenhouse effect. The species com position will probably shift towards more UV-tolerant species, and the food web may evolve towards larger species, which are less sensitive. The biggest problem for aquatic systems as well is that northern ecosystems are not adap ted to high radiation levels. There may therefore be a greater risk of change here than in regions where the UV-B levels are already high.

21 Summary

The global mean temperature rise predicted to result from a doubling of the carbon dioxide concentration in the atmosphere is 1.5-4.5°C. This would result in a replacement of ecosystems on 10-20% of the earth’s land area. How and where such a change would occur is unclear. What is certain is that the most sensitive ecosystems in Sweden are the forest, the mountains and the Baltic Sea. The forest, which comprises one of our basic industries, would have diffi culty adapting to new conditions, while agriculture, which has a shorter rota tion period, could adjust to new crops more quickly. Fishing may have to adjust to a change in species composition, from salmon and cod to perch and roach in the Baltic Sea and from cod and flatfish to sea perch in the Skager- rak and the Kattegat. Climate-dependent economic activities will be affected more negatively by an increase in the variability of the climate than by a slow, limited rise in ave rage temperature and/or precipitation. An overall assessment of the risks of a warmer climate shows that perturbation of the hydrological cycle would have great consequences for vegetation and for the range of ecosystems. Increased UV-B radiation could influence the carbon dioxide assimilation and water uptake capacity of plants. The effectiveness of the protective pig ments that are formed to protect the plants against increased radiation is not known.

22 4. Impact of climate on technical systems Lindman/Bildhuset

E:son

Ake

Photo:

Technical systems in our society are often not designed to adapt to rapid chan ges in external physical conditions. What distinguishes our age is that man has built a remarkable socio-eco nomic system during a period when this was possible. The climate was sta ble enough for the development of an infrastructure that has permitted the maintenance of an advanced society. A drastic change in the climate or extre me weather conditions with the changes this would bring could pose great risks to people in exposed areas, for example in the form of flooding and water shortages resulting in poorer water quality. A sensitivity analysis of the vulnerability of certain technical systems to climate change has been compiled in the Swedish Environmental Protec tion Agency’s future study “Sweden 2021“. This has not involved any research; the purpose has rather been to try to predict what problems might arise in conjunction with a changed or more variable climate, and to deter mine whether any preparedness exists in society to cope with such changes. The areas that have been studied are: corrosion, hydrological systems, geo technical systems and energy systems. Finally, an assessment of the risks has been included.

23 Corrosion

Corrosion is an economic problem of large proportions. A multitude of com plex processes which are affected by physical, chemical and biological fac tors are involved. The general principles are:

1) high temperature, like cold, reduces corrosion 2) small temperature variations have no effect 3) moisture is a crucial factor

Temperature has a complex effect on corrosion: On the one hand, the cor rosion rate increases with the temperature, since the rate of the electrochemi cal and chemical reactions, as well as the rate of diffusion, increases. On the other hand, a higher temperature (at temperatures above 0°C) also leads to faster drying of the moisture film that is present on material surfaces, which results in a shorter time of wetness (the time during which the relative humi dity exceeds 80%). Relative humidity is linked to the temperature, but a chan ge in a degree or two has little or no effect, while a couple of degrees can be of some importance.

A - If the temperature rises, the time of wetness may be affected (TOW = f(T>0°C, RH>80%) 1) the time of wetness is shortened - does not affect corrosion 2) same or longer time of wetness - accelerates corrosion B - If the number of occasions with temperatures <0°C is reduced the corrosion will increase

The variations and the extremes are more important than a small increase in the mean temperature. Corrosion rates are low in subarctic and arctic regions. The time of wet ness, which is linked to corrosion, varies in different parts of Sweden and Europe. In northern Sweden, corrosion is low because temperatures <0°C pre vail during much of the year. If the time with temperatures <0°C decreases, 24 corrosion will increase. South and central Sweden comprise a relatively homogeneous region. However, salt-saturated winds from the sea affect cor rosion along the west coast. Depending on the topography, the affect further inland can be detected up to a distance of between one or two kilometres to several tens of kilometres. Malmo in southern Sweden and Stockholm in central Sweden have the same annual mean values of relatively humidity. Nevertheless, moisture-related degradation is a much more important factor in Malmo than in Stockholm. Most areas in the temperate climate zone have a time of wetness of 2500-5500 h/y (in Stockholm 2900 h/y). Stockholm ’s inner city has a shorter (700 hours) time of wetness than the surrounding countryside, except to the east of Stock holm where the Baltic influences with moisture. Big city’s generates heat, which leads to a higher temperature which in turn leads to higher evaporation and therefore shorter time of wetness. Nevertheless, corrosion is higher in urban areas than in the surrounding countryside, due to the fact that higher air pollution levels also promote corrosion. The air contains pollutants in three states: gases, solid particles and sub stances dissolved in tiny droplets of water or on a liquid film on the surface of the solid particles. Dry deposition of pollutants is worse than acid rain. Acid rain is rinsed off as more rain falls, while the dry-deposited pollutants dissol ve in a condensate film and remain in place for a longer time. Ultraviolet radiation affects all organic material. In combination with acid rain, metals are particularly affected. The corrosion is further exacerbated by the action of sulphur compounds and nitrogen oxides, which together exert a strong synergy effect. When ground-level ozone is present, sulphur com pounds are oxidized to a more aggressive form, which attacks e.g. the zinc that is present in the galvanized surfaces of lampposts, sheet metal on buil dings and cars. Experiments aimed at studying the effect of acid precipitation and UV-B radiation in which plastic materials are either exposed freely to the elements or are protected from the rain and sun have shown clearly that the plastics degrade faster where they are exposed freely. Cost/benefit analyses of organic systems, e.g. paint coatings on wood and metal surfaces, have also shown that degradation of painted wood is accelerated when the surface is freely expos ed. The results also show that corrosion is worse in areas with high pollution levels.

25 Buildings

The climate influences the degradation of building materials on building fagades and roofs. Such degradation is normally the result of interaction with external factors such as the elements (weather and wind), as well as interac tion between different materials and material properties. Knowledge of the influence of climate and environmental factors on the degradation of exterior sheathing materials on buildings is necessary to be able to make reliable predictions of maintenance needs and lifetime. Relevant factors are relative humidity, temperature, precipitation and the presence of pollutants and corrosion products on the surface. The local topo graphy and the combination of wind and precipitation (driving rain) are, in addition to the properties of the material, of great importance. Proper design of buildings is important to prevent the ravages of the elements. Every porous material has a critical moisture content. At temperatures below freezing, the water in the pores freezes and causes mechanical frost burst. This type of damage can increase with a greater number of thaw-freeze-thaw cycles. Wind is the climate factor that is most neglected when planning buildings, despite the fact that it greatly influences our perception of both the indoor and the outdoor environment. Wind gives rise to pressure differentials over diffe rent building surfaces. These pressure differentials can affect the performan ce of e.g. ventilation systems. It may also be necessary to make allowance for wind loads in designing exposed building components. The dispersion of air pollution in relation to buildings and other structures is important both from a health viewpoint and from the viewpoint of their catalytic effects on corrosion. Rain in combination with wind can cause water to penetrate into porous materials. If these wind effects occur frequently, seve re problems can occur due to moisture migration. Moisture penetration in porous materials such as brick walls rarely occurs in the interior of the coun try, but does occur along open stretches of coast. This is primarily an econo mic problem, but can also constitute a health hazard by giving rise to allergic reactions. Air pollution causes corrosion of building materials. Pollutants are depo sited on the surface by dry deposition (adsorption of gases or deposition of particles) or by wet deposition (rain, snow). Nitrogen compounds, which are acidic, attack zinc, limestone, rendering, plastics and wood. In combination 26 with sulphur compounds, they cause corrosion of copper, electrical contacts, etc. Ultraviolet radiation has exhibited a noticeable effect on materials. Ultra violet, radiation, which splits molecular bonds, plus air pollutants which act as catalysts are highly reactive agents. They attack materials that are covered with organic surface coatings. Such materials are common in the infrastruc ture, for example in paints, plastics and jointing mastics. Measured values of ultraviolet radiation on S/90° surfaces are 10-30% higher than on horizontal surfaces and 7-14% higher than on S/45° surfaces. Southern fagades receive 12-40% higher ultraviolet radiation than northern fagades. An increase in ultraviolet radiation could lead to high costs for maintenance of surfaces containing ultravioletsensitive materials.

Hydrological systems Hydropower and dam safety

Ample water resources are of crucial importance for hydroelectric power gene ration. The volume of water available is sensitive to the annual distribution of precipitation and extreme values. The balance between precipitation and evaporation determines how much water is available. These relationships can be determined with great accuracy by mathematical models. The models make it possible to calculate the consequences of a climate change if we know the change in temperature and precipitation. A Nordic climate scenario developed in the Nordic project “Climate Chan ge and Energy Production" (Saelthun et al., in press) predicts an increase in both temperature and precipitation in the entire region. This scenario was used for hydrological simulations for assessment of hydropower production, but the scenario is also relevant for assessments of water resources. The Nordic scenario assumes a temperature increase in Sweden of about 4°C over a period of a hundred years. This is a considerable change, but still not more than the natural range of variation in Sweden ’s annual mean tem perature of about 3.7°C that was measured during the period 1961-1990 (Bergstrom 1994). The hydrological calculations in the Nordic project indicated increases in water volumes and thereby in the potential for hydropower generation in tho

27 Water discharge (m-Vs) NORTHERN SWEDEN Water discharge (m^/s) SOUTHERN SWEDEN

J I F I Ml A I Ml J I J IAI SIOINI D| Jl F1MIA1MI J 1J IA1SIOIMID Figure 5. Change in water discharge between the period 1961-1990 (thick line) and a hundred years later (thin line), mainly due to increased evaporation. se parts of the Nordic countries with the greatest precipitation, which in the case of Sweden is in the westernmost mountains. In Sweden the results also pointed towards a decrease in the available water due to increased evapotranspiration in other parts of the country, especially in the southeastern area, where the climate already today is dry. According to the above climate scenario, a climate change would bring about great changes in water resources. Milder winters will mean shorter periods with snow and thereby less need to store water, since the flow will be more even over the year. This is not particularly surprising in itself, since Swe den has large climatically related variations even with today’s climate. The mild winters of recent years, for example, have reduced the size of the spring flood in southern Sweden and led to increased water runoff (Figure 5). The effect of climate on electricity consumption is another factor that must be taken into account when analyzing the impact of climate change on hydro electric power generation. Calculations indicate a decrease in consumption by about 1.3% per degree of increase in the annual mean temperature in Swe den, due to a reduced heating requirement. This is however a very uncertain figure in a hundred-year perspective, since we cannot foresee the structure of the energy supply system so far into the future. The occurrence of extreme values is usually of greater interest than avera ge values of water discharge. It is usually extreme values that serve as the design parameters for technical systems, regardless of whether it is a ques tion of extreme flows or extreme drought. However, it is more difficult to draw conclusions about extreme water conditions. Taking these reservations into account, the conclusion can be drawn that the spring flows will general 28 ly decrease. Winter flows will become more common, and in some cases may exceed today’s flows. Even if the calculations are subject to strong reservations and cannot be used as a basis for decisions, they have a value as a basis for sensitivity analyses. They show that we must be prepared for rather drastic changes in water flows if a permanent climate change occurs, regardless of the direction of that change. This applies to both extreme values and average conditions. The risk of impact applies to both the average water discharge values and the extreme values that are often used as design parameters for technical systems. Experience from the Swedish hydropower system and physical planning in both regulated and unregulated river valleys show that there is a tendency to underestimate the variations of nature even with today’s climate. The serious ness of the problem has been confirmed by numerous severe floods in the past few decades. This has led to a review of the methods for calculating design flows for dam installations, resulting in a thorough re-evaluation of the safe ty of the hydropower system in this respect (Plow Committee, 1990). Fur thermore, the need for a mapping of risk zones has been pointed out by, among others, the recently concluded government report on dam safety (SOU, 1995:40). This is also an international problem. Recent flooding in the val leys of the Mississippi, Rhine and Meuse rivers are only a few examples. It is worth noting that the 1990 Flow Committee did not consider the time ripe for taking climate predictions into account in formulating the new guidelines for calculation of design flows for dams. Reservation was, however, made for the possibility that a permanent change in the precipitation climate in the futu re may necessitate a review of the guidelines.

Dwellings in the risk zone

In view of the flooding that has occurred on repeated occasions in recent years, the National Board of Housing, Building and Planning initiated a study of the consequences of such flooding for building planning. Among other things, the study examines the consequences of high water flows for the rescue ser vices. The first phase of the work - which has been conducted in cooperation with the National Rescue Services Board, the Swedish Power Association, the river regulation associations and local authorities - has resulted in a

29 Photo: Peter Hanneberg Flooding basis intention rivers, Board been major A Housing, material tion given as thrust material, graphy and ance 30 officials. serve 100-year

Eventually The where set

of one

identified for

level

of rivers.

including of dwellings.

purpose

( building

Voxnan/Edsbyn)

recommendations unregulated is

maps the where changes Then Housing, or is basis risks Building

being

but not 1,000-year

proposal

The this

for has to a

by of virtually

the permits

trial

Planning, collected are serve work map

assessment the traditional

issue Building been

and regulated

river

occurring,

planning is

continued

set

flows. will as

Planning

to along produced any all .

The pertains in (Vindelalven via map show

maps result guideline and

connection general investigation ones.

watercourses. a will new for materials

various Planning questionnaire work

water

except to

showing to It

example be

methodology

the gain

prohibitions one for

carried under particularly level

issuance building with River). the

future and regulated concrete is

methodology

the

to overview at

lines

physical the other

out sent roads identify

building. risks Previously,

measures. leadership

is by against on

background important

experience out

river guidelines

based ping ried Institute, investigated landslides thern on of ching along Swedish of

and the of

all planning map risk

ground to Sweden.

two flooding

using maps, building

National (Umealven out

local building

Sweden.

will on

are on

The areas a

risk of

rivers to

investigations

the

river for Geotechnical

existing

suitable.

aerial

material collapse the of and

for identify which government background

serve for

areas

This has west

along

sites. using

construc

Board

National

example the The below by in

debou

photo all River)

map

coast been issu have

nor

map

car The will The are risk and our our the the

a a

recommendations that are the end result of the entire investigation. The infra structure recommendations should apply to the entire country.

Ground collapse and landslide risks The investigation of ground collapse and landslide risks that has been carri ed out will be methodically coordinated by the National Board of Housing, Building and Planning with the results from the investigation of flooding risks so that recommendations can be coordinated. The risk of ground collapse and landslides must be taken into account at the comprehensive physical planning level in order to mitigate the risks to people and property. The Planning and Building Act and the Act Concerning the Management of Natural Resources exercise risk supervision wherever collapse and landslide risks exist. Ground stability can be improved (reduces the likelihood of collapse and landslides) by reinforcement measures; other wise all activity that would result in loss of life and property or harm to the environment in the event of a collapse or landslide must be discontinued or avoided.

Areas prone to landslides include: 1) areas with dwellings built above steep underwater slopes that are being eroded due to wave activity from passing vessels. 2) areas subject to flooding or prolonged rainfall, making the sediments water-saturated

A potential landslide risk is also posed by the water supply pipes that run across and alongside the river, should they begin to leak. The hydropower- related regulation of river flow helps to mitigate the landslide risk due to heavy flows. Another risk element is the possibility that toxic sediments might be washed into the river, affecting the quality of its water, which is used as drin king water.

Water supply and sewerage

Drinking water supplies in Sweden are based to a large extent on raw water (50-75%) taken from groundwater sources and lakes and subjected to subse

31 quent purification treatment. After treatment the drinking water is distributed to subscribers in water distribution systems consisting primarily of pipes made of steel, cast iron, PVC and polyethylene. The oldest parts of the water dis tribution systems in Stockholm are from the mid-19th century.Wastewater is conducted via stormwater and sanitary sewers to sewage treatment plants. The sewer systems are made primarily of concrete, clay and PVC pipes. Modem water supply and sewer systems have a technical life of between 200 and 300 years.

Factors that affect the piping systemsare: raw water quality external and internal corrosion in the piping system precipitation and flooding pollution from users, land and buildings ground movements lake and groundwater levels spreading root systems

At a higher air temperature, the water temperature of the lakes may rise, resulting in poorer raw water quality. A higher biological activity will follow a higher temperature. Problems can then arise in the form of blue-green algae, which produce toxins and deplete the water of oxygen (roughly a doubling of biological activity for every 10° of temperature increase). The production of algae has a great influence on which technology can be used for treatment. If evaporation increases more than precipitation, this also means that the lakes’ water turnover rate and groundwater recharge will be affected, which can impair the drinking water supply. Consumers will get slightly warmer and therefore not as good-tasting drinking water. This can, however, be count eracted by large investments in the form of new intake pipes for raw water. There are no large margins for increased precipitation in the sewer systems, which are designed to cope with a certain quantity of rainwater. Increased pre cipitation and more intensive rains can cause more frequent basement flood ing and sewage overflow into receiving bodies of water. This also leads to an increase in the quantity of stormwater to sewage treatment plants, resulting in larger total discharges of pollutants. 32 As an effect of soil acidification, external corrosion of metals in the ground will increase, shortening the lifetime of the water distribution networks., A sea level rise of 10 cm means increased seepage into the sewer system in extreme weather situations, especially in harbour areas. At lowlying pla ces, it sometimes happens that the sea level is so high that the water cannot run out. This will increase the load on treatment plants unless targeted invest ments are made in the sewer system. Intrusion of salt water into Lake Mala- ren, which constitute a large water supply to Stockholm and places around the lake, is not likely in the short term, but in a 100-year perspective it could happen due to a decreased level between the lake and the Baltic. Ground movements in conjunction with thawing of the frozen ground or groundwater lowering are directly reflected in a larger number of breaks in water and sewer pipes. If a warmer climate results in reduced frost heave, the effect will be positive, but the reverse is true if thaw-freeze-thaw becomes a more drawn-out process. Land use in the Lake Malaren Valley has a clear influence on water quality in the lake. The drinking water supply is particularly dependent on electricity. Large water works have emergency power stations while there are mobile emer gency power units for local water pumping stations. Water supply problems may arise (in Stockholm) if evaporation increases at the same time as precipitation remains normal. This means that the water left in streams and rivers will be of poorer quality. Lake Malaren ’s summer outflow is about 10 m3/s. Withdrawal in the summer can be 8 m3/s, which is virtually the entire volume. If the inflow to Lake Malaren should decrease, the drinking water supply could thus be affected negatively. Similar problems could arise at other places, especially along the Baltic coast.

Industrial and process waste water

At a higher frequency of precipitation, the quality of raw water to industry could be affected negatively. An increased hydrological load on humic-rich soils around the catchment area for raw water could cause increased leaching of coloured and oxygen-consuming pollutants. Reduced water quality could then cause problems with certain product quality parameters. However, this could probably be counteracted by a more effective and heavier dosage of tre atment chemicals in the raw water treatment stage. Another negative effect is an increase in the temperature of the incoming 33 water. It is then highly likely that bacterial activity in the water will increase. This will lead to growth of deposits in pipelines and probably also increased growth in cisterns. At worst, this could also lead to a higher bacterial content in the end products of industry. A rise in the sea level by about 20 cm by the year 2030 should not have any great effect on industrial plants along the northern coast of the Baltic Sea, given the ongoing rate of land uplift. A possible negative effect is increased problems with discharge of wastewater from outfall pipes that are already low-lying today. It should be possible to counteract this with increased pump capacity. If the problems turn out to be greater than anticipated, it may be necessary to raise the entire sewer system from its current level. Ground thaw does not cause any problems today with subsurface pipes in southern and central Sweden. The problems are greater in the northern parts of the country, but with a warmer climate these problem areas should shrink further as the ground frost period becomes shorter. However, uncer tainty in this respect is great, since a drawn-out thaw-freeze period may also be the case.

Harbour facilities

Land uplift, air pressure, wind direction, wind force and the period of time during which the wind acts in a certain direction are crucial parameters in the design of harbours in the Baltic Sea. In Lake Malaren, additional design fac tors are precipitation, meltwater and evaporation. The water level varies in the Baltic Sea, while the level in Lake Malaren is kept at about 4.14 m abo ve the lock threshold in the Karl-Johan lock in Stockholm. A higher average temperature will lead to greater evaporation from Lake Malaren, but this can be compensated for by reduced discharge. Sea level rise will be compensated for, at least in the short term, by land uplift along the east coast. The rate of land uplift varies from about 0.8 m/100 years along the Gulf of Bothnia to 0.4 m/100 years along the coast of the central Baltic to a trans gression in the southern Baltic of about 0.1 m/100 years. The quays in Stockholm harbour are 2 m high, which means that they are not sensitive to a small change. They could tolerate a change of ± 0.5 m. At high water levels, however, difficulties can sometimes arise even today at ferry berths, especially at fixed berths. If the sea level rises, the height of bridges may pose a problem to pleasure boat traffic. 34 Also in Gothenburg harbour, 0.5 m ± normal water level is acceptable. Water level increases of 1 m have occurred during strong westerly winds, resulting in flooding in the city. A prolonged rise in the mean water level has occurred on certain occasions. This can affect container handling in conjunc tion with strong winds during loading and unloading. Costly alterations would be required if the mean water level were to rise more permanently. Ferry har bours are generally more sensitive to small variations. The height of bridges can pose a problem also on the west coast even with a half a metre rise in the mean water level, since the water level can rise as much as 1.5 m more during severe weather. The consequences of a change in runoff in the upstream river are not known. Harbours are dependent for their operation on reliable supplies of electri city and water and smooth functioning of sewer and transport systems. It is not known today what the effect would be if these services were to be dis rupted. The vulnerability of the Stockholm and Gothenburg harbours to the pro jected climate changes up to the year 2030 is not particularly great.

Geotechnical systems

Roads

Design-basis factors for roads are climate and traffic load. Ground frost is the crucial design-basis factor north of the Lake Malaren Valley today. If the cli mate grows warmer, the ground frost limit will move northward and desig ning for ground frost will become less important. It is estimated that for 20% of the roads where ground frost is the crucial design factor today, traffic will be the design-basis factor in the future. This means that road structures can be made thinner, resulting in a cost reduction of about SEK 100 per metre of new road. In southern Sweden the maintenance costs will be less, while they will presumably increase in central Sweden, since the number of alternations between thaw and frost is expected to increase there. The effect for the coun try as a whole will be limited, since the greatest traffic works are in southern Sweden. The same reasoning applies to the use of studded tyres. Some road deformation can be expected in a warmer climate, but the cost of winter road

35 Photo: Marianne Lillieskold ber hern require the climate, however, the nic water lion try of of on the comes city soil 36

Way An If Acidification the Another Sea the

many

industry,

number of west- nitrogen

national

along per groundwater parts

runs increase snow. country,

days

level the before

up year. salts

when and

of runoff

the

slowly, north with

effect

rise

roads the economy, traps),

bridges

its

the south

roadsides

in ice and roads

in the

comparison country, slippery of roads

is winter

the levels

frost

southern of roads

the

closed

thereby great. other soil

while coasts,

than climate to warmth

contaminants

much has but

and

precipitation in be

and

whereas at hand, roads

Natural

today

this today. the built.

gone a Sweden high also

since

acid

shallower shortening more

change

country will

will made snow declines.

due

on pH, out

rain purification erosion they

not

than have

car may

to will

clearance to

will with is

will rise, will

frost

happen frost

the corrosion.

be Sweden.

the their necessitate This be negative

will mean

increase Germany cause bly upkeep released in build extended. million ment. er also this country cause ground.

deposited damage

increase depth

technical The Total useful the will would

make

costs will decrease if

more make

concrete

lead, less

the country.

construction

not

kronor. consequences The due will

if in lead will

moving salt

them cost will season. have

which,

quantity

snow

impact work in the

Less the

to which life, it nonetheless

outcome

to decrease ultraviolet the

consumption

by go about the more north.

roads a risky pH They drainage

and in

if positive

frost vegetation down some

several This despite

insulation

on the

is the declines. and

season

they SEK

profitable to For

further bound

have

the

is spring in

colder if roads

heave will

in drive flow

presuma radiation, not

problems

the

the probably

reducing effect environ the hundred a 800

will in milder

(orga

fores in

effect along If clear, long north

num flood

velo sout parts on.

will mil

turn

the

on be to

which will make road surfaces, markings and signs age faster than today which in the long run will lead to increased costs. During 1994, the National Road Administration published new rules for road construction, based in part on modem technology and theories and in part on updated climate information. Rules for reinforcement and improve ment of the existing road network will also be issued during 1995-96.

Bridges

If the climate becomes warmer with milder winters and more precipitation, many bridges in the country may suffer erosion problems and need to be streng thened. This is only true, however, provided that water levels and flows in our watercourses are appreciably higher in the future than today, at least periodi cally. Greater variations with more extreme flows at other times of the year may cause erosion problems at more places than today.

Energy systems

The consequences of electric power failures were recently described in “The threat and risk report". These are disaster scenarios, where the main empha sis is on defence aspects, but in some scenarios there are parallels with what might happen as a consequence of greater variability in the climate. This is particularly the case with electric power (SOU 1995:20) and water (SOU 1995:21). The riskreports deals with a chain of events (storm - wet snow - cold - salt coatings on power lines - line breakage - power outages - tree cle arance - fault tracing) which can have great effects on society. Following is a brief summary of events that demonstrate the vulnerability of the systems to changes in climate, particularly as regards activities that are dependent on electricity. In severe weather situations, the electric power system can suffer serious outages caused by storm winds, heavy wet snow with subsequent cold and/or salt coating, causing brief interruptions in the power supply. There is some capacity for automatic washing of salt-coated lines, but to a great extent this must be done manually. Other scenarios are windthrown trees across roads, causing accidents without the alarm and rescue services working; hospitals becomes overloaded. The casualty hospitals have some form of emergency

37 power, but its capacity varies. The district medical centres and old-age homes have no emergency power at all. Water works and sewage treatment plants, energy utilities and industries (refineries, petrochemical plants and the process industry) are particularly sen sitive to losses of power and water supplies. It is important to shut down the plant quickly before any major damage occurs. The risk of fires increases during sudden power outages. Only a few industries have emergency power. Computers and other technical equipment go down when power poles breaks and switching substations are knocked out. The infrastructure collapses. Food distribution, which is dependent on trans port, comes to a halt because petrol pumps don ’t work. Grocery stores beco me dark and cold, freezers and refrigerators don ’t work and food spoils. Rail traffic is down due to salt on the lines, windthrown trees, etc. The major air ports have emergency power for up to 70% of their normal needs, mainly for the lights on the takeoff and landing runways. Telephone stations have their own emergency power generators that work as long as they have fuel, while small stations have a capacity for 6-12 hours’ operation. Radio and TV have to transmit on emergency power.

Electric power - the trunk-line system

The national power transmission grid in Sweden consists of about 15,000 km of lines. Normally, ice loads occur on lines northwards from around 200 kilo metres north of L. Malaren at elevations over 400 m above sea level at tem peratures between 0°C and -2°C. Problems occur along the northern coast as well when the Bothnian Bay has not frozen. Moisture-laden air is then brought in over the land and cooled off. Winds along the whole northern coast cause problems at elevations from 400 to 600 m above sea level. If the temperatu re rises around one degree, the risk of ice load can increase, increasing the risk of power outages. Transmission lines are designed for an ice load of 3 kg/m and conductor. In severe weather, wires in the line can have an ice layer of up to 10 kg/m and conductor. The wires then become so heavy that poles and cross-arms break. Operational disruptions on lines can take weeks to repair. Corrosion of poles, stays and stay anchors will also increase if the tempe rature rises, just as pressure-impregnated wooden poles will suffer greater decay if the decay period per year is extended. The risk of power interrup-

38 tions will increase (pole collapse) and the costs of inspections and repairs will increase. Poles that are impregnated with creosote, as well as old oil and new PVC cables, are affected by ultraviolet radiation. Today glass is used as an insulation material, but in the future composite materials will be used, which are unfortunately more sensitive to ultraviolet radi Bild ation. It is possible to protect against ultraviolet radiation, but it is not known what other effects such protective material may have

on recipient waters and on water Carlsson/Pressens supply and sewage systems. Rolf The national power grid has tree-proof transmission pathways, Foto: so the risk of operational disruptions caused by windthrown trees falling on lines should not increase. At a higher carbon dioxide concentration, the growth rate of trees will increase, which will raise the cost of tree clearance in trans mission pathways. Felling of edge trees and compensation to landowners for these trees will also increase. Tree damage is not an extreme cost, but ope rational disruptions can have serious consequences. Ground damage could increase if the climate gets warmer. A shorter peri od with frost in the ground could impede maintenance and construction work on mires, through which transmission lines are often run today in northern Sweden.

Regional and distribution networks

Power lines and equipment in transformer stations and switching substations are designed to function at extreme temperatures. Regional lines (130-70 kV) and distribution networks (0.4-40 kV) are designed taking into account geo graphic location with respect to wind and ice loads. This means that areas near the west coast are given reinforced insulation to withstand salt coating. Near

39 the west coast, more insulation is used and special equipment is installed for washing of salt-coated plant parts. If the frequency of salt storms (and thunderstorms) increases, the frequen cy of outages in the power grids will increase, which means reduced availa bility. Extensive salt coating can lead to long outages. It is difficult to impro ve existing plants to make them more resistant to salt. The salt problem is worst for the highest lines and the switching substations, which are part of the trunk-line system. A higher wind frequency may pose a problem, since even today salt from the Atlantic is sometimes blown in over the mountain area and deposited far inland. Salt coating may then become a more widespread problem. Power transmission involves some energy losses, which cause heating of lines and equipment. With increased power transmission, it is therefore a nor mal phenomenon that equipment may have to be replaced or reinforced becau se the increased heating causes wear. A moderate increase in the ambient tem perature may accelerate such wear in some cases. Large temperature variations which greatly change the pattern of electrici ty consumption from heating in the wintertime to cooling in the summertime may have a great effect. Changes in electricity consumption are mainly of economic interest and mainly affect the consumer. But it is possible that elec tricity consumption will increase, since the biggest difference in a warmer cli mate is that the need of cooling and air conditioning will increase. More rain is probably not a problem, but wet snow and ice is. The availability of the electric power system will initially be reduced, but probably not to any dras tic extent. In the long run, countermeasures may be taken if the electricity consumers are prepared to pay for them.

Gas

There are four types of gas pipes in the Stockholm town gas network. They were laid during different eras in the more than 140-year history of the gas system. Cast iron pipes jointed and caulked with lead, steel pipes connected using the same jointing technique, steel pipes with welded joints (which con stitute more than half of the gas piping system), and in recent years PE plas tic pipes.

40 Distinguishing characteristics of the different gas line types: - The cast iron pipes are brittle and can crack if they are subjected to stresses, e.g. due to ground movements. - Socketed pipes, of both cast iron and steel, have a tendency to leak at the joints if subjected to ground movements and corrosion. - Steel pipelines corrode where the corrosion protection is inadequate. - PE pipelines are much more resistant to both ground movements and corrosion.

Experience shows that warm winters tend to reduce the frequency of breaks in the cast iron pipes, caused by cracks in the material. It is possible that this means that ground movements, for example those caused by ground frost, will decrease in Stockholm as the temperature rises. It is unclear whether this trend will be affected by a simultaneous increase in quantities of precipitation. Lar ge quantities of snow may act as insulation so that the rising temperature, in combination with the insulating snow cover, may lead to reduced frost depth. If this is true, it should be favourable for the gas lines in terms of both crack ing and joint leaks in the socketed sections. In the scenario with a lower groundwater table and extremely dry periods, subsidence will occur more frequently, which could have a negative effect on the pipes. Corrosion is a slow process that proceeds over a long period of time befo re the damage can be detected by the occurrence of a leak. Increased precipi tation should mean that more of the gas lines will be surrounded by water during certain periods. There is not enough knowledge today of whether this will increase the corrosion rate. The quality of the precipitation, acid rain, is presumably more crucial than the quantity. The cost of leak detection and repairs is about SEK 25 million annually.

Risk management

Certain types of damage caused by climate changes can have great reper cussions on the national economy. In order to get an idea of the importance of the problems, an analysis has been done where an estimate of what is impor 41 tant, for both society and the individual, is weighted against the certainty in the prediction of the evolution of the climate in time, space and intensity (after Saaty in Mason and Mitroff, 1981).

1 corrosion Factors Degree of importance 2 roads/bridges of importance to individual and society for future 3 stormwater/sewage actions Low High 4 harbour facilities 5 water resources I II High 6 energy distribution Certainty 1,2 of 7 dwellings, dam safety assess III IV Low ment 3,4 5,6,7

Box I. High certainty, low importance. Corrosion will most probably increase with a rising mean temperature, regard less of whether precipitation increases or remains as today. However, it does not pose any great safety risk, so it is accorded low importance. The effect on roads can be predicted fairly well, especially as regards assess ments of construction time and salting need. No serious problems are deemed to exist, so the system is assigned low importance from the risk viewpoint. Bridges may be damaged by erosion, but in this stage of a climate change the effects are not deemed to be great, which gives the problem low importance.

Box HI. Low certainty, low importance. The sewer system is sensitive to extreme precipitation as well as a rising sea level. Changes in these parameters would have consequences for low-lying sewers. The difficulty of predicting precipitation makes the level of certain ty in the assessment low. It has also been accorded low importance, butthe problem could be great if bacterial infection enters the picture. Harbour facilities are mainly sensitive to sea level rise. Sea level rise would have impacts of varying severity in different parts of Sweden, but the level differences are small in a shorter perspective.

42 Box IV. Low certainty, high importance. The most important questions are strongly linked to precipitation and evapo ration, but here the uncertainty in the scenarios is great. At higher temperature and unchanged precipitation compared with today, the water supply may be threatened, mainly due to impaired quality. With a different distribution of precipitation during the year, it may be more difficult to cope with overfilling of the water reservoirs. The risk of flooding will incre ase. Extreme variation in precipitation will lead to an increased risk of ground collapse and landslides. Furthermore, a higher mean temperature could lead to disruptions in ener gy distribution due to line breakage caused by wet snow or ice loads.

In deference to the precautionary principle, the technical systems described in box IV should be given priority in the continued work. The long-term planning for these technical systems should provide a pre paredness for future extreme weather situations and climate changes. Furthermore, it is of the utmost importance to advance the current state of knowledge regarding climate change by means of research and investigations. Questions that need to be answered are how weather types and climate can change with time and in intensity, and how different Swedish regions may be affected. Finally, it can be concluded that climate changes will in different ways be of concern to the insurance industry, which is a sector of the economy that works with risk management for the purpose of avoiding costs for future, more or less predictable, losses. Weighing in possible changes in the climate in evaluations of loss risks has become increasingly common practice among foreign insurance companies in particular, but Swedish companies as well should strive to determine limit values for “acceptable" effects of weather and wind. In Europe, five European insurance companies have joined an environ mental alliance under the leadership of the United Nations Environment Pro gramme, UNEP. Environmental risks are difficult to assess, and the possibi lity that the environmental risks have been misjudged is causing concern. Thair policy in general, is that a premium is placed on a good attitude towards environmental issues, which is seen as a sign of responsibility and therefore a good financial strategy for insurance guarantees.

43 Table 1. Summary of potential problems in Sweden caused by climate change. Platform for discussion.

Climate trend Increased temperature Increased temperature Sea level rise Technical systems Unchanged Increased precipitation precipitation

corrosion increased corrosion due increased corrosion to acid particles

hydropower reduced need of water risk of overfilling, dams storage due to more even erosion and dam failure flow during the year

dwellings in the risk water saturation of sedi areas near the coast and zone ments-increased risk of in the lower reaches of flooding and landslides rivers

water resources risk of water shortage due greater pollution load due to increased evaporation to leaching -poorer drinking water quality

stormwater risk of greater quantity of more frequent overflow, water intrusion in low- sewage micro-organisms more bacteria, more lying sewers leaching of nutrients

industrial wastewater increased algal and water intrusion in low- bacterial growth in sewers lying sewers

harbour small rise: no problems facilities higher rise: problems with loading, ferry berths most vulnerable

roads increased precipitation erosion problems along bridges can cause drainage the south coast problems, flooding

energy distribution increased corrosion of heavy wet snow-ice power lines, heavy wet load, increased corrosion snow and ice load in ground with acid groundwater

44 Summary

In the future climate scenario that has been sketched here with higher mean temperature and large local variations in precipitation, it is the extreme valu es that may pose problems for both ecosystems and technical systems. The analysis made of technical systems in this study suggests that corro sion will increase both in a milder and moister climate and in a milder climate with unchanged precipitation. Corrosion is strongly linked to acidification, which means that the stress on our technical systems may be great for a long time to come. With more frequent episodes of extreme precipitation, the flooding risk may increase. As a consequence, the risk of collapse and landslide may also incre ase, and design requirements for dams may have to be modified. Evaporation increases in a warmer climate, which can affect not only the quantity of water that is available but also its quality. Periods with heavy wet snow and ice loads on transmission lines may be more common in a warmer climate. This increases the risk of power outages, which can have far-reaching consequences in an electricity-dependent society. Sea level rise will be offset along the coast of the Bothnian Sea by land uplift, while the Baltic Sea coast and the west coast may experience problems with erosion and navigation in certain harbours. The aggregate results of the sensitivity analysis for technical systems and the possible impact of a milder climate in the future show that no great changes will occur in a shorter time perspective that could entail an increa sed risk to man and society. In a longer perspective, however, it is important to act now to create an emergency preparedness for the effects that may result from climate change. The process must be long-range but should be taken into account in our physical social planning, in our infrastructure planning, and in our design of technical systems. Many of our technical systems are old and will soon have to be replaced or repaired. This should be done with conside ration given to the future climate as an obvious design parameter.

45 References

Ambio Special Issue 1995: Environmental effects of ozone depletion. No 3, May 1995. Andren, A. & Jutman, T. 1995: Oversikliga riskzonkartor: Ett underlag for samhallsplanering. SMHI. Bergstrom, S. 1994: Naturliga variationer och betydelsen av extremhandel- ser. I: Klimatforandringens grander, ett seminarium om drivhuseffekten, Klimatdelegationen, Stockholm. Callaghan, T.V., Somme, L. & Sonesson, M. 1993: Impacts of climate chan ge at high latitudes on terrestrial plants and invertebrates. Directorate for Nature Management. Carlsson, B. & Sanner, H. 1994: Influence of river regulation on runoff to the Gulf of Bothnia. SMHI reports RH, No 9, Norrkoping. Flddeskommitten 1990: Riktlinjer for bestamning av dimensionerande floden for dammanlaggningar. Slutrapportfran Flddeskommitten. Statens Vatten- fallsverk, Svenska Kraftverksforeningen, Sveriges Meteorologiska och Hydrologiska Institut. Forsberg, C. 1992: Will an increased greenhouse impact in Fennoscandia give rise to more humic and coloured lakes? Hydrobiologia, 229:51-58. GRIP (Greenland Ice Core Project) Dansgaard, W., Johnsen, S.J., Clausen, H.B., Dahl-Jensen, D., Gundestrap, N.S., Hammer, C.U. Hvidberg, C.S., Steffensen, J.P., Sveinbjomsdottir, A.E., Jouzel, J., Bond, G. 1993: Evidence for a general instability of past climate from a 250-kyr ice-core record. Nature 364: 218-220. Grip members 1993: Climate instability during the last interglacial period recorded in the GRIP ice core. Nature 364: 203-207. GISP 2(Preenland Ice Sheet Project) Taylor, K.C., Lamorey, G.W., Doyle, G.A., Alley,R.B., Grootes, P.M., May- ewski, P.A., White, J.W.C. & Barlow, L.K. 1993: The ’’flickering switch” of late Pleistocene climate change. Nature 361: 432-436. Kapsner, W.R., Alley, R.B. Shuman, C.A., Anandakrishnan, S., Grootes, P.M. 1995: Dominant influence of atmospheric circulation on snow accumula tion in Greenland over the past 18,000 years. Nature 373: 52-54. Gardenas, A. & Jansson, P-E. 1994: Simulated water balance of Scots pine stands in Sweden for different climate change scenarios. Journal of Hydro logy 166:107-125. 46 Hanell, B. 1988: Postdrainage forest productivity of peatlands in Sweden. Canadian Journal Forest Research 18:1443-1456. Imbiie and Imbrie 1979: Ice Ages: Solving the Mystery. 224 sidor. IPCC (Intergovernmental Panel on Climate Change) Climate Change 1990: The IPCC Scientific Assessment. Cambridge Univer sity Press, 1990. 364 sid. Climate Change 1992: The Supplementary Report to the IPCC Scientific Assessment. 200 sidor. Climate change 1994: Radiative forcing of climate change and An Evalua tion of the IPCC IS92 Emission Scenarios. 339 sidor. Jouzel et al. 1987: Vostok ice core: A continuous isotope temperature record over the last climate cycle (160 000 years). Nature329:403-408. Jouzel et al. 1993: Extending the Vostock ice core record of Palaeoclimate to the penultimate glacial period. Nature 364: 407-412. Karlen, W. 1982: Holocene glacier fluctuations in Scandinavia. STRIAE 78.26-34. Kullman, L. 1989: Recent retrogression of the forest-alpine tundra ecotone (Betula pubescens Ehrh. ssp. tottuosa (Ledeb.) Nyman) in the Scandes Mountains, Sweden. Journalof Bio geography, 16:83-90. Kullman, L. 1993: Tree-limit dynamics of Betula pubescens ssp. tortuosa in relation to climate variability: evidence from central Sweden. Journal of Vegetation Science, 4:765-772. Lindstrom, G., Gardelin, M. & Persson, M. 1994: Conceptual modelling of evapotranspiration for simulations of climate change effects. SMHIRH. No 10: 26 sidor. Lundstrom, U. & Ohman, L-0.1990: Dissolution of feldspars in the presen ce of natural organic solutes. Journalof Soil Science, 41:359-369. Mason, R. and Mitroff, 1.1981: Challenging Strategic Planning Assumptions. J. Wiley & Sons. School of Business Administration. University of Sout hern California. Los Angeles. 324 sid. Mitchell, J.F.B., Manabe, S., Tokioka, T. & Meleshko, V. 1990: Equilibrium climate change. In Houghton J.T. Jenkins, G.T. & Ephraums, J.J. (eds) Cli mate Change, the IPCC Scientific Assessment. Cambridge University Press, Cambridge. 131-172. Pearce, F. 1994: Is broken ocean pump a global warning? New Scientist, mars, 7994.

47 Prentice, I.C., Sykes, M.T. & Cramer, W. 1991: The possible dynamic respon ses of northern forests to greenhouse warming. Global Ecology and Bio geography Letters, 1:129-135. Riskzonkarta over Edsbyn; oversvamningskansliga omraden vid hoga floden i Voxnan. SMHI, Lantmateriet. 1992. Rodhe, H. 1990: The IPCC Report on Climate Change; A Nordic perspecti ve, Nordiska Minister Radet. Saelthun, N.R., Aittoniemi, P., Bergstrom, S., Einarsson, K., Lindstrdm, G., Ohlsson, P-E., Thomsen, T., Vehvilainen, B. & Aamodt, K.O. 1995: Cli mate Change and Energy Production. Final Report. Silvola, J. 1986: Carbon dioxide dynamics in mires reclaimed for forestry in eastern Finland. Ann. Bot. Fenn., 23: 59-67. Sonesson, M. & Hoogesteger, J. 1983: Recent tree-limit dynamics (Betula pubescens Ehrh. spp. tortuosa (Ledeb.) Nyman) in northern Sweden. Nord- icana, 47:47-54. Swedens National Report. Under the UNFCCC. English version of Ds 1994:121. Fritzes Forlag, Stockholm. Sverdrup, H. & Warfvinge, P. 1993: Calculating field weathering rates using a mechanistic geochemical model-PROFILE. Applied Geochemistry, &27J-284. Utan el stannar Sverige. SOU 1995:20. Fritzes Forlag, Stockholm. Vag 94: Teknisk beskrivning for vagkonstruktioner. Vagverket. Staden pa vattnet utan vatten. SOU 1995:21. Fritzes Forlag, Stockholm. Samhallsutbyggnad och skredriskanalys i sodra Gora alvdalen. SGI Varia439. Scientific Assessment of Ozone Depletion: 1994. WMO, Global Ozone Rese arch Monitoring Project - Report No 37. Alvsakerhet: betankande. Utredning om dammsakerhet och hoga floden. SOU 1995:40. Fritzes Forlag, Stockholm.