Content uploaded by Ari Venäläinen
Author content
All content in this area was uploaded by Ari Venäläinen on Nov 06, 2019
Content may be subject to copyright.
PROJECTIONS OF FUTURE CLIMATE
FOR EUROPE, URUGUAY AND CHINA
WITH IMPLICATIONS ON FORESTRY
RAPORTTEJA
RAPPORTER
REPORTS
2019:3
ARI VENÄLÄINEN
KIMMO RUOSTEENOJA
ILARI LEHTONEN
RAPORTTEJA
RAPPORTER
REPORTS
No. 2019:3
PROJECTIONS OF FUTURE CLIMATE FOR EUROPE,
URUGUAY AND CHINA WITH IMPLICATIONS ON
FORESTRY
Ari Venäläinen
Kimmo Ruosteenoja
Ilari Lehtonen
Ilmatieteen laitos
Meteorologiska institutet
Finnish Meteorological Institute
Helsinki 2019
ISBN (paperback) 978-952-336-084-6
ISBN (pdf) 978-952-336-085-3
ISSN 0782-6079
Edita Prima Oy
Helsinki, 2019
Julkaisija
Ilmatieteen laitos
Erik Palménin aukio 1
PL 503,
00101 Helsinki
Julkaisusarjan nimi ja numero
Raportteja 2019:3
Julkaisuaika
2019
Tekijät
Ari Venäläinen, Kimmo Ruosteenoja ja Ilari Lehtonen
Toimeksiantaja
UPM-Kymmene Oy
Nimeke
Malliennusteisiin perustuva tulevaisuuden ilmasto Euroopassa, Uruguayssa ja
Kiinassa sekä ilmastonmuutoksen vaikutuksia metsätalouteen.
Tiivistelmä
Raportti tarkastelee odotettavissa olevia ilmastonmuutoksia neljällä UPM-Kymmenen toiminnan
kannalta keskeisellä maantieteellisellä alueella: Suomessa, Etelä-Saksassa, Uruguayssa ja Itä-Kiinassa.
Lisäksi arvioidaan ilmastonmuutoksen vaikutuksia metsien kasvuun ja käsitellään muutoksesta
mahdollisesti seuraavia kasvua häiritseviä tekijöitä. Ilmastonmuutosarviot perustuvat 28
maailmanlaajuisella ilmastomallilla tehtyihin ajoihin. Arvioitten pohjana käytetään RCP4.5-
skenaariota, joka vastaa kohtalaisen voimakkaita kasvihuonekaasujen päästöjä. Ensisijaisesti
tarkastellaan vuosisadan puolivälin (jakso 2040–2069) ilmastoa verrattuna 1900-luvun lopun (jakso
1971–2000) tilanteeseen.
Kaikki tutkimuksessa mukana olleet mallit ennustavat lämpötilojen nousevan tulevaisuudessa, joskin
lämpiämisen voimakkuus vaihtelee aika paljon mallista toiseen. Monille muille ilmastosuureille, kuten
sademäärälle ja auringonsäteilylle, muutoksen suuntaa ei sen sijaan aina voida varmuudella päätellä.
Tarkastelluilla alueilla sademäärä kuitenkin pääsääntöisesti kasvaa. Poikkeuksia ovat Etelä-Saksa
kesällä ja alkusyksystä, Itä-Kiina loppusyksystä sekä Uruguay eteläisen pallonpuoliskon talvella ja
keväällä, joissa sademäärä vähenee. Sielläkin missä sademäärät jonkin verran kasvavat, korkeamman
lämpötilan aiheuttama haihdunnan voimistuminen lisää kuivuuden riskiä. Joissakin tapauksissa sekä
kovat sateet että kuivat jaksot yleistyvät.
Euroopan metsävarat ovat viime vuosikymmeninä kasvaneet. Erityisesti Pohjois-Euroopan metsät
ovat hyötyneet lämpimämmästä ilmastosta ja hiilidioksidipitoisuuden kohoamisesta. Tulevina
vuosikymmeninä tämä kehitys ei välttämättä jatku ainakaan entistä tahtia, sillä metsien kasvua saattavat
häiritä mm. kuivuus, lisääntyvät metsäpalot ja tuhohyönteisten aiheuttamat vahingot. Pyrittäessä
hillitsemään ilmastonmuutosta metsien merkitys hiilen nieluina on tärkeä. Sen tähden mielipiteenvaihto
metsien tehokkaasta mutta samalla kestävästä käytöstä jatkunee vilkkaana.
Jos ihmiskunnan pyrkimykset hillitä ilmaston muuttumista osoittautuvat tehokkaiksi, on mahdollista,
että tulevat muutokset jäävät selvästi pienemmiksi ja tapahtuvat hitaammin kuin mitä tässä raportissa
tarkastellun RCP4.5-skenaarion perusteella olisi odotettavissa. Tämä kuitenkin edellyttää
maailmanlaajuisten kasvihuonekaasupäästöjen nopeaa leikkaamista. Hillinnän epäonnistuminen
puolestaan johtaisi tässä arvioitua vakavampiin seurauksiin.
Julkaisijayksikkö
Sään ja ilmastonmuutoksen vaikutustutkimus
Luokitus (UDK)
551.515.9, 551.556.1, 551.577.62, 551.583, 632.11, 632.92
Asiasanat
ilmasto, ilmastonmuutos,
ilmastoskenaariot,
ilmastonmuutokseen
sopeutuminen, ilmastonmuutoksen
aiheuttamat riskit, metsätalous
ISSN ja avainnimike
0782-6079 Raportteja
ISBN
978-952-336-084-6 (nide)
978-952-336-085-3 (pdf)
Kieli
englanti, (tiivistelmä myös
ruotsiksi ja suomeksi)
Sivumäärä
67
DOI
https://doi.org/10.35614/isbn.9789523360853
Utgivare
Meteorologiska institutet
Erik Palméns plats 1
PB 503
00101 Helsingfors
Publikationens serie och nummer
Rapporter 2019:3
Datum
2019
Författare
Ari Venäläinen, Kimmo Ruosteenoja och Ilari Lehtonen
Uppdragsgivare
UPM-Kymmene Oy
Rubrik
Den beräknade framtida klimatförändringen i Europa, Uruguay och Kina samt
förändringens inverkan på skogsbruket
Sammandrag
Denna rapport behandlar beräknade klimatförändringar inom fyra verksamhetsområden för
bolaget UPM-Kymmene: Finland, södra Tyskland, Uruguay och östra Kina. Dessutom diskuteras
konsekvenserna av dessa förändringar för skogsbruket, inklusive skogstillväxt, produktivitet och
potentiella störningar som förorsakas av klimatförändringen. Klimatprognoserna baserar sig på
simulationer utförda med 28 globala klimatmodeller. Analyserna baserar sig på RCP4.5-
växthusgas-scenariot som representerar måttligt stora utsläpp. Huvudsakligen granskar vi
prognoser för perioden 2040–2069 (i jämförelse med perioden 1971–2000).
Alla de analyserade modellerna simulerar högre temperaturer för framtiden. Storleken på
uppvärmningen varierar dock ganska mycket bland modellerna. För många andra klimatvariabler,
till exempel nederbörden och solstrålningen, kan även tecknet på den framtida förändringen vara
osäker. I de regionerna som undersökts är det dock mer sannolikt att nederbörden kommer att öka
än minska, med undantag för södra Tyskland på sommaren och under tidig höst, Uruguay på södra
halvklotets vinter och vår samt Kina på senhösten. Den stigande temperaturen ökar avdunstningen
och risken för torka, även i områden med en måttlig ökning i nederbörd. Under vissa årstider
beräknas både intensiva ösregn och torra perioder bli allmännare.
Under de senaste decennierna har skogsresurserna i Europa ökat. Särskilt i Nordeuropa har
skogens tillväxt gynnats av det varmare klimatet och den ökande CO2-koncentrationen i
atmosfären. Under de kommande decennierna kan denna positiva utveckling, åtminstone delvis,
avbrytas på grund av potentiella störningar. Till exempel kan skador orsakas av torka, skogsbränder
och skadedjur. Skogens roll som kolsänka är viktig för att begränsa klimatförändringen. Därför
kommer en livlig diskussion om den mest gynnsamma och hållbara användningen av
skogsresurserna att fortsätta.
Om den globala klimatpolitiken visar sig vara framgångsrik, kan de framtida
klimatförändringarna möjligen vara lindrigare än de som förväntas på basen av RCP4.5-scenariot,
vilka har diskuterats i den här rapporten. Detta kräver dock snabba globala begränsningar av
växthusgasutsläppen.
Publikationsenhet
Forskning av väder och klimatförändringens effekter
Klassificering (UDK)
551.515.9, 551.556.1, 551.577.62, 551.583, 632.11, 632.92
Nyckelord
klimat, klimatscenarier,
anpassning till
klimatförändringen, risker,
skogsbruk, global uppvärmning
ISSN ja och serietitel
0782-6079 Rapporter
ISBN
978-952-336-084-6 (volum)
978-952-336-085-3 (pdf)
Språk
engelska (sammandrag också
på svenska och finska)
Sidantal
67
DOI
https://doi.org/10.35614/isbn.9789523360853
Publisher
Finnish Meteorological Institute
Erik Palménin aukio 1
P.O. Box 503
00101 Helsinki, Finland
Report name and number
Reports 2019:3
Date
2019
Authors
Ari Venäläinen, Kimmo Ruosteenoja and Ilari Lehtonen
Commissioned by
UPM-Kymmene Oy
Title
Projections of future climate for Europe, Uruguay and China with implications on
forestry
Summary
This report deals with projected climatic changes in four areas of operation of the UPM-
Kymmene company: Finland, southern Germany, Uruguay and eastern China. The implications of
the projected changes for forestry, including forest growth and productivity and possible climate
change induced disturbances, are discussed as well. Climate projections have been derived from
the output of 28 global climate models. Analyses focus on the RCP4.5 greenhouse gas scenario
that represents an alternative of moderately large emissions. Mainly, projections calculated for the
period 2040–2069 (relative to 1971–2000) have been examined.
All the models analyzed simulate higher temperatures for the future. However, the degree of
warming varies quite a lot among the models. For many other climate variables, like precipitation
and incident solar radiation, even the sign of the future change can be uncertain. Even so, in the
regions examined mean precipitation is more likely to increase than decrease, except for southern
Germany in summer and early autumn, Uruguay in Southern Hemisphere winter and spring and
China in late autumn. Rising temperatures enhance evaporation and increase drought risks despite
modest increases in precipitation. In some seasons, both the intense rainfall events and dry periods
are projected to become more severe.
In recent decades, forest resources have been increasing in Europe. Especially in Northern
Europe, forests have benefitted from the warmer climate and increased CO2 concentration in the
atmosphere. During the coming decades, this positive development may at least partly be cancelled
due to potentially increasing disturbances for the forest growth. For example, drought, fire and
insect pests may cause damage. The role of forests as a carbon sink is an important aspect in the
context of climate change mitigation activities, and vivid discussion on the most beneficial and
sustainable use of forest resources is foreseen to continue.
If global climate policy proves to be successful, it is possible that future changes in climate will
be weaker than those based on the RCP4.5 scenario discussed in this report. However, this requires
rapid restrictions of the greenhouse gas emissions globally.
Publishing unit
Weather and Climate Change Impact Research
Classification (UDK)
551.515.9, 551.556.1, 551.577.62, 551.583, 632.11, 632.92
Keywords
climate, climate scenarios,
adaptation to climate change,
climate risks, forestry, global
warming
ISSN and series title
0782-6079 Reports
ISBN
978-952-336-084-6 (print)
978-952-336-085-3 (pdf)
Language
English (abstract also in
Swedish and Finnish)
Pages
67
DOI
https://doi.org/10.35614/isbn.9789523360853
PREFACE
There is a growing need for consistent, scientific and forward-looking information on
climate change and its impacts on environment and societies. For a company, it is crucial
to understand how its business model, operations and assets could be affected by physical
climate change and by transitional aspects such as policies, regulation, technologies and
market behaviour.
Understanding the exposures to risks and opportunities of changing climate helps in
building the response and taking actions that help adaptation to possible future scenarios.
At the same time, it reminds of the urgency of mitigation actions.
To get the best possible information relying on the latest scientific knowledge, UPM
engaged with the Finnish Meteorological Institute to study the physical impacts of climate
change in its main areas of operations. The study focused especially on company’s main
raw material source, forest, but takes also into account other aspects such as water
availability and weather extremes.
1
CONTENTS
PREFACE ....................................................................................................................... 0
Executive summary ......................................................................................................... 2
1. Introduction ................................................................................................................. 7
1.1 Global climate change ............................................................................................... 7
1.2 Modelling of future changes ..................................................................................... 8
1.3 Projected changes in global climate ........................................................................ 10
1.4 Climate change and forests ..................................................................................... 14
1.5 Observed changes in forest resources ..................................................................... 14
2. Climate change and its impacts in Europe ................................................................ 16
2.1 Overview of projected climate change in Europe ................................................... 16
2.1.1 Finland .......................................................................................................... 19
2.1.2 Southern Germany ........................................................................................ 26
2.2 Climate change impacts in Europe with focus on forest sector .............................. 30
2.3 Climate change induced risks to forests in Europe ................................................. 32
2.3.1 Varying wind damage risk ............................................................................ 32
2.3.2 Insect pests .................................................................................................... 32
2.3.3 Less soil frost ................................................................................................ 35
2.3.4 Drought leading to higher forest fire risk ...................................................... 37
2.3.5 Snow damage risk ......................................................................................... 39
3. Climate change and impacts in Uruguay .................................................................. 41
3.1 Projected change of climate in Uruguay ................................................................. 41
3.2 Overview of climate change impacts in Uruguay ................................................... 46
4. Climate change and impacts in China ....................................................................... 49
4.1 Projected change of climate in China ..................................................................... 49
4.2 Impacts on the Yangtze River hydrology ............................................................... 53
5. Concluding remarks .................................................................................................. 56
References ..................................................................................................................... 57
Appendix A: Technical information about the model data and the analysis methods .. 65
A.1 Climate model data ......................................................................................... 65
A.2 Processing of the model output ....................................................................... 66
2
Executive summary
Discussion on how forests can be utilized in a sustainable way requires information about
the ongoing climate change and as well, about the impacts of the change. This report
provides estimates for future climatic changes for four key regions that are of interest for
the operations of the UPM-Kymmene company: Finland, Southern Germany, Uruguay and
Eastern China. Climate projections have been derived from the output of 28 global climate
models (GCMs), even though for some climate parameters, data were not available from
all the models. In the model runs, three alternative greenhouse gas scenarios have been
used, RCP8.5 representing very large, RCP4.5 intermediate and RCP2.6 very small
emissions of carbon dioxide and other greenhouse gases. This report mainly examines
projections calculated for the period 2040–2069 (relative to 1971–2000) under the RCP4.5
scenario. Potential impacts of the projected changes are discussed as well.
The principal anticipated changes of climate for the focal areas
If not otherwise stated, the numerical values refer to the multi-model mean changes, the
uncertainty interval of the change following in parentheses.
(a) Finland
• Mean temperature is projected to increase by about 4°C (1 to 6°C) in winter and
slightly over 2°C (1 to 4°C) in summer. Temporal variations of temperature
attenuate in winter but remain virtually unchanged in summer. Thermal growing
seasons lengthen by 3–4 weeks and the effective temperature sums increase by
∼400 degree days.
• Precipitation increases by 15 % (0 to 30 %) in mid-winter and 5 % (-15 to +25 %)
in mid-summer. Consequently, it is nearly certain that mean precipitation increases
in winter but in summer the sign of change is uncertain, even though precipitation
is more likely to increase slightly than decrease. Events of heavy precipitation
intensify. Dry periods are likely to shorten in all seasons apart from summer. Soil
moisture declines in spring while in summer the changes are on average fairly
modest.
• Wind speeds change little; this holds both for the temporal averages and the high
wind speeds.
• Solar radiation received at the surface may decrease in winter and slightly increase
in summer, but the inter-model agreement on the sign of change is low.
• Discharge in the Kymmene river increases in winter by 30–40 % (the best estimate
of change) and decreases in late summer by about 20 %.
(b) Southern Germany
• Mean temperature is projected to increase by about 2.5°C (1 to 4°C) in summer and
slightly less than 2°C (1 to 3°C) in the other seasons. Temporal variations of
temperature attenuate slightly in winter and remain unchanged at other times.
• Precipitation increases by 10 % (-10 to 25 %) in mid-winter and decreases by 5 %
(uncertainty interval from -30 to +20 %) in late summer. The events of heavy
precipitation become more intense, less so in summer than in the other seasons. Dry
periods are likely to lengthen in summer and autumn. Soil moisture declines
substantially in summer and early autumn.
• Wind speeds are anticipated to change rather little.
• Solar radiation increases in summer by 10 % (0 to 20 %). In the other seasons the
increase is smaller and less robust. Relative humidity declines in late summer by 6
percentage points (uncertainty interval from -12 to +2 % points). In other seasons
the drying of air is weaker.
3
(c) Uruguay
• Monthly mean temperatures are projected to increase by about 1.2–1.5°C (0.5 to
2°C). Temporal variations of temperature do not alter significantly.
• Precipitation increases by about 10 % (-15 to +40 %) in the Southern Hemisphere
autumn and decreases by about 5 % (uncertainty from -25 to +15 %) in spring. The
events of heavy precipitation become more intense, especially in the Southern
Hemisphere autumn. Dry periods are likely to lengthen in winter, perhaps also in
spring and summer.
• Wind speeds are most likely to change little.
• Changes in solar radiation and relative humidity are likely to be small as well.
(d) Eastern China
• Mean temperatures are projected to increase by about 2°C (1 to 3°C) throughout the
year. Temporal variations of temperature do not alter significantly.
• Precipitation increases slightly, by 0–10 % (-15 to +25 %), in all seasons except late
autumn. The events of intense precipitation become more severe in all seasons. Dry
periods are more likely to lengthen slightly than shorten in winter, spring and
autumn.
• Wind speeds and relative humidity are likely to change fairly little.
• Solar radiation increases by 2 % (-4 to 8 %) in winter and by 4 % (-2 to 10 %) in
summer. However, this is not a purely climatological phenomenon but also affected
by decreasing air pollution.
It is important to note that changes simulated by the individual models diverge
substantially. Even if it is virtually certain that the mean temperatures rise in all the regions
throughout the year, for many other parameters even the sign of the future change cannot
be established firmly. For example, the direction of the precipitation change is generally
at least to some extent uncertain, with the exception of Finland in winter, where the
likelihood of an increasing precipitation is very high. The present projections correspond
to the RCP4.5 scenario. If climate change mitigation measures are effective, the changes
will be weaker than assessed in this report. Accordingly, planning of future activities must
inevitably be founded on information that is more or less imprecise.
Observed changes in vegetation and forest resources
• When we look the recent changes in vegetation, the measurements made from
satellites indicate greening of the world. The greening trend can be detected
especially in China and India. In China, the greening is caused mainly by increasing
forest area. However, there are also regions on the globe where the forested area
has been declining; for example in Brazil, Democratic Republic of Congo and
Indonesia.
• In Europe, the forested area and also growing stock have increased. During the
recent years, the increase of forested area in the EU-28 countries has been more
than 400,000 ha per year. Concurrently, also the growing stock has increased.
• In Finland, the total increment of forests is currently more than 100 million m3 per
year and the total drain of trees (felling and death due to natural causes) more than
80 million m3.
Climate change impacts in Europe
It is likely that climate change will have both positive and negative effects on wood
production and wood supply.
• The positive impacts are due to warmer temperatures, longer growing seasons and
fertilization caused by the increased CO2 concentration in the atmosphere. Tree
4
growth and productivity are predicted to increase especially at relatively high
latitudes and altitudes where low temperatures currently limit the growth.
• The changes may be positive in near future and negative in the mid and long term.
The possible increase of meteorological disturbances, which likewise expose
forests to biotic damages, may to some extent cancel the climate change-induced
productivity increase.
• Increase in the frequency and severity of summer droughts will have an impact on
forest fire danger. In the Mediterranean region, large forest fires occur almost every
summer, and huge fires are not rare either in the boreal forests of Russia and
Canada. The risk of large fires will increase also in northern Europe. Drought may
also impact negatively especially the growth of Norway spruce.
• Most of insect pests will benefit from climate change because their development is
primarily regulated by air temperature. In Finland, Norway spruce is the most
sensitive tree species to climate change, and increasing drought occurrence may
weaken its defence against insect pests. In addition to already existing insect pests
and fungi, new invasive species can be accidentally introduced by international
plant trade.
• In Finland, snow is one of the most important abiotic disturbance agent reducing
stand quality in forests. Assessments based on climate model simulations indicate
a potentially increasing risk for snow damage in eastern and northern Finland,
particularly in the regions of North Karelia, Kainuu, Koillismaa and Lapland. In
southern and western Finland, on the other hand, snow loads on tree crowns are
projected to decrease.
• Climate change will shorten frozen soil period in northern Europe, making tree
harvesting more challenging in locations having poor bearing capacity.
• The overall effect of climate change on forest resources requires continuous
multidisciplinary research. It is obvious that intense discussion on the level of
sustainable utilization of forest resources related to the EU’s land use, land use
change and forestry (LULUCF) will continue.
Climate change impacts in Uruguay
The impacts of climate change in Uruguay are related to sea level rise and other possible
hazards such as droughts and floods, heatwaves, hails, storms and tornados. The El Nio
- Southern Oscillation phenomenon (ENSO) intensifies inter-annual variations in weather
conditions, with higher precipitation during the El Nio years and longer droughts during
the La Nia years. If climate change influences the frequency and intensity of this
phenomenon, it may cause new challenges to the Uruguay society.
• The tree cover loss between 2001 and 2017 in Uruguay was 327,000 ha whereas
the tree cover gain was 499,000 ha. The tree-covered area has thus increased about
170,000 ha during the about 15–year period. This development can contribute to
Uruguay’s greenhouse gas balance in a positive way, i.e., the role of forests as a
carbon sink increases.
• The largest economic risk caused by sea level rise will be experienced in urbanized
areas like the Maldonado-Punta del Este resort and in Montevideo. Sea level rise
may cause also other effects like flooding and salinization of lowlands and the lower
course of the water flows that discharge along the coast.
• For winter crops, like barley and wheat, the increase of temperature can be harmful
but for summer crops like rice, warming can be positive. As well, warmer
temperatures may benefit grassland production. Conversely, the increased
variability of precipitation is regarded as harmful.
5
Climate change impacts on Yangtze River hydrology in China
• Several published studies indicate that climate change will cause a general increase
in annual streamflow at Yangtze River. The increase is predicted to occur mainly
during the warm season whereas during the cool season the streamflow can even
decrease. The increased fluctuations in rainfall can increase the variability of
maximum streamflows and flooding events.
• Glaciers at the source region of Yangtze River are shrinking. This leads to an
increase of water resources over the short term, but on the long-run the impacts of
melting glaciers can be negative and decline water resources.
• Besides rainfall patterns and the glacial melting, the erosion caused by the sediment
starvation and the sea level rise are influencing the hydrological conditions near the
Yangtze River delta. River damming and soil conservation decrease sediment
discharge, and in conjuction with the sea level rise, this increases the erosion
potential of the delta area. As well, saltwater intrusion and storm surge intensity in
the delta area may increase.
6
Table 1. Climate change projections and impacts of climate change under greenhouse gas
concentration pathways RCP2.6, RCP4.5 and RCP8.5. Information given in items 1 and 2
has been extracted from IPCC. Items 5-9 are derived from the findings of the present
report. In items 3-4, some examples of risks have been selected that the authors regard as
particularly severe. The reported degree of severity may vary to some extent in literature.
RCP8.5
RCP4.5
RCP2.6
1. Global warming relative to
the preindustrial level.
2.7°C by the 2050s
2.1°C by the 2050s
1.7°C by the 2050s
4.2°C by the 2080s
2.6°C by the 2080s
1.8°C by the 2080s
2. Global sea level rise from
the 1990s to 2090s. In brackets
there are likely ranges for
projections.1
63 cm
(45 cm – 82 cm)
47 cm
(32 cm – 63 cm)
40 cm
(26 cm – 55 cm)
3. Impacts on global scale
Significant fraction of plant and
animal species may be extinct.
Living conditions in many
developing countries deteriorate
drastically. Risk of massive
immigration into wealthy countries.
Widespread extinction of species but
less severe than under RCP8.5.
Increasingly severe societal problems
in many developing countries.
Immigration pressure into wealthy
countries is likely to increase.
Regional destructions of
ecosystems. Societal problems
exacerbate in some developing
countries. Immigration
pressure into wealthy countries
may increase.
4. Risks of irreversible
changes (examples)
Greenland and Antarctic Ice Sheet
losing mass. Rapid Arctic summer
sea-ice loss. Amazon forest and
boreal forest dieback
Greenland and Antarctic Ice Sheet
losing mass. Frequent Arctic summer
sea-ice loss.
Occasional Arctic summer
Sea-ice loss.
5. Impacts on Finnish forests
Spruce forests suffer seriously from
drought in southern and central
Finland. Forest fire danger increases
manifold. Increased snow damage
risk in northern and reduced in
southern parts of country. Seriously
increased biotic2 damage risk.
Largely reduced soil frost hampers
forest harvesting.
Spruce forests suffer from drought in
southern and central Finland. Pine and
birch benefit of change. Increased
forest fire danger and biotic damage
risk. Reduced soil frost hampers forest
harvesting.
Excluding spruce, forests
benefit from the higher CO2
concentration and longer
growing season. Slightly
increased forest fire danger
and biotic damage risk.
6. Annual mean temperature
changes in eastern-central
Finland relative to 1971-2000.
About 3.8°C by the 2050s and nearly
6°C by the 2080s. In late-21st
century, annual mean temperatures
similar to those in northern Germany
in recent past.
About 2.8°C by the 2050s and 3.6°C
by the 2080s. In late-21st century,
annual mean temperatures similar to
those in southernmost Sweden in
recent past.
About 2.3°C for both periods.
In late-21st century, mean
temperatures similar to those
of the southern coast of
Finland in recent past.
7. Changes in southern
Germany by the 2050s.
Annual mean warming 2.7°C
relative to 1971-2000. Severe
drought risk in summer.
Annual mean warming 2.0°C relative
to 1971-2000. Drought risk in summer.
Annual mean warming 1.6°C
relative to 1971-2000.
Occasional drought risk in
summer.
8. Changes in Uruguay by the
2050s.
Annual mean warming 1.8°C
relative to 1971-2000. Increasing
risk of floods in Southern
Hemisphere summer and autumn
Annual mean warming 1.3°C relative
to 1971-2000. Minor increase in flood
risk.
Annual mean warming 0.9°C
relative to 1971-2000.
9. Changes in eastern China by
the 2050s.
Annual mean warming 2.7°C
relative to 1971-2000. Increasing
occurrence of floods in Yangtze
river in summer.
Annual mean warming 2.0°C relative
to 1971-2000. Increasing occurrence
of floods in Yangtze river but not as
severe as under RCP8.5.
Annual mean warming 1.6°C
relative to 1971-2000. Some
increase in flood occurrence in
Yangtze river.
1
Estimates are derived from IPCC; much higher estimates have been given in other sources. In the
very recently published draft of IPCC Special Report Ocean and Cryosphere, the estimated rise under
RCP8.5 is about 20 cm higher but under RCP2.6 about the same as given here.
2
For example bark beetle and root rot
7
1. Introduction
1.1 Global climate change
Earth and its atmosphere have been existing for more than four milliard years, and the
climate of our planet has fluctuated all this time. Over the majority of the time, climate has
been rather warm, and no wide ice sheets have occurred. These long warm periods have
been interspersed by cool epochs that typically have lasted tens or hundreds millions years.
Cool climate epochs are characterized by widespread continental glaciation.
Global climate is fundamentally determined by the heat balance of the planet. The
principal source of energy is solar radiation. About 70 % of the incoming solar radiation
is absorbed by the Earth, and the remaining 30 % is reflected back into the space without
being converted into heat. The energy input from the absorption of solar radiation is
balanced by the emission of thermal infrared radiation into the space. In low latitudes
between about 40°S and 40°N, the radiation balance is positive with solar absorption
exceeding the emission of thermal radiation; in latitudes higher than 40° the situation is
vice versa. The surplus of heat is transported from the equatorial areas into polar latitudes
by ocean currents and atmospheric circulation. Local climate in the various regions of the
Earth is strongly influenced by the oceanic and atmospheric circulation. For example, in
northern Europe climate is quite mild compared with the conditions prevailing in these
latitudes on average. This is due to the warm oceanic currents that transport heat into the
north Atlantic area and, in particular, the prevailing westerly winds that advect the heat
into north-western Eurasia.
Thermal infrared radiation emitted by the surface of the Earth does not penetrate into
the space directly, but a majority of the radiation is absorbed and then re-emitted by the
atmosphere. This property of the atmosphere is termed the greenhouse effect. The principal
greenhouse gases are water vapour and carbon dioxide, with methane, ozone, nitrous oxide
and several other gases having a smaller but still nonnegligible importance. Without the
greenhouse effect, the average surface temperature of the Earth would be about -18°C;
thanks to these gases, however, the present global mean temperature is about +14°C, i.e.,
more than 30°C higher than in the absence of the greenhouse effect.
Natural climate changes occurred in the climate history of the Earth have been caused
by several factors: long-term changes in solar radiance, changes in the composition of the
atmosphere, continental drift, volcanic eruptions, etc. In the early times of the existence of
the Earth, solar radiation was substantially weaker than presently. The resulting deficit in
the heath input was compensated by a strong greenhouse effect, caused by the very high
concentration of carbon dioxide in the atmosphere. For the relatively cool climate that has
prevailed during the past few million years, two (potentially complementary) main
explanations have been suggested. First, the Antarctic continent, previously in connection
with the other continents of the Southern Hemisphere, was isolated due to continental drift
about 50 million years ago. This allowed a cold ocean current to begin to circulate around
the continent. The transport of heat into southern high latitudes declined drastically, which
was followed by a gradual glaciation of the Antarctica. The high albedo (the ability to
reflect solar radiation) of the ice-covered surface increased the portion of solar radiation
that was reflected into the space, thus reducing the production of heat by the absorption of
solar radiation. Secondly, the development of the Himalayan mountain range brought
certain minerals, especially calcium silicate, to the surface in contact with the air. Such
minerals can react with carbon dioxide, thus reducing the concentration of that gas in the
atmosphere. These changes took place in time scales of millions of years.
During the latest million years, climatic fluctuations have been strong. Ice ages and
milder interglacial periods have followed one another in time scales of 10 000 to 100 000
8
years. The glacial-interglacial variations are primarily induced by periodic changes in the
orbit and axis of rotation of the Earth. Furthermore, the variations are amplified by
contemporaneous changes in the atmospheric carbon dioxide concentration.
For a reader interested in natural climate variations more deeply, exoteric literature is
available abundantly. In particular, the well-written book of Eronen (1991) can be
recommended.
During the past few centuries and, in particular, during the recent decades, the main
drivers of global climate change have been the human-induced emissions of greenhouse
gases and aerosol precursors. The concentration of carbon dioxide has increased from the
pre-industrial level of 280 ppm (parts per million in volume) to 410 ppm in 2019.
Simultaneously, the concentration of methane has more than doubled. On the other hand,
small aerosol particles, especially sulphates, originating from anthropogenic emissions act
to increase the reflection of solar radiation into the space and thus have a cooling effect.
Until presently, increased aerosol forcing has partially compensated global warming
caused by the strengthening of the greenhouse phenomenon. In the long term, however,
increasing greenhouse gas concentrations will dominate owing to the long atmospheric
lifetime of the greenhouse gas emissions. While aerosol particles are washed out from the
atmosphere within a few weeks, present emissions of carbon dioxide influence the
atmospheric composition for several millenia. The major source of carbon dioxide is the
combustion of fossil fuels, but deforestation and other changes in the land use have some
importance as well.
1.2 Modelling of future changes
Compared to the pre-industrial era, global mean temperature has risen by about one degree.
Future climatic changes can been assessed by examining simulations performed with
climate models. A climate model is a simulation tool that is founded on the physical laws
governing the different components of the climate system (in particular, atmosphere,
oceans, soil, vegetation and cryosphere) and their mutual interactions. The operational
principle of climate models is described in more detail in the textbook chapter of
Ruosteenoja (2011), for instance. Because of limitations in computational capacity, the
models inevitably describe the climate system in an approximative way.
Presently, several tens of different climate models have been developed. Because of
the different approximations employed in the models, simulated future changes diverge
across the models. In order to obtain the most realistic picture of the anticipated future
changes and their uncertainties, it is therefore necessary to study a multitude of models
rather than relying on a single or a few ones. In the calculations of this report, we have
utilized the output of 28 global climate models (GCMs) participating in Phase 5 of the
Coupled Model Intercomparison Project (Table A1 in the Appendix of this report).
Future emissions and the resulting atmospheric concentrations of the various
greenhouse gases cannot be known in advance. Therefore, multiple alternative greenhouse
gas scenarios have been developed, founded on different estimates of the future evolution
of world population, energy consumption, land use, energy production technologies and
other factors that determine the future emissions. In the present report, we examine three
Representative Concentration Pathway (RCP) greenhouse gas scenarios, RCP2.6
representing very small, RCP4.5 moderate and RCP8.5 very large emissions. The number
after the acronym refers to radiative forcing, i.e., the imbalance between the solar radiation
absorbed and the thermal infrared radiation emitted by the Earth. For example, if the
RCP4.5 scenario were realized, the positive (=warming) globally averaged radiative
forcing at the end of the 21st century would be 4.5 W/m2. The philosophy behind the RCP
forcing scenarios is discussed in detail in van Vuuren et al. (2011).
9
The evolution of the emissions and atmospheric concentrations of the most important
greenhouse gas, carbon dioxide, under the three RCP scenarios is shown in Figure 1.1.
Under the RCP8.5 scenario, emissions would continue to increase throughout the 21st
century, ultimately three-folding compared to the level that prevailed in 2000, and the
concentration of CO2 would approach 1000 ppm by 2100. According to the other two
scenarios, global emissions start to decline during this century. If the RCP4.5 scenario is
realized, the CO2 concentration stabilizes close to 540 ppm, a level about double that in
the pre-industrial era. Under the most environmental-friendly RCP2.6 scenario, the
concentrations start to diminish slowly after mid-century. The RCP2.6 scenario is likely
to meet the targets of the Paris Agreement. In addition to carbon dioxide, the RCP
scenarios describe future emissions and atmospheric concentrations of several other
greenhouse gases as well as the precursors of aerosol particles.
Figure 1.1: Temporal evolution of the global emissions (gigatonnes of carbon per year;
left panel) and atmospheric concentrations (parts per million in volume; right panel) of
carbon dioxide in 2000–2100 according to three RCP scenarios; see the legend.
In this report, projections of multiple climate quantities (temperature, precipitation,
etc.) for the target areas (Finland, southern Germany, Uruguay and eastern China) are first
given in multi-model mean changes. The multi-model mean can be regarded as the most
probable or “best estimate” projection for the future. Projected changes are expressed
relative to the baseline-period 1971–2000 mean. Time series of the 30-year running mean
change, covering the time span from the 2000s to the 2080s, are depicted separately for all
three RCP scenarios. The spatial distributions of the change are only shown for the period
2040–2069 under RCP4.5.
In addition to the multi-model mean projections, we present 5 to 95 % uncertainty
intervals of the change for the key variables. According to the model data analyzed here,
there is a 90 % probability that the projected change falls within this interval.
Correspondingly, probabilities for the change being below the lower limit or above the
upper limit of the interval are both 5 %. Of course, as simulations performed with new
model generations become available in the future, the climate change estimates will have
to be modified.
To further elucidate changes in precipitation climate, we examine model-based
changes in two precipitation indices. Heavy precipitation events are considered by
studying the maximum precipitation falling within a single day. For drought stress, we
analyze the lengths of the periods of consecutive dry days, with precipitation < 1
millimeter/day in every day within the period. The exact definition of these indices is given
in Lehtonen and Jylhä (2019). Unlike projections for the other quantities, owing to data
10
availability the projections for the precipitation indices are given for the period 2071–2100
rather than 2040–2069. Since climate change will most likely be more ample for the more
distant than the earlier period, changes in the indices are thus not entirely comparable with
the projections given for the other climate quantities. In many cases, changes in the
precipitation indices vary materially across the models. Therefore, we do not fit any
statistical distribution in the data but visualize the inter-model scatter of the change in the
indices by quantile plots.
For Finland, we likewise examine changes in the thermal growing season (onset,
termination and degree days), changes in strong wind speeds and soil moisture (analyzed
soil moisture data is not available outside of Europe), and projections for the discharge in
the Kymmene river. The thermal growing season, consisting of those days when the daily
mean temperature is above +5°C, is a relevant measure of growing conditions only in areas
of a cool climate.
Additional technical information about the model data and the analysis methods is
presented in the Appendix of this report.
1.3 Projected changes in global climate
The simulated evolution of global-mean warming under the three RCP scenarios is shown
in Figure 1.2. Recall that in this figure (as well as elsewhere in this report) the changes are
expressed relative to the temporal mean of the period 1971–2000, before which global
mean temperature had already increased by about 0.5°C. Under RCP8.5, global warming
would continue throughout the century without any sign of deceleration, the global mean
temperature increasing by about 4°C in 100 years. If the RCP4.5 scenario is realized, the
corresponding increase would be about 2°C and under RCP2.6, slightly above 1°C.
Considering the global warming already taken place before the baseline period 1971–2000,
the latter alternative would correspond to a warming slightly less than 2°C compared to
the pre-industrial level.
Figure 1.2: Projected changes in global mean annual temperature (in °C) for the period
2000–2085 (relative to 1971–2000) under the RCP2.6, RCP4.5 and RCP8.5 scenarios (see
the legend); a mean of the simulations performed with 28 global climate models. Prior to
1971–2000, global mean temperature had increased about 0.5°C compared to the pre-
industrial level.
11
The projections depicted in Figure 1.2 represent a mean over the 28 climate models
examined (Table A1 in the Appendix). In practice, the simulated rate of warming diverges
across the models; this topic will be discussed later in this report when considering regional
climate projections. Anyhow, it is evident that, regardless of the future reductions of
emissions, warming will continue during the next few decennia, and mankind must
endeavour to adjust to the anticipated change. On the other hand, the vigour of climate
policy decisively affects the degree of warming that the next generations of mankind have
to experience during the second half of this century and later on.
In particular, in the long term the RCP8.5 scenario would lead to catastrophically large
changes in climate, environment and society, especially in many under-developed
countries. For example, in wide areas of the world, agricultural production would suffer
severely from overly high temperatures and shortage of water. Worsening living
conditions in the third world would induce a massive migration pressure into wealthier
countries. Migration and actions to prevent it would hazard societal stability in
industrialized countries as well. Also, the sea level rise is substantially more rapid under
RCP8.5 than under the other RCP scenarios, particularly after 2100. Fortunately,
considering the present advances in climate policy, the RCP8.5 scenario seems to be quite
an unlikely alternative for long-term global evolution. On the other hand, the RCP2.6
scenario would require drastic reductions in global emissions as early as in the 2020s
(Figure 1.1), which does not sound very easy at present. Therefore, we have deemed
RCP4.5 the most realistic alternative for the evolution by the mid-century, and the
projections will mainly be presented for this scenario.
The geographical distributions of projected changes in temperature and precipitation
for the mid-21st century are shown in Figures 1.3 and 1.4. Very intense warming is
simulated for northern polar areas in winter; this is related to the partial disappearance of
wintertime ice cover. Besides, land areas with scant or decreasing precipitation warm
rather rapidly, especially in summer. In these areas, latent heat consumed for evaporation
of water does not restrict warming materially. Conversely, in ocean areas south of Iceland,
warming is modest since weakening of the warm ocean current partially cancels the
influence of global warming.
Precipitation is projected to increase in equatorial areas, mainly over oceans, and in
high latitudes especially in winter. Decreasing precipitation totals are simulated for low-
latitude areas at both sides of the equator. All the projected changes in Figures 1.3 and 1.4
correspond to multi-model means, and consequently the projections are subject to inter-
model differences. This topic will be discussed below when presenting projections for
focal areas.
The geographical pattern of change in temperature and precipitation is qualitatively
very similar for all three RCP scenarios and also for the less and more distant future periods
(Figures 5–6 and S4–S7 of Ruosteenoja et al., 2016a), the amplitude of the response being
approximatively proportional to the global mean temperature increase (Figure 1.2). One
can thereby approximate the local warming or precipitation change for any scenario and
time horizon by multiplying the change shown in the map 1.3 or 1.4 by the ratio of global
temperature increase for that period and scenario to the global warming corresponding to
the period 2040–2069 and the RCP4.5 scenario.
An illustrative example: according to Figure 1.3, winter temperatures in Northern
Ostrobotnia in Finland would rise by about 4.0°C by the period 2040–2069 under RCP4.5.
If one is interested to assess the corresponding temperature increase for the 2070s under
RCP2.6, for instance, one should first find the required global temperature increases in
Figure 1.2: 1.27°C for the 2070s under RCP2.6 and 1.61°C for the 2050s under RCP4.5.
The ratio of the former to the latter is 1.27/1.61 = 0.79. By multiplying the above-
12
mentioned warming extracted from the map by this ratio, one obtains an approximation
for the inquired local temperature increase: DT ≈ 0.79 x 4.0°C ≈ 3.2°C.
Figure 1.2 indicates that global mean warming between the periods 1971–2000 and
2040–2069 under RCP4.5 is likely to be about 1.5°C, corresponding to a temperature
increase of close to 2°C when compared with the pre-industrial climate. Accordingly, if
the objective to restrain the global mean temperature increase to 1.5°C relative to the pre-
industrial level were achieved, the geographical pattern of the change would be similar to
that in Figure 1.3 or 1.4 but the amplitude would be somewhat weaker (by a factor of ≈
0.75). Even so, bearing in mind that the global temperature has already increased by about
1°C, avoiding the exceedance of the 1.5°C threshold is an extremely challenging target.
Figure 1.3: Projected change in mean temperature (in °C) in December-February (upper
panel) and June-August (lower panel) from the period 1971–2000 to 2040–2069 under the
RCP4.5 scenario; a mean of the simulations performed with 28 climate models.
13
Figure 1.4: Projected multi-model mean change in precipitation (in %) in December-
February (upper panel) and June-August (lower panel) from the period 1971–2000 to
2040–2069 under RCP4.5.
14
1.4 Climate change and forests
Forest growth conditions are influenced by several climatological factors, e.g, air
temperature, solar radiation and rainfall. In addition, the atmospheric concentration of CO2
has a major impact on forest productivity and forest dynamics. Increasing CO2 behaves as
a fertilizer accelerating growth. Forests can remove large amounts of carbon from the
atmosphere, but as well, large amounts of carbon can be released into atmosphere, e.g.,
during wide forest fires. There are also feedback mechanisms i.e., vegetation and forests
impact meteorological phenomena like the absorption and reflection of solar radiation.
Hence, change of forests to shrubs or grassland, or the opposite, may have substantial
influence on earth’s energy balance. Forests also act as a source of cloud-forming aerosols.
1.5 Observed changes in forest resources
Satellite measurements indicate greening of the world during the past decades. The
greening based on the leaf area index (LAI)
3
is very prominent in India and China (Chen
et al., 2019). In China, 42% of the greening is explained by forests and 32% by croplands
whereas in in India, forests have a smaller contribution. Chen et al. (2019) estimated that
for the EU, 51.4% of vegetated land showed greening. According to Munier et al. (2018),
satellite measurement indicate that LAI of the European forests has been increasing during
the period 1995-2015 (Figure 1.5). The increase is most prominent in northern Europe.
Both Munier et al (2018) and Chen et al. (2019) reported small changes of LAI for
Uruguay. Elsewhere in South-America, like in southern Brazil, an increase of LAI has
been detected whereas in the northern Brazil LAI has decreased.
The forested area has been declining for example in Brazil, Democratic Republic of
Congo and Indonesia, and increasing e.g. in China (FAO, 2016). According to the FAO
(2016) estimate, the global rate of forest loss will probably continue to decelerate in
coming years and gradually level out. The fulfilment of FAO’s prediction depends largely
on political decisions taken by the major contributors like Brazil.
Forests cover about 33 % of Europe and about 73 % of the Finnish land area. The
forest area in Europe (excluding the Russian Federation) is about 215 million hectares, and
about 160 million hectares are available for wood supply. During the period 1990–2015,
the forest area in Europe expanded by 17.5 million ha (Figure 1.6). Finland has the most
extensive forest cover in Europe as 22.2 million hectares are classified as forest area.
3
Leaf area index is defined as the projected area of leaves over a unit area of land surface (m2
m−2)
15
Figure 1.5: The trend of Leaf Area Index for broadleaf, coniferous and evergreen forests
over the 1999–2015 period (Munier et al., 2018). Only areas with significant trend (p-
value < 0.01) are marked by colour.
Figure 1.6: Annual increase in forest area by region in Europe, 1990–2015 and 2005–
2015 (1,000 ha per year) (Forest Europe, 2015).
In Europe, about 66 % of the annual forest increment is harvested, while in Finland
that proportion is about 70 % (LUKE, 2018; Forest Europe, 2015). The growing stock
available for wood supply in Europe (excluding Russian Federation) is more than 29
milliard m3 and the annual increase about 400 million m3 (Forest Europe, 2015) (Figure
16
1.7). In Finland, the annual increment of growing stock has been clearly larger than drain
since the 1970s. In 2017, the estimated increment was 107 million m3 whereas the drain
was 87 million m3 (Figure 1.8).
Figure 1.7: Annual increase in growing stock (million m3 per year) by region, 1990–2015
(Forest Europe, 2015).
The distribution of tree species is changing significantly but slowly. Trees migrate
relatively slowly, primarily into newly suitable habitats, e.g., towards the north and higher
altitudes. Simultaneously, former suitable habitats may become unsuitable, e.g., due to
changed precipitation patterns leading to droughts. The climate change affects forest
ecosystems and individual species. The assessment of the total impacts is not always
straightforward as forests are under management, and the effects of forest management are
difficult to separate from the influence of climate change. Nevertheless, over the past
decades, forest biomass has increased at an accelerating rate (EEA, 2016a; EEA, 2016b).
Figure 1.8: The annual growth and drain of trees (felling and death due to natural
causes) in Finland (Lier et al., 2018).
2. Climate change and its impacts in Europe
2.1 Overview of projected climate change in Europe
When considering the multi-model mean and the mid-century climate under RCP4.5,
climate models project the largest annual mean temperature increase, more than 3°C, for
17
the north-eastern parts of Europe (Figure 2.1(a)). In western Europe and on the coasts of
the Mediterranean Sea, warming would correspondingly be about 2°C or even less. Annual
precipitation is projected to increase in northern and central Europe and to decrease in the
south, the largest local changes being about ±10 % (Figure 2.1(b)).
Figure 2.1: Projected changes in annual mean (a) temperature (in °C), (b) precipitation
(in %), (c) incident solar radiation (in %) and (d) relative humidity (in percentage points)
in Europe from the period 1971–2000 to 2040–2069 under RCP4.5. The positions of the
grid point 62.5°N, 27.5°E and 50°N, 10°E, examined in the more detailed analyses, are
marked by a dot and a triangle in panel (a).
The annual aggregate amount of solar radiation received at the surface appears to
increase in the majority of the continent, in central Europe by up to about 6 % (Figure
2.1(c)). Changes in relative humidity (RH) are fairly modest. In the bulk of Europe, RH
decreases by 1–3 percentage point (Figure 2.1(d)); most strongly in southern European
inland areas. In conjunction with changes in the mean climate, temporal variations of
temperature will attenuate and fluctuations in precipitation amplify, mainly in northern and
north-eastern Europe (Figure 2.2). As will be shown below, in reality all these changes are
subject to substantial seasonal variability and inter-model differences.
18
Figure 2.2: Projected changes in the standard deviation of temporal variations in (a)
temperature and (b) precipitation (both in %) in Europe from the period 1971–2000 to
2040–2069 under RCP4.5.
In contrast to temperature, precipitation and many other climate quantities, changes in
the wind speeds do not appear to be particularly significant. By the mid-century, time mean
wind speeds generally alter less than ±2 % (Figure 2.3). The somewhat larger changes
apparent in southern European mountainous areas in summer are evidently spurious,
induced by the subterranean extrapolation of surface pressure into the standard sea level
(Ruosteenoja et al., 2019).
19
Figure 2.3: Projected seasonal changes (in percent) in the time-mean geostrophic wind
speed from 1971–2000 to 2040–2069 in Europe under RCP4.5 in (a) December-February,
(b) March-May, (c) June-August and (d) September-November. The contour interval is 2
%.
2.1.1 Finland
Multi-model mean projections for the temporal evolution of four key climate quantities in
Finland are shown in Figure 2.4. The grid point considered, 62.5°N, 27.5°E, is located in
eastern central Finland (the position of the point is marked in Figure 2.1(a)) and thus
represents the area of the most intensive forestry in Finland.
Annual mean temperature in Finland (Figure 2.4(a)) increases by 1.6–1.9 times as
rapidly as the global mean temperature (see Figure 1.2). In a qualitative sense, however,
the temporal evolution of the change is very similar. In the early 21st century, temperature
is projected to rise by 0.4–0.5°C per a decade. Near the mid-century and, in particular, the
second half of the present century, the projected temperature increases corresponding to
the three RCP scenarios tend to gradually diverge. According to RCP8.5, warming
continues at nearly a constant rate. Under RCP4.5, warming likewise continues throughout
the 21th century but is less rapid in the second than in the first half of the century. If the
extremely low-emission RCP2.6 scenario were realized, warming would cease after the
2050s. Under the RCP8.5 scenario, annual mean temperature in Finland would increase
by nearly 6°C in hundred years. Under RCP4.5, the corresponding warming would be
20
about 3.6°C and under RCP2.6 only 2.2°C. This demonstrates the potential of vigorous
climate policy to reduce warming in the long run.
Figure 2.4: Projected annual mean changes relative to 1971–2000 in (a) mean
temperature (in °C), (b) precipitation (in %), (c) incident solar radiation (in %) and (d)
relative humidity (in percentage points) at 62.5°N, 27.5°E (central Finland; the position
is marked in Figure 2.1(a)). The time series are shown separately for the RCP2.6, RCP4.5
and RCP8.5 scenarios (see the legend).
When considering all three RCP scenarios but not inter-model scatter, annual
precipitation total is projected to increase by 7–18 % by the late-21st century (Figure
2.4(b)). In this case as well, the business-as-usual RCP8.5 scenario yields the largest
change. In incident solar radiation and relative humidity, changes are fairly small on the
annual-mean level (Figures 2.4(c–d)). The quite rapid increase in the simulated solar solar
radiation prior to the 2020s is partially explained by the decreasing emissions of sulphur
dioxide and other pollutants acting as aerosol precursors in Europe after the 1980s.
21
Figure 2.5: Projected monthly changes in (a) the mean temperature (in °C), (b)
precipitation (in %), (c) incident solar radiation (in %) and (d) relative humidity (in
percentage points) at 62.5°N, 27.5°E (central Finland; the position is marked in Figure
2.1(a)) under RCP4.5 for the period 2040–2069, relative to 1971–2000. Corresponding
changes in the temporal standard deviation of daily-mean temperature (in °C) and
precipitation (in %) are shown in panels (e) and (f). The multi-model mean projections for
individual calendar months (J = January,..., D = December) are denoted by open circles.
Shading shows the 90 % uncertainty intervals for the projection.
The seasonal course of projected changes in multiple climate indicators, along with
an estimate of the inter-model scatter in the change, is depicted in Figure 2.5. Increases in
temperature are simulated by all the models, the warming being larger in winter than in
summer. According to the multi-model mean, temperature would increase (by the mid-
22
century) by about 4°C in mid-winter and slightly more than 2°C in summer and early
autumn. Even so, inter-model differences in the simulated temperature change are
substantial. For winter temperatures, the 90 % probability interval of the increase ranges
from about 1°C to 5–6°C. The corresponding uncertainty interval for the projected
summertime warming is from 1 to 4°C. Note, however, that according to the normality
assumption it is more likely that future warming will fall near the multi-model mean
estimate than close to the edges of the interval.
For precipitation, the most likely projection is an increase of 15 % in winter and of ∼5
% in summer (Figure 2.5(b)). In winter, inter-model agreement on the increasing future
precipitation is high but the uncertainty interval is nevertheless wide, ranging from nearly
0 to 30 %. In other seasons, particularly in summer, there is a nonnegligible likelihood that
precipitation would even decrease; in July, for instance, the uncertainly interval ranges
from -15 % to +25 %. Solar radiation is likely to diminish from November to March and
increase in summer and early autumn, even though the inter-model scatter is large (Figure
2.5(c)). The signal-to-noise ratio is likewise low for relative humidity projections; the most
likely alternative is that RH would decrease by about 2 % in spring and summer and remain
virtually unchanged in autumn and winter (Figure 2.5(d)).
Climate change affects moisture conditions in the soil as well. Figure 2.6 shows the
seasonal cycle of the near-surface soil moisture change for central Finland. A very
significant drying is apparent in spring; this is related to an earlier melting of snow and
soil frost in the future. In the summer months, the decline of soil moisture is fairly modest.
The response in soil moisture is a residual of the influences of increasing precipitation and
the strengthening evaporation induced by higher temperatures.
Figure 2.6: Multi-model mean monthly responses (J = January, F = February, ...) in near-
surface (the uppermost 10 cm layer) soil moisture (in percentage points) for three future
time spans (2010–2039, 2040–2069 and 2070–2099, relative to 1971–2000; see the
legend) at 62.5°N, 27.5°E under the RCP4.5 scenario.
The temporal fluctuations of temperature will attenuate in winter (Figure 2.5(e)). This
indicates that the most extreme low wintertime temperatures will become much warmer.
In the mean temperature and particularly in the mildest temperatures, warming is not as
large. Accordingly, in the future the Finnish wintertime temperature climate is going to be
less variable than in the recent past. In summer, no drastic changes in the variability are to
23
be anticipated; relatively cold and warm temperatures become warmer broadly at an equal
rate.
In precipitation, temporal variations are likely to intensify (Figure 2.5(f)). Considering
the multi-model mean, in winter the increase in the variability is of the similar magnitude
as the increase in the time-mean precipitation, in summer somewhat larger. Although the
90 % probability interval of the change intersects the zero line, an increase in the variability
is far more likely than a decrease.
Simultaneous increases in the mean precipitation and its variability imply that heavy
precipitation events will intensify. This tendency is even more distinctly apparent (in part,
due to the more distant projection period) in Figure 2.7, which shows the long-term
averaged change in the 24-hour precipitation total of the wettest day of the season.
According to the multi-model median, the maximum 1-day precipitation would increase
by 12–17 %, depending on the season. Inter-model scatter is certainly fairly large in this
quantity as well, but virtually all the models examined simulate at least some increase in
extreme precipitation. In summer, for instance, the interquartile interval of the increase in
this precipitation index is from 11 to 23 %, in winter from 10 to 15 %. Some individual
models project even far more intense increases.
Figure 2.7: Model-derived seasonal changes (in percent) in the maximum one-day
precipitation R1d (left panel) and the maximum number of consecutive dry days CDD (with
precipitation < 1 mm/day; right panel) at 62.5°N, 27.5°E (central Finland; the position is
marked in Figure 2.1(a)) under RCP4.5 for the period 2071–2100, relative to 1971–2000.
The coloured bars show the 25th to 75th percentile intervals of the change derived from
the responses of 21 models. The corresponding 10th to 90th percentile intervals are
depicted by whiskers and the minimum and maximum responses among the models by
black dots. The black lines within the bars stand for the multi-model medians. The
probability distributions are given separately for four seasons: December-February
(DJF), March-May (MAM), June-August (JJA) and September-November (SON).
The wetter future climate in Finland also manifests itself in changes in the number of
consecutive dry days (Figure 2.7). A majority of the models simulate a shortening of dry
periods in all seasons except summer, even though in this index inter-model agreement is
not as high as in the above-discussed heavy-precipitation index. In summer, the number of
models simulating shorter and longer dry periods in the future is virtually equal.
24
The persistently continuing warming results in warmer and longer growing seasons.
The thermal growing season (i.e., the period with daily mean temperatures above 5°C; see
the appendix) appears to lengthen by 10–15 days both in spring and autumn (from 1971–
2000 to 2040–2069 under RCP4.5; Figure 2.8).
Figure 2.8: Projected multimodel-mean changes in the onset (left) and termination dates
(right) of the thermal growing season for the period 2040–2069 under RCP4.5, relative to
1971–2000. Base temperature is 5°C (re-drawn from Ruosteenoja et al., 2016b).
Figure 2.9: Baseline values (years 1971–2000; at the top) and future projections (periods
2040–2069 and 2070–2099; bottom) for the temperature sum of the growing season under
RCP4.5. Base temperature is 5°C (re-drawn from Ruosteenoja et al., 2016b).
25
Concurrently, the temperature sum of the growing season increases by several
hundreds of degree days. In the mid-century, the average growing degree day sum in
southern Finland would be approximately as large as it was in Poland or East Germany in
the late-20th century (Figure 2.9). By the end of the 21st century, temperature sums would
further increase to some extent.
As it was shown in Figure 2.3, temporally averaged wind speeds are not expected to
change substantially. According to Figure 2.10, the same conclusion holds for strong winds
(note that in this case the projection is given for the RCP8.5 scenario). Strong wind speeds,
corresponding to the 99th percentile of the frequency distribution, may increase by 0–2 %
in summer and autumn, and in the other two seasons the projected change is even more
negligible. Moreover, the proportion of westerly winds is anticipated to increase slightly
at the cost of easterly winds.
Figure 2.10: Projected changes in strong geostrophic wind speeds (in percent) in Northern
Europe from the period 1961–2005 to 2040–2069 under RCP8.5: (a) winter, (b) spring,
(c) summer and (d) autumn. The changes are shown for the geostrophic wind speed
corresponding to the 99th percentile of the frequency distribution.
Climate change also influences hydrological conditions. As an example of this, the
simulated monthly discharge of the Kymmene river in south-eastern Finland is depicted in
Figure 2.11. In the climate that prevailed in the turn of the millennium (note that the
baseline period is here 1981–2010 rather than 1971–2000), the time-mean discharge was
nearly uniform throughout the year.
In the future, winters become milder and wetter, and consequently an increasing
proportion of the larger precipitation is received in the form of rain rather than snow. This
makes the wintertime discharges to increase. Owing to the shallower winter snow cover,
springtime floods gradually weaken and ultimately disappear. In summer, precipitation
increases slightly but in the warmer climate evaporation of water intensifies as well; this
reduces water resources, and summertime discharges decrease accordingly. Considering
the period 2040–2069, mid-winter discharges are projected to increase by more than 30 %
and late-summer discharges to decrease by about 20 %. For further information about
26
future hydrological conditions in selected Finnish watersheds, including extreme
conditions and the impact of inter-model differences, the reader can consult the report of
Veijalainen et al. (2018).
Figure 2.11: (a) Simulated temporally averaged monthly (J = January,..., D = December)
discharge (in m³s-1) at Anjalankoski on the Kymmene river (60.7°N, 26.8°E) under RCP4.5
for the periods 1981–2010, 2020–2049, 2040–2069 and 2070–2099 (see the legend). (b)
Projected changes for the three future periods. Re-drawn using the data given in Table L2
of Veijalainen et al. (2018).
In Finland, the projected climatic change hereby turns out to be strongest in winter:
precipitation increases substantially, solar radiation declines to some extent and the
temperature increase is stronger than in the other seasons. Moreover, temporal variations
in temperature tend to attenuate. In summer, the most important change is the temperature
increase, albeit weaker than in winter. Even so, warming leads to a significant lengthening
of the thermal growing season and an increase in the effective temperature sum.
2.1.2 Southern Germany
Compared to Finland, projected warming in southern Germany is distinctly weaker, only
slightly exceeding the corresponding global average (Figure 2.12(a)). Moreover, the
temperature increase is distributed quite evenly in the various seasons, with a peak in
summer rather than in winter (Figure 2.13(a)). Again, the modelling uncertainty in
warming is fairly large, the interval ranging from less than 1°C to 3–4.5°C. In annual
precipitation, there is a minor increase, which is the residual of an increasing trend in
winter and spring and a decreasing trend in summer and early autumn (Figures 2.12(b) and
2.13(b)). Nonetheless, inter-model agreement on the sign of the future precipitation
response is low in all seasons.
27
Figure 2.12: Projected annual mean changes relative to 1971–2000 in (a) mean
temperature (in °C), (b) precipitation (in %), (c) incident solar radiation (in %) and (d)
relative humidity (in percentage points) at 50°N, 10°E (southern Germany; the position is
marked in Figure 2.1(a)). The time series are shown separately for the RCP2.6, RCP4.5
and RCP8.5 scenarios (see the legend).
Incident solar radiation is anticipated to increase and relative humidity to be reduced
in all seasons apart from winter (Figures 2.13(c)-(d)). This is also reflected in the annual
mean changes (Figures 2.12(c)-(d)). In late summer, the increase in solar radiation is quite
material, about 10 % according to the multi-model mean, with an uncertainty interval from
0 to 22 %. Note that southern Germany belongs to the area where the increase in the annual
sum of incident radiation is largest in Europe (Figure 2.1(c)).
28
Figure 2.13: Projected monthly changes in (a) the mean temperature (in °C), (b)
precipitation (in %), (c) incident solar radiation (in %) and (d) relative humidity (in
percentage points) at 50°N, 10°E (southern Germany; the position is marked in Figure
2.1(a)) under RCP4.5 for the period 2040–2069, relative to 1971–2000. Corresponding
changes in the temporal standard deviation of daily-mean temperature (in °C) and
precipitation (in %) are shown in panels (e) and (f). The multi-model mean projections for
individual calendar months (J = January,..., D = December) are denoted by open circles.
Shading shows the 90 % uncertainty intervals for the projection.
In central Europe, both the decreasing precipitation and enhancing evaporation,
induced by the higher temperatures and increasing solar radiation, act to reduce soil
moisture in summer. This leads to a substantial drying in the soil, particularly in late
summer (Figure 2.14). As a consequence, such dry-soil episodes that occur once per
decade in the late-20th century climate will be experienced nearly every fourth year during
the period 2040–2069 (Ruosteenoja et al., 2018).
29
Figure 2.14: Multi-model mean monthly responses (J = January, F = February, ...) in
near-surface (the uppermost 10 cm layer) soil moisture (in percentage points) for three
future time spans (2010–2039, 2040–2069 and 2070–2099, relative to 1971–2000; see the
legend) at 50°N, 10°E under the RCP4.5 scenario.
The temporal variability of temperature is projected to change fairly little, and the
inter-model agreement on the sign of change is low (Figure 2.13(e)). Even so, it is likely
that in winter the very coldest temperatures would warm somewhat more than the mean
and maximum temperatures.
Figure 2.15: Model-derived seasonal changes (in percent) in the maximum one-day
precipitation R1d (left panel) and the maximum number of consecutive dry days CDD (with
precipitation < 1 mm/day; right panel) at 50°N, 10°E (southern Germany; the position is
marked in Figure 2.1(a)) under RCP4.5 for the period 2071–2100, relative to 1971–2000.
The coloured bars show the 25th to 75th percentile intervals of the change derived from
the responses of 21 models. The corresponding 10th to 90th percentile intervals are
depicted by whiskers and the minimum and maximum responses among the models by
black dots. The black lines within the bars stand for the multi-model medians. The
probability distributions are given separately for four seasons: December-February
(DJF), March-May (MAM), June-August (JJA) and September-November (SON).
According to the multi-model mean estimate, temporal variability in precipitation
would increase throughout the year, slightly so even in summer and early autumn when
the mean precipitation is projected to decrease. Consequently, precipitation conditions will
30
become more variable in summer and autumn; both the maximum 1-day precipitation and
the length of dry periods increase in a majority of models (Figure 2.15). In the winter and
spring, the largest 1-day precipitation totals likewise increase, as a matter of fact more
pronouncedly than in summer, but most likely there is fairly little change in the lengths of
dry periods.
Accordingly, in Germany climatic changes with negative impacts tend to accumulate
for summer. Declining summer precipitation, in conjunction with rising temperatures,
increasing solar radiation and lower relative humidities, lead to an exacerbating drought
risk. This is explicitly visible in the decrease of soil moisture and an increase in the length
of dry periods.
2.2 Climate change impacts in Europe with focus on forest sector
In Europe climate change is characterized by increasing precipitation in the northern parts
of the continent and decreasing precipitation in the southern parts. During winter season
warming is more pronounced at high latitudes than southern Europe whereas during the
warm season also the Mediterranean region is expected to experience substantially higher
temperatures (Figure 1.3). Dry regions will become drier particularly in summer (Figures
1.4 and 2.14). Sea level will continue to rise, and extreme events, like heat waves, heavy
precipitation and drought, will occur more frequently and be more intense in the future.
The estimated impacts of climate change for different regions as compiled by the
European Environment Agency (EEA) (2017) are shown in Figure 2.16. The estimated
impacts can mainly be regarded as negative; however, in the Boreal region there are some
positive impacts like increasing potential for forest growth and increase in crop yields.
When we look the impacts of climate change on forestry, both positive and negative
impacts on forest structure, growth, composition, productivity and functioning are to be
expected, depending on the type and geographical location of forest (EEA, 2016a). As a
whole, in Europe it is likely that climate change will have a positive effect on wood
production and wood supply due to warmer temperatures, longer growing season and
increased CO2 concentration in the atmosphere. Tree growth and productivity are predicted
to increase at high latitudes and altitudes. In other regions, changes may be positive in the
beginning but negative in the mid and long term. For example, it is likely that
Mediterranean regions will experience higher rates of tree mortality and forest fires when
temperatures and the frequency of droughts increase (e.g., Kellomäki et al., 2008; Reyer
et al., 2014). The possible increase of disturbances caused by weather phenomena like
forest fires, wind damages and forest pests may cancel at least to some extent the climate
change-induced productivity increase (Reyer et al., 2017a). The overall effect of climate
change on forest resources requires still new and multidisciplinary research. Especially, a
more thorough inclusion of weather disturbances into impact assessments requires
substantial amount of research.
31
Figure 2.16: Summary of the impacts of estimated climate change at different regions in
Europe (edited from EEA, 2017).
In Finland, climate change is predicted to increase forest growth especially in the
northern parts of the country (Kellomäki et al, 2018). However, for Norway spruce in
southern and central Finland, the shortage of water may have severe negative impacts on
growth. To summarize, the differences between tree species are as follows: birch is
benefitting most but spruce is suffering most seriously from the projected change, and pine
is mainly benefiting but not as much as birch.
According to the energy and climate strategy, the use of Finnish forests could be up
to 80 million m3 year-1. Forests are an important carbon sink, and large amounts of carbon
is stored into wooden biomass and understory. When aiming at limiting global warming
to below 2°C, the current sinks of CO2 should be increased (IPCC, 2018). Finland is a
highly forested country, and the national welfare depends largely on forest industry. The
sustainable level of utilization of forest resources is currently discussed widely. This is
also related to the EU’s land use, land use change and forestry (LULUCF) regulation. The
Finnish Climate Panel (Kalliokoski et al., 2019) reviewed six different models used for the
estimation of carbon balance in forests and found out that the models produced very
different predictions for the carbon balance, and none of the models was able to produce
32
definitively reliable predictions for the future development of forests in Finland. More
research is still needed in order to supplement our knowledge in this essential issue.
2.3 Climate change induced risks to forests in Europe
Observed changes in forest growth and also predictions are rather positive, especially for
northern Europe. However, the weather-related disturbances may also have serious
negative impacts on forests, and in this chapter we review the most important disturbance
agents.
2.3.1 Varying wind damage risk
Wind is the dominant abiotic cause of forest damages in Europe. In the past few decades,
wind storms have damaged a significant amount of timber and caused large economic and
ecological losses in forestry (Schelhaas 2008; Seidl et al., 2014; Reyer et al. 2017a). In
Finland, strong winds have damaged over 24 million m3 of timber during different winter
and summer storms since 2000 (e.g. Zubizarreta-Gerendiain et al., 2017). The increasing
amount of damages in the European forests may at least partially be explained by the
increasing volume of growing stock and changes in forest structure (e.g., age and tree
species). These are related to changes in forest management practices. The projections of
future climate indicate only quite modest changes in extreme wind speeds for northern
Europe (Figure 2.10; Ruosteenoja et al., 2019). Nevertheless, decreasing soil frost may
increase wind damage risk (section 2.3.3).
2.3.2 Insect pests
Changes in the frequency and severity of pest and disease outbreaks are very likely in the
future. Climate change is expected to make conditions more favourable for many insect
pests. An extensive summary of the impacts of climate change on the Finnish forest insect
pests was recently published by Asikainen et al. (2019), and this chapter is largely based
on that report.
In Europe, bark beetles have, on average, destroyed 2.8 million cubic metres of wood
annually during 1950–2000 (Schelhaas et al., 2003). The most significant insect pest in the
European forests is the European spruce bark beetle (Ips typographus) (e.g., Christiansen
and Bakke, 1988). This species lives generally in wind-damaged Norway spruces, but after
extensive wind damage or drought, intensive spruce bark beetle outbreaks may exist
(Marini et al., 2013, 2017). During these kinds of outbreaks, the spruce bark beetles can
also attack healthy spruce trees.
In Finland, spruce bark beetle has been traditionally a univoltine (one generation per
summer) species existing in low numbers, but within the recent years, it has become more
abundant. Particularly, the spruce bark beetle populations increased after the warm
summer of 2010 (Siitonen and Pouttu, 2014). This was partly because there were a lot of
wind-damaged trees in forests after the severe thunderstorms of the summer of 2010 (Viiri
et al., 2011). Secondly, drought stress had weakened the defence of spruces against the
insect pests. Thirdly, as a result of the exceptionally warm thermal growing season in 2010,
spruce bark beetles were able to produce a second generation during the same year (Pouttu
and Annila, 2010). As the next summer was equally warm, and windstorm Tapani in
December 2011 caused additional extensive damage in the forests, conditions for the
growth of spruce bark beetle populations continued to be favourable.
Increase in summer temperatures is favourable for spruce bark beetles because the
species can generate two generations during one summer if the growing degree day (GDD)
sum exceeds approximately 1500 °C days. In the 20th century, annual GDD sums in
33
Southern Finland varied in typical summers mainly between 1300 and 1400 °C days and
only occasionally exceeded the threshold of 1500 °C days. In Southern Sweden, for
instance, the thermal growing season is longer and the GDD sums are thus higher.
Consequently, spruce bark beetle is there a bivoltine (two generations per summer) species
(Öhrn et al., 2014).
Already during the early 21st century, in Southern Finland the GDD sums have tended
to exceed the threshold of 1500 °C days in many years. The summer of 2018 was record
warm, and the highest GDD sums in Southern Finland reached 1900 °C days,
corresponding to typical values in Poland (Wypych et al., 2017). GDD sums exceeded
1500 °C days up in the north to Northern Savonia and Southern Ostrobothnia. Climate
projections indicate that this kind of GDD sums would be typical at some time during the
second half of the 21st century (Ruosteenoja et al., 2011, 2016; Figure 2.9). Moreover, as
the climate warming continues, it will soon become very unlikely to have a thermal
growing season that is cool according to current climate statistics.
Figure 2.17 illustrates that already in the current climate spruce bark beetle is capable
to produce two generations per year every second summer in Southern Finland. In mid-
century, annual GDD sum is projected to exceed 1500 °C days in 8–9 summers out of ten
in the south and about 5–7 summers out of ten in the central parts of Finland. In the late
21st century, spruce bark beetle is expected to produce second generation occasionally
even in Lapland, and in the south the species would clearly become bivoltine.
Figure 2.17:. The probability (in percent) of the annual growing degree day sum exceeding
1500 °C days in different regions of Finland during the periods 2010–2039, 2040–2069
and 2070–2099 under the RCP4.5 and RCP8.5 scenarios. Adapted from Asikainen et al.
(2019).
In other parts of Europe, the population density of spruce bark beetles has recently
increased as well. In Sweden, windstorm Gudrun damaged 70 million cubic metres of
wood in January 2005 (Bengtsson and Nilsson, 2007), and in the subsequent spruce bark
beetle outbreak additional 4 million cubic metres of wood were destroyed (Långström et
al., 2009). In Poland, spruce bark beetles have destroyed annually on average 1 million
cubic metres of wood during the 1980s and 1990s (Grodzki, 1999). At the moment, the
situation is worst in the Czech Republic where a series of years favourable for spruce bark
beetle has occurred since 2003 (Krejzar, 2018). An unprecedented spruce bark beetle
34
outbreak emerged in 2017, and as a result, the timber market in the Czech Republic has
collapsed since bark beetle damage has exceeded the annual need for timber in the country,
and there are not enough forest machinery and resources to fell even all the damaged trees.
In the future, spruce bark beetle is expected to benefit from climate change in Europe
(Seidl et al., 2008, 2009; Hlásny et al., 2011). The development of the species is strictly
regulated by air temperature (Annila, 1969). Traditionally, the species has been univoltine
in Southern Finland, for instance, and at higher elevations in Central Europe as well while
in most of Central Europe it has been bivoltine. Already in the near future in Central
Europe, significant three-generation regimes are projected to appear. By the end of the
21st century, the three-generation regime is expected to occur over all the existing
coniferous stands in the Czech Republic, for instance (Hlásny et al., 2011).
In addition to the spruce bark beetle, many other insect pests are expected to benefit
from climate change (Asikainen et al., 2019). The damage caused by the common pine
shoot beetle (Tomicus piniperda) and the large pine weevil (Hylobius abietis) are both
expected to increase. The populations of the European pine sawfly (Neodiprion sertifer)
may increase in eastern and northern Finland due to rising winter temperatures and in
southern and western Finland due to increasing summer drought (Virtanen et al., 1996;
Nevalainen et al., 2015).
Heterobasidion species are globally the most important wood-decay fungus for
conifers. Woodward et al. (1998) estimated that within the European Union, damage
caused by Heterobasidion species accounts for 800 million euros annually. In Finland, two
species of Heterobasidion occurs, Heterobasidion annosum and Heterobasidion
parviporum. Heterobasidion annosum is considered to be economically the most
important forest pathogen in the Northern Hemisphere. It causes root and butt rot diseases
on pine trees. In Finland, it is most common in southeastern parts of the country. In our
country, Heterobasidion parviporum is a more common species. It causes root and butt rot
on spruces. In Southern Finland, approximately 15–20% of spruces suffers from the root
rot. Heterobasidion parviporum occurs sporadically also in Northern Finland, although
root and butt rot diseases of conifers are there caused mainly by other decay fungi than
Heterobasidion (Müller et al., 2018). Increasing temperatures increase the spore formation
of the Heterobasidion parviporum, and it is expected that the share of infected spruces will
increase in the future (Pukkala et al., 2005; Müller et al., 2014). In addition, the projected
loss of soil frost may increase root damage in forest harvesting, making trees more
vulnerable for wood-decay fungus. Heterobasidion annosum is likewise expected to
become more abundant at least in Southern Finland. In Northern Finland, Heterobasidion
annosum is presently nearly absent for an unknown reason (Müller et al., 2018).
One classic insect pest in Central Europe is the nun moth (Lymantria monacha).
Between the years 1853 and 1863, caterpillars of the nun moth destroyed 147 million cubic
metres of wood in Russia and East Prussia (Bejer, 1988). Most of this area was afterwards
converted into agricultural land. The abundance of the nun moth has varied cyclically from
a decade to decade without a clear connection to weather conditions (Haynes et al., 2014).
However, the northern distribution limit of the nun moth is restricted by the minimum
winter temperatures as the eggs of the species do not survive in temperatures colder than
approximately -30 °C (Fält-Nardmann et al., 2018). Before the 1990s, temperatures colder
than this threshold were common enough so that in Finland small populations of the nun
moth occurred only in the southwestern archipelago. However, after the 1990s, the
abundance of the nun moth in Finland has increased approximately 100-fold, and the
species is nowadays common widely in the southern parts of the country (Leinonen et al.,
2017). It is estimated that the northern distribution limit of the nun moth can still shift
approximately 300 km northwards if the mean temperature increases by additional 5 °C
(Fält-Nardmann et al., 2018). Forest damages caused by the nun moth have been already
35
documented in Estonia (Voolma et al., 2014), and minor damages have been reported also
in the southwestern archipelago of Finland (Heino and Pouttu, 2014).
A close relative to the nun moth, the gypsy moth (Lymantria dispar), is also a well-
known insect pest. While the nun moth damages mainly conifers, especially spruce forests,
caterpillars of the gypsy moth feed on deciduous trees. The gypsy moth is a widespread
species in Central Europe and nowadays common also in the Baltic states. It is moreover
an introduced species in North America where it has caused extensive damage in oak
forests (Elkinton and Liebhold, 1990; Weseloh, 2003; McManus and Csóka, 2007). In
Eurasia, outbreaks of the gypsy moth occur approximately once in every ten years
(McManus and Csóka, 2007; Hlásny et al., 2016). In Finland, before the 2010s the gypsy
moth had occurred only as an extremely rare migrant but during the recent years, a few
populations might have already appeared at the southern coast. Overwintering eggs of the
gypsy moth are even more vulnerable to extreme cold temperatures than the eggs of the
nun moth but. Nonetheless, in the future, winters are projected to became mild enough for
the formation of gypsy moth populations in Southern Finland (Fält-Nardmann et al., 2018;
Neuvonen et al., 2018).
To conclude, most of insect pests benefit from climate change because their
development is closely regulated by air temperature. In the Finnish flora, Norway spruce
is the tree species most sensitive to climate change and increasing drought occurrence may
weaken its defence against insect pests. In addition to already existing insect pests and
fungi, new invasive species can be accidentally introduced by international plant trade
(Lilja et al., 2011; Hantula et al., 2014).
2.3.3 Less soil frost
Soil frost is affected, e.g., by the soil properties and water content. The most important
meteorological factors controlling soil frost are air temperature and the snow depth. Snow
cover is an efficient insulator, and in Finland, for instance, soil frost penetrates typically
deeper in the western than eastern parts of the country although winters are usually milder
in the west. This is because snow cover tends to be thicker in the east. Due to global
warming mean temperature is projected to increase, but in many areas snow cover is
simultaneously projected to decrease and this might partially cancel out the impact of
increasing temperature on soil frost.
Most recently the impact of climate change on soil frost conditions in Finland has
been studied by Lehtonen et al. (2019). Based on that study, the mean annual number of
days with modelled soil frost thickness exceeding 20 cm in Finland is shown in Figure
2.18, separately for mineral soils (clay or silt) and peatlands. The results are shown for the
reference period 1981–2010 and for the future periods of 2021–2050 and 2070–2099 under
the RCP4.5 and RCP8.5 scenarios. According to these results, the soil frost season on
mineral soils is projected to shorten by about one month from 1981–2010 to 2021–2050.
By the end of century, the shortening is most likely about two months under the RCP4.5
scenario and approximately three months if the RCP8.5 scenario realizes. In Southern
Finland, this means that the length of the soil frost season would decrease by more than
50% on average. On peatlands soil frost does not penetrate as deep as on mineral soils, and
in the late 21st century peatlands in Southern Finland are expected to remain virtually
unfrozen in most of winters.
The loss of soil frost is expected to hamper wintertime logging. The bearing capacity
of forest sites is clearly higher during frozen than unfrozen conditions. In addition, small
forest truck roads having light foundations do not bear heavy timber trucks in wet road
sections unless the soil is frozen (Kaakkurivaara et al., 2015). Operations in poorly bearing
conditions increase rut formation on forest floor and tend to cause damage to tree roots
36
and stems (Sirén et al., 2013; Pohjankukka et al., 2016). Moreover, fuel consumption in
the harvesting increases. This holds also for fuel consumption in timber transportation if
the truck roads are in poor condition (Svenson and Fjeld, 2016).
Figure 2.18: Mean annual number of soil frost days (soil frost depth at least 20 cm) on
mineral soils and on peatlands during the periods 1981–2010, 2021–2050 and 2070–2099
as evaluated based on six climate model simulations.
37
Harvesting conditions are particularly difficult on unfrozen drained peatlands because
of their inherently low ground-bearing capacity. Thus, on these sites logging is generally
conducted during winter when the soil is frozen (Ala-Ilomäki et al., 2011). On the other
hand, a more intensive utilization of peatland forests has the largest potential in increasing
the wood harvesting. This might be difficult as peatland forests are expected to lose soil
frost during the forthcoming decades.
The results of Lehtonen et al. (2019) are in accordance with earlier studies (e.g.,
Venäläinen et al., 2001; Kellomäki et al., 2010): the effect of increasing temperature
mostly outweighs the impact of decreasing snow cover when considering expected
changes in soil frost conditions. In particular the soil frost season is projected to shorten
as in a warmer climate the soil freezes, on average, later in autumn and melts earlier in
spring. Nevertheless, according to Kellomäki et al. (2010), in most winters the maximum
soil frost depth on mineral soils would in mid-21st century still exceed 50 cm even in
Southern Finland. On the other hand, from January to March the soil could be almost
unfrozen for half of the time. In central and northern parts of the country, the typical soil
frost depth on mineral soils in late winter would still be around one meter in the late 21st
century (Kellomäki et al., 2010).
In snow-free surfaces, like in forest truck roads if they are kept snow-free, the soil
frost depth is approximately proportional to the square root of the cumulative frost sum
(e.g., Gregow et al., 2011). In these kinds of environments, the increasing temperatures
thus straightforwardly lead to diminishing soil frost.
2.3.4 Drought leading to higher forest fire risk
Increase in the frequency and severity of summer droughts will have an impact on forest
fire danger. In the Mediterranean region, large forest fires occur almost every summer, and
huge fires are not rare either in the boreal forests of Russia and Canada (e.g., Flannigan et
al., 2009; Mei et al., 2011; Vivchar, 2011; Gonçalves and Sousa, 2017).
Also, in Finland, numerous wildfires occur every year, but the average size of the fires
is small, approximately 0.5 ha. Consequently, the annual burned area, on average, is
nowadays historically low (Wallenius, 2011). This is threatening biodiversity as fire is a
natural phenomenon in the process of forest regeneration (e.g., Esseen et al., 1997). The
small average size of fires in Finland is due to the effectiveness of fire suppression. Fire
survey flights contribute to the early detection of fires, and the dense forest road network
aids fire fighters to reach and suppress the fires (e.g., Lehtonen et al., 2016a). In rural areas,
the suppression of fires is largely conducted by local volunteer fire departments. Moreover,
the Finnish landscape is characterized by numerous lakes and swamps creating natural
obstacles for the fires. On the other hand, many other boreal regions are characterized by
large homogenous forest areas. However, large fires are still possible in Finland. Recently,
in environmental conditions similar to Finland, large forest fires have occurred in Sweden
in the summers of 2018 and in 2014. In 2014, a single fire in Västmanland burned 14 000
ha of forest (Länsstyrelsen i Västmanlands län, 2015).
Fire danger is often assessed with fire danger indices. One of the most-widely used
index system is the Canadian forest-fire weather index (FWI) system (Van Wagner, 1987).
The FWI system was used, e.g., by Groenemeijer et al. (2016) in estimating the impact of
climate change on forest fire risk throughout Europe during the 21st century. In the
southern parts of the continent, the meteorological fire danger is projected to increase
significantly already by the mid-21st century (Figure 2.19). Towards the end of the
century, it is much more likely that fire danger increases than decreases also in the north.
The results of another study (Lehtonen et al., 2016a) indicated that the annual burned area
could substantially increase in Finland if the connection between the fire danger and the
38
fire activity remains similar as currently. On the other hand, as the burned area in Finland
has been small during the recent decades, even a single conflagration could burn as much
forest as all the fires have burned during the last 10 or 20 years in total. Hence, it is
uncertain if the increasing fire danger indeed results in large fires, but the probability for
a fire that could escape for a much larger fire than experienced during the recent decades
is expected to increase. Nevertheless, the variations in the occurrence of fires cannot be
predicted by climate forcing alone (Bowman et al., 2009). Other aspects, such as human
behaviour should be taken into account as the large majority of fires, generally more than
90%, is caused by human activities (Ganteaume et al., 2013).
Daily probability of Fire Weather Index > 20
Reference
Period
1971-
2000:
Projected
Changes
2021–
2050:
RCP4.5
RCP8.5
Change in daily probability (percentage points)
Figure 2.19: Multi-model mean daily probability (%) of the fire weather index (FWI) value
exceeding 20 during 1971–2000 (top). In the bottom row, multi-model mean change in the
daily probability of FWI value exceeding 20 (in percentage points) is shown under the
RCP4.5 (left) and the RCP8.5 scenarios (right). White dots denote a change significant at
the 5% level. Adapted from Groenemeijer et al. (2016).
In regions suffering from aridity, like the Mediterranean basin, vegetation productivity
may act as a limiting factor for fires (Migliavacca et al., 2013). In this kind of regions, the
fire activity can thus be estimated to increase less than could be deducted purely on the
basis of fire danger indices. However, even when taking into account the ecosystem
39
functioning, in absolute terms the burned area is still predicted to increase in Europe
particularly in the Mediterranean basins, in the Balkan regions and in Eastern Europe
(Migliavacca et al., 2013). In relative terms, the increase in burned area is in Southern
Europe most likely approximately 15% from the late 20th century to the mid-21st century.
According to that study, for the same period in Central Europe, the change is more than
50% and in Northern Europe approximately 100%, although in hectares the increase is
projected to be largest in the south.
In accordance with the above-mentioned study, Turco et al. (2018) recently estimated
that taking into account the productivity alterations under changing climatic conditions
roughly halve the fire-intensifying signals in the Mediterranean Europe. They estimated
the future burned area under the 1.5, 2 and 3 °C global warming scenarios and found out
that the higher the warming level is, the larger is the increase of the burned area, ranging
from approximately 40% to 100% across the scenarios.
2.3.5 Snow damage risk
Crown snow load consists of snow and rime attached tightly to tree crowns and other
structures. The accumulation of snow on tree branches is dependent on meteorological
conditions which are further modified by topography. Typical forms of snow damage
under extreme snow-loading include stem breakage and bending or leaning of stems, but
trees can also be uprooted if the soil is unfrozen (Petty and Worrell, 1981; Valinger et al.,
1994; Nykänen et al., 1997). As in the case of wind damage, snow-damaged trees can
furthermore disrupt power transmission by bending over or leaning on power lines. In
addition, snow-damaged trees are susceptible to insect attacks and other kinds of
consequential damage (Schroeder and Eidmann, 1993; Schlyter et al., 2006).
In Finland, snow is one of the most important abiotic disturbance agents reducing
stand quality in forests. According to a survey conducted by the Finnish Forest Research
Institute from 2009–2013, snow damage had occurred on 7% of the productive forest land
(Korhonen et al., 2017). At a European level, estimates of the amount of timber damaged
by snow during a typical year vary from 1 million m3 to 4 million m3 (Nykänen et al.,
1997; Schelhaas et al., 2003).
It has been long known that forests at high altitudes are most prone to snow damage
(e.g., Heikinheimo, 1920). In Finland, the weight of snow loads on trees tends to increase
approximately linearly with the terrain elevation (Jalkanen and Konôpka, 1998). In
Northern Europe, snow damages are common already in areas located higher than 100
metres above the sea level whereas in Central Europe, altitudes of 500–900 metres are
generally associated with the highest incidence of snow damage (Nykänen et al., 1997).
The region of Kainuu is considered the most vulnerable area for snow damage in Finland
(Solantie, 1994). Other high-risk regions include North Karelia and Lapland.
The impact of climate change on the snow-load risk can be estimated by using snow
load models driven by climate model data. Groenemeijer et al. (2016) estimated the impact
of climate change on several meteorological and hydrological hazards in Europe during
the ongoing century. In estimating the future changes in heavy snow loads, they applied a
snow load model developed and used operationally at the Finnish Meteorological Institute
(Lehtonen et al., 2014). The climate data used in snow-load calculations originated from
six different model simulations. Based on their results, heavy snow loads occur in Europe
most frequently in the mountainous regions and in north-eastern Europe (Figure 2.20).
Future projections for the mid-21st century indicate slightly decreasing probability for
heavy snow loads over most of Europe. However, in northern Scandinavia, Finland and
north-western Russia, the probability for heavy snow loads is widely expected to increase.
These projected changes are, nevertheless, rather small and statistically not significant.
40
Lehtonen et al. (2016b) studied the impact of climate change on heavy snow loads
focusing only on Finland. Despite of some limitations in the study, their results largely
corresponded with those by Groenemeijer et al. (2016), implying potentially increasing
risk for snow damage in eastern and northern Finland, particularly in the regions of North
Karelia, Kainuu, Koillismaa and Lapland. In southern and western Finland, on the other
hand, snow loads on tree crowns were projected to decrease.
Reference
Period
1971-
2000:
Annual probability (%)
Projectef
Changes
2021–
2050:
RCP4.5
RCP8.5
Change in annual probability (percentage points)
Figure 2.20: Multi-model mean annual probability (%) of snow load exceeding 20 kg m-2
during 1971–2000 (top). In the bottom row, multi-model mean change in the annual
probability (in percentage points) is shown under the RCP4.5 (left) and the RCP8.5
scenarios (right). White dots denote areas with a change significant at the 5% level.
Adapted from Groenemeijer et al. (2016).
41
3. Climate change and impacts in Uruguay
3.1 Projected change of climate in Uruguay
In Uruguay, projected changes in climate are fairly moderate. Temperature increase is
likely to be smaller than the global mean increase of temperature (see Figures 3.3(a) and
1.2). According to the multi-model mean projection under RCP4.5, monthly mean
temperatures increase by 1.2–1.5°C by the period 2040–2069 (Figure 3.4(a)). The
corresponding inter-model uncertainty interval ranges from less than 0.5°C to 1.8–2.4°C.
Figure 3.1: Projected changes in annual mean (a) temperature (in °C), (b) precipitation
(in %), (c) incident solar radiation (in %) and (d) relative humidity (in percentage points)
in Uruguay and its adjacent areas from the period 1971–2000 to 2040–2069 under
RCP4.5. The position of the grid point 32.5°S, 57.5°W examined in the more detailed
analyses is marked by a dot in panel (a).
For precipitation, the best-estimate projection is an increase of about 10 % in Southern
Hemisphere summer and autumn (from December to June), while from July to November
precipitation totals remain nearly unchanged. This leads to an increase of about 6 % in
annual precipitation (Figure 3.1(b)). Considering the inter-model differences, however, the
sign of change cannot be firmly established in any season. The uncertainty interval of the
change ranges from about -20 % to +40 % in autumn and from -30 % to +20 % in spring.
42
Figure 3.2: Projected changes in the standard deviation of temporal variations in (a)
temperature and (b) precipitation (both in %) in Uruguay and its adjacent areas from the
period 1971–2000 to 2040–2069 under RCP4.5
In incident solar radiation and relative humidity, the multi-model mean changes are
negligibly small throughout the year (Figures 3.1(c)–(d) and 3.4(c)–(d)). The inter-model
uncertainty covers an interval from approximately -4 to +4 % (-3 to +4 percentage points)
for solar radiation (humidity). Wind speeds are projected to increase by about 2 % in
seasons other than the Southern Hemisphere winter (Figure 3.5). Moreover, according to
the model simulations the proportion of easterly winds increases.
The standard deviation of temperature in Uruguay does not alter significantly (Figures
3.2(a) and 3.4(e)). This indicates that the cold and warm temperature extremes will rise
approximatively at a rate that is equal to the trend in the time-mean temperature (Figure
3.4(a)).
For precipitation, changes in temporal variability are slightly larger than the
corresponding changes in temporal means (Figures 3.2(b) vs. 3.1(b); Figures 3.4(f) and
(b)). In accordance with this, the maximum one-day precipitation will increase, mostly so
in the December-February and March-May seasons; in these seasons, the inter-model
agreement on the sign of change is also quite high (Figure 3.6, left panel). Heavy
precipitation events are likely to intensify in spring as well, while in winter the direction
of change is ambiguous. The lengths of dry periods change fairly little in the first half of
the year but are likely to increase in the Southern Hemisphere winter and spring (Figure
3.6, right panel). Note, however, that in this quantity the inter-model differences are large.
As the territory of Uruguay is relatively small, the projected climatic changes are
generally fairly uniform across the country (Figures 3.1, 3.2 and 3.5). Even so, somewhat
larger increases in precipitation and its temporal variability are projected for north-eastern
Uruguay than elsewhere.
To summarize, the anticipated climate change in Uruguay is rather gentle compared
to many other areas of the world. One reason for this is that the Southern Hemisphere is
predominantly covered by oceans. The specific heat capacity of water is large, and in
addition, vertical mixing acts to transfer heat deeper into the ocean. Accordingly, oceans
absorb heat effectively, thus stabilizing climate. In addition, over water surfaces latent heat
consumed for evaporation tends to curtail the temperature increase.
43
Figure 3.3: Projected annual mean changes relative to 1971–2000 in (a) mean
temperature (in °C), (b) precipitation (in %), (c) incident solar radiation (in %) and (d)
relative humidity (in percentage points) at 32.5°S, 57.5°W (western Uruguay; the position
is marked in Figure 3.1(a)). The time series are shown separately for the RCP2.6, RCP4.5
and RCP8.5 scenarios (see the legend).
44
Figure 3.4: Projected monthly changes in (a) the mean temperature (in °C), (b)
precipitation (in %), (c) incident solar radiation (in %) and (d) relative humidity (in
percentage points) at 32.5°S, 57.5°W (western Uruguay; the position is marked in Figure
3.1(a)) under RCP4.5 for the period 2040–2069, relative to 1971–2000. Corresponding
changes in the temporal standard deviation of daily-mean temperature (in °C) and
precipitation (in %) are shown in panels (e) and (f). The multi-model mean projections for
individual calendar months (J = January,..., D = December) are denoted by open circles.
Shading shows the 90 % uncertainty intervals for the projection.
45
Figure 3.5: Projected seasonal changes (in percent) in the time-mean geostrophic wind
speed in Uruguay from 1971–2000 to 2040–2069 under RCP4.5 in (a) December-
February, (b) March-May, (c) June-August and (d) September-November. The contour
interval is 2 %.
Figure 3.6: Model-derived seasonal changes (in percent) in the maximum one-day
precipitation R1d (left panel) and the maximum number of consecutive dry days CDD (with
precipitation < 1 mm/day; right panel) at 32.5°S, 57.5°W (western Uruguay; the position
is marked in Figure 3.1(a)) under RCP4.5 for the period 2071–2100, relative to 1971–
2000. The coloured bars show the 25th to 75th percentile intervals of the change derived
from the responses of 21 models. The corresponding 10th to 90th percentile intervals are
depicted by whiskers and the minimum and maximum responses among the models by
black dots. The black lines within the bars stand for the multi-model medians. The
probability distributions are given separately for four seasons: December-February
(DJF), March-May (MAM), June-August (JJA) and September-November (SON).
46
3.2 Overview of climate change impacts in Uruguay
Climate change is predicted to have various impacts in Latin America. The impacts vary
from location to location. Heat extremes will increase, sea level will continue to rise and
the volume of tropical mountain glaciers continues to decrease. Probability of coral reef
bleaching is increasing. There is a risk of Amazon rainforest degradation (Reyer et al.,
2017b). At some regions, climate change may influence negatively on agricultural yields
but, on the other hand, the change may increase, e.g., potential for higher rice yields (Reyer
et al., 2017b). El Niño and La Niña
4
events influence a lot on the weather in Latin America.
In Argentina, Paraguay and Uruguay, rainfall is typically increasing during El Niño. La
Niña indicates typically cooler and dryer than normal weather. The influence of climate
change on the frequency of El Niño and La Niña events is still an open question. However,
there are studies indicating that an extreme El Niño may become more frequent (e.g. Wang
et al., 2017).
In Uruguay, the most harmful climate phenomena are related to events such as
droughts, floods, frosts, heat waves, hail and squall lines. The impact of climate change on
agriculture can vary depending on the crop. For example, for winter crops, like barley and
wheat an increase of temperature can be harmful whereas for summer crops, like rice,
warming can be positive. Increased temperature may benefit grassland production, but
precipitation deficiencies or increased variability is harmful. This is why the potential
increase in climate variability, e.g., in the frequency of El Niño and La Niña events is
important.
Uruguayan coastal resources are vulnerable to global climate change due to the rising
sea level. Under the RCP2.6 scenario, the sea level would rise on Atlantic Coast of Latin
America around 20 to 50 centimeters until the end of this century. Under RCP8.5, the sea-
level rise would be 50 to 90 centimeters (Reyer et al., 2017b); note, however, that these
estimates will be updated by the IPCC. Hareau et al. (1999) estimated that the largest
economic risk caused by sea level rise would be experienced in urbanized areas like the
Maldonado-Punta del Este resort and in Montevideo. Sea level rise may also cause
flooding and salinization of lowlands and reduce course of the water flows that discharge
along the coast. Especially the Santa Lucía river is vulnerable as it is the source of drinking
water for Montevideo (Hareau et al., 1999).
Climate change adaptation and mitigation measures are actively implemented in
Uruguay. The Fourth National Communication was prepared by the Ministry of Housing,
Land-Use Planning and Environment (MVOTMA) within the framework of the National
Climate Change Response System (SNRCC). SNRCC and the National Environment
System coordinate climate change and environment actions. SNRCC has also developed a
National Climate Change Policy (PNCC) which provides a long-term strategic framework
to integrate and strengthen the approach to mitigation and adaptation. This policy identifies
strategic priority actions and specific institutional and capacity building measures to create
an enabling environment for adaptation planning. PNCC was adopted by the National
Environment Cabinet in April 2017. In PNCC, nine key adaptation strategic areas and
objectives were defined (Table 1) that provided a good basis for the future activities to be
taken.
4
El Niño Southern Oscillation (ENSO) is a natural phenomenon. Ocean and atmospheric
conditions in the tropical Pacific Ocean fluctuate between El Niño (warm conditions) and a
drop in temperature (La Niña). The fluctuations are rather irregular, appearing
approximately every three to six years. Intensive phase of each event may last for about a
year.
47
Table 1. The key climate change adaptation strategic areas and objectives in Uruguay
(National Climate Change Policy, 2017).
To summarize: In addition to sea-level rise, the largest risks of climate change in
Uruguay are related to possible hazards such as droughts and floods, heatwaves, hails,
storms and tornados. The El Nio - Southern Oscillation phenomenon (ENSO) further
increases inter-annual variability with higher precipitation during the El Nio years and
more severe droughts during the La Nia years. If climate change influences the frequency
and intensity of these phenomena, it may cause new challenges to the Uruguayn society.
According to the Global Forest Watch (2019), the tree cover loss in Uruguay between
2001 and 2017 was 327,000 ha whereas the tree cover gain between 2001 and 2012 was
499,000 ha (Figure 3.7; Hansen et al., 2013). The tree covered area has thus increased by
about 170,000 ha during the about 15-year period. If the development continues, it also
contributes to Uruguay’s greenhouse gas balance in a positive way, i.e., the role of forests
as a carbon sink increases.
48
Figure 3.7: Tree cover change in Uruguay. Red colour indicates areas of loss and blue
those of gain. Source: Hansen/UMD/Google/USGS/NASA, accessed through Global
Forest Watch.
49
4. Climate change and impacts in China
4.1 Projected change of climate in China
According to the multi-model mean projection, annual mean temperature and precipitation
total are both anticipated to increase over the entire territory of China, the changes being
strongest in the western and northern provinces (Figure 4.1(a)–(b)). Solar radiation
increases by 2–4 % in wide areas of eastern China (Figure 4.1(c)). This phenomenon can
at least partially be attributed to cleaning of air due to a reduction of sulphur dioxide and
other emissions producing aerosol particles. Changes in relative humidity are virtually
negligible (Figure 4.1(d)).
Figure 4.1: Projected changes in annual mean (a) temperature (in °C), (b) precipitation
(in %), (c) incident solar radiation (in %) and (d) relative humidity (in percentage points)
in China from the period 1971–2000 to 2040–2069 under RCP4.5. The position of the grid
point 32.5°N, 120°E examined in the more detailed analyses is marked by a dot in panel
(a).
50
Figure 4.2: Projected changes in the standard deviation of temporal variations in (a)
temperature and (b) precipitation (both in %) in China from the period 1971–2000 to
2040–2069 under RCP4.5.
Figure 4.3: Projected annual mean changes relative to 1971–2000 in (a) mean
temperature (in °C), (b) precipitation (in %), (c) incident solar radiation (in %) and (d)
relative humidity (in percentage points) at 32.5°N, 120°E (eastern China; the position is
marked in Figure 4.1(a)). The time series are shown separately for the RCP2.6, RCP4.5
and RCP8.5 scenarios (see the legend).
51
Focusing on the target point located near the eastern coast of China (the position is
given in Figure 4.1(a)), the best estimate for the annual-mean temperature increase is about
2°C and for the precipitation increase 5 % (by the period 2040–2069 under RCP4.5; see
Figure 4.3 (a)-(b)).
Figure 4.4: Projected monthly changes in (a) the mean temperature (in °C), (b)
precipitation (in %), (c) incident solar radiation (in %) and (d) relative humidity (in
percentage points) at 32.5°N, 120°E (eastern China; the position is marked in Figure
4.1(a)) under RCP4.5 for the period 2040–2069, relative to 1971–2000. Corresponding
changes in the temporal standard deviation of daily-mean temperature (in °C) and
precipitation (in %) are shown in panels (e) and (f). The multi-model mean projections for
individual calendar months (J = January,..., D = December) are denoted by open circles.
Shading shows the 90 % uncertainty intervals for the projection.
Compared to the other seasons, warming may be slightly stronger in late summer and
early autumn; however, considering the inter-model scatter, it is likely that the temperature
52
increase falls between about 1 and 3°C throughout the year (Figure 4.4(a)). Precipitation
is most likely to increase by 0–10 % in all seasons except late autumn, but the uncertainty
interval is wide and the sign of change remains uncertain (Figure 4.4(b)). The sign of
change cannot either be established for solar radiation and relative humidity (Figures
4.4(c)-(d)), even though it is more likely that solar radiation increases than decreases. Wind
speeds may slightly increase in summer and decrease in winter; according to the multi-
model mean, by about 2 % (Figure 4.5).
The standard deviation of temperature presumably does not change notably in eastern
China (Figure 4.2(a)), even though there is some modelling uncertainty in this estimate as
well (Figure 4.4(e)). Consequently, the temporal variations of temperature are not likely
to amplify or attenuate significantly, and the extremely low and high temperatures will rise
approximately at a similar rate as the mean temperature (Figures 4.1 (a) and 4.4(a)).
Figure 4.5: Projected seasonal changes (in percent) in the time-mean geostrophic wind
speed in China from 1971–2000 to 2040–2069 under RCP4.5 in (a) December-February,
(b) March-May, (c) June-August and (d) September-November. The contour interval is 2
%.
The temporal variability of precipitation is more likely to increase than decrease
(Figures 4.2(b) and 4.4(f)). In conjunction with increasing time-mean precipitation, this
indicates that extreme precipitation events will become more intense. According to the
multi-model median, the maximum one-day precipitation totals would increase by 10–15
%, depending on the season (Figure 4.6, left panel). The inter-model agreement on the
intensification is largest in spring and summer when more than 90 % of all the models
inspected project stronger maximum 1-day precipitation for the future. As another
53
indication of the more extreme precipitation climate in the future, dry periods are expected
to lengthen in all seasons apart from summer (Figure 4.6(right panel)). Nonetheless, for
this quantity inter-model agreement on the direction of change is lower than for the
maximum one-day precipitation.
In the catchment area of the Yangtze river, both the mean temperature and annual
precipitation are likely to increase moderately. Accordingly, the increasing supply of water
by precipitation is partially compensated by increasing evaporation of water due to higher
temperatures, resulting in fairly modest changes in the time-mean discharge in the Yangtze
river. Even so, as strong precipitation events intensify and, probably, dry periods between
them simultaneously lengthen, temporal variations in the discharge evidently increase.
This topic will be discussed in the next subsection.
Figure 4.6: Model-derived seasonal changes (in percent) in the maximum one-day
precipitation R1d (left panel) and the maximum number of consecutive dry days CDD (with
precipitation < 1 mm/day; right panel) at 32.5°N, 120°E (eastern China; the position is
marked in Figure 4.1(a)) under RCP4.5 for the period 2071–2100, relative to 1971–2000.
The coloured bars show the 25th to 75th percentile intervals of the change derived from
the responses of 21 models. The corresponding 10th to 90th percentile intervals are
depicted by whiskers and the minimum and maximum responses among the models by
black dots. The black lines within the bars stand for the multi-model medians. The
probability distributions are given separately for four seasons: December-February
(DJF), March-May (MAM), June-August (JJA) and September-November (SON).
4.2 Impacts on the Yangtze River hydrology
The Yangtze River is 6,380 km long and thus the longest river in Asia and the third-longest
in the world (Figure 4.7). It is the sixth-largest river by discharge volume in the world. It
drains one-fifth of the land area of China, and the river basin is the home for nearly one-
third of the population of the country. According to the observations, the annual
streamflow at Datong station located near the tidal limit of the river has been slightly
increasing during the period 1960–2000 (Figure 4.8). The glaciers in the source region of
the Yangtze River are shrinking. This leads to an increase of water resources over the short
term, but the long-run impacts of melting glaciers can be negative and decline water
resources (e.g. Liu et al., 2009; Shen et al., 2009; Piao et al., 2010).
54
Figure 4.7: Map of the Yangtze River basin (edited from Gu et al., 2015).
Figure 4.8: Annual runoff at Yangtze River at the Datong station during 1960–2000 (Piao
et al. 2010).
Precipitation is projected to increase during the coming decades (Figures 4.1, 4.3 and
4.4), accompanied with an increase of the variability of precipitation (Figure 4.2). A
number of studies dealing with the impact of climate change on the hydrological regime
of the Yangtze River has been published, and these studies indicate a general increase in
the annual streamflow at the Yangtze River. The increased variability of rainfall means
more intense short-period rainfall events and maximum stream flows. This can increase
flooding events (Wang & Zhang, 2011; Gu et al., 2015; Yu et al., 2018; Gu et al., 2018).
55
In Figure 4.9 is given, as an example, one study (Yu et al., 2018) on how monthly
discharges could change at the Datong station. According to this study, in the 2080s the
warm season discharge would be larger and the cold season discharge smaller, compared
with the 1970–1999 period. However, the inter-model variability is large. As well, the
study of Gu et al. (2015) indicates that discharge during the dry winter season may
decrease.
Figure 4.9: An estimate of the current and projected future monthly discharge in the 2080s
at Datong station for the RCP4.5 (left panels) and RCP8.5 (right panels) scenarios based
on 27 global climate models (GCMs). The red dotted line indicates the median of the model
run results and continuous dark grey line is the baseline used in the study. Dark and light
grey shaded areas indicate where 50% and 80% of model runs fall, respectively. The figure
is edited from Yu et al. (2018).
Besides changes in rainfall patterns and glacial melting, the other factors influencing
the hydrological conditions near the Yangtze River delta are the erosion caused by the
sediment starvation and the sea level rise. River damming and soil conservation decrease
sediment discharge, and this in conjunction with sea level rise increases erosion potential
in the delta area (Yang et al., 2017). In the recent past decades, sea level rise has been
approximately 3 mm/year (Figure 4.10).
Figure 4.10: Mean sea level anomalies on the Chinese coast in 1980–2015 compared with
the 1975–1993 mean (edited form Cheng&Chen, 2017).
During the coming decades, the sea-level will continue to rise. The estimates of the
magnitude of the rise vary from study to study. In the case of global warming of 2°C, the
global sea level rise is estimated to be slightly over 20 cm (90% confidence limits 15–33
cm). If warming is 4°C, the sea level would rise more than 60 cm (90% confidence limits
39–124 cm) (Jevreva et al., 2016). Globally, the sea level rise varies from location to
RCP4.5
RCP8.5
Month
Month
25-75%
10-90%
1970-1999
median
Discharge (m3s-1)
Discharge (m3s-1)
sea level anomaly
linear regression
56
location, but according to Jevereva et al. (2016), for the Chinese coast the values are close
to global mean. Sea level rise will increase coastal erosion, saltwater intrusion and storm
surge intensity in the delta area. As well, the arrival time of tidal bores becomes earlier
under sea-level rise and the height of bores and the velocities at the surface and bottom
layers may increase (Wang et al, 2018).
5. Concluding remarks
This report deals with projected climatic changes in four areas of operation of the UPM-
Kymmene company: Finland, southern Germany, Uruguay and eastern China. The
implications of the projected changes for forestry, including forest growth and productivity
and possible climate change induced disturbances are discussed as well. In the
Introduction, the main natural and anthropogenic factors inducing changes in climate are
examined, along with the methods used to assess the future changes. The main quantitative
findings of the work are listed in the Executive summary in the beginning of the report and
are therefore not replicated here.
In many cases, inter-model differences in the simulated responses are fairly large.
Although all the models analyzed simulate higher temperatures for the future, the degree
of warming varies quite a lot among the models. For many other climate variables, e.g.,
precipitation, even the sign of the future change is usually more or less uncertain. Even so,
in the regions examined mean precipitation is more likely to increase than decrease, with
the exception of southern Germany in summer and early autumn, Uruguay in Southern
Hemisphere winter and spring and China in late autumn. Rising temperatures act to
enhance evaporation, and consequently drought risks may exacerbate despite modest
increases in precipitation. Actually, in many regions both the risks of drought and floods
are increased owing to the larger fluctuability of precipitation. In some seasons both the
intense rainfall events and dry periods are projected to become more severe.
In general, forest resources have been increasing in Europe. Especially in Northern
Europe, forests have benefitted of the warmer climate and increased CO2 concentration in
the atmosphere. During the coming decades, this positive development may at least partly
be cancelled due to potentially increasing disturbances. For example, drought, fire and new
pests may harm the forests. The role of forests as a carbon sink is an important aspect in
the context of climate change mitigation activities. The key question is what climate smart
forestry is; how forest can be utilized in a sustainable way but simultaneously maintaining
or increasing their role as carbon sinks. Vivid discussion on the amount of sustainable use
of forests is foreseen to continue in Finland but also elsewhere in Europe.
If global climate policy proves to be successful, it is possible that future changes in
climate will be weaker than those based on the intermediate RCP4.5 scenario discussed in
this report. On the other hand, in that case rapid restrictions of the greenhouse gas
emissions are required globally, which evidently substantially influences the prerequisites
of operations in heavy industry.
57
References
Ala-Ilomäki, J., Högnäs, T., Lamminen, S. and Sirén, M. 2011. Equipping a conventional
wheeled forwarder for peatland operations. International Journal of Forest Engineering,
22, 7–13.
Annila, E. 1969. Influence of temperature upon the development and voltinism of Ips
typographus L. (Coleoptera, Scolytidae). Annales Zoologici Fennici, 6, 161–207.
Asikainen, A., Viiri, H., Neuvonen, S., Nevalainen, S., Lintunen, J., Laturi, J., Uusivuori,
J., Venäläinen, A., Lehtonen, I. and Ruosteenoja, K. 2019. Ilmastonmuutos ja
metsätuhot – analyysi ilmaston lämpenemisen seurauksista Suomen osalta. The Finnish
Climate Change Panel, Helsinki, 76 pp. [in Finnish]
Bejer, B. 1988. The nun moth in European spruce forests. In: Berrymann, A.A. (ed.)
Dynamics of forest insect populations, patterns, causes, and implications. Plenum Press,
New York, NY, pp. 211–231.
Bengtsson, A. and Nilsson, C. 2007. Extreme value modelling of storm damage in Swedish
forests. Natural Hazards and Earth System Sciences, 7, 515–521.
Bowman, D., Balch, K., Artaxo, P., Bond, W., Carlson, J., Cochrane, M., D’Antonio, C.,
DeFries, R., Doyle, J., Harrison, S., Johnston F., Keeley, J., Krawchuk, M., Kull, C.,
Marston, J., Moritz, M., Prentice, C., Roos, C., Scott, A., Swetnam, T., Der Werf, G.
and Pyne, S. 2009. Fire in the Earth system. Science, 324, 481–484.
Chen, C., Park, T., Wang, X., Piao, S., Xu, B., Chaturvedi, R., Fuchs, R., Brovkin, V.,
Ciais, P., Fensholt, R., Tømmervik, H., Bala, G., Zhun, Z., Nemani, R. and Myneni, R.
2019. China and India lead in greening of the world through land-use management.
Nature Sustainability, 2(128), p.122–129, https://doi.org/10.1038/s41893-019-0220-7.
Cheng, H.-Q. and Chen, J.-Y. 2017. Adapting cities to sea level rise: A perspective from
Chinese deltas. Advances in Climate Change Research, 8(2017), 130-136.
Christiansen, E. and Bakke, A. 1988. The spruce bark beetle of Eurasia. In: Berryman,
A.A. (ed.) Dynamics of forest insect populations, patterns, causes, and implications.
Plenum Press, New York, NY, pp. 480–505.
EEA, 2016a. EEA Report No 5/2016, European forest ecosystems, State and trends.
European Environment Agency (doi:10.2800/964893).
EEA, 2016b. Mapping and assessing the condition of Europe's ecosystems: progress and
challenges, EEA Technical report No 3/2016, European Environment Agency.
EEA, 2017. EEA Report No 1/2017, Climate change, impacts and vulnerability in Europe
2016. An indicator-based report. European Environment Agency
(doi:10.2800/534806).
Elkinton, J.S. and Liebhold, A.M. 1990. Population dynamics of gypsy moth in North
America. Annual Review of Entomology, 35, 571–596.
Eronen, M. 1991. Jääkausien jäljillä. Tähtitieteellinen yhdistys Ursa, Helsinki, 271 s.
Esseen, P.A., Ehnström, B., Ericson, L. and Sjöberg, K. 1997. Boreal forests. Ecological
Bulletins, 46, 16–47.
Fält-Nardmann, J.J.J., Ruohomäki, K., Tikkanen, O.-P. and Neuvonen, S. 2018. Cold
hardiness of Lymantria monacha and L. dispar (Lepidoptera: Erebidae) eggs to extreme
winter temperatures: implications for predicting climate change impacts. Ecological
Entomology, 43, 422–430.
Flannigan, M., Stocks, B., Turetsky, M. and Wotton, M. 2009. Impacts of climate change
on fire activity and fire management in the circumboreal forest. Global Change Biology,
15, 549–560.
Food and Agriculture Organization of The United Nations (FAO), 2016. GLOBAL
FOREST RESOURCES ASSESSMENT 2015, How are the world’s forests changing?
Second edition.
FOREST EUROPE, 2015. State of Europe’s Forests 2015.
58
Ganteaume, A., Camia, A., Jappiot, M., San-Miguel-Ayanz, J., Long-Fournel, M. and
Lampin, C. 2013. A review of the main driving factors of forest fire ignition over
Europe. Environmental Management, 51, 651–662.
Global Forest Watch, 2019. Tree cover gain in Uruguay compared to other areas. Accessed
on 29.3.2019 from www.globalforestwatch.org.
Gonçalves, A.C. and Sousa, A.M.O. 2017. The fire in the Mediterranean region: a case
study of forest fires in Portugal. In: Fuerst-Bjeliš, B. (ed.) Mediterranean identities:
environment, society, culture. InTech, Rijeka, pp. 305–335.
Gregow, H., Ruosteenoja, K., Juga, I., Näsman, S., Mäkelä, M., Laapas, M. and Jylhä, K.
2011. Lumettoman maan routaolojen mallintaminen ja ennustettavuus muuttuvassa
ilmastossa (Modelling and predictability of soil frost depths of snow-free ground under
climate-model projections). Finnish Meteorological Institute, Reports 2011:5, 45 pp.
[in Finnish with English abstract]
Grodzki, W. 1999. Phytosanitary situation in Polish mountain forests in 1998 and forecast
for 1999. In: Forster, B., Knizek, M. and Grodzki, W. (eds.) Methodology of forest
insect and disease survey in Central Europe. Proceedings of the Second Workshop of
the IUFRO Working Party 7.03.10, 20–23 April 1999, Sion-Châteauneuf, Switzerland.
Swiss Federal Institute for Forest, Snow and Landscape Research, Birmensdorf,
Switzerland, pp. 30–35.
Groenemeijer, P., Vajda, A., Lehtonen, I., Kämäräinen, M., Venäläinen, A., Gregow, H.,
Becker, N., Nissen, K., Ulbrich, U., Morales Nápoles, O., Paprotny, D. and Púčik, T.
2016. Present and future probability of meteorological and hydrological hazards in
Europe. European Severe Storms Laboratory, 165 s.
Gu, H., Yu, Z., Wang, G., Wang, J., Ju, Q., Yang., C and Fan., C. 2015. Impact of climate
change on hydrological extremes in the Yangtze River Basin, China. Stoch Environ Res
Risk Assess 29:693–707, DOI 10.1007/s00477-014-0957-5
Gu, H., Yu, Z., Yang, C. and Ju, Q. 2018. Projected changes in hydrological extremes in
the Yangtze River basin with an ensemble of regional climate simulations. Water, 10,
1279; doi:10.3390/w10091279.
Hansen, M. C., P. V. Potapov, R. Moore, M. Hancher, S. A. Turubanova, A. Tyukavina,
D. Thau, S. V. Stehman, S. J. Goetz, T. R. Loveland, A. Kommareddy, A. Egorov, L.
Chini, C. O. Justice, and J. R. G. Townshend. 2013. High-Resolution Global Maps of
21st-Century Forest Cover Change. Science 342 (15 November): 850–53. Data
available on-line from:http://earthenginepartners.appspot.com/science-2013-global-
forest. Accessed through Global Forest Watch on [date]. www.globalforestwatch.org
Hantula, J., Müller, M.M. and Uusivuori, J. 2014. International plant trade associated risks:
Laissez-faire or novel solutions. Environmental Science & Policy, 37, 158–160.
Hareau, A., Hofstadter, R. and Saizar, A., 1999. Vulnerability to climate change in
Uruguay: potential impacts on the agricultural and coastal resource sectors and response
capabilities. Clim Res 12: 185–193.
Haynes, K., Allstadt, A. and Klimetzek, D. 2014. Forest defoliator outbreaks under climate
change: effects on the frequency and severity of outbreaks of five pine insect pests.
Global Change Biology, 20, 2004–2018.
Heikinheimo, O. 1920. Suomen lumituhoalueet ja niiden metsät. Metsätieteellisen
koelaitoksen julkaisuja 3. [in Finnish]
Heino, E. and Pouttu, A. 2014. Metsätuhot vuonna 2013. Metlan työraportteja, 295, 24–
25. [in Finnish]
Hlásny, T., Zajíčková, L., Turčáni, M., Holuša, J. and Sitková, Z. 2011. Geographical
variability of spruce bark beetle development under climate change in the Czech
Republic. Journal of Forest Science, 57, 242–249.
59
Hlásny, T., Trombik, J., Holuša, J., Lukášová, K., Grendár, M., Turčáni, M., Zúbrik, M.,
Tabakovič-Tošič, M., Hirka, A., Buksha, I., Modlinger, R., Kacprzyk, M. and Csóka,
G. 2016. Multi-decade patterns of gypsy moth fluctuations in the Carpathian Mountains
and options for outbreak forecasting. Journal of Pest Science, 89, 413–425.
IPCC, 2013. Climate Change 2013. The physical science basis. Contribution of Working
Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate
Change. Cambridge University Press, Cambridge, U.K., 1535 pp. [Stocker, T.F., D.
Qin, G.-K. Plattner, M. Tignor, S.K. Allen, J. Boschung, A. Nauels, Y. Xia, V. Bex and
P.M. Midgley (eds.)].
IPCC. 2018. Global Warming of 1.5 °C http://www.ipcc.ch/report/sr15/
Jalkanen, R. and Konôpka, B. 1998. Snow-packing as a potential harmful factor on Picea
abies, Pinus sylvestris and Betula pubescens at high altitude in northern Finland.
European Journal of Forest Pathology, 28, 373–382.
Jevrejeva, S., Jackson, L.P., Riva, R.E., et al., 2016. Coastal sea level rise with warming
above 2 °C. Proc. Natl. Acad. Sci. U. S. A. 113 (47), 13342–13347.
Kaakkurivaara, T., Vuorimies, N., Kolisoja, P. and Uusitalo, J. 2015. Applicability of
portable tools in assessing the bearing capacity of forest roads. Silva Fennica, 49, article
id 1239.
Kalliokoski, T., Heinonen, T., Holder, J., Lehtonen, A., Mäkelä, A., Minunno, F.,
Ollikainen, M., Packalen, T., Pelroniemi, M., Pukkala, T., Salminen, O., Schelhaas, M.-
J., Seppälä, J. and Kanninen, M. 2019. Skenaarioanalyysi metsien kehitystä kuvaavien
mallien ennusteiden yhtäläisyyksistä ja eroista. Suomen Ilmastopaneeli, Raportti
2/2019.
Kellomäki, S., Peltola H., Nuutinen T., Korhonen K. T., & Strandman, H. 2008. Sensitivity
of managed boreal forests in Finland to climate change, with implications for adaptive
management. Philosophical Transactions of the Royal Society B: Biological Sciences,
Volume 363, Issue 1501: 2339–2349. http://dx.doi.org/10.1098/rstb.2007.2204.
Kellomäki, S., Maajärvi, M., Strandman, H., Kilpeläinen, A. and Peltola, H. 2010. Model
computations on the climate change effect on snow cover, soil moisture and soil frost
in the boreal conditions over Finland. Silva Fennica, 44, 213–233.
Kellomäki, S., Strandman, H., Heinonen, T., Asikainen, A., Venäläinen, A. and Peltola,
H., 2018. Temporal and Spatial Change in Diameter Growth of Boreal Scots Pine,
Norway Spruce, and Birch under Recent-Generation (CMIP5) Global Climate Model
Projections for the 21st Century. Forests 2018, 9, 118; doi:10.3390/f9030118
Korhonen, K.T., Ihalainen, A., Ahola, A., Heikkinen, J., Henttonen, H.M., Hotanen, J.-P.,
Nevalainen, S., Pitkänen, J., Strandström, M. and Viiri, H. 2017. Suomen metsät 2009–
2013 ja niiden kehitys 1921–2013. Luke’s series focusing on natural resources and the
bioeconomy 59/2017. Natural Resources Institute Finland, Helsinki, 86 pp.
Krejzar, T. 2018. Bark beetle outbreak in the Czech Republic: challenges and solutions.
European Network INTEGRATE seminar on managing bark beetle impacts on forest,
Poland 25–27 June 2018.
Långström, B., Lindelöw, Å., Schroeder, M., Björklund, N. and Öhrn, P. 2009. The spruce
bark beetle outbreak in Sweden followinf the January-storms in 2005 and 2007.
Proceedings of a workshop on Insects and Fungi in Storm Areas. Štrbské Pleso,
Slovakia, 15–19 September 2008.
Länsstyrelsen i Västmanlands län. 2015. Skogsbranden i Västmanland 2014. En
dokumentation utgiven av Länsstyrelsen i Västmanlands län. Länstyrelsen i
Västmanlands län, Västerås, 31 pp. [in Swedish]
Lehtonen, I., Hoppula, P., Pirinen, P. and Gregow, H. 2014. Modelling crown snow loads
in Finland: a comparison of two methods. Silva Fennica, 48, article id 1120.
60
Lehtonen, I., Venäläinen, A., Kämäräinen, M., Peltola, H. and Gregow, H. 2016a. Risk of
large-scale fires in boreal forests of Finland under changing climate. Natural Hazards
and Earth System Sciences, 16, 239–253.
Lehtonen, I., Kämäräinen, M., Gregow, H., Venäläinen, A. and Peltola, H. 2016b. Heavy
snow loads in Finnish forests respond regionally asymmetrically to projected climate
change. Natural Hazards and Earth System Sciences, 16, 2259–2271.
Lehtonen, I., Venäläinen, A., Kämäräinen, M., Asikainen, A., Laitila, J., Anttila, P. and
Peltola, H. 2019. Projected decrease in wintertime bearing capacity on different forest
and soil types in Finland under a warming climate. Hydrology and Earth System
Sciences, 23, 1611–1631.
Lehtonen, I., and Jylhä, K., 2019. Tendency towards a more extreme precipitation climate
in the Coupled Model Intericomparison Project Phase 5 models. Atmospheric Science
Letters, doi:10.1002/asl.895.
Leinonen, R., Pöyry, J., Söderman, G. and Tuominen-Roto, L. 2017. Suomen
yöperhosyhteisöt muutoksessa – valtakunnallisen yöperhosseurannan keskeisiä
tuloksia 1993–2012. Baptria, 42, 74–92.
Lier, M., Korhonen,K.T., Packalen, T., Aarne, M., Tuomainen, T., Viitanen, J., Mutanen,
A., Vaahtera, E. and Hyvärinen, J. (LUKE), 2018. Finland’s forests,
https://www.luke.fi/en/natural-resources/forest/forest-resources-and-forest-
planning/finnish-forests-in-europe-in-2015/
Lilja, A., Rytkönen, A., Hantula, J., Müller, M., Parikka, P., Kurkela, T. 2011. Introduced
pathogens found on ornamentals, strawberry and trees in Finland over the past 20 years.
Agricultural and food science, 20, 74–85.
Liu, S., Zhang, Y., Zhang, Y. and Ding, Y., 2009. Estimation of glacier runoff and future
trends in the Yangtze River source region, China, Journal of Glaciology, Vol. 55, No.
190.
Luomaranta, A., K. Ruosteenoja, K. Jylhä, H. Gregow, J. Haapala, and A. Laaksonen,
2014. Multimodel estimates of the changes in the Baltic Sea ice cover during the present
century. Tellus A, 66, 22617, doi: http://dx.doi.org/10.3402/tellusa.v66.22617.
Marini, L., Økland, B., Jönsson, A.M., Bentz, B., Carroll, A., Forster, B., Grégoire, J.-C.,
Hurling, R., Nageleisen, L.M., Netherer, S., Ravn, H.P., Weed, A. and Schroeder, M.
2017. Climate drivers of bark beetle outbreak dynamics in Norway spruce forests.
Ecography, 40, 1426–1435.
McManus, M. and Csóka, G. 2007. History and impact of gypsy moth in North America
and comparison to recent outbreaks in Europe. Acta Silvatica & Lignaria Hungarica, 3,
47–64.
Mei, L., Xue, Y., de Leeuw, G., Guang, J., Wang, Y., Li, Y., Xu, H., Yang, L., Hou, T.,
He, X., Wu, C., Dong, J. and Chen, Z. 2011. Integration of remote sensing data and
surface observations to estimate the impact of the Russian wildfires over Europe and
Asia during August 2010. Biogeosciences, 8, 3771–3791.
Migliavacca, M., Dosio, A., Camia, A., Hobourg, R., Houston-Durrant, T., Kaiser, J.W.,
Khabarov, N., Krasovskii, A.A., Marcolla, B., San Miguel-Ayanz, J., Ward, D.S. and
Cescatti, A. 2013. Modeling biomass burning and related carbon emissions during the
21st century in Europe. Journal of Geophysical Research: Biogeosciences, 118, 1732–
1747.
Munier, S., Carrer, D., Planque, C., Camacho, F., Albergel, C. and Calvet, J.-C., 2018.
Satellite Leaf Area Index: Global Scale Analysis of the Tendencies Per Vegetation Type
Over the Last 17 Years. Remote Sens., 10, 424; doi:10.3390/rs10030424.
Müller, M.M., Henttonen, H.M., Penttilä, R., Kulju, M., Helo, T. and Kaitera, J. 2018.
Distribution of Heterobasidion butt rot in northern Finland. Forest Ecology and
Management, 425, 85–91.
61
Müller, M.M., Sievänen, R., Beuker, E., Meesenburg, H., Kuuskeri, J., Hamberg, L. and
Korhonen, K. 2014. Predicting the activity of Heterobasidion parviporum on Norway
spruce in warming climate from its respiration rate at different temperatures. Forest
Pathology, 44, 325–336.
Neuvonen, S., Kullberg, J., Kämäräinen, M., Lehtonen, I., Nevalainen, S., Siljamo, P. and
Venäläinen, A. 2018. Havununna ja lehtinunna – tulevaisuuden metsätuholaisiin on
syytä varautua ennakolta. Luonnonvara- ja biotalouden tutkimus, 44/2018, 30–39. [in
Finnish]
Nevalainen, S., Sirkiä, S., Peltoniemi, S. and Neuvonen, S. 2015. Vulnerability to pine
sawfly damage decreases with site fertility but the opposite is true with Scleroderris
canker damage; results from Finnish ICP Forests and NFI data. Annals of Forest
Science, 72, 909–917.
Nykänen, M.-L., Peltola, H., Quine, C.P., Kellomäki, S. and Broadgate, M. 1997. Factors
affecting snow damage of trees with particular reference to European conditions. Silva
Fennica, 31, 193–213.
Petty, J.A. and Worrell, R. 1981. Stability of coniferous tree stems in relation to damage
by snow. Forestry, 54, 115–128.
Piao, S., Ciais, P., Huang, Y., Shen, Z., Peng, S., Li, J., Zhou, L., Liu, H., Ma., Y., Ding,
Y., Friedlingstein, P., Liu, C., Tan, K., Yu, Y., Zhang, T. and Fang, J., 2010. The
impacts of climate change on water resources and agriculture in China. Nature 467, 43-
51, doi:10.1038/nature09364.
Pohjankukka, J., Riihimäki, H., Nevalainen, P., Pahikkala T., Ala-Ilomäki, J., Hyvönen,
E, Varjo, J. and Heikkonen, J. 2016. Predictability of boreal forest soil bearing capacity
by machine learning. Journal of Terramechanics, 68, 1–8.
Pouttu, A. and Annila, E. 2010. Kirjanpainajalla kaksi sukupolvea kesällä 2010.
Metsätieteen aikakauskirja, 4/2010, 521–523.
Pukkala, T., Möykkynen, T., Thor, M., Rönnberg, J. and Stenlid, J. 2005. Modeling
infection and spread of Heterobasidion annosum in even-aged Fennoscandian conifer
stands. Canadian Journal of Forest Research, 35, 74–84.
Reyer, C., Lasch-Born P., Suckow F., Gutsch M., Murawski A., & Pilz T. 2014.
Projections of regional changes in forest net primary productivity for different tree
species in Europe driven by climate change and carbon dioxide. Annals of Forest
Science, Volume 71, Issue 2: 211–225. http://dx.doi.org/10.1007/s13595-013-0306-8
Reyer C., Bathgate S., Blennow K., Borges JG., Bugmann H., Delzon S., Faias S.P.,
Garcia-Gonzalo J., Gardiner, B., Gonzalez-Olabarria J.R., Gracia C., Hernández J.G.,
Kellomäki, S., Kramer K., Lexer M.J., Lindner, M., van der Maaten, E., Maroschek M.,
Muys B., Nicoll B., Palahi M., Palma J. HN., Paulo J.A., Peltola H., Pukkala T.,
Rammer W., Ray D., Sabaté S., Schelhaas M.J., Seidl R., Temperli C., Tomé M.,
Yousefpour R., Zimmermann N.E., Hanewinkel M. 2017a. Are forest disturbances
amplifying or cancelling out climate change-induced productivity changes in European
forests? Environmental Research Letters 12(3), 034027.
Reyer, C., Adams, S., Albrech, T., Baarsch, F., Boit, A., Trujillo, N., Cartsburg, M.,
Coumou, D., Eden, A., Fernandes, E., Langerwisch, F., Marcus, R., Mengel, M., Mira-
Salama, D., Perette, M., Pereznieto, P., Rammig, A., Reinhard, J., Robinson, A., Rocha,
M., Sakshewski, B., Schaeffer, M., Schleussner, C.-F., Serdeczny, O. and Thornicke,
K., 2017b. Climate change impacts in Latin America and the Caribbean and their
implications for development. Reg Environ Change (2017) 17:1601–1621, DOI
10.1007/s10113-015-0854-6.
Ruosteenoja, K., Räisänen, J., and Pirinen, P. 2011. Projected changes in thermal seasons
and the growing season in Finland. International Journal of Climatology, 31, 1473–
1487.
62
Ruosteenoja, K., 2011. Miten ja miksi ilmasto muuttuu? Ilmastonmuutos käytännössä –
hillinnän ja sopeutumisen keinoja, A. Virtanen and L. Rohweder, Eds., Gaudeamus,
chap. 3, 69–108.
Ruosteenoja, K., Jylhä, K., Kämäräinen, M., 2016a. Climate projections for Finland under
the RCP forcing scenarios. Geophysica, 51, 17–50.
Ruosteenoja, K., J. Räisänen, A. Venäläinen, and M. Kämäräinen, 2016b. Projections for
the duration and degree days of the thermal growing season in Europe derived from
CMIP5 model output. International Journal of Climatology, 36, 3039-3055,
doi:10.1002/joc.4535.
Ruosteenoja, K., Markkanen, T., Venäläinen, A., Räisänen, P. and Peltola, H., 2018.
Seasonal soil moisture and drought occurrence in Europe in CMIP5 projections for the
21st century. Climate Dynamics, 50, 1177-1192, doi:10.1007/s00382-017-3671-4.
Ruosteenoja, K., A. Venäläinen, and T. Vihma, 2019. Projected changes in European and
North Atlantic seasonal wind climate derived from CMIP5 simulations. Submitted to
journal.
Schelhaas, M.-J., Nabuurs, G.-J. and Schuck, A. 2003. Natural disturbances in the
European forests in the 19th and 20th centuries. Global Change Biology, 9, 1620–1633.
Schelhaas, M.-J., 2008. Impacts of natural disturbances on the development of European
forest resources: application of model approaches from tree and stand levels to large-
scale scenarios. Dissertationes Forestales, 56, 28 p. https://doi.org/10.14214/df.56
Schlyter, P., Stjernquist, I., Bärring, L, Jönsson, A.M. and Nilsson, C. 2006. Assessment
of the impacts of climate change and weather extremes on boreal forests in northern
Europe, focusing on Norway spruce. Climate Research, 31, 75–84.
Schroeder, L.M. and Eidmann, H.H. 1993. Attacks of bark- and wood-boring Coleoptera
on snow-broken conifers over a two-year period. Scandinavian Journal of Forest
Research, 8, 257–265.
Shen, Y., Wang, G., Wang, G., Pu, J. and Wang, X., 2009. Impacts of climate change on
glacial water resources and hydrological cycles in the Yangtze River source region, the
Qinghai-Tibetan Plateau, China: A Progress Report. Sciences in Cold and Arid
Regions, , 1(6): 0475–0495
Seidl, R., Rammer, W., Jäger, D. and Lexer, M.J. 2008. Impact of bark beetle (Ips
typographus L.) disturbance on timber production and carbon sequestration in different
management strategies under climate change. Forest Ecology and Management, 256,
209–220.
Seidl, R., Schelhaas, M.-J., Lindner, M. and Lexer, M. 2009. Modelling bark beetle
disturbanes in a large scale forest scenario model to assess climate change impacts and
evaluate adaptive management strategies. Regional Environmental Change, 9, 101–
109.
Seidl, R., Schelhaas, M.-J. J., Rammer, W., and Verkerk, P. J., 2014. Increasing forest
disturbances in Europe and their impact on carbon storage. Nature Climate Change,
4(9), 806–810. https://doi.org/10.1038/nclimate2318
Siitonen, J. and Pouttu, A. 2014. Kirjanpainajatuhot Rörstrandin vanhojen metsien
suojelualueella sekä ympäröivissä talousmetsissä Sipoossa. Metsätieteen aikakauskirja,
3/2014, 183–193.
Sirén, M., Ala-Ilomäki, J., Mäkinen, H., Lamminen, S. and Mikkola, T. 2013. Harvesting
damage caused by thinning of Norway spruce in unfrozen soil. International Journal of
Forest Engineering, 24, 60–75.
Solantie, R. 1994. Effects of weather and climatological background on snow damage of
forests in Southern Finland in November 1991. Silva Fennica, 28, 203–211.
63
Svenson, G. and Fjeld, D. 2016. The impact of road geometry and surface roughness on
fuel consumption of logging trucks. Scandinavian Journal of Forest Research, 31, 526–
536.
Turco, M., Rosa-Cánovas, J.J., Bedia, J., Jerez, S., Montávez, J.P. Llasat, M.C. and
Provenzale, A. 2018. Exacerbated fires in Mediterranean Europe due to anthropogenic
warming projected with non-stationary climate-fire models. Nature Communications,
9, article number 3821.
Taylor, K. E., R. J. Stouffer, and G. A. Meehl, 2012. An overview of CMIP5 and the
experiment design. Bulletin of the American Meteorological Society, 93, 485-498,
doi:10.1175/BAMS-D-11-00094.1.
Valinger, E., Lundqvist, L. and Brandel, G. 1994. Wind and snow damage in a thinning
and fertilisation experiment in Pinus sylvestris. Scandinavian Journal of Forest
Research, 9, 129–134.
Van Wagner, C.E. 1987. Development and structure of the Canadian Forest Fire Weather
Index System. Forestry Technical Report 35, Canadian Forest Service, Ottawa, ON, 37
pp.
van Vuuren, D. P., et al., 2011. The representative concentration pathways: an overview.
Climatic Change, 109, 5-31, doi:10.1007/s10584-011-0148-z.
Veijalainen, N., K. Ruosteenoja, J. Uusikivi, A. Mäkelä, and B. Vehviläinen, 2018.
Ilmastonmuutos ja virtaamien muuttuminen Kemi-, Kymi- ja Lieksanjoen alueilla
(climate change and changes in discharges in Kemi-, Kymi- and Lieksanjoki areas).
Tech. Rep. 27/2018, Finnish Environment Institute, 43 pp.
Venäläinen, A., Tuomenvirta, H., Heikinheimo, M., Kellomäki, S., Peltola, H., Strandman,
H. and Väisänen, H. 2001. Impact of climate change on soil frost under snow cover in
a forested landscape. Climate Research, 17, 63–72.
Viiri, H., Ahola, A., Ihalainen, A., Korhonen, K.T., Muinonen, E., Parikka, H. and
Pitkänen, J. 2011. Kesän 2010 ukkosmyrskyt ja niistä seuraava hyönteistuhoriski.
Metsätieteen aikakauskirja 3/2011, 221–225.
Virtanen, T., Neuvonen, S., Nikula, A., Varama, M. and Niemelä, P. 1996. Climate change
and the risks of Neodiprion sertifer outbreaks on Scots pine. Silva Fennica, 30, 169–
177.
Vivchar, A. 2011. Wildfires in Russia in 2000–2008: estimates of burnt areas using the
satellite MODIS MCD45 data. Remote Sensing Letters, 2, 81–90.
Voolma, K., Nilson, T. and Pilt, E. 2014. Okkalainelane Euroopas, Saaremaal ja
satelliidipildil. Eesti Mets, 2, 38–43. [in Estonian]
Wallenius, T. 2011. Major decline in fires in coniferous forests – reconstructing the
phenomenon and seeking for the cause. Silva Fennica, 45, 139–155.
Wang, S. and Zhang, Z., 2011. Effects of climate change on water resources in China.
Clim Res, 47, 77-82. doi: 10.3354/cr00965.
Wang, Q.-S., Pan, C.-H. and Zhang G..Z., 2018. Impact of and adaptation strategies for
sea-level rise on Yangtze River Delta. Advances in Climate Change Research 9, 154-
160. doi.org/10.1016/j.accre.2018.05.005.
Wang, G., Cai, W., Gan, B., Wu, L., Santoso, A., Lin, X., Chen, Z. and McPhade, M.,
2017. Continued increase of extreme El Niño frequency long after 1.5 °C warming
stabilization, Nature Climate Change 7, p. 568, DOI: 10.1038/NCLIMATE3351.
Weseloh, R.M. 2003. People and the gypsy moth: a story of human interactions with an
invasive species. American Entomologist, 49, 180–190.
WMO, 1989. Calculation of monthly and annual 30-year standard normals. World
Meteorological Organization, Geneva, 11 pp., WCDP-No. 10, WMOTD/ No. 341.
64
Woodward, S., Stenlid, J., Karjalainen, R. and Hüttermann, A. 1998. Heterobasidion
annosum. Biology, ecology, impact and control. CAB International, Wallingford, 589
pp.
Wypych, A., Sulikowska, A., Ustrnul, Z., and Czekierda, D. 2017. Variability of growing
degree days in Poland in response to ongoing climate changes in Europe. International
Journal of Biometeorology, 61, 49–59.
Yang, H., Yang, S., Xu, K., Wu, H., Shi, B., Zhu, Q., Zhang, W. and Yang, Z., 2017.
Erosion potential of the Yangtze delta under sediment starvation and climate change.
Scientific reports, 7:10535. DOI:10.1038/s41598-017-10958-y.
Yu, Z., Gu, H., Wang, J., Xia, J. and Lu, B., 2018. Effect of projected climate change on
the hydrological regime of the Yangtze River basin, China. Stoch Environ Res Risk
Assess, 32:1-16. Doi:10.1007/s00477-017-1391-2.
Zubizarreta-Gerendiain A, Pukkala T, Peltola H., 2017. Effects of wind damage on the
optimal management of boreal forests under current and changing climatic conditions.
Canadian Journal of Forest Research 47(2):246-256. doi:10.1139/cjfr-2016-0226.
Öhrn, P., Långström, B., Lindelöw, Å. and Björklund, N. 2014. Seasonal flight patterns of
Ips typographus in southern Sweden and thermal sums required for emergence.
Agricultural and Forest Entomology, 16, 1–23.
65
Appendix A: Technical information about the model data and the analysis methods
A.1 Climate model data
In this report, estimates for future climatic changes were inferred from simulations
performed with 28 global climate models (Table A1). The output data of the models have
been downloaded from the international data archive hosted by the Coupled Model
Intercomparison Project (CMIP) (Taylor et al., 2012). The model ensemble is nearly the
same that has been utilized in the elaboration of the Fifth Assessment Reports of the
Intergovernmental Panel on Climate Change (IPCC). More detailed information about the
models is available in Table 9.A.1 of IPCC (2013).
Table A1. Global climate models used in creating the climate projections. The first and
second columns state the model acronym and the country of origin; the EC-EARTH model
has been developed by a consortium of multiple European countries. The remaining
columns show the availability of model data under the RCP4.5 scenario for temperature,
precipitation and incident solar radiation (T+P+S), relative humidity (RH), near-surface
soil moisture (SOILM), geostrophic wind (VG), daily standard deviations of temperature
and precipitation and precipitation indices (STDEV) and the growing season (GROW).
Model
Country
T+P+S
RH
SOILM
VG
STDEV
GROW
MIROC5
Japan
X
X
X
X
X
X
MIROC-ESM
Japan
X
X
X
X
X
X
MIROC-ESM-CHEM
Japan
X
X
X
X
MRI-CGCM3
Japan
X
X
X
X
X
X
BCC-CSM1-1
China
X
X
X
X
X
X
INMCM4
Russia
X
X
X
X
X
X
NorESM1-M
Norway
X
X
X
X
X
X
NorESM1-ME
Norway
X
X
X
HadGEM2-ES
U.K.
X
X
X
X
X
X
HadGEM2-CC
U.K.
X
X
X
X
X
X
MPI-ESM-LR
Germany
X
X
X
X
X
MPI-ESM-MR
Germany
X
X
X
X
X
CNRM-CM5
France
X
X
X
X
X
X
IPSL-CM5A-LR
France
X
X
X
X
X
X
IPSL-CM5A-MR
France
X
X
X
X
X
X
CMCC-CM
Italy
X
X
X
X
X
CMCC-CMS
Italy
X
X
X
X
X
GFDL-CM3
U.S.A.
X
X
X
X
X
X
GFDL-ESM2M
U.S.A.
X
X
X
X
X
X
GISS-E2-R
U.S.A.
X
X
X
X
GISS-E2-H
U.S.A.
X
X
X
NCAR-CCSM4
U.S.A.
X
X
X
X
X
NCAR-CESM1-CAM5
U.S.A.
X
X
X
NCAR-CESM1-BGC
U.S.A.
X
X
X
CanESM2
Canada
X
X
X
X
X
X
ACCESS1-0
Australia
X
X
X
X
X
X
ACCESS1-3
Australia
X
X
X
EC-EARTH
Europe
X
X
X
X
The future evolution of the time-mean surface air temperature, precipitation and solar
radiation was simulated by all of the 28 models. For the other quantities examined, by
contrast, data were not available from the entire model ensemble. Accordingly, near
surface soil moisture was analyzed from 26 models and relative humidity and the growing
season length from 23 models. Calculating the temporal variations of temperature and
precipitation and the precipitation indices requires data at a daily level, which was
available from 21 models only (Table A1). Daily data was likewise needed in constructing
wind speed projections.
66
The true near-surface wind speeds are influenced, in addition to meteorological
conditions, by the properties of the underlying surface. In forests, for instance, winds are
consistently far weaker than over an open land or sea. In some models, the modelled
surface conditions vary quite arbitrarily in the course of the model run, and thereby, it is
hard to obtain reliable projections for wind speed changes directly from the model-
simulated wind data. Therefore, we have focused on the so-called geostrophic wind speeds
that can be derived from the horizontal distribution of surface pressure. Geostrophic winds
are relevant only in extratropical latitudes, and they mainly reflect winds related to large-
scale weather systems.
Model simulations were forced by the observation-derived “historical” greenhouse-
gas concentrations up to the year 2005, after which the concentrations were adopted from
the selected RCP scenario. We analyzed model output data up to the year 2099 or 2100,
which constitute the termination years for most of the model runs.
Many climate models provide data from multiple parallel runs for the individual RCP
scenarios. Parallel runs are forced by identical greenhouse gas and aerosol concentrations
but the initial conditions employed in the run diverge. Using data from parallel runs helps
to reduce the uncertainty induced by internal natural variability of the climate system.
Information about the parallel-run data available for the 28 models is given in Table 1 of
Ruosteenoja et al. (2016).
A.2 Processing of the model output
The computational grid varies among the 28 climate models. Therefore, prior to
calculating the multi-model statistics, all model data were interpolated bi-linearly onto a
common 2.5 * 2.5 degree latitude-longitude grid. Climate change projections are given for
tridecadal periods. The 30-year averaging period, consistent with the recommendations of
WMO (1989) to calculate climatological standard normals, is a reasonable compromise;
such a period rather representatively contains years with different weather conditions, but
climate does not change too much within the period.
Future changes in the climate variables were calculated relative to the baseline-period
1971–2000 mean. In calculating the multi-model means and standard deviations for the
simulated changes, all 28 models were weighted equally, with the exception that no
individual research centre was given more than two “votes”. Accordingly, halved weight
coefficients were given for MIROC-ESM, MIROC-ESM-CHEM, NCAR-CESM1-CAM5
and NCAR-CESM1-BGC, while the remaining models were weighted by unity. Daily data
were analyzed from a more limited set of models in which no centre is represented more
than twice (Table A1).
After calculating the standard deviations representing the scatter of the changes
simulated by the various model runs, the 90 % uncertainty intervals for the change were
calculated by using the normality approximation.
Temporal standard deviations of temperature and precipitation were first calculated
separately for all 12 calendar months from a time series containing 30 * (28 to 31) days
for the individual months. This was done separately for all 21 models using only one
parallel run. Then, changes relative to 1971–2000 were determined both in the absolute
and percentage terms, and these were used to calculate both the multi-model mean and
inter-model scatter for the STD change.
The thermal growing season begins in spring when daily mean temperature rises
“permanently” above 5°C. More precisely, after the onset date there has to be less summed
daily-mean temperature anomaly below than above this threshold temperature (see Figure
2 of Ruosteenoja et al., 2016b). The termination of the growing season in autumn is defined
analogously. The effective temperature sum (also termed growing degree days) of the
growing season is calculated by summing all the deviations above 5°C in the daily mean
67
temperature from the onset till the termination of the growing season. Growing degree
days give a reasonable picture of the thermal conditions of the growing season for plants
adjusted to live in the boreal climate zone (e.g., in Finland). For heat-loving plants growing
in warm climate, the concept is not relevant since those plants require temperatures far
higher than 5°C for their growth.
If the reader is interested in further details of the analysis methods, the following
references provide additional information: Ruosteenoja et al. (2016a) for the time-mean
changes in temperature, precipitation and solar radiation, including the multi-model mean
changes as well as the uncertainty intervals; Ruosteenoja et al. (2018) for soil moisture;
Lehtonen and Jylhä (2019) for the precipitation indices; Ruosteenoja et al. (2019) for the
geostrophic wind speeds; Ruosteenoja et al. (2016b) for growing season projections;
Veijalainen et al. (2018) for hydrological changes in selected Finnish rivers.
FINNISH METEOROLOGICAL INSTITUTE
Erik Palménin aukio 1
P.O. Box 503
FI-00560 HELSINKI
tel. +358 29 539 1000
WWW.FMI.FI
FINNISH METEOROLOGICAL INSTITUTE
REPORTS 2019:3
ISSN 0782-6079
ISBN 978-952-336-084-6 (paperback)
ISBN 978-952-336-085-3 (pdf)
https://doi.org/10.35614/isbn.9789523360853
Edita Prima Oy
Helsinki 2019
