AN ENSEMBLE OF REGIONAL
CLIMATE CHANGE SCENARIOS FOR
THE NORDIC COUNTRIES
Erik Kjellström, Martin Drews, Jens
Hesselbjerg Christensen, Jan Erik Haugen,
Hilde Haakenstad and Igor Shkolnik
A changing climate in the Nordic region
Climate change in Northern Sweden:
Comparing 2071-2100 vs 1961-1990 (SRES A1B)
Lind & Kjellström, 2008
A changing wind climate in the Nordic
region?
DJF MAM JJA SON
/media/ces/Kjellstrom_Erik_CES_2010.pdf
st
c
o
ve
r
(%
)
8 x 8
y = -17.1Ln(x) + 67
R2 = 0.82
F
o
re
st
c
o
ve
r
(%
)
44
y = -4.2x + 65
R2 = 0.94
30
40
50
6
0 2 4 6
l ti it
F
o
re
st
c
o
ve
r
(%
)
2 2
Hypothetical aggregation error
by upscaling non-linear relationships
Observed from hypothetical exampleTheo etical under inning (Rastetter, 1992)
Spatial scale – Dominant cells
Conclusions - scale
• “Scale” has been on the (land use
/media/loftslag/Kok_1-scenarios-lecture-1.pdf
-11:30
Practical examples + conclusions
• Exploratory scenario development – SAS approach
• Group model building - Fuzzy Cognitive Maps
• Normative scenario development - Backcasting
Conclusions
LECTURE 2
Scenario development
In practice
Content
Lecture 2: scenario development in practice
•Story-And-Simulation approach
•Fuzzy Cognitive Mapping
•Backcasting
A Project goal - exploration vs/media/loftslag/Kok_2-scenarios-lecture-2.pdf
Manage- Basic Purchase Road Road net- Const- Acquisi- Govern- Com- Participati- Value Miscel- Total 2009
ment and road of equip- system work post- ruction tion of ment missi- on of the added laneous cost
operating manage- ment develop- poned-, of land grants ons European tax over- level
costs ment ment comprehen- build- and com- Regional De- heads
1991- 1991- sive- and ings pensation
/media/loftslag/Traffc-maintenance_expenditures.pdf
of winter mean
temperature in Helsinki: 1961-2008 vs. 2010
Only best (multi-model average) estimates are shown here
Shift to right:
mild winters more
common, cold winters
less common
Narrower distribution
with a higher peak:
a slight decrease in
variability, i.e., largest
warming of coldest
winters
-12 4
Note: probability ~ area
How frequently at least as
mild as in winter 2007-08
(+1.3°C)?
1961
/media/ces/RaisanenJouni_CES_2010.pdf
228± 4 224± 3 216± 2 212± 1 209± 0.5 206± 0.5
Vs MI [km3] 71.3 66.8 63.8 57.7 56.8 55.0 54.4
Vs MII [km3] 71.3 64.7 63.7 57.9 57.1 55.3 54.7
Table 4. Mean annual temperature (T ) at Hólar in Hornafjörður and
precipitation (P ) at Fagurhólsmýri, averaged over the years from t1
to t2. Vo is the volume at the year t2, from Table 3. Vs is the volume
of a stable spin-up glacier corresponding
/media/ces/Adalgeirsdottir-etal-tc-5-961-2011.pdf
the model. Usually, the model
is run within a graphical user interface (GUI) where many
tasks have been automated.
STEP 4 (calibration and validation). This step is con-
cerned with the process of analysing the model that was
constructed during the previous step, first by calibrating
the model, and then by validating its performance against
independent field data. Finally, the reliability of model
/media/loftslag/Refsgaard_etal-2007-Uncertainty-EMS.pdf
% of respondents have felt this
way for a number of years. While comparable studies on both sides of the At-
lantic indicate that on average people have negative feelings about the images they
associate with climate change (see Lorenzoni et al., 2006), an OST/MORI survey
in the UK (2004)2 also found that 62% of respondents described climate change
as a “fairly bad thing” or a “very bad thing”’ (vs. 10% who
/media/loftslag/Lorenzoni_Pidgeon_2006.pdf