of the area
and mean elevation for the ice-free and glaciated parts of each elevation band or grid cell. The
areas are denoted by ai and gi, and the elevations by zi and yi, for the ice-free and the ice-covered
areas, respectively. At the end of each hydrological year, the hydrological model will provide
a simulated value for the total mass balance of each glacier group within the watershed, DVa
/media/ces/ces-glacier-scaling-memo2009-01.pdf
!
!
0
20
40
60
80
100
120
140
jan feb mar apr mai jun jul aug sep okt nov des
m
m
/
m
å
n
e
d
55550 Hafslo
60
80
100
120
140
jan feb ar apr ai j j l t
m
m
/
m
å
n
e
d
55550 Hafslo 54130 Lærdal
Ensemble approach for probabilistic
hydrological projections
Catchment
/media/ces/Lawrence_Deborah_CES_2010.pdf
distribution and parameter estimation methods
The Generalized Extreme Value (GEV) distribution (Jenkinson, 1955) is adopted to model the
flood frequency distribution at each site, from the AMF series:
Qi(D;T ) =
ei +
ai
ki
(1 [ ln(1 1=T )]ki) if ki 6= 0
ei ailn( ln(1 1=T )) if ki = 0
(6)
where ei is the location parameter, ai is the scale parameter and ki is the shape parameter.
The method of probability
/media/vedurstofan/utgafa/skyrslur/2014/VI_2014_001.pdf
nostaa,
m utta
oh ijuoksutu
ksia tulee
Sähkönvas tus kasvaa
-> energ iahäv iö ita
Muuntajien
elinikä lyhenee
Jääkannen
m uodostaminen
h idastuu
Very likely,
the
probab ility
that the next
decade is
warmer is
90% .
Ilm iö
1 .1 – korkeammat
läm pötilat etenkin talvella
Skenario
1. Lä mpimäpi i lmasto
O ma luokitteluOma luokit te lu
Nykyiset t ai
tuleva t
varautumiskahdo
llisuudet
/media/ces/Keranen_Jaana_CES_2010.pdf
of
complex interdependencies, the effort to solve one aspect may
create other problems.
Complex problem:
A problem with many relationships between parts that give rise to
collective behaviour of the system.
Complex system approach
A broad term encompassing a research approach to problems in
many diverse disciplines including computer science, AI, biology,
sociology, etc.
Common elements
/media/loftslag/Kok_1-scenarios-lecture-1.pdf
cia
l
sy
ste
m
):
th
re
e
m
ai
n
gr
ou
ps
o
fthreat
s
ar
e
at
th
e
scene
:
cl
im
at
e,
so
ci
o
-e
co
n
o
m
ic
s
an
d
gover
nanc
e
1:
M
ai
nl
y
ep
ist
em
ic
pl
us
som
e
onto
logi
ca
l
2:S
cenari
o
Pa
rti
ci
pa
to
ry
pr
oc
es
s
(PP
)
imp
ortan
tt
o
de
al
w
ith
un
certaint
y
Pa
rti
cip
at
o
ry
pr
o
ce
ss
(PP
)im
po
rta
n
tt
o
de
al
w
ith
u
n
ce
rta
in
ty
:
Pa
rti
ci
pa
to
ry
go
al
se
tti
ng
:c
o
n
/media/loftslag/VanderKeur_etal-2008-Uncertainty_IWRM-WARM.pdf
distribution and parameter estimation method
The Generalized Extreme Value (GEV) distribution (Jenkinson, 1955) was adopted to model the
flood frequency distribution from the AMF series:
Qi(D;T ) =
ei +
ai
ki
(1 [ ln(1 1=T )]ki) if ki 6= 0
ei ailn( ln(1 1=T )) if ki = 0
(7)
where ei is the location parameter, ai is the scale parameter and ki is the shape parameter.
The method of probability weighted
/media/vedurstofan/utgafa/skyrslur/2015/VI_2015_007.pdf
model errors discussed in the previous section, a model time-series
of 2-m temperature or 10-m wind speed, Mi(t), interpolated to the i-th station location, can be
linearly transformed such that the mean square error compared with the local station time-series
is minimised. Generally, the corrected time-series is then given by
M˜i(t) = ai Mi(t)+bi ; (2)
where at each station location, the correction
/media/vedurstofan/utgafa/skyrslur/2014/VI_2014_005.pdf
Simulatio
n
mode
l
(scientific
)
Im
pr
o
ve
m
o
de
l
(quality
)
Fa
rm
er
s,
Fa
rm
busines
s
advisor
sNO
P
IN
D
Interview
s
NO
P
IN
D Qu
es
tio
n
n
ai
re
–
verificatio
n
NO
P
OT
:Researcher
s
M
KA/
M
M
(re
nements
)
KA
M
Has
e
I
Simulatio
n
mode
l
(scientific
;
discussio
n
su
pp
o
rt
)
Im
pr
o
ve
m
o
de
l
(quality
,
acceptance
)
Farmers
,loca
l
authorities
,
association
s
IN
D Reques
t
fo
rdat
a
N
/media/loftslag/Hare-2011-ParticipatoryModelling.pdf