is the constant terrain-following temperature
lapse rate, and L is the constant atmospheric temperature lapse rate. The integral in (5) can then be
explicitly evaluated, giving
p(h(x;y);z) = p0
T0
T0 +LT h
g
LT R
T0 +LT h
T0 +LT h+Lz
g
LR
: (7)
The values of all atmospheric constants are given in Table 2. Based on the ideal gas law, air density
is then given by
r(h(x;y);z) = p(h(x;y);z)
RT (h(x;y
/media/vedurstofan/utgafa/skyrslur/2013/2013_001_Nawri_et_al.pdf
-scale Category
P
e
r
c
e
n
t
a
g
e
o
f
C
a
t
e
g
o
r
y
F
a
t
a
l
i
t
i
e
s
Permanent Homes
Casualties and Timing
Casualties and Time of Day
150
200
250
I
n
d
e
x
V
a
l
u
e
Fatalities
0
50
100
Overnight Morning Early Afternoon Late Afternoon Late Evening
I
n
d
e
x
V
a
l
u
e
Injuries
Nocturnal Tornadoes
7
8
9
10
R
a
t
i
o
N
i
g
h
t
t
o
O
t
h
e
r
T
i
m
e
s
0
1
2
3
4
5
6
F0 F1 F2 F3 F4
F
/media/loftslag/Tornado_Impacts_-_FMI_Presentation.pdf
of land origin ▲•
Undetermined or unknown x
Table 3.3
Form of ice (Fa Fb Fc Fp Fs)
Element Floe size Symbo
l
Pancake ice - 0
Small ice cake; brash ice < 2 m 1
Ice cake 2-20 m 2
Small floe 20-100 m 3
Medium floe 100-500 m 4
Big floe 500 m-2 km 5
Vast floe 2-10 km 6
Giant floe > 10 km 7
Fast ice - 8
Icebergs, growlers or
floebergs
- 9
Undetermined or
unknown
- x
- 5 -
Annex I
Sample ice charts
from
/media/hafis/frodleikur/ice-chart_colour-code-standard.pdf
is ice flux and b is mass balance. For (small) changes
in glacier geometry with respect to a datum (often steady) state, perturbations in ice thickness,
flux and mass balance will satisfy
¶(Dh)
¶t
+
¶(Dq)
¶x
= Db or
¶(Dh)
¶t
+~ (D~q) = Db : (2)
Changes in mass balance are the driving factor of glacier changes in climate change simu-
lations. If the datum glacier is initially comparatively close
/media/ces/ces-glacier-scaling-memo2009-01.pdf
was represented on the native grids of each individual
model. Therefore, the monthly means of the modelled radiation were first interpolated
onto a common 2.5 x 2.5 degree grid, and 30 year running means were applied to smooth
the influence of random interannual variability. Thereafter, anomalies from the baseline
period mean were calculated.
2
Fig. 2. Percentage change of incident global solar
/media/ces/CES_D2.4_solar_CMIP3.pdf
)+Ewi1( ˜X2) (5)
⇔ EUi1 = pi1[vi1(Xi1)+wi1(X2)] + (1 − pi1)[vi1(Xi1)+wi1(X2)] (6)
here vi1(·) represents the utility from the first mover’s own gain. We assume constant relative
risk aversion for the function vi1(·) to represent the risk preferences of agent i as mover 1:
vi( ˜Xi1)=
˜X1−rii1
1 − ri
(7)
Agent i is risk neutral if ri = 0, risk averse if ri > 0 and risk loving if ri < 0.8 Subjects
/media/loftslag/Public-Choice-2012---Teyssier---Inequity-and-risk-aversion-in-sequential-public-good-games.pdf