Jökull - 01.01.2005, Blaðsíða 143

Mass balance of Vatnajökull outlet glaciersreconstructed back to 1958
Table 3: Percent r2 model results. Models are: 1 upper-air model, 2 deWoul and Hock (2005), 3 deRuyter
deWildt et al. (2003). For Model 1, period of record is N years, critical direction is φ′. Boldface r2 are signifi-
cant at 99 percent, italic r2 at 95 percent. Period of record for Model 2 is N except 11 years for Eyjabakkajökull,
for Model 3 was not stated. Significance levels were not stated for Models 2,3. – icelandic
Model 1 Model 2 Model 3
Glacier N φ′ bw bs bw bs bw bs
1 Tungnaárjökull 8 182◦ 82 61 55 62 74 66
2 Köldukvíslarjökull 9 217◦ 88 55 60 45 42 79
3 Dyngjujökull 7 215◦ 51 84 83 69 59 85
4 Brúarjökull 9 161◦ 41 73 49 35
5 Eyjabakkajökull 6 152◦ 93 74 41 44 18 58
W. Brúarjökull 46 67
E. Brúarjökull 42 81
riod of days over which averages are formed. If the
period of forming f
w
is extended into May, for in-
stance, it will be lower because f decreases in May
(Fig. 1); in the regression of Equation (4), however,
the coefficient αw will tend to be increased to match
the observed bw as well as possible with the smaller
f
w
. Varying z2 has little effect because T appears lin-
early in Equation (5) and T (z) is roughly linear, so
that γs will be adjusted during the regression by an
amount proportional to the product of the mean lapse
rate dT/dz and the difference in z2 to match the ob-
served bs as well as possible.
Sensitivities of calculated bw and bs to hypothet-
ical warming ∆T or increased precipitation ∆F are
shown in Table 4. For ∆F , only about 10 percent of
the increase of bn is due to bs becoming less negative.
Also shown in Table 4 are sensitivities from two other
models.
Glacier to glacier correlation r(bw) of bw
(Table 2) is consistent with the variation of
the critical direction φ′ (Table 3, Fig. 4).
Figure 4. Vatnajökull outlet glaciers 1-5 and NCEP-
NCAR Reanalysis gridpoint (open circle). Unsigned
number identifies glacier (see Table 1). Signed num-
ber is calculated 1958–2003 cumulative mass balance
∫
b∗
n
in water equivalent meters. Wind vane shows di-
rection of critical 850-hPa wind at the gridpoint, not
of local wind at the glacier. – icelandic
When the difference in φ′ between two glaciers is
small (such as between glaciers 2 and 3 or glaciers 4
JÖKULL No. 55 143