Jökull - 01.01.2009, Qupperneq 8

Guðmundsson et al.
Figure 4. Comparison of physi-
cal and empirical melt rate mod-
els for at G500 during the sum-
mer 2001. (a): EBM, DDM1 and
DDM2. (b): Residuals. The val-
ues in (a-b) have been filtered
by a three-day moving average.
(c): Correlation of daily values
of EBM with DDM1 and DDM2.
(d): Correlation of daily values of
Mc with R and Hd + Hl. The
daily correlation in (c-d) is cal-
culated separately for each of the
months from May to September.
– Árangur reynslubundinna lík-
ana metinn með orkubúskapar-
líkani.
latter part of the ablation season (Figure 4d). Our data
include two years with high melting rates in Septem-
ber related to spells of strong wind speeds and high
temperatures (Figure 3).
In May and early June, temperatures typically
fluctuated close to 0 !C at G1100, but some abla-
tion was caused by solar radiation penetrating into the
glacier surface despite the high albedo. During the
summers 2001–2005, the albedo declined rapidly at
the lower station (G500), until reaching an approx-
imately constant value of 0.07 (Figure 3d) between
May 20 to June 11, depending on the year, when the
ice surface was exposed.
It is evident from our field work that the low
albedowas due to sand and dirt blown over the surface
that reached only to an elevation !100 m above the
8 JÖKULL No. 59