Ice-Covered Lakes Environment - Freezing of Lakes and the Evolution of their Ice Cover - page 263

with new model simulations. Therefore it is preferable to consider a general case: the
sensitivity of lake ice seasons to climate change.
This sensitivity to air temperature is quite clear. Precipitation is a more dif cult

question, since the ice thickness cycle depends on the phase and timing of precipitation in
addition to the total. Local snow accumulation and air temperature may have weak
connections since in longer winters more precipitation comes down in the solid phase, but
in general air temperature and snow accumulation can be considered independent pre-
dictors of ice thickness. However, indirect conditions exist in that air temperature must be
low for solid precipitation to ground and the thicker the ice the more snow is needed for
snow-ice growth. Thus, if the total ice thickness were to decrease under climate warming,
the buoyancy of ice would be less and

flooding events would become more common. In
large lakes the mobility of ice is an additional question related to wind climate.
Figure 4.11 a gives insight into the in

fl

uence of snowfall on the ice thickness. In the
model a constant snowfall rate is assumed in a lake district where the ice thickness varies
between 60 and 110 cm. If the rate of snowfall is less than 0.75 mm day − 1 , increasing
snowfall decreases the ice thickness due to insulation. At 0.25 mm day − 1 , doubling the
rate decreases the ice thickness by about 20 cm, and 20 % increase consequently would
give 4 cm decrease. But if the rate of snowfall is more than 0.75 mm day − 1 , increasing
snowfall increases the ice thickness due to snow-ice formation. At 1.0 mm day − 1 , dou-
bling the rate increases the ice thickness by about 20 cm, and 20 % increase consequently
would give 4 cm increase.

fl

8.4.2 Analytic Modelling

Analytic models were derived for the relation between temperature and ice phenology,
thickness and coverage (Sects. 4.3 and 7.1 ). In the

first order modelling approach the
surface heat balance is taken in the linear form (Sect. 4.1 ). Then a constant change ʔ T a can

be employed for the air temperature level to examine the in fl uence of an assumed climate

change. This first-order approach is quite appropriate for examining the general variations

in the ice season, since the simpli

cations in the surface heat budget are largely cancelled,
i.e. the freezing date estimate is rather crude but we get better estimates for the change of
the freezing date for a given climate change scenario. As a matter of fact, the parameters of
the linearized heat balance depend on the wind speed, humidity and cloudiness, and
therefore we can also examine the in fl uence of systematic changes in these quantities on

the ice season.
Whether a lake freezes or not, the condition for the length of the cold season (T

≤ °

C)
was given by Eq. ( 7.10 ) based on a parabolic curve for the winter air temperature. The
climate change would shift the minimum temperature up by

0

ʔ T a and then length of the

cold season would be changed to