Ice-Covered Lakes Environment - Freezing of Lakes and the Evolution of their Ice Cover - page 262

After ice cover formation, the main factors controlling the ice growth rate are the snow
cover, heat

fluxes at the upper and lower boundaries, and the thickness and structure of the
accumulated ice. With this regard, an early appearance of snow cover can play a crucial
role in decelerating the ice growth due to the low heat conductivity of snow. But high
snow accumulation may lead to formation of superimposed ice that in turn tends to
increase the resulting total ice thickness. Thermal strati

fl

cation under lake ice is stable and
water velocities are weak, so that heat transfer at the ice lower boundary is normally weak.
As was shown in Sect. 4.3 , a simple formula for the ice growth is provided by a semi-
empirical law h = F(S; a*, b*), where the freezing-degree-days S is the driving force and
a* and b* are the tuning parameters. In Eq. ( 4.44 ) the theoretical values (a, b) are shown,
and in semi-empirical modi cations, to fit with observations of ice thickness, usually ½

a < a* < a and b* * b. A climatic change would change the freezing-degree-days that is

easy to evaluate, but first of all the coef cient a* is the critical parameter and sensitive to

snow accumulation.
The heat balance is fundamentally different during ice melting from that during lake
cooling and ice growth (Jakkila et al. 2009). The stable strati cation of atmospheric

surface layer strongly reduces the sensible and latent heat exchange at the ice-air interface,
and heat for melting is provided mostly by solar radiation. This fact implies also a strong
in

uence of the surface albedo and, consequently, of snow layer in ice melting. Liquid
precipitation is another major factor to deteriorate and melt the ice sheet. Solar radiation
and rain also explain the fact that the timing of ice breakup is coherent at spatial scales of
hundreds of kilometres (Magnuson et al. 2000). There is a degree of lake dependence via
ice thickness, since the timing of breakup depends on how much there is ice to melt. The
strong role of solar radiation also implies that the timing of ice breakup has much less
variability between different years than in the case of freeze-up. Melting of ice follows
spring warming and breakup date is reproducible by simple correlation models.
The sequence of an ice season, from

fl

first freezing to final break-up is not a simple cycle

but there may be melt refreeze events in between. In general, in milder climate the

-

probability of melt freeze events during ice season is larger, but this question has not

been examined in detail. Ice formation and melting are not symmetric processes but ice is
self-protecting. Once an ice cap has formed, it is dif

−

cult to melt it since solar radiation
level is low, and lake heat storage is largely insulated by weak mixing conditions in lake
water. Also albedo feedback from ice formation keeps the absorption of solar radiation
low by ice cover.
The results from Chap. 4 and Sect. 7.1 can be examined for the impact of climate
change on the ice season. The expected climate change at year 2100, as we understand it
now from the IPCC (Intergovernmental Panel for Climate Change) scenario, is predicted
to increase the air temperature by ʔ P (>0 or <0) in

the lake ice zone. 3 The scenario varies from place to place and changes with time, along

ʔ T a > 0 and to change precipitation by

3 In Finland the predicted changes in the temperature and precipitation are 3 - 5 ° C and 10 - 20 %,
respectively (Finnish Meteorological Institute web site www.fmi. , August 2014).