Geoscience Reference
In-Depth Information
grow to reach several metres in height and thickness. This process happens very slow and
lasts many years to decades (French 1990 ). Ice wedges (and other ice bodies) are the reason
for the often oversaturated ice contents in permafrost soils and can be seen as long-term
storages of both energy and water in the climate system.
However, the presence of ice wedges may also lead to abrupt changes that occur on the
landscape scale when large ice bodies in the soil, after a period of relatively slow and
constant warming, collapse in sudden events, e.g., due to intense rain events that bring high
heat inputs into the ground. This might lead to considerable change in the landscape in the
form of severe soil subsidence, opening of channels and coastal erosion. Belonging to these
phenomena are the so-called thermokarst lakes that develop when formerly stable per-
mafrost thaws at the top due to perturbation (e.g., a fire event), and soil subsidence and
melting ground ice lead to the formation of a lake (see, e.g., French 2007 ). This describes a
particular process of wetland formation (see also Sect. 3.3 ). On the other hand, these
thermokarst lakes may also drain by catastrophic outflow following lake tapping due to the
expansion of adjacent basins or truncation by coastal retreat (see, e.g., Mackay 1988 ;
Walker 1978 ; Romanovskii et al. 2000 ). Cycles of slow build-up of ice masses in the
ground and relatively short-term collapses in conjunction with the implied morphological
changes have happened ever since. Yet, since the atmosphere is warming, and since the
atmospheric moisture transport from mid- to high northern latitudes as well as precipitation
and circulation patterns is believed to change with anthropogenic climate change, these
events might become more abundant in the future. This again has implications for the
carbon cycle, as erosion events always bring formerly bound carbon back into the cycle.
3.2 Representation of Permafrost Processes in ESMs
The climate modelling community has a long history in systematic model intercomparison
through the climate model intercomparison projects (CMIPs; Meehl et al. 2000 ). Results
from CMIPs provide a good overview of the respective state of ESM model accuracy and
performance. Koven et al. ( 2012 ) analysed the performance of ESMs from the most recent
CMIP5 exercise over permafrost areas. They found that the CMIP5 models have a wide
range of behaviours under the current climate, with many failing to agree with fundamental
aspects of the observed soil thermal regime at high latitudes. This is partially related to the
fact that most of these models do not include permafrost-specific processes, not even the
most basic process of freezing and melting of soil water. Moreover, the land surface
parameterizations used in GCMs usually do not adequately resolve the soil conditions
(Walsh et al. 2005 ), which often rely on either point measurements or information derived
from satellite data.
Although a good understanding of many permafrost-related hydrological processes
exists at the point and hillslope scales, this knowledge had not been adequately or sys-
tematically incorporated even into process-based mesoscale hydrological models
(V ¨r ¨smarty et al. 1993 ) for a long time. Models on point/hillslope scales were generally
constrained to one-dimensional domains of vertical extent only (Riseborough et al. 2008 ),
which usually could not be upscaled to larger scales due the complexity of physical
interactions in permafrost regions. Also, Bolton ( 2006 ) has identified a lack of process-
based hydrology models that adequately simulate the soil moisture dynamics at the
watershed scale and also include a realistic land-atmosphere exchange in permafrost-
dominated regions. But, such models are required to bridge the gap between the point/
hillslope scale understanding and the scale of RCMs and GCMs by capturing the hydro-
logical behaviour and variation in individual watersheds. Recent developments started to
Search WWH ::




Custom Search