HR: 09:45h
AN: C51C-08    [Abstracts]
TI: Influcence of Initial Conditions on Permafrost Localization
AU: Endrizzi, S
EM: Stefano.endrizzi@ing.unitn.it
AF: Dipartment of Civil and Environmental Engineering, University of Trento, Via Mesiano 77, Trento, 38050, Italy
AU: Dall'Amico, M
EM: matteo.dallamico@ing.unitn.it
AF: Dipartment of Civil and Environmental Engineering, University of Trento, Via Mesiano 77, Trento, 38050, Italy
AU: * Cordano, E
EM: emanuele.cordano@ing.unitn.it
AF: Dipartment of Civil and Environmental Engineering, University of Trento, Via Mesiano 77, Trento, 38050, Italy
AU: Gruber, S
EM: stephan.gruber@geo.uzh.ch
AF: Department of Geography, University of Zurich, Winterthurerstr. 190, Zuerich, 8057, Switzerland
AU: Rigon, R
EM: riccardo.rigon@ing.unitn.it
AF: Dipartment of Civil and Environmental Engineering, University of Trento, Via Mesiano 77, Trento, 38050, Italy
AB: Permafrost is a long-term phenomenon and often formed by freezing conditions over hundreds f year or millennia. Modeling its distribution and degradation is challenging because, due to its high thermal inertia, it responds slowly to temperature forcing. Nevertheless a proper assessment of permafrost presence is very relevant to any study which involves climate change and process oriented models have been recently applied to detect the permafrost presence in alpine areas. The objective of this paper is the localization and analysis of permafrost distribution using the model GEOtop (Rigon at al., 2006) in the case of an alpine basin. GEOtop is a distributed hydrological model that jointly solves the water and energy budgets within a catchment and considers the dependencies of temperature and water content on soil thermal and hydraulic dynamics. One of the main difficulties, besides having the appropriate model, lies in the derivation of the initial conditions of temperature and ice content in the ground. Especially during phase change, the response of temperatures to surface signals is null, and, if averaged in the area, slow. Furthermore, surface and ground temperature could be very different from air temperature, and therefore it is difficult to properly initialize the temperature profile of the model domain. In fact, an unrealistic temperature initialization may take several hundred years (of simulated time) to equilibrate with surface conditions. In this case, it becomes difficult (if not impossible) to distinguish between real transient effects represented by a model and the delayed response to an initialization that did not correspond to real conditions. Current methodologies used in climate and hydrological models tend to use simplified initial conditions of both ground temperature and ice content, e.g. constant temperature profile and zero ice content, and then spin the model for several years until a steady condition has been set. This approach, however, is highly computationally expensive and eventually does not assure to gain a realistic profile. In the present work, the initial conditions for the temperature are provided by the analytical solutions of the linearized heat equation, taking into account the mean annual air temperature (MAAT), as derived from accurate kriging of 15 years of hourly temperatures from nine meteo station, and the geothermal gradient . The initial conditions for the ice content take into account the topographical considerations and the temperature profile previously computed. The sensitivity of the model, with respect to various initial conditions, has been tested in the Adamello Presanella Group (Rhaetian Alps, Italy). The results are verified by a DC resistivity survey, GIS technique using updated glacier map with little ice age expansion and empirical topoclimatic approaches in order to infer the correlation between permafrost inferior limit and initial conditions.
DE: 0475 Permafrost, cryosphere, and high-latitude processes (0702, 0716)
DE: 0798 Modeling
SC: Cryosphere [C]
MN: 2007 Fall Meeting