University of ZurichZurich Open Repository and Archive

Winterthurerstr. 190

CH-8057 Zurich

http://www.zora.uzh.ch

Year: 2008

The importance of snow cover evolution in rock glaciertemperature modeling

DallAmico, M; Endrizzi, S; Rigon, R; Gruber, S

DallAmico, M; Endrizzi, S; Rigon, R; Gruber, S. The importance of snow cover evolution in rock glaciertemperature modeling. In: 9th International Conference on Permafrost, Fairbanks, Alaska, 29 June 2008 - 03 July2008, 57-58.Postprint available at:http://www.zora.uzh.ch

Posted at the Zurich Open Repository and Archive, University of Zurich.http://www.zora.uzh.ch

Originally published at:9th International Conference on Permafrost, Fairbanks, Alaska, 29 June 2008 - 03 July 2008, 57-58.

DallAmico, M; Endrizzi, S; Rigon, R; Gruber, S. The importance of snow cover evolution in rock glaciertemperature modeling. In: 9th International Conference on Permafrost, Fairbanks, Alaska, 29 June 2008 - 03 July2008, 57-58.Postprint available at:http://www.zora.uzh.ch

Posted at the Zurich Open Repository and Archive, University of Zurich.http://www.zora.uzh.ch

Originally published at:9th International Conference on Permafrost, Fairbanks, Alaska, 29 June 2008 - 03 July 2008, 57-58.

57

The Importance of Snow Cover Evolution in Rock Glacier Temperature Modeling

Matteo DallAmicoDepartment of Civil and Environmental Engineering, University of Trento, Italy

Stefano EndrizziDepartment of Civil and Environmental Engineering, University of Trento, Italy

Riccardo RigonDepartment of Civil and Environmental Engineering, University of Trento, Italy

Stephan GruberDepartment of Physical Geography, University of Zurich, Switzerland

IntroductionThe snow cover evolution is one of the crucial factors

affecting the thermal and hydraulic regime of rock glaciers (Mittaz et al. 2000), as snow strongly controls soil energy balance through its high albedo and insulating properties. Therefore, accurate modeling of the snowpack is absolutely necessary to reliably describe soil temperatures. The importance of accurate snow modeling entails the use of sophisticated models based on the solution of the snow energy balance and, consequently, on a good parameterization of radiation and turbulent fluxes (e.g., Jordan 1991). An advance or delay in estimating the time of snow disappearance would cause a strong error in the calculation of the energy balance at the soil surface, altering the ground heating or freezing and, therefore, affecting the soil temperature profile for the whole summer.

The goal of this work is to simulate and discuss the rock glacier snow evolution in order to analyze the influence of the snow cover and accumulation/melting time on the temperature regime of the active layer of a rock glacier.

Modeling Features and Case StudyThe model used in the simulation is GEOtop (Rigon et

al. 2006), a distributed physically-based model which jointly solves the energy and water balance of soil (Bertoldi et al. 2006) and snow (Zanotti et al. 2004), and accounts for the geotechnical parameters of unsaturated soils affecting slope stability (Simoni et al. 2007). The model has been improved recently to include a correct treatment of frozen soil (Endrizzi et al. 2008) and to model snow with a multilayer scheme capable of describing snow metamorphism and water circulation and refreezing in the snowpack (Endrizzi 2007).

Commonly, in alpine climates the soil exchanges heat directly with the atmosphere only in a short time window, roughly spanning from June to October, whereas during winter and early spring, heat transfer between soil and atmo-sphere is mediated by the snowpack. Consequently, the heat flux reaching the soil surface is strongly reduced due to high snow albedo, which reduces net energy input, and to snow insulating properties, which cause heat conduction to be very small below the upper snow layers. In fact, the snow energy balance equation can be written as follows (Oke 1990):

where the terms in the left-hand side (LHS) represent the heat storage rate in the snowpack due to sensible heat (∆SQS) and to latent heat (∆QM, melting/refreezing and rain on snow). In the right-hand side (RHS), Rn is the net all-wave radiation, P is the sensible heat flux supplied by precipitation, H and L are, respectively, the sensible and latent heat fluxes exchanged between the surface (be it snow or soil) and the atmosphere, and G is the heat flux reaching the soil surface acting as soil energy input. When the ground is snow-free, the LHS in equation (1) is null, and G is equal to the net energy flux exchanged with the atmosphere. On the other hand, for snow covered ground, G is proportional to the temperature gradient at the snow-soil interface, namely:

where K is the snow-soil averaged thermal conductivity calculated as a harmonic mean, Tsn is snow temperature in the layer close to the soil surface, TS is the soil surface temperature, and Dsn and DS are the depths of the snow and surface layer, respectively.

Investigated siteSimulations have been carried out on the active rock

glacier Murtèl (Upper Engadin, Swiss Alps: 46°26′N, 9°49.5′E, 2670 m a.s.l., 15° slope with NW aspect) in which the oldest temperature time series of Alpine Permafrost has been measured (Vonder Mühll & Haeberli 1990, Hoelzle et al. 1999). Input data include incoming shortwave radiation (both direct and diffuse), incoming longwave radiation, air temperature, wind speed and direction, air pressure, and precipitation.

Simulations and ResultsThe simulation spans a period of two hydrological years

beginning from October 1997. As the first snowfall normally occurs in November, this choice allows the avoidance of the problem of determining the initial condition of snow on the surface. Most of the parameters used by the snow model of GEOtop were simply taken from literature, for example, snow reflectance and snow thermal and hydraulic properties. As only total precipitation was available, the calibration was reduced to the definition of the threshold air temperatures above (below), where precipitation is considered to occur as rain (snow).QS QM Rn P H L G [W/m2] (1)

Gsn K Tsn TS12 (Dsn DS )

[W/m2] (2)

QS QM Rn P H L G [W/m2] (1)

Gsn K Tsn TS12 (Dsn DS )

[W/m2] (2)

NiNth iNterNatioNal CoNfereNCe oN Permafrost

58

As can be seen in Figure 1, the model proves to simulate well both the snow depth and the time when snow is completely ablated. The heat flux reaching the soil surface clearly depends on snow presence. When soil is snow free, the flux is of the order of 50 W/m2, but it can drop by an order of magnitude or more when snow is present.

A delay (anticipation) in the estimation of the snow cover complete ablation date may lead to an underestimation (overestimation) of the ground surface temperature and of the temperature profile of the layers below. For example, in Figure 2 the temperature behavior at the soil surface and at 55 cm depth during the snow melting period is reported, considering a “proper” snow simulation (full grey line) and a “poor” delayed snow simulation (dotted grey line). The surface temperature increases as the snow is melted, and the delay between the two scenarios is disappear after few days. At 55 cm depth, instead, the delay in the temperature evolution is still visible after one month, indicating that the error in snow model will propagate and increase as we go deeper in the soil.

ConclusionsThe work shows that the model is capable of reproducing

the evolution of the snow cover and the temperatures in the active layer of the rock glacier. Snow evolution, together with the thermal and hydraulic parameters (DallAmico et al. sub-mitted), is a crucial process to take into consideration when the thermal regime of an active layer is to be modeled. A prop-er representation of the snow evolution can provide the right time window of direct soil exposure to solar radiation and, in turn, a reliable quantification of the soil energy fluxes. Con-versely, a poor representation may lead to significant errors that propagate and increase the deeper we go in the ground.

ReferencesBertoldi, G., Rigon, R. & Over, T.M. 2006. Impact of

watershed geomorphic characteristics on the energy and water budgets. J. Hydrometeorology 7: 389-403.

DallAmico, M., Endrizzi, S., Rigon, R. & Gruber, S. Submitted. Modelling the thermal regime of a rock glacier active layer using GEOtop. Proceedings of the Ninth International Conference on Permafrost, Fairbanks, Alaska, June 29–July 3, 2008.

Endrizzi, S. 2007. Snow cover modeling at local and distrib-uted scale over complex terrain. Ph.D. dissertation. Dept. of Civil and Environmental Engineering, Uni-versity of Trento, Italy.

Endrizzi, S., Rigon, R. & DallAmico, M. 2008. A soil freeze/thaw model through the soil water characteristic curve. Extended Abstracts, Ninth International Conference on Permafrost, Fairbanks, Alaska, June 29–July 3, 2008.

Hoelzle, M., Wegman, M. & Krummenacher, B. 1999. Miniature temperature dataloggers for mapping and monitoring of permafrost in high mountain areas: first experience from the Swiss Alps. Permafrost and Periglacial Processes 10: 113-124

Jordan R. 1991 A one-dimensional temperature model for a snow cover. Technical documentation for SNTHERM 89. CRREL, Hanover, NH, USA.

Mittaz, C., Hoelzle, M. & Haeberli, W. 2000. First results and interpretation of energy-flux measurements of Alpine permafrost. Annals of Glaciology 31: 275-280.

Oke, T.R. 1990. Boundary Layer Climates. Routledge.Rigon, R., Bertoldi, G. & Over, T.M. 2006. GEOtop: A

distributed hydrological model with coupled water and energy budgets. J. of Hydromet. 7: 371-388.

Simoni, S., Zanotti, F., Bertoldi, G. & Rigon, R. 2007. Modelling the probability of occurrence of shallow landslides and channelized debris flows using GEOtop-FS. Hydrological. Processes.

Vonder Mühll, D. & Haeberli, W. 1990. Thermal character-istics of the permafrost within an active rock glacier (Murtèl/Corvatsch, Grisons, Swiss Alps). Journal of Glaciology 36(123): 151-158.

Zanotti F., Endrizzi, S., Bertoldi, G. & Rigon, R. 2004. The GEOTOP snow module. Hydrological. Processes 18: 3667-3679. doi:10.1002/hyp.5794.

Figure 1. Simulated vs. measured snow depth and energy flux input to the ground in the Murtèl rock glacier.

Figure 2. The error in temperature profile depends on snow modeling and becomes bigger the deeper in the ground. “Proper” and “Poor” refer to real measures and delayed modeling, respectively