The application of Regional Climate Model output for the simulation of high-mountain permafrost scenarios

https://doi.org/10.1016/j.gloplacha.2006.07.006Get rights and content

Abstract

This study investigates the possibilities and limitations of using Regional Climate Model (RCM) output for the simulation of alpine permafrost scenarios. It focuses on the general problem of scale mismatch between RCMs and impact models and, in particular, the special challenges that arise when driving an impact model in topographically complex high-mountain environments with the output of an RCM. Two approaches are introduced that take into account the special difficulties in such areas, and thus enable the use of RCM for alpine permafrost scenario modelling. Intended as an initial example, they are applied at the area of Corvatsch (Upper Engadine, Switzerland) in order to demonstrate and discuss the application of the two approaches, rather than to provide an assessment of future changes in permafrost occurrence.

There are still many uncertainties and inaccuracies inherent in climate and impact models, which increase when driving one model with the output of the other. Nevertheless, our study shows that the use of RCMs offers new and promising perspectives for the simulation of high-mountain permafrost scenarios.

Introduction

High-mountain environments in general, and the alpine cryosphere in particular, are affected most seriously by climate change (Haeberli and Beniston, 1998, Smith and Riseborough, 2002, Diaz et al., 2003, Harris et al., 2003, Kääb et al., 2005). The thawing of alpine permafrost can have a severe impact on the stability of rock walls and debris slopes and on engineered structures (Haeberli, 1992, Haeberli et al., 1997, Davies et al., 2001, Harris et al., 2001, Kääb et al., 2005). The extremely hot and dry summer of 2003 in the European Alps led to higher incidence of rockfall events, probably from warm permafrost (Schiermeier, 2003, Gruber et al., 2004a). This might be indicative of the aforementioned permafrost sensitivity. At the same time, human settlements, infrastructural development and tourism in high-mountain areas such as the Alps are expanding, which intensifies the permafrost-related hazard potential.

In order to simulate the distribution and evolution of permafrost in topographically complex high-mountain areas, the process-based model TEBAL (Topography and Energy Balance) has been developed (Stocker-Mittaz et al., 2002, Gruber et al., 2004b). TEBAL simulates the energy fluxes between the atmosphere and the ground surface based on daily meteorological input data. It is well-suited for modelling the impact of climate change on mountain permafrost (Stocker-Mittaz et al., 2002). However, to date, alpine energy balance models such as TEBAL have only been driven with past or current climate data sets. Scenario simulations have been carried out only in the form of ad hoc scenarios with a very general use of climate model output (Stocker-Mittaz et al., 2002). The use of scenario time series from climate model output has been neglected so far for alpine permafrost modelling. Nevertheless, General Circulation Models (GCMs) have been used for impact studies of lowland permafrost (Anisimov et al., 1997, Stendel and Christensen, 2002). However, the horizontal resolution of a GCM (in the order of 300 km) is not able to describe explicitly the spatial distribution of the climate variables for a local site in areas of complex topography such as the Alps. Therefore, the technique of regional climate modelling (RCM) is required in such regions (Christensen et al., 1998, Giorgi et al., 2001, Frei et al., 2003, Vidale et al., 2003). This technique uses output from GCMs or re-analysis data (global analysis of the state of the atmosphere such as ERA-40 (Kallberg et al., 2004); Re-Analysis from the European Centre for Medium-Range Weather Forecasts (ECMWF)) to provide initial conditions and time-dependent lateral meteorological boundary conditions to drive high-resolution RCM simulations for selected time-slices of the global model run (e.g. Dickinson et al., 1989, Giorgi, 1990, Giorgi and Mearns, 1991). Giorgi et al. (2001) and Denis et al. (2002) showed that the simulation of the spatial patterns of precipitation and temperature over complex terrain is generally improved with the increasing resolution obtained using such regionalisation techniques. The added value of RCMs is also particularly evident for precipitation in the Alps (Frei et al., 2006).

In order to deal with the impact of changing atmospheric conditions on alpine permafrost and to assess the potential of permafrost related hazards, it is essential to establish a link between the latest developments in cryospheric and atmospheric science. Studies dealing with the use of regional climate models to drive impact models have already been carried out in other scientific disciplines, mainly in hydrology (e.g. Arnell et al., 2003, Wood et al., 2004, Kleinn et al., 2005) and agriculture science (e.g. Mearns et al., 1999, Mearns et al., 2001). A problem that occurs primarily when driving an impact model with climate model output is the spatial scale mismatch between the two model types (RCM 50 km, alpine permafrost model 25 m). However, the use of RCM output in topographically complex high-mountain environments presents additional challenges. These include:

  • (1)

    the low vertical resolution of an RCM that cannot represent high-mountain ranges adequately;

  • (2)

    the horizontal shifts of precipitation pattern (at grid boxes scale) that are observed along mountain ranges (Kleinn, 2002);

  • (3)

    the high spatial variability of the climate variables, and

  • (4)

    the reduced availability of observation data in remote mountain areas.

An additional problem from an impact modeller's point of view is the technical handling of climate model output data (data format, rotated grids, etc.), a topic not discussed in this paper.

This study aims at investigating the possibilities and limitations of the application of RCM output for local high-mountain permafrost scenario simulations, and presents two possible approaches which take into account the special challenges mentioned above. The approaches are applied in the Corvatsch region (Upper Engadine, Switzerland) in order to illustrate and discuss their potential, rather than to make an assessment of the potential future changes in the permafrost occurrence for this region.

The RCM Climate High Resolution Model (CHRM), the permafrost model TEBAL (Topographical Energy Balance), the observation data and the study site are presented in Section 2 of this paper. Section 3 introduces the two approaches. The example of the application at the Corvatsch test site is given in Section 4, followed by a discussion in Section 5, and conclusions and perspectives in Section 6.

Section snippets

Climate High Resolution Model (CHRM)

The CHRM limited area model developed at the Institute for Atmospheric and Climate Science, ETH Zurich (IAC-ETHZ), derives from the former numerical weather prediction model, HRM, of the German and Swiss meteorological service (Majewski, 1991). The HRM model has been modified by Lüthi et al. (1996) and Vidale et al. (2003) to provide an RCM. The computational grid of the model is a regular latitude/longitude grid with a rotated pole, a spatial resolution of 0.5° (about 56 km) and 20 vertical

Factors to consider

Currently, the horizontal resolution of a multi-decadal RCM simulation, as must be used for climate change studies, is limited to about 1/6° to 1/2° (approx. 20–50 km), due to limited current computational resources (e.g. Räisänen et al., 2004). However, the spatial resolution required to obtain information about alpine permafrost is a few decametres (25 m with the DHM25 digital terrain model). This is necessary, since elevation, slope and aspect change within very short distances in the

A first example of an application for the Corvatsch regions (eastern Swiss Alps)

In order to illustrate the proposed approaches, we provide in the following an application intended as a first example for the test region Corvatsch in the Swiss Alps (see Fig. 1).

Discussion

In this paper we have introduced two strategies that enable the use of output from regional climate models over topographically complex high-mountain environments.

Of the two approaches proposed, the delta approach is the simpler and faster one. It can include only a quantitative change (e.g. increase of temperature) but not a qualitative one (e.g. increase of variability). If a change in the variability ought to be considered the bias approach must be applied. Nevertheless, the calculation of a

Conclusion and perspectives

Impact modelling by using output from climate models is an interdisciplinary research field full of promise. Due to the high complexity of climate models and the specialized knowledge needed to handle these data, good collaboration between impact modellers and climate modellers is of critical importance.

As discussed, the amount of information gained from the application of one single RCM run to one single impact model is minimal. Many inherent uncertainties and errors of the models and

Acknowledgements

This study was made possible thanks to the European Cooperation in the field of Scientific and Technical Research (COST-Action 719). We are deeply grateful to Christoph Schär for allowing access to CHRM output, and for a very fruitful and friendly collaboration. Further thanks are extended to Stephan Gruber who allowed us to use TEBAL, and to Frank Paul and Wilfried Haeberli for their useful comments. We gratefully acknowledge the constructive remarks of the two reviewers (Ch. Hauck, and

References (78)

  • O.B. Christensen et al.

    Very high-resolution regional climate simulations over Scandinavia — present climate

    J. Clim.

    (1998)
  • J.H. Christensen et al.

    PRUDENCE employs new methods to assess European climate change

    Eos, Trans.

    (2002)
  • U. Damrath

    Verifikation von Wettervorhersagen

    Promet

    (2002)
  • M.C.R. Davies et al.

    The effect of rise in mean annual temperature on the stability of rock slopes containing ice-filled discontinuities

    Permafr. Periglac. Process.

    (2001)
  • B. Denis et al.

    Downscaling ability of one-way nested regional climate models: the Big-Brother experiment

    Clim. Dyn.

    (2002)
  • H.F. Diaz et al.

    Climate variability and change in high elevation regions: past, present and future

    Clim. Change

    (2003)
  • R.E. Dickinson et al.

    A regional climate model for western United States

    Clim. Change

    (1989)
  • C. Frei et al.

    Daily precipitation statistics in Regional Climate Models: evaluation and intercomparison for the European Alps

    J. Geophys. Res.

    (2003)
  • C. Frei et al.

    Future change of precipitation extremes in Europe: An intercomparison of scenarios from regional climate models

    J. Geophys. Res.

    (2006)
  • F. Giorgi

    Simulation of regional climate using a limited area model nested in a general circulation model

    J. Clim.

    (1990)
  • F. Giorgi et al.

    Approaches to the simulation of regional climate change: a review

    Rev. Geophys.

    (1991)
  • F. Giorgi et al.

    Regional Climate Information - Evaluation and Projections. Chapter 10, Climate Change - Third Assessment Report of the Intergovernmental Panel on Climate Change (IPCC)

  • M. Gruber et al.

    Surface temperatures in steep alpine rock faces - a strategy for regional-scale measurement and modelling

  • S. Gruber et al.

    Permafrost thaw and destabilization of alpine rock walls in the hot summer of 2003

    Geophys. Res. Lett.

    (2004)
  • S. Gruber et al.

    Rock-wall temperatures in the Alps: modelling their topographic distribution and regional differences

    Permafr. Periglac. Process.

    (2004)
  • W. Haeberli

    Glacier and permafrost signals of 20th-century warming

    Ann. Glaciol.

    (1990)
  • W. Haeberli

    Construction, environmental problems and natural hazards in periglacial mountain belts

    Permafr. Periglac. Process.

    (1992)
  • W. Haeberli et al.

    Climate change and its impact on glaciers and permafrost in the Alps

    Ambio

    (1998)
  • W. Haeberli et al.

    Murgänge

  • W. Haeberli et al.

    Slope stability problems related to glacier shrinkage and permafrost degradation in the Alps

    Eclogae Geol. Helv.

    (1997)
  • S. Hanson et al.

    Installation of a shallow borehole network and monitoring of the ground thermal regime of a high alpine discontinuous permafrost environment, Eastern Swiss Alps

    Norw. J. Geogr.

    (2005)
  • S.A. Harris et al.

    Thermal regimes beneath coarse blocky materials

    Permafr. Periglac. Process.

    (1998)
  • C. Harris et al.

    Scaled physical modelling of mass movements on thawing slopes

    Permafr. Periglac. Process.

    (2001)
  • Hauck, C., 2001. Geophysical methods for detecting permafrost in high mountains. PhD thesis, VAW-ETH Zurich....
  • C.D. Hewitt et al.

    Ensembles-based Predictions of climate changes and their impacts

    Eos, Trans.

    (2004)
  • E. Heyer

    Witterung und Klima

  • O. Humlum

    Active layer thermal regime 1991–1996 at Qeqertarsuaq, Disko Island, Central West Greenland

    Arc. Alp. Res.

    (1998)
  • M. Hoelzle et al.

    Simulating the effects of mean annual air temperature changes on permafrost distribution and glacier size. An example from the Upper Engadin, Swiss Alps

    Ann. Glaciol.

    (1995)
  • M. Hoelzle et al.

    Miniature temperature dataloggers for mapping and monitoring of permafrost in high mountain areas: First experience from the Swiss Alps

    Permafr. Periglac. Process.

    (1999)
  • Cited by (66)

    • Global permafrost simulation and prediction from 2010 to 2100 under different climate scenarios

      2022, Environmental Modelling and Software
      Citation Excerpt :

      Through experiments based on the performance, the LRM model adopted in this research has the Nagelkerke R2 of 0.942 and the corrected percentage of 97.0%, which shows a good fitting degree of the constructed model (Coon et al., 2016). Meanwhile, compared to previous model, such as the models based on topographic conditions (Imhof, 1996; Lewkowic and Bonnaventure, 2011), various temperature indexes (Hoelzle, 1992; Ishikawa and Hirakawa, 2000; Li and Wu, 2005; Zhao et al., 2007; Cheng et al., 2012) and RCM (Salzmann et al., 2007), the LRM model in this research adopts various factors including topography, location, LUCC and climate status. Furthermore, besides the climate factors, the LUCC factors in this research are acquired in different periods and under various climate scenarios.

    • Streamflow response to projected climate changes in the Northwestern Upper Indus Basin based on regional climate model (RegCM4.3) simulation

      2019, Journal of Hydro-Environment Research
      Citation Excerpt :

      The effect of model bias on simulated change suggests a sophisticated bias correction method before the application for climate change impact assessment. The main advantage of bias correction method adopted in the present study is that it includes the possible changes in the variability of data (e.g. temperature and precipitation) simulated by the RCMs (Salzmann et al., 2007). This approach has been applied successfully in few previous studies over the major tributaries on Indus Basin i.e. the Jhelum and Kunar Rivers (Mahmood and Babel, 2012, 2014; Mahmood and Jia, 2016; Mahmood et al., 2016).

    • Permafrost activity and atmospheric warming in the Argentinian Andes

      2018, Geomorphology
      Citation Excerpt :

      Thus, under global warming, rock glaciers might also play an important role as an important transporter of sediment. Numerous studies have investigated past and likely future effects of atmospheric warming on glaciers in mountains around the world (Kääb et al., 2007), but the possible impacts on rock glaciers have been largely neglected (Salzmann et al., 2007). Yet rock glaciers have drawn more research attention in recent decades (Jones et al., 2018).

    View all citing articles on Scopus
    1

    Now: Federal Office of Meteorology and Climatology, Zurich, Switzerland.

    2

    Now: NCAS CGAM, University of Reading, UK.

    View full text