We will therefore apply an ensemble approach for Arctic downscaling of global climate scenarios. Based on EC-Earth global scenario simulations, dynamical downscaling will be carried out by means of RCAO including the LPJ-GUESS vegetation model (Smith et al. 2001).
Expected feedbacks (involving surface albedo, water vapor and release of greenhouse gases from Arctic soil) will be explored for their significance on regional Arctic and global climate. To continue to even local scales, we will carry out statistical downscaling (including dynamic scales due to varying landscape elements) of the Abisko and Torneträsk area.
To complement the statistical approach, a double-nesting technique will be used in RCA (atm. component of RCAO) in order to provide local climate simulations over the Abisko and Torneträsk regions at ~5 km. In this approach, a large RCA domain at 25 km will first downscale either analyses or GCM climate simulations. Results from this RCA simulation will then be applied as boundary conditions to force a smaller RCA domain over the area of interest at ~5 km resolution.
These chain of experiments will be extended by using additional global scenarios to provide better probability statistics.
References
Döscher, R., K. Wyser, H.E.M. Meier, M. Qian, R. Redler, 2009: Quantifying Arctic Contributions to Climate Predictability in a regional coupled Ocean-Ice-Atmosphere Model. Climate Dynamcis, DOI: 10.1007/s00382-009-0567-y, accepted for publication.
Smith, B., Prentice, I.C. & Sykes, M.T. 2001. Representation of vegetation dynamics in modelling of terrestrial ecosystems:
comparing two contrasting approaches within European climate space. Global Ecology and Biogeography 10: 621-637.