High resolution present and future climate and surface mass balance of Svalbard modelled by the regional climate model MAR

• Main Author: Lang, Charlotte
• Other Authors: Erpicum, Michel, Fettweis, Xavier
• Format: Doctoral or Postdoctoral Thesis
• Language: English
• Published: ULiège - Université de Liège 2015
• Subjects: Svalbard; surface mass balance; regional climate model; Physical; chemical; mathematical & earth Sciences; Earth sciences & physical geography; Physique; chimie; mathématiques & sciences de la terre; Sciences de la terre & géographie physique; Arctic; Greenland; albedo
• Online Access: https://orbi.uliege.be/handle/2268/187914; https://orbi.uliege.be/bitstream/2268/187914/1/charlotte_lang_thesis_final.pdf

This worked aimed to simulate the climate and surface mass balance of Svalbard at high spatial resolution with the regional climate model MAR. First, simulations of the present (1979 - 2013) climate and surface mass balance were performed at a spatial resolution of 10 km. MAR was evaluated over Svalbard by comparing its outputs to measurements of temperature and precipitation measurement and outputs of other modelling products fo the surface mass balance. The results of the MAR simulations were then analysed over 1979 - 2013 and shoed a stability of the (negative) SMB opposed to the recent melt records observed in Greenland. This stability was attributed to a recent change in summer atmospheric circulation damping the Arctic warming over Svalbard. A future projection over 1980 - 2100 was then performed with MAR forced by MIROC5 and the RCP8.5 scenario and showed that the future SMB decrease is projected to be mainly driven by the albedo decrease (related to the expansion of the ablation area) through the increase of the net shortwave radiation absorbed by the surface.Finally, an online downscaling technique has been implemented in MAR to allow the surface modeule SISVAT, computing the surface mass balance and its components, to run at a resolution twice as high as the atmospheric module. This method, based on near-surface temperature and humidity corrections on a subgrid, allows SMB outputs at a resolution twice as high with only 25% more computation time.