Ocean surface heat flux variability in the Barents Sea M. Årthun a, ⁎, C. Schrum a,b a Geophysical Institute, University of Bergen, Allégaten 70, N-5007 Bergen, Norway b Norwegian Institute for Water Research, Thormøhlensgate 53 D, N-5006 Bergen, Norway abstract article info Article history: Received 6 September 2009 Received in revised form 30 June 2010 Accepted 5 July 2010 Available online 14 July 2010 Keywords: Barents Sea HAMSOM Heat fluxes Sea-ice A 40 year (1958–1997) hindcast simulation from the regional coupled ice–ocean model HAMSOM is used to study climate relevant processes in the Barents Sea and their interannual to decadal variability. Compared to observations the model captures the variability in temperature and ice extent in a satisfying manner. The heat input through the Barents Sea Opening (BSO) is effectively lost through intense atmosphere–ocean heat exchange within the Barents Sea. Correlation analysis suggests that heat transport through the BSO leads the Barents Sea heat content by 1–10 months, while the heat content leads the air–sea heat fluxes with 1– 5 months. Averaged over the period the advected heat input is 32 TW, augmented by 79 TW of shortwave radiation and reduced by 113 TW through longwave radiation and latent and sensible heat loss. Including the sensible heat loss at the ice–ocean boundary yields an oceanic heat loss in the Barents Sea of 40 TW. Cooling of Atlantic Water is very efficient just east of the BSO, and contributes to 50% of the total heat loss. Significant positive trends in both heat transport through the BSO and solar radiation, combined with a reduction in seasonal ice cover cause increased oceanic heat loss. Excess heat still enters the Barents Sea and a significant warming is observed in the northern areas. Sea-ice acts as an effective insulator against oceanic heat loss resulting in 4 TW of net heat input at the sea- ice surface. This heat flux balances the ice–ocean heat budget and the corresponding ice melt compensates for net ice production at the ice–ocean interface and ice advection into the Barents Sea. © 2010 Elsevier B.V. All rights reserved. 1. Introduction The Barents Sea is one of the largest shallow shelf seas adjacent to the Arctic Ocean, and is one of two pathways in which Atlantic Water (AW) reaches the Arctic Ocean (e.g. Nansen, 1906; Schauer et al., 2002; Rudels et al., 2004). Thus, the climatic variability in the Barents Sea is of major importance to the Arctic region as a whole and in understanding present and future climate changes. The physical oceanographic conditions depend mainly on the AW inflow between Fugløya and Bjørnøya, often called the Barents Sea Opening (BSO, Fig. 1), and the inflow of Arctic Water from the Kara Sea and Arctic Ocean (Loeng et al., 1997). Based on two months of current measurements in 1978, Blindheim (1989) calculated the net inflow through the BSO to be 1.9 Sv (1 Sv ≡ 10 6 m 3 /s). Recent measurements (2003–2005, Skagseth, 2008) yield a mean net transport of 1.1 Sv withlarge seasonal (Ingvaldsen et al., 2004) and interannual variations (Skagseth et al., 2008), to which 0.5–1 Sv from the Norwegian Coastal Current (Schauer et al., 2002; Gammelsrød et al., 2009) should be added. The seasonal variability is related to local atmospheric forcing (Ådlandsvik and Loeng, 1991; Ingvaldsen et al., 2002, 2004), generally causing a higher inflow during winter than summer. Variability in AW inflow on interannual to decadal time- scales may in turn be connected to large scale atmospheric variability like the North Atlantic Oscillation/Arctic Oscillation (Dickson et al., 2000; Furevik, 2001; Ingvaldsen et al., 2003). Although some of the inflowing water recirculates within the Bear Island Trough (Skagseth, 2008), most of it passes through the Barents Sea and exits between Novaya Zemlya and Frans Josef Land, here designated as the Barents Sea Exit (BSX, Fig. 1) following Gammelsrød et al. (2009). The water then enters the Arctic Ocean through the St. Anna Trough (Schauer et al., 2002). Combining models and observa- tions, Gammelsrød et al. (2009) estimated the net volume flux toward the Arctic Ocean to 2.0 ± 0.6 Sv between 1991 and 1992. Due to intense cooling of the inflowing water on its way across the Barents Sea, heat transports are small at the BSX, and it cannot be ruled out that the Barents Sea is a heat sink rather than a heat source for the Arctic Ocean egammelsrod08. Oceanic heat loss during winter can reach over 500 W/m 2 , cooling the eastward flowing water by 4–5 °C within a couple of months (Häkkinen and Cavalieri, 1989). According to estimates based on atmospheric observations and oceanic heat budgets, about half of the heat loss in the entire Nordic Seas takes place in the Barents Sea (Simonsen and Haugan, 1996). At high latitudes, sea-ice modifies the atmosphere–ocean interac- tion. Sea-ice is also intimately linked to the climatic variability in the Barents Sea through the ice-albedo feedback and brine release during freezing. Wintertime observations show that the Barents Sea ice Journal of Marine Systems 83 (2010) 88–98 ⁎ Corresponding author. Tel.: +47 55582889. E-mail address: marius.arthun@gfi.uib.no (M. Årthun). 0924-7963/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.jmarsys.2010.07.003 Contents lists available at ScienceDirect Journal of Marine Systems journal homepage: www.elsevier.com/locate/jmarsys