On the accuracy of the measured and modelled surface latent and sensible heat flux in the interior of the Greenland Ice Sheet

Abstract. The latent (LHF) and sensible (SHF) heat fluxes are key components of the surface mass and energy balance in the accumulation area of the Greenland Ice Sheet, making them critical for accurate sea level projections. While Eddy-Covariance(EC) systems provide accurate measurements of the turbulent surface transport of mass and energy in the low and mid-latitudes, frequent stable boundary layer conditions in polar regions introduce uncertainties in the EC method. In addition, as EC measurements are sparse, it is critical to characterise biases in the more common bulk fluxes obtained from automatic weather stations and climate models in polar areas. In this study, we present an intercomparison of three independent EC systems at the EastGRIP site at ∼2700 m a.s.l on the Greenland Ice Sheet to assess the accuracy of LHF and SHF measurements. A comparison of the fluxes by the three systems demonstrates excellent agreement, with a correlation (r) of 0.97 to 0.98, an absolute bias of 0.2 W m^-2, an RMSE between 1.2 W m^-2 and 1.5 W m^-2 and slopes between 1.01 and 1.16 for the LHF, and r = 0.98, an absolute bias of less than 0.5 W m^-2, an RMSE between 1.6 and 1.9 W m^-2, and slopes of 1.0 for the SHF. A comparison of the validated EC fluxes against the bulk method highlights the sensitivity to the site-specific roughness length z_0,m and the limitation of common parameterisations of the humidity and temperature roughness lengths z_0,q and z_0,t. Using improved values for z_0,m, z_0,q and z_0,t, recomputed bulk fluxes are compared to fluxes simulated by regional climate models MAR, RACMO2.3p2 and RACMO2.4p1. We find an overall good agreement of the summer turbulent flux magnitudes, while all evaluated models simulate stronger near-surface temperature gradients during winter compared to observations from automatic weather stations, leading to consistently larger modelled SHF and LHF values in winter.