Comparison of Longwave Atmospheric Fluxes and Cloud Radiative Forcing over Barrow, Alaska and Eureka, Nunavut

The modulation of the surface energy budget in the Arctic is controlled in large part by clouds and water vapor, the properties of which make up some of the most uncertain parameters in climate modeling. Despite their importance, most studies of Arctic clouds rely on satellite-based retrievals, which in many cases have not been properly validated. Therefore, accurate, ground-based measurements of atmospheric and cloud properties are invaluable. We present surface measurements made at the Atmospheric Radiation Measurement (ARM) site at Barrow, Alaska (71° N, 156° W) and the Canadian Network for the Detection of Arctic Change (CANDAC) site at Eureka, Nunavut (80° N, 86° W). These measurements are used to examine atmospheric variables between 2006 and 2008 with particular emphasis on longwave downwelling (LW) cloud radiative forcing, LW all-sky, and LW clear-sky fluxes. A method is developed by which flux calculations are derived from Atmospheric Emitted Radiance Interferometers (AERIs) with the aid of radiative transfer models. Flux results are shown to agree well with independently derived flux measurements from a pyrgeometer at Barrow. A comparison of the measured fluxes between the two sites show considerable differences. Longwave cloud radiative forcing is larger at Barrow than at Eureka primarily due to larger cloud fraction. An evaluation of comparable fluxes from the NCEP-NCAR Reanalysis I dataset (NNR) demonstrates that the all-sky and clear-sky fluxes from the NNR are biased low compared to the measurements, although the dataset is well correlated to the measurements. NNR does not accurately capture the magnitude of the seasonal cycle in LW cloud radiative forcing. Despite this, it is shown that NNR is able to capture the approximate relative difference in annual-average LW cloud radiative forcing between Barrow and Eureka.