Investigating arctic cloud and radiative properties associated with the large-scale climate variability through observations, reanalysis, and mesoscale modeling

This dissertation examines two decades of Arctic cloud cover data and the variability in Arctic clouds with relation to changes in sea ice using observational and reanalysis data, as well as a state-of-the-art mesoscale model. Decadal length Arctic cloud cover data are examined because of the inherent differences within these measurements that have not been explored in previous research. Cloud cover data are analyzed from regions poleward of 60°N from several sources of visual surface observations including surface remotely sensed measurements at two locations, two spaced-based passive remotely sensed datasets (Advanced Very High Resolution Radiometer Polar Pathfinder extended (APPx) and Television Infrared Observation Satellite Operational Vertical Sounder (TOVS) Polar Pathfinder (TPP)), and one reanalysis dataset (European Center for Medium-Range Weather Forecasting Reanalysis (ERA-40)) are compared. The passive remotely sensed data are sensitive to surface type. Cloud amounts from the APPx and TPP decrease with increases in sea ice concentrations. In comparison to the surface remotely sensed measurements over sea ice, the APPx and TPP cloud amounts are consistently low. The ERA-40 output cloud cover not contain a sharp decrease from water to ice surfaces, and compares reasonably with the remotely sensed surface measurements over sea ice. During the northern hemisphere winter at land stations, the TPP and ERA-40 cloud amounts are similar. This is most likely a result of the ERA-40 model using TOVS irradiances as input data. The APPx and surface cloud amounts are similar during all seasons, but they are not in precise agreement with the TPP/ERA-40 values. Cloud amounts from the ERA-40 are also most similar to surface measurements in regions where radiosonde data are used as input.;Cloud radiative forcing calculated from the ERA-40 output is examined with relation to sea ice concentrations using 20 years of data. The radiative effect of clouds varies linearly with sea ice concentrations during the winter and spring. This relationship is most statistically significant in the North Atlantic region, but statistically significant relationships also occurring the northern Pacific. Statistically significant correlations do not occur during the summer months. By calculating differences in cloud amount during low and high sea ice concentration summers, greater cloud cover amounts occur with decreases in sea ice in the Arctic poleward of the Pacific at the 80 percent statistical significant level. In October, clouds are varying with relation to sea ice near the sea ice edge. One-month lag relationships are calculated to examine if the cloud radiative forcing terms are changing before or after changes in sea ice concentration. Changes in the longwave radiative forcing of clouds occurs before changes in sea ice concentrations and surface temperatures in the North Atlantic region. Cloud radiative forcing, sea ice concentrations, and surface temperatures are interrelated in this region, and may be forced by the same physical mechanism.;The response of Arctic clouds and surface radiative properties is examined using the polar version of the Weather Research and Forecasting (WRF) regional model over the Laptev Sea. WRF is run for four Septembers and Octobers with anomalously low and high sea ice concentrations. Differences in the surface radiative forcing, cloud radiative forcing, cloud properties and the surface heat budget are examined for the composite low and high years. In both months, there are more clouds during low sea ice years. WRF produces more low-level liquid cloud amount during years without sea ice. The increase in clouds during low sea ice years corresponds with an increase in downwelling longwave radiation, and hence longwave cloud radiative forcing. Increases in downwelling longwave radiation during low sea ice years are canceled by the increased amount of upwelling longwave radiation, which is a result of warmer surface skin temperatures. In September, the decrease in surface albedo associated with sea ice retreat/melt results in an increased net surface radiation during low sea ice years. In October, the changes in net surface radiation are not statistically significant. After the Arctic solar night begins, during times with no sea ice, large latent and sensible heat upward surface fluxes aids in the deepening of the boundary layer and preventing the formation of the typical Arctic inversion. In WRF, the increases in cloud water liquid content and downwelling longwave radiation, in low sea ice years, seems to be a result of increased open water, while the changes in the boundary layer are the result of changes in the surface radiative fluxes.