Evaporation From the Southern Ocean Estimated on the Basis of AIRS Satellite Data

Abstract

Evaporation plays an important role in the global water and energy cycles and, hence, in climate change. Evaporation over the Southern Ocean, where the Antarctic sea ice coverage has a large annual cycle, is poorly quantified. In this study, daily evaporation is estimated for the Southern Ocean with a sea-ice-specific algorithm, using surface temperature and air humidity from National Aeronautics and Space Administration's Atmospheric Infrared Sounder (AIRS), and wind speeds from Modern-Era Retrospective Analysis for Research and Applications, Version 2 (MERRA-2), reanalysis during 2003-2016. An uncertainty of 34% was found in the evaporation product. The results indicate that annual evaporation has considerable interannual and regional variability, but with a decreasing trend during the study period over most of the Southern Ocean. There are, however, areas where evaporation has increased, specifically in the Ross Sea in winter and summer, with smaller positive trends in spring and fall. Overall, the changes in the difference between the surface specific humidity and the air specific humidity, and to a much lesser extent in the wind speed, are the main drivers for the changes in evaporation throughout the year. During spring and fall months, changes to the sea ice cover, which alter the surface specific humidity, are the main drivers for the change, but in summer and winter the main driver is the air-specific humidity. Air masses originating from the Antarctic continent (south) are associated with cold and dry conditions, which increase evaporation, whereas air masses from lower latitudes in the Southern Ocean (north) are associated with warm and moist conditions, decreasing evaporation. Comparisons with other reanalysis evaporation products produce similar trends, although annual averages differ.

Arctic Cloud Response to a Perturbation in Sea Ice Concentration: The North Water Polynya

“…Although available data sources do not provide a consistent picture, consistently positive surface turbulent flux retrievals from MERRA‐2 and ERA5 over winter sea ice suggest a poor representation of sea ice and snow cover on sea ice, and subsequent limitations in near‐surface estimates of temperature and specific humidity are likely producing skewed surface fluxes. Moreover, the surface turbulent flux evolution found within the satellite data set is consistent with previous studies (Boisvert et al., 2012, 2015, 2020; Yao & Tang, 2003) and is used below in subsequent analysis of the cloud response to the polynya.…”

Southern Ocean Precipitation Characteristics Observed From CloudSat and Ground Instrumentation During the Macquarie Island Cloud & Radiation Experiment (MICRE): April 2016 to March 2017

“…Because of the sparseness of surface observations, studies requiring regional knowledge of SO precipitation characteristics typically rely on reanalyses or satellite-based products. Unfortunately, there are large differences between various reanalyses and between different satellite precipitation products in the phase (liquid versus ice), frequency and rate of precipitation, as well as total accumulation over the SO (Behrangi et al, 2014(Behrangi et al, , 2016Boisvert et al, 2020;Manton et al, 2020). While reanalyses are constrained by observations and often run at finer resolutions than climate models, they use many of the same cloud and precipitation parameterizations used in climate models and suffer from many of the same biases over the SO as climate models (Lang et al, 2018;Naud et al, 2014).…”

The Earth‐Observing Aqua Satellite Mission: 20 Years and Counting

“…Key components of the water cycle are all included in Aqua‐derived data products. For instance, SSTs have been derived from both MODIS data (e.g., Kilpatrick et al., 2015 ) and AMSR‐E data (e.g., Nielsen‐Englyst et al., 2018 ), evaporation from AIRS data (e.g., Boisvert et al., 2020 ), evapotranspiration from MODIS data (e.g., Mu et al., 2011 ; Nishida et al., 2003 ), water vapor from AIRS data (e.g., Irion et al., 2018 ; Worden et al., 2019 ), precipitation from AMSR‐E data (e.g., Joseph et al., 2009 ), and soil moisture from AMSR‐E data (e.g., Bhagat, 2015 ; Jackson et al., 2009 ).…”