The fluctuation of Antarctic sea ice has strengthened since 2014 and hit the historical low several times, which is different from the past. The rapid change of sea ice can touch off the abnormal variations of heat-moisture exchange between atmosphere and ocean, break the balance between the atmospheric and the marine system. Cloud anomalies occur in this process, which is capable of controlling the sea-ice growth and melting processes, through its influences on the surface energy budget, via reducing shortwave radiation and transmitting longwave radiation. Sea ice prediction in the polar region becomes very valuable for environmental protection, social security, and economic development, in the context of dramatic changes in the polar climate. However, the sea ice prediction model behaves not well due to a lack of knowledge of cloud processes in the polar climate system.
Aim to solve this problem, the short-term and long-term influence mechanism of Antarctic winter clouds on sea ice is investigated. The study on the short-term forcing of winter clouds on sea ice is mainly to investigate the forcing of winter Antarctic clouds on sea ice in 2011, and their contribution to the area and thickness of summer sea ice in 2012. The results show that the cloud anomaly in winter is in agreement with the convection and the advection related to the large-scale atmospheric circulation. The negative cloud anomaly in winter 2011 may be caused by the atmospheric pressure anomalies. The longwave radiation from the surface is released into space in the less-cloud region, which cools and freezes the surface due to heat loss and increase the sea-ice area and thickness. Although this extra sea ice has experienced drifting, there is still much sea ice that survived in the following summer, which makes the summer sea ice show a positive anomaly.
Based on the study of the influence mechanism of cloud to sea ice seasonally, this paper further investigates the long-term coupling mechanism of the Antarctic winter cloud to sea ice. The results show that atmosphere forces sea ice in two ways: in the low-troposphere, the atmosphere can directly force sea ice dynamics and thermodynamics through large-scale atmospheric circulation. The wind related to the large-scale atmospheric circulation can drive sea ice dynamically, and also generate or melt sea ice by carrying warm or cold air. On the other hand, clouds related to the large-scale atmospheric circulation can force sea ice by the radiative effect. In the Antarctic winter, the large-scale atmospheric circulation is mainly controlled by wave-3 and SAM. In the low-troposphere, the large-scale atmospheric circulation controlled by wave-3 and SAM determines the wind direction, air temperature, and moisture. Under the influence of the Antarctic and Andes topography, a negative cloud anomaly with the same phase of sea ice anomaly is formed. The radiation effect of the negative cloud anomaly on the surface weakens the direct impact of large-scale atmospheric circulation on sea ice. In the mid-troposphere, the formation of clouds is mainly related to the atmospheric circulation anomaly controlled by wave-3, and the combined action of advection and convection results in the anomalies of middle clouds. However, in the mid- and upper troposphere, the radiative effects of cloud mode on the surface promote the direct forcing of low-tropospheric large-scale atmospheric circulation on sea ice. This result reveals, for the first time, the different effects of cloud cover in the different level troposphere on the sea ice distribution.
Based on the study of the short-term and long-term influence mechanism of clouds on sea ice, it is found that clouds have a significant radiation impact on sea ice, and sea ice has a good record of this impact, which is reflected in many Antarctic regions. Besides, the trace of robust cloud forcing to sea ice can retain for seasonal time scales. It shows that clouds have great potential to predict sea ice. Based on these understandings, a linear Markov sea-ice prediction model is designed. We chose to define the coupled Antarctic climate system with many variables: cloud cover, sea ice concentration, and other climate factors (including geopotential height, wind vector, and air temperature in different levels of the troposphere). The multivariate empirical orthogonal function of these variables is taken as the component of the model. The prediction ability of the model is evaluated by cross-validation. A series of sensitivity experiments are carried out to determine the best model. Although this kind of statistical model can't simulate nature, it has better practical value and application space before the birth of an ideal climate dynamic prediction model.
With the increasing demand for sea ice and climate prediction in polar. Our ability to understand and quantitatively simulate the cloud-air-sea ice coupling process is valuable to the sea ice, synoptic, and climate simulation and prediction.