Snow on the Antarctic sea ice is a crucial component of the cryosphere. In response to the dynamic and highly heterogeneous Antarctic snow during the sea ice melting season, this study employed a combined multi-source data and deep learning method to accurately retrieve snow depth on Antarctic sea ice. Initially, we integrate multiple datasets, including satellite remote sensing, geospatial information, and meteorological data. Subsequently, a Convolutional Neural Network (CNN) is utilized to construct a snow depth retrieval model (PSDCNN-5_7 model). Compared to snow depth measurements from Alfred Wegener Institute (AWI) snow buoys, the PSDCNN-5_7 model outperforms existing algorithms, exhibiting a deviation of only −3.38 cm. The uncertainty of the snow depth caused by the model input is only 1.64 cm. In West Antarctica, snow depth is more affected by snowfall (SF), 2-m air temperature (T2m), and sea ice velocity (SIV). Conversely, in East Antarctica, snow depth is primarily influenced by SIV. The proposed approach accurately retrieves snow depth on Antarctic sea ice and facilitates the derivation of long-term variations and trends in snow depth, contributing to a better understanding of the relationship between sea ice, snow, and climate change.

The characteristics of sea ice leads (SILs) in the Weddell Sea are an important basis for understanding the mechanism of the atmosphere–ocean system in the Southern Ocean. In this study, we derived the sea ice surface temperature (IST) of the Weddell Sea from MODIS thermal images and then generated a daily SIL map for 2015 and 2022 by utilizing the iterative threshold method on the optimised MOD35 cloud-masked IST. The results showed that SIL variations in the Weddell Sea presented remarkable seasonal characteristics. The trend of the SIL area exhibited an initial rise followed by a decline from January to December, characterised by lower values in spring and summer and higher values in fall and winter. SILs in the Weddell Sea were predominantly concentrated between 70~78°S and 60~30°W. The coastal spatial distribution density of the SILs exceeded that of offshore regions, peaking near the Antarctic Peninsula and then near Queen Maud Land. The SIL variation was mainly influenced by dynamical factors, and there were strong positive correlations between the wind field, ocean currents, and sea-ice motion.

Dome A is the summit of the Antarctic plateau, where the Chinese Kunlun inland station is located. Due to its unique location and high altitude, Dome A provides an important observatory site in analyzing global climate change. However, before the arrival of the Chinese Antarctic expedition in 2005, near-surface air temperatures had not been recorded in the region. In this study, we used meteorological parameters, such as ice surface temperature, radiation, wind speed, and cloud type, to build a reliable model for air temperature estimation. Three models (linear regression, random forest, and deep neural network) were developed based on various input datasets: seasonal factors, skin temperature, shortwave radiation, cloud type, longwave radiation from AVHRR-X products, and wind speed from MERRA-2 reanalysis data. In situ air temperatures from 2010 to 2015 were used for training, while 2005–2009 and 2016–2020 measurements were used for model validation. The results showed that random forest and deep neural network outperformed the linear regression model. In both methods, the 2005–2009 estimates (average bias = 0.86 °C and 1 °C) were more accurate than the 2016–2020 values (average bias = 1.04 °C and 1.26 °C). We conclude that the air temperature at Dome A can be accurately estimated (with an average bias less than 1.3 °C and RMSE around 3 °C) from meteorological parameters using random forest or a deep neural network.

The rectifier effect refers to the exchange of CO2 near the boundary layer of the atmosphere. This effect is an important part of the carbon cycle because it affects the vertical distribution of CO2 and indirectly affects global CO2 distribution. However, the intensity of this effect is difficult to measure directly. A differential absorption LiDAR (DIAL) system can observe the rectifier effect because this device can accurately measure planetary boundary layer height (PBLH) and CO2 profile concentration (Quan et al., 2014; Guo et al., 2016). Accordingly, we conducted experiments in Huainan, China using a DIAL system for CO2 detection developed by our group. We selected two cases, namely, summer and winter, for the analysis. Firstly, we calculated CO2 profile, average CO2 concentration and PBLH. Secondly, we analysed the rectifier effect. Results showed a negative correlation between PBLH and the rectifier effect. Simultaneously, we observed a special phenomenon in which CO2 concentration is lower near the ground than in high altitudes. This phenomenon may be explained by the activities of plants.

Methane (CH4) is the second most important anthropogenic greenhouse gas that contributes to global warming. The global warming potential of CH4 is 72 times that of CO2 per molecule. However, the uncertainties of the sinks and sources of CH4 remain large. Unfortunately, range-resolved atmospheric CH4 concentrations have been rarely observed, thereby hindering the accurate understanding of some of the key features of carbon cycle. Differential absorption Lidar (DIAL) is a powerful and promising means of obtaining range-resolved CH4 concentrations and is helpful for estimating the emissions of anthropogenic and natural CH4 sources, as well as uncovering previously unknown aspects of the carbon cycle. Prior to developing a CH4-DIAL, a comprehensive preliminary study on the selection of on-line wavenumber must be performed to guarantee a high signal-to-noise ratio (SNR) of the differential signal and minimize the errors due to interference from other gases. This study aims to find the optimal selection of on-line wavenumbers in terms of atmospheric factors and detection scenarios. After roughly estimating the on-line wavenumbers using an evaluation index, the 1K temperature error and 1 hPa pressure error were separately and qualitatively analyzed. Then, the weighting function of the potential wavenumbers based on pressure was analyzed. Results show that 6076.938 cm(-1) is the most suitable wavenumber for vertical range-resolved measurements. The errors caused by uncertainties in temperature and pressure are only 0.33%/K and 0.11%/hPa when 6076.938 cm(-1) is selected as the on-line wavenumber. In addition, our simulation experiments indicated that the influences of H2O and CO2 can be neglected if a reasonable pair ofon-line and off-line wavenumbers was selected. Moreover, we demonstrated the relationship between on-line wavenumber and precision in different background concentrations for horizontal detection, which determines the optimal on-line wavenumbers for horizontal measurements under different circumstances. (C) 2019 Published by Elsevier Ltd.