Application of Machine Learning for Simulation of Air Temperature at Dome A
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,...
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-02-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/14/4/1045 |
| _version_ | 1827652870748504064 |
|---|---|
| author | Xiaoping Pang Chuang Liu Xi Zhao Bin He Pei Fan Yue Liu Meng Qu Minghu Ding |
| author_facet | Xiaoping Pang Chuang Liu Xi Zhao Bin He Pei Fan Yue Liu Meng Qu Minghu Ding |
| author_sort | Xiaoping Pang |
| collection | DOAJ |
| description | 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. |
| first_indexed | 2024-03-09T21:08:13Z |
| format | Article |
| id | doaj.art-750ffcd2aaeb448abc21401d51be6023 |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-09T21:08:13Z |
| publishDate | 2022-02-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-750ffcd2aaeb448abc21401d51be60232023-11-23T21:56:06ZengMDPI AGRemote Sensing2072-42922022-02-01144104510.3390/rs14041045Application of Machine Learning for Simulation of Air Temperature at Dome AXiaoping Pang0Chuang Liu1Xi Zhao2Bin He3Pei Fan4Yue Liu5Meng Qu6Minghu Ding7Chinese Antarctic Center of Surveying and Mapping, Wuhan University, Wuhan 430079, ChinaChinese Antarctic Center of Surveying and Mapping, Wuhan University, Wuhan 430079, ChinaSchool of Geospatial Engineering and Science, Sun Yat-sen University, Zhuhai 519000, ChinaChinese Antarctic Center of Surveying and Mapping, Wuhan University, Wuhan 430079, ChinaChinese Antarctic Center of Surveying and Mapping, Wuhan University, Wuhan 430079, ChinaChinese Antarctic Center of Surveying and Mapping, Wuhan University, Wuhan 430079, ChinaKey Laboratory of Polar Science of Ministry of Natural Resources, Polar Research Institute of China, Shanghai 200136, ChinaState Key Laboratory of Severe Weather and Institute of Polar Meteorology, Chinese Academy of Meteorological Sciences, Beijing 100081, ChinaDome 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.https://www.mdpi.com/2072-4292/14/4/1045Dome Aair temperatureskin temperaturelinear regressionrandom forest modeldeep learning |
| spellingShingle | Xiaoping Pang Chuang Liu Xi Zhao Bin He Pei Fan Yue Liu Meng Qu Minghu Ding Application of Machine Learning for Simulation of Air Temperature at Dome A Remote Sensing Dome A air temperature skin temperature linear regression random forest model deep learning |
| title | Application of Machine Learning for Simulation of Air Temperature at Dome A |
| title_full | Application of Machine Learning for Simulation of Air Temperature at Dome A |
| title_fullStr | Application of Machine Learning for Simulation of Air Temperature at Dome A |
| title_full_unstemmed | Application of Machine Learning for Simulation of Air Temperature at Dome A |
| title_short | Application of Machine Learning for Simulation of Air Temperature at Dome A |
| title_sort | application of machine learning for simulation of air temperature at dome a |
| topic | Dome A air temperature skin temperature linear regression random forest model deep learning |
| url | https://www.mdpi.com/2072-4292/14/4/1045 |
| work_keys_str_mv | AT xiaopingpang applicationofmachinelearningforsimulationofairtemperatureatdomea AT chuangliu applicationofmachinelearningforsimulationofairtemperatureatdomea AT xizhao applicationofmachinelearningforsimulationofairtemperatureatdomea AT binhe applicationofmachinelearningforsimulationofairtemperatureatdomea AT peifan applicationofmachinelearningforsimulationofairtemperatureatdomea AT yueliu applicationofmachinelearningforsimulationofairtemperatureatdomea AT mengqu applicationofmachinelearningforsimulationofairtemperatureatdomea AT minghuding applicationofmachinelearningforsimulationofairtemperatureatdomea |