Version 5 2025-05-04, 12:31Version 5 2025-05-04, 12:31
Version 4 2025-04-21, 07:04Version 4 2025-04-21, 07:04
Version 3 2025-04-21, 06:58Version 3 2025-04-21, 06:58
Version 2 2025-04-19, 14:26Version 2 2025-04-19, 14:26
Version 1 2025-04-19, 14:19Version 1 2025-04-19, 14:19
dataset
posted on 2025-05-04, 12:31authored byTao Li, Lixing Wang, Zihan Qiu, Philippe Ciais, Steven J. Davis, Zhu Deng, Yufei Zhao, Glen P. Peters, Piyu KePiyu Ke, Matthew W. Jones, Robbie M. Andrew, Ye Hao, Taochun Sun, Robert B. Jackson, Xiaoting Huang, Pierre Friedlingstein, Chenxi Lu, Duo CUi, Zhu Liu
By collecting over two million records of near-real-time electricity generation, traffic, and industrial output since 2019, and employing machine learning to simulate the nonlinear, time-varying relationships between emissions and predictors (i.e., temperature and time surrogates), we constructed a daily CO2 emissions dataset for 13 countries and regions globally, spanning the period from 1970 to 2024. This dataset encompasses four sectors: power, industry, residential, and transport (including ground transport and aviation).