Articles | Volume 15, issue 10
Geosci. Model Dev., 15, 4225–4237, 2022
https://doi.org/10.5194/gmd-15-4225-2022
Geosci. Model Dev., 15, 4225–4237, 2022
https://doi.org/10.5194/gmd-15-4225-2022
Model evaluation paper
01 Jun 2022
Model evaluation paper | 01 Jun 2022

A comparative analysis for a deep learning model (hyDL-CO v1.0) and Kalman filter to predict CO concentrations in China

Weichao Han et al.

Model code and software

Deep learning model (hyDL-CO v1.0) to predict CO in China Zhe Jiang https://doi.org/10.5281/zenodo.5913013

Download
Short summary
We present an application of a hybrid deep learning (DL) model on prediction of surface CO in China from 2015 to 2020, which utilizes both convolutional neural networks and long short-term memory neural networks. The DL model performance is better than a Kalman filter (KF) system in the training period (2005–2018). Furthermore, the DL model demonstrates good temporal extensibility: the mean bias and correlation coefficients are 95.7 ppb and 0.93 in the test period (2019–2020) over eastern China.