Deep learning-based air temperature mapping by fusing remote sensing, station, simulation and socioeconomic data

REMOTE SENSING OF ENVIRONMENT

卷:240

Article

Shen, Huanfeng^1,4,5  Jiang, Yun¹  Li, Tongwen¹  Cheng, Qing²  Zeng, Chao¹  Zhang, Liangpei^3,4 

[1] Wuhan Univ, Sch Resource & Environm Sci, Wuhan 430079, Peoples R China

[2] Wuhan Univ, Sch Urban Design, Wuhan 430079, Hubei, Peoples R China

[3] Wuhan Univ, State Key Lab Informat Engn Surveying Mapping & R, Wuhan 430079, Peoples R China

[4] Collaborat Innovat Ctr Geospatial Technol, Wuhan 430079, Peoples R China

[5] Wuhan Univ, Key Lab Geog Informat Syst, Minist Educ, Wuhan 430079, Peoples R China

关键词: Air temperature;    Land surface temperature;    Deep learning;    Remotely sensed data;    Assimilation data;    Socioeconomic data;   

DOI  :  10.1016/j.rse.2020.111692

来源: Elsevier

【 摘 要 】

Air temperature (Ta) is an essential climatological component that controls and influences various earth surface processes. In this study, we make the first attempt to employ deep learning for Ta mapping mainly based on space remote sensing and ground station observations. Considering that Ta varies greatly in space and time and is sensitive to many factors, assimilation data and socioeconomic data are also included for a multi-source data fusion based estimation. Specifically, a 5-layers structured deep belief network (DBN) is employed to better capture the complicated and non-linear relationships between Ta and different predictor variables. Layer-wise pre-training process for essential features extraction and fine-tuning process for weight parameters optimization ensure the robust prediction of Ta spatio-temporal distribution. The DBN model was implemented for 0.01 degrees daily maximum Ta mapping across China. The ten-fold cross-validation results indicate that the DBN model achieves promising results with the RMSE of 1.996 degrees C, MAE of 1.539 degrees C, and R of 0.986 at the national scale. Compared with multiple linear regression (MLR), back-propagation neural network (BPNN) and random forest (RF) method, the DBN model reduces the MAE values by 1.340 degrees C, 0.387 degrees C and 0.222 degrees C, respectively. Further analysis on spatial distribution and temporal tendency of prediction errors both validate the great potentials of DBN in Ta estimation.

【 授权许可】

Free