Publications

Su, PF; Abera, T; Guan, YL; Pellikka, P (2023). Image-to-Image Training for Spatially Seamless Air Temperature Estimation With Satellite Images and Station Data. IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING, 16, 3353-3363.

Abstract
Air temperature at approximately 2 m above the ground (T-a) is one of the most important environmental and biophysical parameters to study various earth surface processes. T-a measured from meteorological stations is inadequate to study its spatio-temporal patterns since the stations are unevenly and sparsely distributed. Satellite-derived land surface temperature (LST) provides global coverage, and is generally utilized to estimate T-a due to the close relationship between LST and T-a. However, LST products are sensitive to cloud contamination, resulting in missing values in LST and leading to the estimated T-a being spatially incomplete. To solve the missing data problem, we propose a deep learning method to estimate spatially seamless T-a from LST that contains missing values. Experimental results on 5-year data of mainland China illustrate that the image-to-image training strategy alleviates the missing data problem and fills the gaps in LST implicitly. Plus, the strong linear relationships between observed daily mean T-a (T-mean), daily minimum T-a (T-min), and daily maximum T-a (T-max) make the estimation of T-mean, T-min, and T(max )simultaneously possible. For mainland China, the proposed method achieves results with R-2 of 0.962, 0.953, 0.944, mean absolute error (MAE) of 1.793 degrees C, 2.143 degrees C, and 2.125 degrees C, and root-mean-square error (RMSE) of 2.376 degrees C, 2.808 degrees C, and 2.823 degrees C for T-mean, T-min, and T-max, respectively. O

DOI:
10.1109/JSTARS.2023.3256363

ISSN:
2151-1535