Reconstructing long-term global satellite-based soil moisture data using deep learning method
(2023) In Frontiers in Earth Science 11.- Abstract
Soil moisture is an essential component for the planetary balance between land surface water and energy. Obtaining long-term global soil moisture data is important for understanding the water cycle changes in the warming climate. To date several satellite soil moisture products are being developed with varying retrieval algorithms, however with considerable missing values. To resolve the data gaps, here we have constructed two global satellite soil moisture products, i.e., the CCI (Climate Change Initiative soil moisture, 1989–2021; CCIori hereafter) and the CM (Correlation Merging soil moisture, 2006–2019; CMori hereafter) products separately using a Convolutional Neural Network (CNN) with autoencoding approach,... (More)
Soil moisture is an essential component for the planetary balance between land surface water and energy. Obtaining long-term global soil moisture data is important for understanding the water cycle changes in the warming climate. To date several satellite soil moisture products are being developed with varying retrieval algorithms, however with considerable missing values. To resolve the data gaps, here we have constructed two global satellite soil moisture products, i.e., the CCI (Climate Change Initiative soil moisture, 1989–2021; CCIori hereafter) and the CM (Correlation Merging soil moisture, 2006–2019; CMori hereafter) products separately using a Convolutional Neural Network (CNN) with autoencoding approach, which considers soil moisture variability in both time and space. The reconstructed datasets, namely CCIrec and CMrec, are cross-evaluated with artificial missing values, and further againt in-situ observations from 12 networks including 485 stations globally, with multiple error metrics of correlation coefficients (R), bias, root mean square errors (RMSE) and unbiased root mean square error (ubRMSE) respectively. The cross-validation results show that the reconstructed missing values have high R (0.987 and 0.974, respectively) and low RMSE (0.015 and 0.032 m3/m3, respectively) with the original ones. The in-situ validation shows that the global mean R between CCIrec (CCIori) and in-situ observations is 0.590 (0.581), RMSE is 0.093 (0.093) m3/m3, ubRMSE is 0.059 (0.058) m3/m3, bias is 0.032 (0.037) m3/m3 respectively; CMrec (CMori) shows quite similar results. The added value of this study is to provide long-term gap-free satellite soil moisture products globally, which helps studies in the fields of hydrology, meteorology, ecology and climate sciences.
(Less)
- author
- Hu, Yifan ; Wang, Guojie ; Wei, Xikun ; Zhou, Feihong ; Kattel, Giri ; Amankwah, Solomon Obiri Yeboah ; Hagan, Daniel Fiifi Tawia and Duan, Zheng LU
- organization
- publishing date
- 2023-02
- type
- Contribution to journal
- publication status
- published
- subject
- keywords
- data reconstruction, deep learning, long-term, satellite-based, soil moisture
- in
- Frontiers in Earth Science
- volume
- 11
- article number
- 1130853
- publisher
- Frontiers Media S. A.
- external identifiers
-
- scopus:85148209899
- ISSN
- 2296-6463
- DOI
- 10.3389/feart.2023.1130853
- language
- English
- LU publication?
- yes
- id
- ccbdc626-6b89-4e4e-bb2e-385e8f3d9e04
- date added to LUP
- 2023-03-06 10:12:43
- date last changed
- 2023-03-06 10:12:43
@article{ccbdc626-6b89-4e4e-bb2e-385e8f3d9e04,
abstract = {{<p>Soil moisture is an essential component for the planetary balance between land surface water and energy. Obtaining long-term global soil moisture data is important for understanding the water cycle changes in the warming climate. To date several satellite soil moisture products are being developed with varying retrieval algorithms, however with considerable missing values. To resolve the data gaps, here we have constructed two global satellite soil moisture products, i.e., the CCI (Climate Change Initiative soil moisture, 1989–2021; CCI<sub>ori</sub> hereafter) and the CM (Correlation Merging soil moisture, 2006–2019; CM<sub>ori</sub> hereafter) products separately using a Convolutional Neural Network (CNN) with autoencoding approach, which considers soil moisture variability in both time and space. The reconstructed datasets, namely CCIr<sub>ec</sub> and CM<sub>rec</sub>, are cross-evaluated with artificial missing values, and further againt in-situ observations from 12 networks including 485 stations globally, with multiple error metrics of correlation coefficients (R), bias, root mean square errors (RMSE) and unbiased root mean square error (ubRMSE) respectively. The cross-validation results show that the reconstructed missing values have high R (0.987 and 0.974, respectively) and low RMSE (0.015 and 0.032 m<sup>3</sup>/m<sup>3</sup>, respectively) with the original ones. The in-situ validation shows that the global mean R between CCI<sub>rec</sub> (CCI<sub>ori</sub>) and in-situ observations is 0.590 (0.581), RMSE is 0.093 (0.093) m<sup>3</sup>/m<sup>3</sup>, ubRMSE is 0.059 (0.058) m<sup>3</sup>/m<sup>3</sup>, bias is 0.032 (0.037) m<sup>3</sup>/m<sup>3</sup> respectively; CM<sub>rec</sub> (CM<sub>ori</sub>) shows quite similar results. The added value of this study is to provide long-term gap-free satellite soil moisture products globally, which helps studies in the fields of hydrology, meteorology, ecology and climate sciences.</p>}},
author = {{Hu, Yifan and Wang, Guojie and Wei, Xikun and Zhou, Feihong and Kattel, Giri and Amankwah, Solomon Obiri Yeboah and Hagan, Daniel Fiifi Tawia and Duan, Zheng}},
issn = {{2296-6463}},
keywords = {{data reconstruction; deep learning; long-term; satellite-based; soil moisture}},
language = {{eng}},
publisher = {{Frontiers Media S. A.}},
series = {{Frontiers in Earth Science}},
title = {{Reconstructing long-term global satellite-based soil moisture data using deep learning method}},
url = {{http://dx.doi.org/10.3389/feart.2023.1130853}},
doi = {{10.3389/feart.2023.1130853}},
volume = {{11}},
year = {{2023}},
}