Home About Browse Search
Svenska


Deep transfer learning of global spectra for local soil carbon monitoring

Shen, Zefang and Ramirez-Lopez, Leonardo and Behrens, Thorsten and Cui, Lei and Zhang, Mingxi and Walden, Lewis and Wetterlind, Johanna and Shi, Zhou and Sudduth, Kenneth A. and Song, Yongze and Catambay, Kevin and Rossel, Raphael A. Viscarra (2022). Deep transfer learning of global spectra for local soil carbon monitoring. ISPRS Journal of Photogrammetry and Remote Sensing. 188 , 190-200
[Research article]

[img] PDF
3MB

Abstract

There is global interest in spectroscopy and the development of large and diverse soil spectral libraries (SSL) to model soil organic carbon (SOC) and monitor, report, and verify (MRV) its changes. The reason is that increasing SOC can improve food production and mitigate climate change. However, 'global' modelling of SOC with such diverse and hyperdimensional SSLs do not generalise well locally, e.g. at a field scale. To address this challenge, we propose deep transfer learning (DTL) to leverage useful information from large-scale SSLs to assist local modelling. We used one global, three country-specific SSLs and data from three local sites with DTL to improve the modelling and localise the SOC estimates in individual fields or farms in each country. With DTL, we transferred instances from the SSLs, representations from one-dimensional convolutional neural networks (1DCNNs) trained on the SSLs, and both instances and representations to improve local modelling. Transferring instances effectively used information from the global SSL to most accurately estimate SOC in each site, reducing the root mean square error (RMSE) by 25.8% on average compared with local modelling. Our results highlight the effectiveness of DTL and the value of diverse, global SSLs for accurate local SOC predictions. Applying DTL with a global SSL one could estimate SOC anywhere in the world more accurately, rapidly, and cost-effectively, enabling MRV protocols to monitor SOC changes.

Authors/Creators:Shen, Zefang and Ramirez-Lopez, Leonardo and Behrens, Thorsten and Cui, Lei and Zhang, Mingxi and Walden, Lewis and Wetterlind, Johanna and Shi, Zhou and Sudduth, Kenneth A. and Song, Yongze and Catambay, Kevin and Rossel, Raphael A. Viscarra
Title:Deep transfer learning of global spectra for local soil carbon monitoring
Series Name/Journal:ISPRS Journal of Photogrammetry and Remote Sensing
Year of publishing :2022
Volume:188
Page range:190-200
Number of Pages:11
Publisher:ELSEVIER
ISSN:0924-2716
Language:English
Publication Type:Research article
Article category:Scientific peer reviewed
Version:Published version
Copyright:Creative Commons: Attribution-Noncommercial-No Derivative Works 4.0
Full Text Status:Public
Subjects:(A) Swedish standard research categories 2011 > 2 Engineering and Technology > 207 Environmental Engineering > Remote Sensing
(A) Swedish standard research categories 2011 > 4 Agricultural Sciences > 401 Agricultural, Forestry and Fisheries > Soil Science
Keywords:Soil organic carbon, Visible -near-infrared spectra, Transfer learning, Deep learning, Spectral library
URN:NBN:urn:nbn:se:slu:epsilon-p-117328
Permanent URL:
http://urn.kb.se/resolve?urn=urn:nbn:se:slu:epsilon-p-117328
Additional ID:
Type of IDID
DOI10.1016/j.isprsjprs.2022.04.009
Web of Science (WoS)000797216600002
ID Code:28231
Faculty:NJ - Fakulteten för naturresurser och jordbruksvetenskap
Department:(NL, NJ) > Dept. of Soil and Environment
(S) > Dept. of Soil and Environment
Deposited By: SLUpub Connector
Deposited On:03 Jun 2022 07:16
Metadata Last Modified:03 Jun 2022 07:21

Repository Staff Only: item control page

Downloads

Downloads per year (since September 2012)

View more statistics

Downloads
Hits