A global soil organic carbon estimation service

Spectroscopy, particularly near-infrared (NIR) and visible-near-infrared (Vis-NIR) spectroscopy(Shen et al., 2022; Shen & Viscarra Rossel, 2021), has become a valuable tool for predicting soil organic carbon (SOC) due to its rapid, non-destructive, and accurate nature. Here are some key insights from recent studies on this application:

This web app integrates data science pipelines for the preprocessing, modelling, and prediction on spectra data to estimate organic carbon content in soil.

I was the lead developer of this platform. I developed a R package for spectroscopic modelling for soil organic carbon, data processing pipelines in Python, and backend APIs with Django and Django REST framework. I also developed early version of the frontend, and contributed to the latest frontend. The app was deployed to AWS.

This project will be publicly released soon.

References

2022

P&RS

Deep transfer learning of global spectra for local soil carbon monitoring

Zefang Shen, Leonardo Ramirez-Lopez, Thorsten Behrens, and 8 more authors

ISPRS Journal of Photogrammetry and Remote Sensing, 2022

Bib HTML

@article{shen2022deep,
  title = {Deep transfer learning of global spectra for local soil carbon monitoring},
  author = {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 Baumann, Philipp and others},
  journal = {ISPRS Journal of Photogrammetry and Remote Sensing},
  volume = {188},
  pages = {190--200},
  year = {2022},
  publisher = {Elsevier},
}

2021

Sci. Rep.

Automated spectroscopic modelling with optimised convolutional neural networks

Zefang Shen, and RA Viscarra Rossel

Scientific Reports, 2021

Bib HTML

@article{shen2021automated,
  title = {Automated spectroscopic modelling with optimised convolutional neural networks},
  author = {Shen, Zefang and Viscarra Rossel, RA},
  journal = {Scientific Reports},
  volume = {11},
  number = {1},
  pages = {208},
  year = {2021},
  publisher = {Nature Publishing Group UK London},
}