Preprints
https://doi.org/10.5194/egusphere-2024-1360
https://doi.org/10.5194/egusphere-2024-1360
04 Jun 2024
 | 04 Jun 2024

NN-TOC v1: global prediction of total organic carbon in marine sediments using deep neural networks

Naveenkumar Parameswaran, Everardo González, Ewa Burwicz-Galerne, Malte Braack, and Klaus Wallmann

Abstract. Spatial predictions of total organic carbon (TOC) concentrations and stocks are crucial for understanding marine sediments’ role as a significant carbon sink in the global carbon cycle. In this study, we present a geospatial prediction of TOC concentrations and stocks at a 5 x 5 arc minute grid scale, using a deep learning model — a novel machine learning approach based on a new compilation of over 22,000 global TOC measurements and a new set of predictors, such as seafloor lithologies, grain size distribution, and an alpha-chlorophyll satellite data. In our study, we compared the predictions and discuss the limitations from various machine learning methods. Our findings reveal that the neural network approach outperforms methods such as k Nearest Neighbors and random forests, which tend to overfit to the training data, especially in highly heterogeneous and  complex geological settings. We provide estimates of mean TOC concentrations and total carbon stock in both continental  shelves and deep sea settings across various marine regions and oceans. Our model suggests that the upper 10 cm of oceanic sediments harbors approximately 171 Pg of TOC stock and has a mean TOC concentration of 0.68 %. Furthermore, we introduce a standardized methodology for quantifying predictive uncertainty using Monte Carlo dropout and present a map of information gain, that measures the expected increase in model knowledge achieved through in-situ sampling at specific locations which is  pivotal for sampling strategy planning.

Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims made in the text, published maps, institutional affiliations, or any other geographical representation in this preprint. The responsibility to include appropriate place names lies with the authors.
Naveenkumar Parameswaran, Everardo González, Ewa Burwicz-Galerne, Malte Braack, and Klaus Wallmann

Status: final response (author comments only)

Comment types: AC – author | RC – referee | CC – community | EC – editor | CEC – chief editor | : Report abuse
  • CEC1: 'Comment on egusphere-2024-1360', Juan Antonio Añel, 20 Jun 2024
    • AC1: 'Reply on CEC1', Naveenkumar Parameswaran, 21 Jun 2024
      • CEC2: 'Reply on AC1', Juan Antonio Añel, 21 Jun 2024
  • RC1: 'Comment on egusphere-2024-1360', Taylor Lee, 24 Jun 2024
    • RC2: 'Reply on RC1', Taylor Lee, 24 Jun 2024
    • AC2: 'Reply on RC1', Naveenkumar Parameswaran, 25 Jul 2024
  • RC3: 'Comment on egusphere-2024-1360', Sarah Paradis, 30 Jul 2024
    • AC3: 'Reply on RC3', Naveenkumar Parameswaran, 25 Oct 2024
Naveenkumar Parameswaran, Everardo González, Ewa Burwicz-Galerne, Malte Braack, and Klaus Wallmann

Data sets

Dataset for the Global Prediction Of Total Organic Carbon In Marine Sediments Using Deep Neural Networks (nn-toc) Naveenkumar Parameswaran et al. https://zenodo.org/records/11186224

Interactive computing environment

Global Prediction Of Total Organic Carbon In Marine Sediments Using Deep Neural Networks (nn-toc). Naveenkumar Parameswaran et al. https://doi.org/10.3289/SW_3_2024

Naveenkumar Parameswaran, Everardo González, Ewa Burwicz-Galerne, Malte Braack, and Klaus Wallmann

Viewed

Total article views: 622 (including HTML, PDF, and XML)
HTML PDF XML Total BibTeX EndNote
396 158 68 622 18 20
  • HTML: 396
  • PDF: 158
  • XML: 68
  • Total: 622
  • BibTeX: 18
  • EndNote: 20
Views and downloads (calculated since 04 Jun 2024)
Cumulative views and downloads (calculated since 04 Jun 2024)

Viewed (geographical distribution)

Total article views: 617 (including HTML, PDF, and XML) Thereof 617 with geography defined and 0 with unknown origin.
Country # Views %
  • 1
1
 
 
 
 

Cited

Latest update: 08 Dec 2024
Download
Short summary
Our research uses deep learning to predict organic carbon stocks in ocean sediments, crucial for understanding their role in the global carbon cycle. By analyzing over 22,000 samples and various seafloor characteristics, our model gives more accurate results than traditional methods. We estimate the top 10 cm of ocean sediments hold about 171 petagrams of carbon. This work enhances carbon stock estimates and helps plan future sampling strategies to better understand oceanic carbon burial.