Preprints
https://doi.org/10.5194/gmd-2021-426
https://doi.org/10.5194/gmd-2021-426
Submitted as: model description paper
14 Feb 2022
Submitted as: model description paper | 14 Feb 2022
Status: a revised version of this preprint was accepted for the journal GMD and is expected to appear here in due course.

A map of global peatland extent created using machine learning (Peat-ML)

Joe R. Melton1, Ed Chan2, Koreen Millard3, Matthew Fortier1, R. Scott Winton4,5, Javier M. Martín-López6, Hinsby Cadillo-Quiroz7, Darren Kidd8, and Louis V. Verchot6 Joe R. Melton et al.
  • 1Climate Research Division, Environment and Climate Change Canada, Victoria, B.C., Canada
  • 2Climate Research Division, Environment and Climate Change Canada, Toronto, ON, Canada
  • 3Geography and Environmental Studies, Carleton University, Ottawa, ON, Canada
  • 4Institute of Biogeochemistry and Pollutant Dynamics, ETH Zurich, 8092 Zurich, Switzerland
  • 5Department of Surface Waters, Eawag, Swiss Federal Institution of Aquatic Science and Technology, 6047 Kastanienbaum, Switzerland
  • 6Agroecosystems and Sustainable Landscapes Program, Alliance Bioversity-CIAT, Cali, Colombia
  • 7School of Life Sciences, Arizona State University, Tempe, AZ 85287, USA
  • 8Natural Values Science Services, Department of Natural Resources and Environment, Tasmania, Australia

Abstract. Peatlands store large amounts of soil carbon and freshwater, constituting an important component of the global carbon and hydrologic cycles. Accurate information on the global extent and distribution of peatlands is presently lacking but is needed by Earth System Models (ESMs) to simulate the effects of climate change on the global carbon and hydrologic balance. Here, we present Peat-ML, a spatially continuous global map of peatland fractional coverage generated using machine learning techniques suitable for use as a prescribed geophysical field in an ESM. Inputs to our statistical model follow drivers of peatland formation and include spatially distributed climate, geomorphological and soil data, along with remotely-sensed vegetation indices. Available maps of peatland fractional coverage for 14 relatively extensive regions were used along with mapped ecoregions of non-peatland areas to train the statistical model. In addition to qualitative comparisons to other maps in the literature, we estimated model error in two ways. The first estimate used the training data in a blocked leave-one-out cross-validation strategy designed to minimize the influence of spatial autocorrelation. That approach yielded an average r2 of 0.73 with a root mean squared error and mean bias error of 9.11 % and −0.36 %, respectively. Our second error estimate was generated by comparing Peat-ML against a high-quality, extensively ground-truthed map generated by Ducks Unlimited Canada for the Canadian Boreal Plains region. This comparison suggests our map to be of comparable quality to mapping products generated through more traditional approaches, at least for boreal peatlands.

Joe R. Melton et al.

Status: closed

Comment types: AC – author | RC – referee | CC – community | EC – editor | CEC – chief editor | : Report abuse
  • CEC1: 'Comment on gmd-2021-426', Juan Antonio Añel, 01 Mar 2022
    • AC1: 'Reply on CEC1', Joe Melton, 10 Mar 2022
  • RC1: 'Comment on gmd-2021-426', Anonymous Referee #1, 06 Apr 2022
    • AC2: 'Authors' reply to Reviewers (Julie Loisel and #2)', Joe Melton, 29 Apr 2022
  • RC2: 'Comment on gmd-2021-426', Anonymous Referee #2, 19 Apr 2022
    • AC2: 'Authors' reply to Reviewers (Julie Loisel and #2)', Joe Melton, 29 Apr 2022
  • AC2: 'Authors' reply to Reviewers (Julie Loisel and #2)', Joe Melton, 29 Apr 2022

Status: closed

Comment types: AC – author | RC – referee | CC – community | EC – editor | CEC – chief editor | : Report abuse
  • CEC1: 'Comment on gmd-2021-426', Juan Antonio Añel, 01 Mar 2022
    • AC1: 'Reply on CEC1', Joe Melton, 10 Mar 2022
  • RC1: 'Comment on gmd-2021-426', Anonymous Referee #1, 06 Apr 2022
    • AC2: 'Authors' reply to Reviewers (Julie Loisel and #2)', Joe Melton, 29 Apr 2022
  • RC2: 'Comment on gmd-2021-426', Anonymous Referee #2, 19 Apr 2022
    • AC2: 'Authors' reply to Reviewers (Julie Loisel and #2)', Joe Melton, 29 Apr 2022
  • AC2: 'Authors' reply to Reviewers (Julie Loisel and #2)', Joe Melton, 29 Apr 2022

Joe R. Melton et al.

Data sets

A map of global peatland extent created using machine learning (Peat-ML) Joe R. Melton, Ed Chan, Koreen Millard, Matthew Fortier, R. Scott Winton, Javier M. Martín-López, Hinsby Cadillo-Quiroz, Darren Kidd, and Louis V. Verchot https://doi.org/10.5281/zenodo.5794336

Model code and software

Code for A map of global peatland extent created using machine learning (Peat-ML) Joe R. Melton, Ed Chan, Matthew Fortier https://gitlab.com/jormelton/peatlandmachinelearning

Joe R. Melton et al.

Viewed

Total article views: 875 (including HTML, PDF, and XML)
HTML PDF XML Total BibTeX EndNote
623 227 25 875 7 15
  • HTML: 623
  • PDF: 227
  • XML: 25
  • Total: 875
  • BibTeX: 7
  • EndNote: 15
Views and downloads (calculated since 14 Feb 2022)
Cumulative views and downloads (calculated since 14 Feb 2022)

Viewed (geographical distribution)

Total article views: 781 (including HTML, PDF, and XML) Thereof 781 with geography defined and 0 with unknown origin.
Country # Views %
  • 1
1
 
 
 
 
Latest update: 20 May 2022
Download
Short summary
Peat-ML is a high-resolution global peatland extent map generated using machine learning techniques. Peatlands are important in the global carbon and water cycles but their extent is poorly known. We generated Peat-ML using drivers of peatland formation including climate, soil, geomorphological, and vegetation data, and train the model with regional peatland maps. Our accuracy estimation approaches suggest Peat-ML is of comparable or higher quality than other available peatland mapping products.