Using Machine Learning to Predict Inland Aquatic CO 2 and CH 4 Concentrations and the Effects of Wildfires in the Yukon‐Kuskokwim Delta, Alaska

Ludwig, Sarah M., Natali, Susan M., Mann, Paul, Schade, John D., Holmes, Robert M., Powell, Margaret, Fiske, Greg and Commane, Roisin (2022) Using Machine Learning to Predict Inland Aquatic CO 2 and CH 4 Concentrations and the Effects of Wildfires in the Yukon‐Kuskokwim Delta, Alaska. Global Biogeochemical Cycles, 36 (4). e2021GB007146. ISSN 0886-6236

Global Biogeochemical Cycles - 2022 - Ludwig - Using Machine Learning to Predict Inland Aquatic CO2 and CH4 Concentrations.pdf - Published Version

Download (4MB) | Preview
Official URL:


Climate change is causing an intensification in tundra fires across the Arctic, including the unprecedented 2015 fires in the Yukon-Kuskokwim (YK) Delta. The YK Delta contains extensive surface waters (∼33% cover) and significant quantities of organic carbon, much of which is stored in vulnerable permafrost. Inland aquatic ecosystems act as hot-spots for landscape CO2 and CH4 emissions and likely represent a significant component of the Arctic carbon balance, yet aquatic fluxes of CO2 and CH4 are also some of the most uncertain. We measured dissolved CH4 and CO2 concentrations (n = 364), in surface waters from different types of waterbodies during summers from 2016 to 2019. We used Sentinel-2 multispectral imagery to classify landcover types and area burned in contributing watersheds. We develop a model using machine learning to assess how waterbody properties (size, shape, and landscape properties), environmental conditions (O2, temperature), and surface water chemistry (dissolved organic carbon composition, nutrient concentrations) help predict in situ observations of CH4 and CO2 concentrations across deltaic waterbodies. CO2 concentrations were negatively related to waterbody size and positively related to waterbody edge effects. CH4 concentrations were primarily related to organic matter quantity and composition. Waterbodies in burned watersheds appeared to be less carbon limited and had longer soil water residence times than in unburned watersheds. Our results illustrate the importance of small lakes for regional carbon emissions and demonstrate the need for a mechanistic understanding of the drivers of greenhouse gasses in small waterbodies.

Item Type: Article
Additional Information: Funding information: I would like to acknowledge that our research work took place on the traditional land of the Yup'ik, who have stewarded this land through many generations. Our work would not have been possible without the support of the Yukon Delta National Wildlife Refuge, U.S. Fish and Wildlife Service. This study was supported with funding from a National Aeronautics and Space Administration FINESST grant (80NSSC19K1301) to S.M.L, and National Science Foundation grants (NSF-1044610 and NSF-1561437) to S.M.N.
Subjects: F800 Physical and Terrestrial Geographical and Environmental Sciences
G400 Computer Science
Department: Faculties > Engineering and Environment > Geography and Environmental Sciences
Depositing User: John Coen
Date Deposited: 14 Apr 2022 10:07
Last Modified: 26 Sep 2022 08:00

Actions (login required)

View Item View Item


Downloads per month over past year

View more statistics