Using Machine Learning to Predict Inland Aquatic CO2 and CH4 Concentrations and the Effects of Wildfires in the Yukon‐Kuskokwim Delta, Alaska

• Published in: Global biogeochemical cycles 2022-04, Vol.36 (4), p.n/a
• Main Authors: Ludwig, Sarah M., Natali, Susan M., Mann, Paul J., Schade, John D., Holmes, Robert M., Powell, Margaret, Fiske, Greg, Commane, Roisin
• Format: Article
• Language: English
• Subjects: Aquatic ecosystems; Arctic; Carbon; Carbon dioxide; Carbon dioxide concentration; Carbon dioxide emissions; Carbon emissions; Climate change; Composition; Dissolved organic carbon; Edge effect; Emissions; Environmental conditions; fire; Image classification; Imagery; Inland waters; lake; Lakes; Landscape; Learning algorithms; Machine learning; Methane; Moisture content; Nutrient concentrations; Organic carbon; Organic matter; Permafrost; Properties; Soil water; Surface water; Tundra; Water chemistry; Watersheds; Wildfires
• ISSN: 0886-6236; 1944-9224
• DOI: 10.1029/2021GB007146

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. Key Points Waterbody size and shape have strong nonlinear effects on CO2, with the highest concentrations in small, complex, waterbodies Dissolved organic carbon quantity and composition are the most important drivers of CH4 concentrations in waterbodies Wildfires increase the sensitivity of waterbody CO2 and CH4 concentrations to degraded permafrost in upstream watersheds