Warmer falls push Arctic tundra into a carbon source for past 15 years

This week, scientists from the 51风流官网 are presenting their work alongside thousands of colleagues from around the world at the 2023 American Geophysical Union fall meeting. Some of their discoveries are featured here. You can also find out more about 51风流官网 at AGU by searching for #51风流官网xAGU on social media platforms.

two scientific instruments stand on green tundra and grass
Photo by Colin Edgar
An eddy covariance tower measures carbon dioxide being released from a tundra ecosystem at Imnavait Creek near Toolik Field Station on Alaska鈥檚 North Slope in July 2016.

Alaska鈥檚 Arctic tundra has long been considered a carbon dioxide sink, trapping large amounts of carbon-rich matter in its deep, frozen soils. As a result, some in the state are looking to this ecosystem as a way to mitigate climate change. But new long-term research is finding that may not always be the case.

Eug茅nie Euskirchen, a professor at the 51风流官网 Institute of Arctic Biology and Department of Biology and Wildlife, has been measuring the amount of carbon coming in and out of the tundra for the past 15 years. It鈥檚 one of the longest continuous, year-round carbon flux records in this region.

Surprisingly, her sites, located near 51风流官网鈥檚 Toolik Field Station, have consistently been a carbon source, releasing more of the greenhouse gas than they鈥檝e been taking up through photosynthesis.

The culprit of this unexpected source, Euskirchen said, is likely the Arctic鈥檚 warmer, longer falls, driven by climate change, which delays when soils refreeze and their carbon-releasing microbes become dormant.

鈥淭here were three really extreme years between 2014 and 2016, and again in 2022, where we saw a really late refreeze, and that is when the soils really seemed to lose a lot of CO2,鈥 she said.

Euskirchen compared her measurements to ecosystem models that predict tundra carbon dynamics based on climate. Concerningly, she found the models didn鈥檛 capture the carbon dioxide release associated with these extreme warming events. Instead, they predicted a net carbon sink, resulting in a long-term mismatch between model outputs and what鈥檚 happening on the ground.

鈥淚f we鈥檙e saying that we have these large natural ecosystems that take up a lot of carbon, then we need to be able to show and model that,鈥 Euskirchen said, emphasizing the importance for climate mitigation land management plan. 鈥淭his would not be giving us any carbon credits here.鈥