Rivers and streams emit much more CO2 at night than during the day

EPFL scientists have found that streams and rivers – or “running waters” – emit on average four times more CO2 at night than during the day. That matters because most estimates of global CO2 emissions are made using measurements taken during the day, thereby introducing a potential bias into the calculation of the global carbon cycle.
© 2021 EPFL

Researchers are starting to get a clearer picture of the role that running waters play in the global carbon cycle. A team of scientists at EPFL’s Stream Biofilm and Ecosystem Research Laboratory (SBER) – which had already shown that mountain streams and rivers emit more CO2 (the main greenhouse gas) than previously thought – has now made an important new discovery together with international colleagues. In an article appearing today in Nature Geoscience, they show that CO2 emissions from running waters are substantially higher at night than during the day. This indicates that calculations of how much CO2 is released by these waters to the atmosphere has been biased too low, leading to incorrect estimates of their contribution to the global carbon cycle.

Researchers had long thought that running waters were less important for global carbon fluxes than the oceans, for instance. Hovever, stream and river networks receive massive amounts of terrestrial organic carbon which they decompose with the ultimate production of respiratory CO2. Calculating just how much of this CO2 is released to the atmosphere is very difficult, due to the complexity of the networks that drain the continents. Until now, researchers have been basing their estimates primarily on measurements taken manually during the day.

And that’s where the SBER scientists spotted the calculation bias. They found that 90% of existing measurements were made between 8am and 4pm, and, by comparing these measurements with data collected continuously by automated sensors, that CO2 emissions reached their peak during this 8am to 4pm window just 10% of the time.

The scientists used automated sensors from 66 rivers around the world – including in previously underrepresented areas such as the Congo, the Amazon, the Arctic and global mountains – to supplement diurnal measurements with measurements taken at night. Thanks to their meticulous approach, they found that CO2 emissions were an average of 27% - or about 4 times - higher in the nighttime hours. These results show how using automated sensor networks and big environmental data can unravel yet unseen dynamics of stream ecosystems, adds Tom battin, SBER’s director.

«Much of the CO2 produced in rivers and streams is absorbed by photosynthesis during the day, thereby reducing the amount that gets released into the atmosphere.»      Lluís Gómez-Gener

“This difference is due to several factors,” says Lluís Gómez-Gener, an SBER scientist and one of the study’s lead author. “The main one relates to photosynthesis – much of the CO2 produced in rivers and streams is absorbed by photosynthesis during the day, thereby reducing the amount that gets released into the atmosphere.”

Other factors that explain fluctuations in CO2 emissions over the course of 24 hours include the vegetation, canopy shading, altitude, temperatures, slope and turbulence of the water. The biggest swings in diurnal versus nocturnal emissions were found to occur in temperate forests and mountain prairies. Large fluctuations were also consistently observed during the summer and in open canopy reaches, whereas reduced amplitudes were observed in streams with closed canopies. That’s because heat and sunlight are key drivers of photosynthesis, and therefore of CO2 fixation.

The scientists’ findings once again underscore the highly complex, interconnected nature of streams and rivers with carbon fluxes, and the need to develop a more comprehensive understanding of the mechanisms involved. In this case in particular, their work shows that the role of running waters in the global carbon cycle needs to be better analyzed and quantified through accurate measurements using novel technologies and approaches.