How quickly do streams lose their natural fizz?...by Lou Maurice and Gareth Farr

Lou testing the field equipment during a wet and
 windy November morning

Carbon dioxide (CO2) is a well known greenhouse gas that is produced both naturally and by human activity. It is important to be able to measure the amounts of CO2 that enter the atmosphere as this enables climate change scientists to put better data into their models, helping them to improve predictions of future climate changes.


Interestingly, streams and rivers are one of the many natural sources of CO2. Biological activity, mainly respiration, produces CO2 in soils, streams and rivers, which is eventually lost into the atmosphere.  In order to calculate the amount of CO2 that is lost from streams, the rate of gas loss from the water needs to be measured.  It is this process that Lou Maurice and I investigated using tracer tests.

Yes it is as cold and wet as it looks!
 Gareth collects elevation survey data to
calculate the slope of a stream

We chose an upland area of South Wales as our study area, not just because of its remote beauty and excellent country pubs, but because it offered numerous streams of varying slope all on the same geology: important factors for our experiment.

The first step was to measure the flow of water in the stream. We did this using a method called ‘salt dilution’, which involves adding small amounts of salt upstream at a known and constant rate, and measuring the peak concentration downstream. After a few calculations, we can compute the velocity of the stream.

Lou collects a sample using a syringe and special
metal gas sample container. the water is injected
 from the bottom to ensure any air bubbles are removed,
then they are taken to the BGS labs in Wallingford
where they are processed   

To measure the rate at which CO2 would be lost from the stream, we injected an inert gas at a constant rate into the top of the stream section at the same time as the salt. The inert gas is sulphur hexafluoride (SF₆) and acts in a similar way to CO2, so is a good ‘proxy’ for the CO2 in the stream. We were then able to measure how much of the inert SF₆ gas was lost from the stream as the water flowed downhill.  It is then possible to calculate the rate at which CO2 gas would be lost over that section of stream.  We repeated this experiment in both low flow conditions in the summer and high flow conditions in the late autumn to see how changes in flow could affect the loss of CO2 from streams and rivers.

Our initial results look promising and we hope to be able to show differences between CO2 loss from streams with steep and shallow gradients, under both high and low flow conditions. In the future we hope to be able to ‘scale up’ these findings to calculate CO2 loss from streams and rivers across the UK.

We need to understand how different flow conditions affect the loss of CO2 from streams so we visited the same stream sections in low flow conditions e.g. summer (left) and then again in high flow conditions e.g. autumn (right)