Paving the way for future methods to measuring soil carbon content

"We are certainly increasing our understanding of how to improve the accuracy of soil carbon measurements, in particular how to reduce the impact of water in the soil", says Bo Stenberg, soil researcher at the Swedish University of Agricultural Sciences (SLU) and project leader of a methods development project for the accounting of soil carbon and other soil properties directly in the field.

Together with his international team, Bo Stenberg tries to find ways to use in-field proximal sensing more efficiently, in other words – without measurement disturbance caused by different conditions in the soil. The method is key to making carbon farming work, since it can generate detailed soil information on a large number of sampling points, taking account of the large spatial variation that exists in the soil. 

"Proximal sensing, in our project, is a fancy notion for a rather straight-forward thing, namely to go out in the field and scan the soil with a handheld or vehicle mounted sensor. By making the very surface of the soil dry, or at least dryer, we can improve the accuracy of the measurements. Also the surface structure can be improved to more resemble conditions in the lab", says Bo Stenberg.

The potential of satellite data is in combination with other data

Johanna Wetterlind, another researcher at SLU, is working with a similar project, but focussing on improving satellite data collection – in other words, remote sensing. Particularly, the project is looking into the extent to which soil moisture, soil texture, salinity and plant residues in the field affect the carbon measurements.

Satellites are perfect for cheap and efficient collection of information over larger areas. But for soil analyses, such as the amount of carbon, it comes with a lot of challenges. For instance clear, bare and preferably dry soil is needed to get reliable data. However, depending of region, clouds can be a major problem, and soil is often moist or covered with crops.

"We need to overcome these difficulties for it to be fruitful. Methods to select satellite information when the best possible conditions apply, like dry bare soil for instance, are part of the solution. Other ways to overcome the difficulties are to combine the satellite data with data from other data collection methods. Without sufficient information on the ground, it will be very difficult to get reliable information from satellites to find out how much carbon there is in the soil. With good measurements from proximal sensing there is an opportunity to make models that can be used together with satellite data", says Johanna Wetterlind.

Uncertainties still large

There are currently ongoing initiatives related to carbon farming both in the political sphere and in the business sphere. However, there are parts of the puzzle that are missing for it to work out properly.

"There is still a bit to go until we have reliable measurements of soil carbon that can be used to verify soil management benefitting carbon farming. The change in soil carbon content needs to be greater than the uncertainties of the measurements to have relatively accurate measurements, but we are not there yet", says Johanna Wetterlind.

"There is a risk that the market and the politics is moving a bit too fast. Another uncertainty is that you would need about 30 years before you can observe any changes in the carbon content of the soil. So, the question is if we will we pay farmers for their carbon farming activities before, during, or after this period", says Bo Stenberg.

Results in about a year

Having patience, and waiting for results indicating the potential of methods like remote and proximal sensing for soil carbon measurement, seems like the preferable way to go before kicking off carbon farming at full scale.

"Our project ends in a little more than one year, and the aim is to suggest a “best practice” for using satellite data for carbon accounting, and to advice policymakers on the possibilities and limitations of the techniques for that purpose", says Johanna Wetterlind.

Johanna Wetterlind, researcher at the Swedish University of Agricultural Sciences (SLU), works in a project that evaluates the potential of remote sensing for measuring soil carbon content. Photo: Private.

Bo Stenberg, researcher at the Swedish University of Agricultural Sciences (SLU), leads a project focused on evaluating the potential of in-field proximal sensing for measuring soil carbon content. Photo: Anna-Karin Johnson, SLU.

Satellite image over cloudy fields from sentinel-2. By using sentinel-2 images, it is possible to estimate the amount of carbon in the field. But clouds and variable soil conditions can disturb the estimations. Johanna Wetterlind is part of a project team that look into how to minimise the impact of these disturbances. Photo: Sentinelhub.

Researchers use a handheld sensor to scan the soil and measure the carbon content.  Photo: Private

FACTS BOX

Read more about the projects

ProbeField, the project Bo Stenberg is leading, focused on proximal sensing.

STEROPES, the project Johanna Wetterlind works in, focused on remote sensing.

Carbon farming

According to a report from the European Parliament 2021 carbon farming “refers to sequestering and storing carbon and/or reducing greenhouse gas emissions at farm level. It offers significant but uncertain mitigation potential in the EU, can deliver co-benefits to farmers and society, but also carries risks that need to be managed.”

Remote sensing

According to a definition by Campbell (2002) “Remote sensing is the practice of deriving information about the Earth’s land and water surfaces using images acquired from an overhead perspective, using electromagnetic radiation in one or more regions of the electromagnetic spectrum, reflected or emitted from the Earth’s surface.”

The remote sensing referred to in the text is satellite images, primarily from Sentinel-2, which has multispectral sensors measuring reflected light in the visible, near infrared, and short wave infrared wavelengths.

Read more about the satellite Sentinel-2

Proximal sensing

Proximal sensing involves the use of sensors in close proximity (≤ 2 m) to the soil or crop, such as on a tractor or harvester or ground-based robot. It can also be handheld. The definition is not very rigid when it comes to contact. A proximal sensor could actually be invasive to some extent, but should not involve extractions or addition of chemicals.

Contact information:

ProbeField: Project Coordinator: Bo Stenberg: bo.stenberg@slu.se 

STEROPES: Project Communication Representative: Johanna Wetterlind Johanna.Wetterlind@slu.se