Combining Multiple Techniques: Benefits of Unified Measurements
Researchers are no strangers to using multiple measurement techniques to cover the many scales of spatial and temporal variation experienced in the field. On the Eosense blog, we’ve surveyed a number of publications by scientists who need multiple measurement techniques to answer their core research questions, ranging from quantifying aqueous CO2 fluxes to measuring the spatial variability of N2O in agricultural systems.
Recently, I had the pleasure of presenting at the European Geophysical Union annual meeting in Vienna in a session called “Production and transport of gases in the soil: measurements and modelling.” In my presentation, I looked at how two different techniques that measure soil gas flux, the gradient and chamber methods, can be used in unison to yield more information.
All of the major greenhouse gases (CO2, N2O, CH4) are produced, consumed, and stored in the soil profile and ultimately the balance of these production, consumption and storage processes determine what is emitted from the soil to the atmosphere. In order to measure this production and consumption, researchers most often use either chamber methods, or the flux gradient approach.
Chambers are commonly used and widely recognized as a good method to measure soil to atmosphere gas fluxes. However, they only offer information about what is passing across the soil-atmosphere boundary and don’t tell researchers what has happened within the soil profile that ultimately resulted in that soil-atmosphere flux. Alternatively, the gradient method offers depth-resolved information about what production and consumption look like in the soil profile, but is strongly affected by the researcher’s estimate of soil gas diffusivity which has been traditionally very difficult to estimate in situ. In our research, we wanted to demonstrate how the chamber and gradient based methods could be used in unison to address the downfalls of using each method in isolation.
In order to do this, we simulated the production and diffusion of gas (CO2 in this case) in the soil profile. In the model, a chamber was used to measure gas flux, and gas concentration sensors were used to monitor the in-soil gas concentration. Using Fick’s second law and a numerical optimization routine we tried to recover information about the soil gas diffusivity and the depth dependent soil gas production. Fundamentally what we are trying to achieve with this approach is to use the chamber measurements to reduce the uncertainty in the diffusivity estimates that are subsequently used to estimate the gradient based fluxes.
Simulated results suggested that using the methods (simulated methods, for now) in unison allowed for a good estimate of the soil gas diffusivity, and as a result allowed us to estimate the depth dependent soil gas production with good accuracy (see the figure below or view the whole PICO presentation). Further simulations across a range of diffusivity and production rates showed that the inverse estimates of the diffusivity and production in the upper soil layers were good; however, in the deeper soils where gradients are typically smaller the model had a harder time estimating the true value due to increased relative fitting errors.
While this work only scratches the surface of the opportunities of using multiple approaches, the results we presented at EGU 2016 show that the use of these two methods in unison does help overcome the downfalls of using either method in isolation. More work remains to be done to ensure that the method is robust, and to make the numerical optimization used in this presentation accessible to researchers who have, or want to use, these types of combined measurement techniques.
If you’re doing similar research, have questions, or are interested in applying these approaches in the future at your field sites, please get in touch! We’d love to feature your work in upcoming blog articles, and are interested in collaboration on projects that are applying multiple measurement approaches to answer questions about spatial and temporal variability of greenhouse gas emissions.