To Automate or Not to Automate
eosAC among the strawberry plants. Photo: Los Huertos Lab
When you think of measuring soil gas flux, you might first think of static chambers and gas chromatography. Static chambers, a traditional field sampling system, appear to be the least expensive option on the market and offer the benefits of a well-blazed trail.
Automated systems are relatively new in the soil gas flux world and appear pricey at a first glance, which may deter some from switching up their sampling methods. Depending on the objectives of the research being done and the sampling regime, these automated systems can end up paying for themselves over time. When finer-scale temporal resolution is needed to understand diurnal, periodic or event-driven fluxes, automated systems allow continuous sampling on fine time scales. These fine time scales would not be possible with vials and syringes, which ultimately give the resolution needed to make confident conclusions. Real-time, highly resolved data can also reduce the time spent second-guessing whether the data obtained is capturing an actual flux event or whether it is due to an error.
Automated systems collect measurements on site, allowing you to check your data in the field, and take it home for in-depth analysis once you leave. Since you have the data at your disposal immediately, the data analysis process can happen whenever you are ready to start.
Here is a real world example of Dr. Marc Los Huertos of Pomona College in California, who was deciding between static chambers/gas chromatography or an automated chamber and analyzer system, and what his reasoning was for choosing the system that he did.
Dr. Los Huertos and his team are studying the relationship between greenhouse gas (CO2, CH4, N2O) flux in agricultural soils and how these fluxes vary in relation to the farming practices employed and the diversity of the landscape. They have a total of 27 field sites, all of which are strawberry fields located in the Salinas and Pajaro Valleys in California. Their furthest field site in Davenport, CA was 50 miles away from their field season headquarters in Prunedale, CA.
At one of the Los Huertos team’s field sites, CA. Photo: Los Huertos Lab
One of Dr. Los Huertos’ main priorities was data quality and reliability, which had been an issue in the past when using the non-continuous vial and syringe method. He found the sparse datasets generated through manual sampling made it difficult to separate natural variability from errors introduced in the handling and lab analysis of gas chromatograph samples. This is critical when trying to identify if an outlier is a real measure of site activity, or simply a bad measurement.
Time is usually a main constraint when it comes to the field season and sampling, with student training being a major component. It was decided that the straight-forward analysis built-in to an automated system would allow the researchers to spend more time interpreting data, rather than trying to extract it from manual samples. The eosAC–Picarro automated system was straightforward for undergraduate students to set up and start collecting data on their own in the field after only a few days.
When comparing the two systems, Dr. Los Huertos was keen to ensure the majority of the work occurred in the field. While both systems have a fieldwork component, the automated system cuts out any processing of samples and shipping/handling involved with glass vials, and reduces the amount of time between when data collection finishes and data processing commences.
Benefits to the Los Huertos Team
*Processing times for static system & gas chromatography (GC) are based on an estimated 1.5 hours per measurement. This includes sample analysis (GC set up, QC checks, maintenance, etc) and data processing (reformatting, rate estimations, using HMR and QA/QC processing).
The single biggest benefit of the automated system to the Los Huertos team has been how quickly they have been able to start analyzing their field data. They managed to avoid having to drive around with a carload of glass vials, which are fragile and can be easily-contaminated, as well as cut their time down by avoiding processing samples in the lab. Instead of having to transport every sample back to the lab for analysis, they have been able to immediately start processing and analyzing their datasets.
Dr. Los Huertos’ team collected 15 measurements daily over their 24 day field campaign, for a total of 360 flux measurements. By using the eosAC system, they were able to avoid the excessive processing time of manual sampling, which at an estimated 1.5 hours per measurement would have taken them several months to finish processing. Additionally, data processing and analysis were able to occur simultaneously during field sampling, furthering their time savings.
Whether you are looking to build your own chamber or going with one already on the market, considering the time and data cost of any soil gas sampling method is important to factor to consider. Resource availability will be a big factor in deciding which method is best for your research objectives. For research where short-term temporal resolution is key, automated systems collect far more data points in the same time frame, resulting in a much more robust dataset overall. Important ecological processes are likely to be missed when samples are taken infrequently.
These considerations are even more important depending on the gas species of interest. Nitrous oxide (N2O) emissions, for example, are particularly variable over time and are often driven by episodic patterns (ie: rain events, fertilizer applications, thawing). If your manual sampling schedule is too infrequent, your ability to capture nitrous oxide emissions after an episodic event may be limited.
Hopefully after reading this, you will have an improved idea of how your needs and resources can help you choose the right method for you. The method you ultimately choose will depend on the objectives of your research and the resources available, whether that is time, labour, or cost.
Many thanks to Dr. Los Huertos and his student Neha Vaingankar for their help and cooperation in writing this case study.
View/download the PDF here.