Overcoming barriers in long-term, continuous monitoring of soil CO₂ flux: A low-cost sensor system

Abstract. Soil CO₂ flux (F_s) is a carbon cycling metric crucial for assessing ecosystem carbon budgets and global warming. However, global F_s datasets often suffer from low temporal-spatial resolution, as well as from spatial bias. F_s observations are severely deficient in tundra and dryland ecosystems due to financial and logistical constraints of current methods for F_s quantification. In this study, we introduce a novel, low-cost sensor system (LC-SS) for long-term, continuous monitoring of soil CO₂ concentration and flux. The LC-SS, built from affordable, open-source hardware and software, offers a cost-effective solution (~USD700), accessible to low-budget users, and opens the scope for research with a large number of sensor system replications. The LC-SS was tested over ~6 months in arid soil conditions, where fluxes are small, and accuracy is critical. CO₂ concentration and soil temperature were measured at 10-min intervals at depths of 5 and 10 cm. The LC-SS demonstrated high stability and minimal maintenance requirements during the tested period. Both diurnal and seasonal soil CO₂ concentration variabilities were observed, highlighting the system's capability of continuous, long-term, in-situ monitoring of soil CO₂ concentration. In addition, F_s was calculated using the measured CO₂ concentration via the gradient method and validated with F_s measured by the flux chamber method using the well-accepted LI-COR gas analyzer system. Gradient method F_s was in good agreement with flux chamber F_s, highlighting the potential for alternative or concurrent use of the LC-SS with current methods for F_s estimation. Leveraging the accuracy and cost-effectiveness of the LC-SS (below 10 % of automated gas analyzer system cost), strategic implementation of LC-SSs could be a promising means to effectively increase the number of measurements, spatially and temporally, ultimately aiding in bridging the gap between global F_s uncertainties and current measurement limitations.