A new study conducted by researchers at Sungkyunwan University in Suwon, South Korea, and Hanyang University in Seoul, South Korea, proposes a technique for accurately predicting carbon dioxide (CO2) concentrations in flow fields with temperature gradients (1). The new approach combines machine learning with single-laser path absorption spectrum measurements using simulated data (1).
Concept depicting the issue of carbon dioxide emissions and its impact on nature in the form of a pond in the shape of a co2 symbol located in a lush forest | Image Credit: © Gan - stock.adobe.com
Conventional methods for measuring CO2 concentrations in flow fields relies on tunable diode laser absorption spectroscopy (1). This method calculates concentrations based on the ratio of two integrated absorbances from spectral lines with different temperature dependencies (1). However, this approach has limitations: because of the presence of temperature gradients, significant deviations from actual concentrations often emerge, leading to inaccuracies (1).
Because of these challenges, the research team proposed a new method that leverages the entire absorption feature of CO2 (1). The researchers decided that their model should be data-driven and predictive, with the idea that it would then be able to compensate for the impact of nonuniform temperature fields on the absorption spectrum (1).
Using simulated data helped the researchers construct and program the model, given the impracticality of collecting a comprehensive data set covering all possible temperature gradient conditions (1). To validate their model, the researchers conducted experiments using real-world data, comparing its performance to the traditional two-line method. In all test cases, the newly proposed model outperformed the conventional approach, demonstrating its superior predictive capabilities in the presence of temperature gradients (1).
Additionally, a gradient-weighted regression activation mapping analysis confirmed that the model harnesses both peak intensities and changes in the shape of absorption lines for prediction, further highlighting the effectiveness of the method (1).
Currently, significant interest in climate science has triggered many research endeavors that all seek to combat climate change concerns. An approach for detecting CO2 concentrations accurately helps in this context. This new method proposed by the research team resolves the longstanding challenges posed by temperature gradients, paving the way for more precise and reliable environmental data, which is key to finding solutions to preserve the environment and our planet.
(1) Choi, J.; Bong, C.; Yoo, J.; Bak, M.-S. Carbon Dioxide Concentration Estimation in Nonuniform Temperature Fields Based on Single-Pass Tunable Diode Laser Absorption Spectroscopy. Appl. Spectrosc. 2023, ASAP. DOI: 10.117700037028231191782
This article was written with the help of artificial intelligence and has been edited to ensure accuracy and clarity. You can read more about our policy for using AI here.
PNNL and BaySpec Launch Compact Mass Spectrometry System for Rapid Narcotics Detection
July 8th 2025The U.S. Department of Energy’s Pacific Northwest National Laboratory’s (PNNL) VaporID, which is a newly developed portable air sampling system incorporating a miniaturized mass spectrometer (MS), can detect trace levels of fentanyl, methamphetamine, cocaine, and even explosives like TNT with great accuracy.
Radar and Soil Spectroscopy Boost Soil Carbon Predictions in Brazil’s Semi-Arid Regions
July 7th 2025A new study published in Geoderma demonstrates that combining soil spectroscopy with radar-derived vegetation indices and environmental data significantly improves the accuracy of soil organic carbon predictions in Brazil’s semi-arid regions.
Advancing Deep Soil Moisture Monitoring with AI-Powered Spectroscopy Drones
July 7th 2025A Virginia Tech study has combined drone-mounted NIR hyperspectral imaging (400 nm to 1100 nm) and AI to estimate soil moisture at root depths with remarkable accuracy, paving the way for smarter irrigation and resilient farming.