Accuracy and Sensitivity Analysis of Open-Path FTIR Quantification of N2O and CO2 Emitted from Agricultural Fields.

Open-path Fourier transform infrared spectroscopy (OP-FTIR) combined with the inverse dispersion techniques (e.g., backward Lagrangian stochastic model, bLS) has been increasingly applied to measure trace gas emissions from agricultural fields (e.g., greenhouse gas emissions) because this technique possesses better spatial and temporal representatives than the chamber methods. The gas fluxes calculated by the bLS model rely on the discrepancy between upwind and downwind concentrations determined by OP-FTIR. OP-FTIR, however, is subject to the potential challenges (i.e. the interference of water vapor, the diurnal variations in air temperature, and the changing path length) that substantially influence the accuracy and precision of the FTIR-determined gas concentrations, and only a few studies showed the credible method to quantify gas concentrations using OP-FTIR. In this study, we developed the methodology that is capable of quantifying CO₂ and N₂O concentrations in the atmosphere accurately and also evaluated the sources of errors that influence the gas quantification using OP-FTIR. Our results showed that we can mitigate the interference of water vapor in gas quantifications by restricting the spectral window that is used to calculated gas concentrations. The increasing air temperature reduced the accuracy of FTIR-determined gas concentrations because the infrared characteristics of gases are sensitive to the changing environmental temperature (i.e. temperature broadening effect). The bias of N_­2O concentration increased to 10% when air temperature increased from 15°C to 30°C. The adjustable path length is one of the advantages of OP-FTIR that allows measuring gas emissions from the various sizes of the hot spot (e.g., physical path length = 50-200 m). However, the bias increased with the decreasing path length, and the shortest path (50 m) led to 7% and 29% biases in CO₂ and N₂O concentrations, respectively.