Quantifying fugitive gas emissions from an oil sands tailings pond with open-path Fourier transform infrared measurements

Fugitive emissions from tailings ponds contribute significantly to facility emissions in the Alberta oil sands, but details on chemical emission profiles and the temporal and spatial variability of emissions to the atmosphere are sparse, since flux measurement techniques applied for compliance monitoring have their limitations. In this study, openpath Fourier transform infrared spectroscopy was evaluated as a potential alternative method for quantifying spatially representative fluxes for various pollutants (methane, ammonia, and alkanes) from a particular pond, using vertical-fluxgradient and inverse-dispersion methods. Gradient fluxes of methane averaged 4.3 g m−2 d−1 but were 44 % lower than nearby eddy covariance measurements, while inversedispersion fluxes agreed to within 30 %. With the gradient fluxes method, significant NH3 emission fluxes were observed (0.05 g m−2 d−1, 42 t yr−1), and total alkane fluxes were estimated to be 1.05 g m−2 d−1 (881 t yr−1), representing 9.6 % of the facility emissions. Copyright statement. The works published in this journal are distributed under the Creative Commons Attribution 4.0 License. This license does not affect the Crown copyright work, which is re-usable under the Open Government Licence (OGL). The Creative Commons Attribution 4.0 License and the OGL are interoperable and do not conflict with, reduce or limit each other. © Crown copyright 2021

S1. Methane mole fractions, vertical profiles, and gradient fluxes S1.1 Calibration of retrieved CH4 mole fraction from OP-FTIR The amplitude of spectra for all the three paths varied substantially over the study period, especially for the top path.
As a proxy for the spectral amplitude, the signal-to-noise ratio (SNR) of the CH4 fitting was used. The CH4, NH3, CH3OH and HCHO mole fraction for all three paths when this SNR dropped fast, or stayed below 10 were flagged.
3% and 13% of the measurements from bottom and top path were flagged and invalidated from further mole fraction gradient and flux calculations.
Since CH4 mole fraction was also continuously measured by cavity ring-down spectroscopy (CRDS) at four heights during the study, the measurements at 4m were compared to CH4 mole fraction retrieved from the FTIR bottom path to calibrate the retrieved CH4 mole fraction from three paths of this OP-FTIR system.
Each CRDS in this study was calibrated before and after the campaign, and CH4 mole fraction from three CRDS at the same height was well compared (r 2 > 0.96, slope=0.98 -1.01, intercept = 0.01 -0.02 ppm). Therefore, CH4 mole fraction retrieved from FTIR all three paths were calibrated by the linear relationship in Fig  In the analysis of methane vertical profile below, all the mole fractions measurements (half-hour averages) were taken from the Picarro G2204 at 4, 8, 18, and 32m. There are 271 half-hours in total when the wind was from the pond. About 83% of the half-hour periods when the wind was from the pond direction, the CH4 vertical profiles are similar to Fig. S4. Within this 83% of periods, some profiles are close to linear, and others are not strict decreasing trend with height. For the rest of 17% of half-hour periods, the CH4 vertical profiles are closer to logarithmic (Fig.   S5). Therefore, CH4 vertical profiles are considered linear over the entire period for calculating gradient flux with OP-FTIR measurement.
In addition, those half-hour periods when logarithmic relationship is better than linear to describe the vertical profile are mainly (65%) associated with wind speed greater than 6 m s -1 (Fig. S6). For the majority of the time (85%) when the wind was from the pond, wind speed was less than 6 m s -1 (Fig. S6).   where z is the height for which flux is calculated ( Thompson and Pinker, 1981).
_ , is a function of stability (z/L) and is calculated with eq. (5) and (6) in the main text. The gradient flux of CH4 with logarithmic vertical profile is calculated with eq. (S2) and the area-weighted average flux from the pond sectors is 4.1 gm -2 d -1 ,which is 19% greater than the gradient flux calculated with linear vertical profile.
Beside top-bottom paths of CH4 mole fractions gradient, middle-bottom paths of gradient can also be used to calculate CH4 gradient fluxes. The results are summarised in the first row of Table S1 to compare to gradient fluxes with top-bottom paths CH4 gradients. The area-weighted averaged fluxes with middle-bottom paths is 29% lower than the area-weighted averaged fluxes with top-bottom paths (Table S1).   The half-hour IDM fluxes with these two approaches agree well (slope = 0.9, r 2 = 0.92). The sector-area-weightaveraged IDM fluxes with two approaches are also within 20% difference. The interquartile ranges overlap (Table   S1).     Table S1. Compared to gradient flux results with our approach modified Bowen ratio, CH4, NH3 and total alkane fluxes with the slant path flux-gradient method are 24%, 25%, and 30% smaller. Tables  Table S1 Summary of CH4 IDM fluxes with two Meteorol., 20, 250-254, 1981.