Determining the Source of Water from the Combustion of Dangerous Fossil Fuels by Oxygen Isotopes.

[NNadir]

The paper I'll briefly discuss in this post is this one: Measurements of Combustion-Derived Water Vapor Isotopic Composition from Different Fossil Fuels Meng Xing, Junji Cao, Zhoufeng Wang, Qiyuan Wang, Wenwu Cai, Jie Tian, Jianjun Li, and Weiguo Liu, Environmental Science & Technology 2026 60 (3), 2615-2626

Oxygen has three stable isotopes, overwhelmingly dominated by the doubly magic isotope ¹⁶O. (Eight is a "magic number" in nuclear physics, isotopes have a magic number of protons or neutrons or both are unusually stable. sup]16O has 8 neutrons and 8 protons. ) The other two stable isotopes are sup]17O and sup]18O, which respectively account for roughly 0.04% and 0.2%, depending on source, as noted in the paper.

I have always wondered how much of the water in the atmosphere comes from the combustion of dangerous fossil fuels, and the paper suggests an answer.

From the paper's introduction:

Water vapor (H₂Ov) is an essential component of the Earth’s atmosphere, playing critical roles in climate regulation, weather patterns, and the water cycle. Its sources primarily come from natural processes such as ocean evaporation and terrestrial evapotranspiration (natural water vapor sources: NWV). However, during the fossil fuels (e.g., coal, petroleum, natural gas) combustion process, in addition to emitting substantial amounts of CO₂, they also generate significant amounts of water vapor as a byproduct (combustion-derived water vapor sources: CDWV). (1,2) Additionally, the combustion of plants or organic matter during natural events such as wildfires and volcanic eruptions can also generate water vapor. As reported by the Sixth Assessment Report of the Intergovernmental Panel on Climate Change (IPCC AR6), globally speaking, statistical data indicate that during the 2010–2019 period, fossil fuel combustion emitted an average of 9.6 PgC yr^-1 of CO₂. (3) Based on Gorski et al.’s (6) assumption and our results (see Sections 4.2 and 4.3), the average molar ratio of H₂O to CO₂ emission is 1.5; thus, the total of combustion water vapor emissions is ∼21.6 Pg yr^-1. At the global scale, the long-term average water vapor content in the global atmosphere is estimated at 12,700–13,000 Pg. (4,5) Thus, the annually generated CDWV (21.6 Pg yr^-1) from human activities accounts for only 0.17% of the global atmospheric reservoir. However, in densely populated areas, the portion of CDWV can exceed 10% in the planetary boundary layer. (6,7) Enhanced atmospheric water vapor content in megacities can trigger a cascade of adverse effects, including exacerbated air pollution, weakened solar radiation, increased frequency of extreme weather events, and enhanced greenhouse effects. (7−10) Consequently, accurately quantifying the contribution of CDWV to total atmospheric moisture is critical for assessing anthropogenic impacts on the hydrological cycle. Current observational methods, such as monitoring absolute humidity, relative humidity, vapor pressure, or dew point temperature, cannot reliably distinguish the atmospheric water vapor contributions from NWV and anthropogenic activities (primarily CDWV). (11) However, the distinct isotopic signatures between NWV and CDWV offer a promising solution. (6,12)

Generally, the isotopic composition of NWV undergoes fractionation through hydrological and biogeochemical processes, including evaporation, transpiration, atmospheric transport, mixing, biological utilization, and atmospheric chemical reactions. (13) In contrast, CDWV originates from the oxidation of H atoms in fossil fuels with atmospheric molecular oxygen (O₂). Consequently, the δ²Hv of the resultant water vapor molecules preserves the original isotopic imprint of fossil fuel-bound hydrogen, while their δ¹⁸Ov is jointly governed by the isotopic characteristics of O atoms in both the fuel matrix and atmospheric O₂. During biological synthesis, organisms exhibit preferential utilization of lighter hydrogen isotopes (¹H over ²H), a process termed biosynthetic discrimination. This metabolic selectivity results in deuterium atom depletion (δ²H values more negative) in fossil fuels relative to the Vienna Standard Mean Ocean Water (VSMOW) reference. (14,15) Consequently, CDWV inherits this ²H-depleted signature. In contrast, the δ¹⁸Ov of CDWV predominantly reflects the isotopic inheritance from atmospheric O₂, which exhibits a characteristic δ¹⁸O value of +23.9‰, (16) and results in its signature being markedly enriched compared to NWV...

The oxygen in the combustion gases partially result from atmospheric oxygen, and the authors discuss this issue extensively.

Several figures in the text show this point, mixing ratios, but I will not have the time to elaborate on this point from the paper.

Here, however is one figure from the text:



The caption:

Figure 1. Variation ranges of water vapor δ¹⁸O and δ²H in the troposphere (data from Wei et al. (18)), upper troposphere/lower stratosphere (data from Galewsky and Samuels-Crow; (19) Sayres et al.; (20) Yang and Yoshimura (11)), theoretical CDWV (data from Gorski et al.; (6) Fiorella et al. (1)), and reported CDWV in references (data from Gorski et al.; (6) Xing et al. (7)) (a); the same as (a) but for the variation ranges of the d-excessv values (b). Error bars represent one standard deviation of the mean, with the sample sizes (n) provided in parentheses. Theoretical CDWV composition and upper troposphere/lower stratosphere data merely indicate the ranges of variation. In panel (b), blue shading indicates combustion-derived water vapor sources, and red shading indicates natural water vapor sources.


Tables from the text:







An additional figure from the text:



The caption:

Figure 5. Distribution ranges of δ¹⁸Ov, δ²Hv, and d-excess_v values from CDWV of different fossil fuels and vehicle operation conditions. The dashed line represents the reference δ¹⁸O value of +23.9‰ for atmospheric O2. (24) The dash-dotted line represents the reference δ²H value for the fuels, ranging from −280‰ to −85‰. (6,22,23,59−61) The background water vapor values of δ¹⁸Ov, δ²Hv, and d-excess_v are statistically derived from the data in this study. Error bars represent one standard deviation of the mean. Notably, the d-excess_v values are positive for both background water vapor and idling gasoline vehicle (2.7‰ and 2.6‰, respectively), whereas they are negative in all other cases.

The technique used for measurements is one with which I have limited familiarity although I know that it is used at the Mauna Loa Carbon Dioxide Observatory, Cavity Ring Down Spectroscopy, which is based on the decay of emissions generated by exposure of a gas to laser light.

The authors rightly discuss the limitations of the work and avenues of work building upon it.

Have a pleasant Sunday, even if you are, as I am, in the extreme cold regions of the country.