Abstract
The hydroxyl radical (OH) is a powerful oxidant in the troposphere controlling the atmospheric lifetimes of many short-lived climate forcers such as methane. In this study, the GFDL-AM4.1 model is used to investigate the meteorological impacts on OH and the methane budget and lifetime over 1980–2017. Driven by meteorological reanalyses from the National Centers for Environmental Prediction and the Modern-Era Retrospective analysis for Research and Applications (Version 2), our model gives an 11.2 Tg yr⁻¹ difference in the derived methane emissions and 0.24 years difference in the estimated methane lifetime. The results suggest that meteorology affects the mean OH concentrations but not OH trend, while the latter shows strong correlation with reactive nitrogen emissions. Despite the relatively small difference in global tropospheric OH (~2%) induced by different meteorological forcings, the difference is much larger regionally, leading to a difference of 8 Tg yr⁻¹ in tropical methane emissions.
Generated Summary
This research investigates the impact of meteorological forcing on hydroxyl radical (OH) concentrations and its implications for the methane budget and lifetime. The study utilizes the full-chemistry version of the Geophysical Fluid Dynamics Laboratory (GFDL) new-generation Atmospheric Model, version 4.1 (AM4.1), driven by meteorological reanalyses from the National Centers for Environmental Prediction (NCEP) and the Modern-Era Retrospective analysis for Research and Applications, Version 2 (MERRA-2). The research aims to understand how uncertainties in meteorology affect OH distribution, trends, and variability, and consequently, the methane budget and lifetime. The methodology involves comparing simulations nudged to different reanalysis products to investigate meteorology-driven uncertainties in OH. The study also examines the influence of uncertainties in OH induced by uncertainties in meteorology on the budget and lifetime of methane. The simulations were conducted from 1980-2017, and the findings provide insights into the sensitivity of methane emissions and lifetime to different meteorological forcings.
Key Findings & Statistics
- The hydroxyl radical (OH) is a powerful oxidant in the troposphere, and its global abundance and distribution are primarily controlled by emissions of short-lived climate forcers (SLCFs) and modulated by meteorology.
- The study finds an 11.2 Tg yr⁻¹ difference in derived methane emissions and a 0.24-year difference in the estimated methane lifetime due to different meteorological forcings.
- The model gives an 11.2 Tg yr⁻¹ difference in the derived methane emissions and 0.24 years difference in the estimated methane lifetime.
- The results suggest that meteorology affects the mean OH concentrations but not the OH trend, with the latter showing a strong correlation with reactive nitrogen emissions.
- Despite the relatively small difference in global tropospheric OH (~2%) induced by different meteorological forcings, the difference is much larger regionally, leading to a difference of 8 Tg yr⁻¹ in tropical methane emissions.
- The climatological mean of tropospheric mean fields is summarized in Table S3.
- Compared to NCEP, MERRA generally simulates slightly higher temperature in most regions, but lower temperature in the upper troposphere at high latitudes.
- Significant differences exist in tropospheric specific humidity (Q). For example, MERRA simulates much higher Q in the low latitudes of the Northern Hemisphere (by up to 20%), whereas lower Q in the low latitudes over Southern Hemisphere (by up to 10%).
- The differences in wind reanalysis between MERRA and NCEP induce different ocean evaporation and mixing, leading to different humidity and deep convection.
- Compared to AMIP, both NCEP and MERRA simulate higher ozone photolysis rate (J03) over tropics but lower ozone photolysis rate over middle and higher latitudes due to different clouds simulated in the model.
- MERRA simulates much higher OH concentrations (by 4%-8%) than NCEP especially over the tropics, where strong solar radiation and convection occur.
- The largest difference occurs over the tropics, with >10% difference over tropical Africa, Southeast Asia, tropical Pacific Ocean, tropical South America, and Southern Cone.
- A strong linear dependence of global mean tropospheric OH concentrations on Jo3QSN exists in all three simulations, with R² = 0.8.
- The methane emissions derived from concentration-driven simulations are based on mass balance, that is, the annual change of methane burden equals to the methane sources minus sinks. Since the differences in OH levels between MERRA and AMIP are small, the differences in derived methane emissions between MERRA and AMIP are also small, with an average of −1.8 Tg yr−1.
- However, much larger differences exist in OH levels between MERRA and NCEP, leading to an average of +11.2 ± 3.5 Tg yr−1 difference in derived methane emissions.
- Specifically, during the methane stabilization period (i.e., 1999–2006), the difference in derived methane emissions between MERRA and NCEP varies from 8.5 to 15.0 Tg yr−1, with an average of 11.9 Tg yr−1.
- In the emission-driven simulations, we produce similar simulated methane concentrations by constraining methane emissions with surface methane observations. Therefore, the emissions derived for the MERRAe simulation are higher to offset the impacts from the higher simulated OH concentrations with MERRA.
- The differences in OH concentrations between NCEPe and MERRAe range from 0.4% to 4.9%, with a global average of 2.2% and the tropical average of 2.4%.
- As a result, methane emissions derived for MERRAe are about 5%-54% higher than NCEPe over most of the tropical region, which is about 8 Tg yr¯¹ higher over 30°S-30°N, and 3 Tg yr¯¹ (<5%) higher mainly over 30°-90°N.
- With similar methane burdens but different OH concentrations in the NCEPe and MERRAe experiments, the tropospheric methane lifetimes range from 9.9 to 10.2 years with a difference of 0.24 years.
- The annual mean estimated emissions from emission-driven simulations are about 7 Tg yr¯¹ lower than those from concentration-driven simulations.
Other Important Findings
- Compared to NCEP, MERRA generally simulates slightly higher temperature in most regions.
- Compared to AMIP, both NCEP and MERRA simulate higher ozone photolysis rate (J03) over the tropics.
- MERRA simulates higher tropospheric OH concentrations than NCEP, especially over the tropics.
- The differences in surface and tropospheric OH are most significant in the tropics.
- All three simulations show an overall increasing trend in OH concentrations.
- The increases in tropical NOx emissions dominate the increasing trend in OH.
- The important role of LNOx emissions on OH and methane lifetime is not only demonstrated in GFDL-AM4 model but also in other models.
Limitations Noted in the Document
- The study uses a three-dimensional numerical model, and the results are based on model simulations, which have inherent uncertainties.
- The study acknowledges the potential for differences in modeled meteorology due to different reanalysis products, leading to differences in large-scale dynamical features.
- The study notes the challenges in building confidence in derived methane emissions from inverse modeling with prescribed OH fields due to large differences in OH levels over the tropics.
- The study focuses on the global and regional methane budget, and the results may not fully capture the complexities of local-scale processes.
- The study highlights the need for better constraints on NOx emissions, especially LNOx, to reduce model uncertainty in OH estimates.
Conclusion
The study’s findings highlight the critical role of hydroxyl radical (OH) in the troposphere and its impact on the methane budget and lifetime. The research demonstrates that meteorological forcing significantly affects OH concentrations, particularly in the tropics, leading to substantial variations in methane emissions. The differences in derived methane emissions between the two meteorological datasets were significant, emphasizing the sensitivity of the methane budget to meteorological uncertainties. The study’s results suggest that meteorology affects the mean OH concentrations but not the OH trend, while the latter shows strong correlation with reactive nitrogen emissions. The strong linear dependence of global mean tropospheric OH concentrations on the factors influencing OH concentrations indicates the importance of accurately representing these factors in models. The study underscores the need for better constraints on NOx emissions to reduce model uncertainty in OH estimates. The findings align with previous research indicating that the interannual variability in tropospheric OH is highly sensitive to NOx emissions. The study’s results also suggest that interannual variability of global OH is highly correlated with NOx emissions. The research also reveals that differences in OH concentrations can have significant impacts on estimated regional methane emissions. With such large differences in OH levels over the tropics, it is a great challenge to build confidence in derived methane emissions from inverse modeling with prescribed OH fields. The study’s outcomes contribute to a deeper understanding of the complex interplay between meteorology, OH, and the methane cycle, and they emphasize the need for more accurate representation of OH and its drivers in atmospheric models to better estimate the methane budget and its implications for climate change. The study underscores the critical role of OH in estimating the methane budget, no matter whether a top-down inversion or bottom-up approach is used, highlighting the need to better constrain OH abundances and variations.