Subgrid processes in global climate models are represented by parameterizations which are a major source of uncertainties in simulations of climate. In recent years, it has been suggested that machine‐learning (ML) parameterizations based on high‐resolution model output data could be superior to traditional parameterizations. Currently, both traditional and ML parameterizations of subgrid processes in the atmosphere are based on a single‐column approach, which only use information from single atmospheric columns. However, single‐column parameterizations might not be ideal since certain atmospheric phenomena, such as organized convective systems, can cross multiple grid boxes and involve slantwise circulations that are not purely vertical. Here we train neural networks (NNs) using non‐local inputs spanning over 3 × 3 columns of inputs. We find that including the non‐local inputs improves the offline prediction of a range of subgrid processes. The improvement is especially notable for subgrid momentum transport and for atmospheric conditions associated with mid‐latitude fronts and convective instability. Using an interpretability method, we find that the NN improvements partly rely on using the horizontal wind divergence, and we further show that including the divergence or vertical velocity as a separate input substantially improves offline performance. However, non‐local winds continue to be useful inputs for parameterizating subgrid momentum transport even when the vertical velocity is included as an input. Overall, our results imply that the use of non‐local variables and the vertical velocity as inputs could improve the performance of ML parameterizations, and the use of these inputs should be tested in online simulations in future work.

This study evaluates formaldehyde (HCHO) over the U.S. from 2006 to 2015 by comparing ground monitor data from the Air Quality System (AQS) and a satellite retrieval from the Ozone Monitoring Instrument (OMI). Our comparison focuses on the utility of satellite data to inform patterns, trends, and processes of ground-based HCHO across the U.S. We find that cities with higher levels of biogenic volatile organic compound (BVOC) emissions, including primary HCHO, exhibit larger HCHO diurnal amplitudes in surface observations. These differences in hour-to-hour variability in surface HCHO suggests that satellite agreement with ground-based data may depend on the distribution of emission sources. On a seasonal basis, OMI exhibits the highest correlation with AQS in summer and the lowest correlation in winter. The ratios of HCHO in summer versus other seasons show pronounced seasonal variability in OMI, likely due to seasonal changes in the vertical HCHO distribution. The seasonal variability in HCHO from satellite is more pronounced than at the surface, with seasonal variability 20–100% larger in satellite than surface observations. The seasonal variability also has a latitude dependency, with more variability in higher latitude regions. OMI agrees with AQS on the interannual variability in certain periods, whereas AQS and OMI do not show a consistent decadal trend. This is possibly due to a rather large interannual variability in HCHO, which makes the small decadal drift less significant. Temperature also explains part of the interannual variabilities. Small temperature variations in the western U.S. are reflected with more quiescent HCHO interannual variability in that region. The decrease in summertime HCHO in the southeast U.S. could also be partially explained by a small and negative trend in local temperatures.

Following the Hunga Tonga–Hunga Ha'apai (HTHH) eruption in January 2022, significant reductions in stratospheric hydrochloric acid (HCl) were observed in the Southern Hemisphere mid‐latitudes during the latter half of 2022, suggesting potential chlorine activation. The objective of this study is to comprehensively understand the loss of HCl in the aftermath of HTHH. Satellite measurements and a global chemistry‐climate model are employed for the analysis. We find strong agreement of 2022 anomalies between the modeled and the measured data. The observed tracer‐tracer relations between nitrous oxide (N₂O) and HCl indicate a significant role of chemical processing in the observed HCl reduction, especially during the austral winter of 2022. Further examining the roles of chlorine gas‐phase and heterogeneous chemistry, we find that heterogeneous chemistry emerges as the primary driver for the chemical loss of HCl, and the reaction between hypobromous acid (HOBr) and HCl on sulfate aerosols is the dominant loss process.

The 2019 to 2020 Australian summer wildfires injected an amount of organic gases and particles into the stratosphere unprecedented in the satellite record since 2002, causing large unexpected changes in HCl and ClONO ₂ . These fires provided a novel opportunity to evaluate heterogeneous reactions on organic aerosols in the context of stratospheric chlorine and ozone depletion chemistry. It has long been known that heterogeneous chlorine (Cl) activation occurs on the polar stratospheric clouds (PSCs; liquid and solid particles containing water, sulfuric acid, and in some cases nitric acid) that are found in the stratosphere, but these are only effective for ozone depletion chemistry at temperatures below about 195 K (i.e., largely in the polar regions during winter). Here, we develop an approach to quantitatively assess atmospheric evidence for these reactions using satellite data for both the polar (65 to 90°S) and the midlatitude (40 to 55°S) regions. We show that heterogeneous reactions apparently even happened at temperatures at 220 K during austral autumn on the organic aerosols present in 2020 in both regions, in contrast to earlier years. Further, increased variability in HCl was also found after the wildfires, suggesting diverse chemical properties among the 2020 aerosols. We also confirm the expectation based upon laboratory studies that heterogeneous Cl activation has a strong dependence upon water vapor partial pressure and hence atmospheric altitude, becoming much faster close to the tropopause. Our analysis improves the understanding of heterogeneous reactions that are important for stratospheric ozone chemistry under both background and wildfire conditions.

The ocean is a reservoir for CFC-11, a major ozone-depleting chemical. Anthropogenic production of CFC-11 dramatically decreased in the 1990s under the Montreal Protocol, which stipulated a global phase out of production by 2010. However, studies raise questions about current overall emission levels and indicate unexpected increases of CFC-11 emissions of about 10 Gg ⋅ yr⁻¹ after 2013 (based upon measured atmospheric concentrations and an assumed atmospheric lifetime). These findings heighten the need to understand processes that could affect the CFC-11 lifetime, including ocean fluxes. We evaluate how ocean uptake and release through 2300 affects CFC-11 lifetimes, emission estimates, and the long-term return of CFC-11 from the ocean reservoir. We show that ocean uptake yields a shorter total lifetime and larger inferred emission of atmospheric CFC-11 from 1930 to 2075 compared to estimates using only atmospheric processes. Ocean flux changes over time result in small but not completely negligible effects on the calculated unexpected emissions change (decreasing it by 0.4 ± 0.3 Gg ⋅ yr⁻¹). Moreover, it is expected that the ocean will eventually become a source of CFC-11, increasing its total lifetime thereafter. Ocean outgassing should produce detectable increases in global atmospheric CFC-11 abundances by the mid-2100s, with emission of around 0.5 Gg ⋅ yr⁻¹; this should not be confused with illicit production at that time. An illustrative model projection suggests that climate change is expected to make the ocean a weaker reservoir for CFC-11, advancing the detectable change in the global atmospheric mixing ratio by about 5 yr. Significance Manufactured CFC-11 is depleting the Antarctic ozone layer. CFC production has been strictly controlled by the Montreal Protocol, but emission estimates are very sensitive to choices of lifetimes, which are often assumed as constant over time. We employ a hierarchy of models to study the effect of the ocean on the time-dependent uptake and release of atmospheric CFC-11. The ocean is a sink for CFC-11 and significantly affects its total lifetime and hence the emission inferred from concentration data of past decades. This has not been explicitly included in international ozone assessments. We show that, as anthropogenic production ceases, ocean fluxes become more important, suggesting a need for further studies with high-resolution global models linking atmospheric chemistry and ocean processes.