Ammonia (NH₃), the most prevalent alkaline gas in the atmosphere, plays a significant role in PM_2.5 formation, atmospheric chemistry, and new particle formation. This paper reviews quantification of [NH₃] through measurements, satellite-remote-sensing, and modeling reported in over 500 publications towards synthesizing the current knowledge of [NH₃], focusing on spatiotemporal variations, controlling processes, and quantification issues. Most measurements are through regional passive sampler networks. [NH₃] hotspots are typically over agricultural regions, such as the Midwest US and the North China Plain, with elevated concentrations reaching monthly averages of 20 and 74 ppbv, respectively. Topographical effects dramatically increase [NH₃] over the Indo-Gangetic Plains, North India and San Joaquin Valley, US. Measurements are sparse over oceans, where [NH₃] ≈ a few tens of pptv, variations of which can affect aerosol formation. Satellite remote-sensing (AIRS, CrIS, IASI, TANSO-FTS, TES) provides global [NH₃] quantification in the column and at the surface since 2002. Modeling is crucial for improving understanding of NH₃ chemistry and transport, its spatiotemporal variations, source apportionment, exploring physicochemical mechanisms, and predicting future scenarios. GEOS-Chem (global) and FRAME (UK) models are commonly applied for this. A synergistic approach of measurements↔satellite-inference↔modeling is needed towards improved understanding of atmospheric ammonia, which is of concern from the standpoint of human health and the ecosystem.

The concentrations of atmospheric ammonia ([NH₃]) have been observed to be increasing over the United States in the last decade, especially in Eastern United States. It is important to understand this temporal trend and variation due to the role of NH₃ in particle formation and its ecological effects. Here the long‐term trend of [NH₃] over the United States is investigated using GEOS‐Chem, a global 3‐D tropospheric chemistry model, and is corroborated with empirical evidence from the Ammonia Monitoring Network. Model simulations, consistent with observations, show increase in [NH₃] over the United States from 2001 to 2016, with magnitude largest in the East (~5% to 12%/year) and smallest in the West (~0% to 5%/year). Reasons for this are examined, and evidence for the role of decreasing SO₂ and NO_x emissions in increasing [NH₃] is provided. The contributions of meteorology and NH₃ emission changes to the [NH₃] increase appear to be small during the period. Our sensitivity study suggests that decreasing SO₂ and NO_x emissions over the United States owing to stringent regulations explain about 2/3 and 1/3 of the increase in [NH₃], respectively. This effect is different for various NH₃ and SO₂ and NO_x regimes. Given the continued reduction of SO₂ and NO_x emissions due to U.S. regulations mainly aimed at PM_2.5 reduction, the present results are important towards better assessing the environmental impact of emission controlling policies.

Atmospheric particles can act as cloud condensation nuclei (CCN) and modify cloud properties and precipitation and thus indirectly impact the hydrological cycle and climate. New particle formation (NPF or nucleation), frequently observed at locations around the globe, is an important source of ultrafine particles and CCN in the atmosphere. In this study, wintertime NPF over the Northeastern United States (NEUS) is simulated with WRF-Chem coupled with a size-resolved (sectional) advanced particle microphysics (APM) model. Model-simulated variations in particle number concentrations during a 2-month period (November–December 2013) are in agreement with corresponding measurements taken at Pinnacle State Park (PSP), New York, and Appalachian State University (APP), North Carolina. We show that, even during wintertime, regional nucleation occurs and contributes significantly to ultrafine-particle and CCN number concentrations over the NEUS. The model shows that, due to low biogenic emissions during this period, wintertime regional nucleation is solely controlled by inorganic species and the newly developed ternary ion-mediated nucleation scheme is able to capture the variations in observed particle number concentrations (ranging from ∼200 to 20 000 cm⁻³) at both PSP and APP. Total particle and CCN number concentrations dramatically increase following NPF events and have the highest values over the Ohio Valley region, where elevated [SO₂] is sustained by power plants. Secondary particles dominate particle number abundance over the NEUS, and their fraction increases with altitude from ≳85 % near the surface to ≳95 % in the upper troposphere. The secondary fraction of CCN also increases with altitude, from 20 %–50 % in the lower boundary layer to 50 %–60 % in the middle troposphere to 70 %–85 % in the upper troposphere.

Cloud condensation nuclei (CCN) number concentrations are an important aspect of aerosol–cloud interactions and the subsequent climate effects; however, their measurements are very limited. We use a machine learning tool, random decision forests, to develop a random forest regression model (RFRM) to derive CCN at 0.4 % supersaturation ([CCN0.4]) from commonly available measurements. The RFRM is trained on the long-term simulations in a global size-resolved particle microphysics model. Using atmospheric state and composition variables as predictors, through associations of their variabilities, the RFRM is able to learn the underlying dependence of [CCN0.4] on these predictors, which are as follows: eight fractions of PM_2.5 (NH₄, SO₄, NO₃, secondary organic aerosol (SOA), black carbon (BC), primary organic carbon (POC), dust, and salt), seven gaseous species (NO_x, NH₃, O₃, SO₂, OH, isoprene, and monoterpene), and four meteorological variables (temperature (T), relative humidity (RH), precipitation, and solar radiation). The RFRM is highly robust: it has a median mean fractional bias (MFB) of 4.4 % with ≈96.33 % of the derived [CCN0.4] within a good agreement range of $-\mathrm{60}\mathit{\%}<\text{MFB}<+\mathrm{60}\mathit{\%}$ and strong correlation of Kendall's τ coefficient ≈0.88. The RFRM demonstrates its robustness over 4 orders of magnitude of [CCN0.4] over varying spatial (such as continental to oceanic, clean to polluted, and near-surface to upper troposphere) and temporal (from the hourly to the decadal) scales. At the Atmospheric Radiation Measurement Southern Great Plains observatory (ARM SGP) in Lamont, Oklahoma, United States, long-term measurements for PM_2.5 speciation (NH₄, SO₄, NO₃, and organic carbon (OC)), NO_x, O₃, SO₂, T, and RH, as well as [CCN0.4] are available. We modify, optimize, and retrain the developed RFRM to make predictions from 19 to 9 of these available predictors. This retrained RFRM (RFRM-ShortVars) shows a reduction in performance due to the unavailability and sparsity of measurements (predictors); it captures the [CCN0.4] variability and magnitude at SGP with ≈67.02 % of the derived values in the good agreement range. This work shows the potential of using the more commonly available measurements of PM_2.5 speciation to alleviate the sparsity of CCN number concentrations' measurements.