Putian Zhou

Isoprene-derived oxygenated organic molecules (IP-OOM) drive new particle formation in the upper troposphere (Shen et al., 2024; Curtius et al., 2024), but the mechanisms that produce them, and the role of photochemistry in shaping their distribution, remain poorly constrained. These products are both hard to quantify, owing to isobaric and isotopic interferences that conventional time-of-flight mass spectrometers cannot resolve, and hard to interpret, as their formation pathways cannot be established from the measured product distribution alone. We address both with a combined measurement and modelling framework, which we demonstrate through chamber and flow reactor experiments. On the measurement side, we optimise a high-resolution Q Exactive Plus Orbitrap mass spectrometer, coupled to a bromide multi-scheme chemical ionisation inlet (the Br–-MION2-Orbitrap), by tuning three operating parameters that change individual-species signals by 10–90 %. Benchmarked side by side against a Br–-MION2-APi-TOF in chamber experiments, it agrees to within a factor of two for the oxygenated oxidation products while resolving isobaric interferences the time-of-flight instrument cannot. This comparison also shows that the Orbitrap’s apparent non-linearity at low concentrations, sometimes regarded as an intrinsic limitation, is rather an artefact of post-acquisition data processing: averaging that assigns zero intensity to non-detections biases the signal low near the detection limit. Disregarding the non-detections gives a conservative upper-bound detection limit of ∼1.3×105 cm−3, comparable to that of the APi-TOF. On the modelling side, we extend MARFORCE from simple inorganic chemistry to comprehensive organic mechanisms of hundreds of species coupled to transport, and add a continuous, spatially distributed OH source; the same model can also run as a well-mixed box model for chamber experiments. Applying the framework to the OH oxidation of 1,2–ISOPOOH in the flow reactor, we detect several abundant products absent from the current isoprene mechanism, and find that photolysis of the hydroperoxide-bearing products, a previously underappreciated process, shapes the product distribution. These experiments show how pairing molecular-level measurement with a model that resolves chemistry and transport provides a quantitative and transferable basis for studying the photochemical oxidation of isoprene and other volatile organic compounds relevant to upper-tropospheric aerosol formation.

A new open-source chemistry and aerosol box model BoxMART (Box Model for Atmospheric Research and Teaching) has been developed, aiming for easy and flexible configuration for various simulation cases, as well as suitable to teach and learn complex chemistry and aerosol processes. The model is written in Fortran and stored as a gitlab repository. The chemistry module is handled by KPP (Kinetic PreProcessor) which converts the chemistry scheme in an adapted format to Fortran code. The aerosol module currently includes coagulation, condensational and dissolutional growth, evaporation and aerosol thermodynamic processes. The MOSAIC (Model for Simulating Aerosol Interactions and Chemistry) model has been implemented to calculate the non-equilibrium dynamic gas-particle partitioning of non-volatile and semivolatile species, including H₂SO₄, HNO₃, HCl, NH₃, and organic compounds to multiple particle size bins. The current ion system includes H⁺, NH₄⁺, Na⁺, SO₄^2–, NO₃^–, Cl^– and H₂O, where pH value in each size bin is also calculated. The particle size distribution is discretized with a fully stationary method. In order to provide good examples for evaluating BoxMART, as well as future model development, a number of benchmark simulation cases from previous studies ranging from simple to complicated cases, have been revised in details and simulated to verify different processes.

Aerosol particles larger than roughly 50–100 nm in diameter are climatically important because they can act as cloud condensation nuclei (CCN), making their global number concentrations essential for understanding aerosol–cloud interactions. However, observationally constrained, long-term global datasets of particle number concentrations in this size range remain scarce. In this investigation, we present a global dataset of ground-level particle number concentrations for the period 2003–2024, produced by combining in situ observations with a machine-learning approach. The dataset includes two variables: the number concentrations for particles larger than 100 nm (N100) and larger than 50 nm (N50), provided at 0.75° × 0.75° spatial resolution and daily temporal resolution. To generate this dataset, we trained an eXtreme Gradient Boosting (XGB) model using measurements from 62 in situ stations as targets and reanalysis variables as predictors, enabling a data-driven representation of particle number concentrations at the global scale. We evaluated the dataset against independent observations from 12 additional stations. At 2/3 of these stations, the dataset shows good performance, capturing the median concentrations within a factor of 1.5 from the observations. Furthermore, we describe the main characteristics of the dataset in terms of global spatial patterns, temporal variability, and seasonal cycles, and demonstrate its ability to capture long-term trends in particle number concentrations, including both increasing and decreasing tendencies reported in the literature. This work provides the first observation-constrained, machine-learning-based global dataset of N50 and N100 at daily resolution over two decades, bridging the gap between sparse measurements and computationally expensive process-based models. The dataset, publicly available at https://doi.org/10.5281/zenodo.20202080, offers a valuable resource for evaluating model simulations, improving CCN-related parameterizations, and supporting weather and climate studies without the need for explicit knowledge of the aerosol particle microphysics.

A new open-source chemistry and aerosol box model BoxMART (Box Model for Atmospheric Research and Teaching) has been developed, aiming for easy and flexible configuration for various simulation cases, as well as suitable to teach and learn complex chemistry and aerosol processes. The model is written in Fortran and stored as a gitlab repository. The chemistry module is handled by KPP (Kinetic PreProcessor) which converts the chemistry scheme in an adapted format to Fortran code. The aerosol module currently includes coagulation, condensational and dissolutional growth, evaporation and aerosol thermodynamic processes. The MOSAIC (Model for Simulating Aerosol Interactions and Chemistry) model has been implemented to calculate the non-equilibrium dynamic gas-particle partitioning of non-volatile and semivolatile species, including H2SO4, HNO3, HCl, NH3, and organic compounds to multiple particle size bins. The current ion system includes H+, NH4+, Na+, SO42–, NO3–, Cl– and H2O, where pH value in each size bin is also calculated. The particle size distribution is discretized with a fully stationary method. In order to provide good examples for evaluating BoxMART, as well as future model development, a number of benchmark simulation cases from previous studies ranging from simple to complicated cases, have been revised in details and simulated to verify different processes.

Aerosol and cloud uncertainty dominates uncertainty in climate prediction (Masson-Delmotte 2021). Aerosol Cloud Interactions (ACIs) are mediated through cloud condensation nuclei (CCN), aerosol particles of large enough size and hygroscopicity to act as seeds upon which cloud droplet can form. Cloud droplet number concentration (CDNC) is determined by both the number of available CCN and the water vapour supersaturation these CCN experience. Variability and uncertainty in CCNC are primarily important in CCN-limited regimes where the CDNC is limited by the number of available CCN. These conditions prevail in much of the boundary layer and lower free troposphere (Rosenfeld et al. 2014). The number concentration of CCN is highly variable, both globally and locally (Schmale et al. 2018), depending on the abundance, size distribution and chemical composition of primary and secondary aerosols from both anthropogenic and biogenic sources. Therefore, CCN and related aerosol observations in different environments are needed to evaluate their representation in global models.Here we use long-term in-situ observations of CCN number concentrations, particle number size distributions, and particle chemical composition from ground-stations to evaluate CCN representation in the TM5 chemical transport model. We evaluate the default version of TM5 alongside representation of some of the sources of aerosol uncertainty in the model, represented by one-at-a-time sensitivity studies of uncertain aerosol input parameters including emissions of sea salt and biomass burning aerosol, secondary organic aerosol precursors and removal by wet and dry deposition. The model was run for 2017 and 2018, and so ground stations with relevant publicly available observations over at least 9 months of this period are chosen for evaluation including US Department of Energy (DOE) Atmospheric Radiation Measurement (ARM) site at Oklahoma, USA (Southern Great Plains) and ARM mobile facility deployment during the Layered Atlantic Smoke Interactions with Clouds (LASIC) campaign on Ascension Island (Andrews et al. 2025), SMEAR II station in Hyytiälä, Finland (Kulmala 2023) and Kennaook/Cape Grim Baseline Air Pollution Station in Tasmania (Keywood 2018).ReferencesAndrews, E., Zabala, I., Carrillo-Cardenas, G., et al. 'Harmonized aerosol size distribution, cloud condensation nuclei, chemistry and optical properties at 10 sites', Scientific Data, 12: 937.10.1038/s41597-025-04931-y 2025.Keywood, M., Ward, J., Derek, N., GAW-WDCA. 'Cloud_condensation_nuclei_number_concentration at Kennaook / Cape Grim Baseline Air Pollution Station, data hosted by EBAS at NILU'.10.48597/Z8RB-RDC9 2018.Kulmala, M., Petäjä, T. "Particle_number_concentration at Hyytiälä, data hosted by EBAS at NILU." In.https://doi.org/10.48597/RHAH-5H7M 2023.Masson-Delmotte, V., P. Zhai, A. Pirani, S.L. Connors, C. Péan, S. Berger, N. Caud, Y. Chen, L. Goldfarb, M.I. Gomis, M. Huang, K. Leitzell, E. Lonnoy, J.B.R. Matthews, T.K. Maycock, T. Waterfield, O. Yelekçi, R. Yu, and B. Zhou. "IPCC Climate Change 2021: The Physical Science Basis." In.: Cambridge University Press.10.1017/9781009157896 2021.Rosenfeld, D., Andreae, M.O., Asmi, A., et al. 'Global observations of aerosol-cloud-precipitation-climate interactions', Reviews of Geophysics, 52: 750-808.https://doi.org/10.1002/2013RG000441 2014.Schmale, J., Henning, S., Decesari, S., et al. 'Long-term cloud condensation nuclei number concentration, particle number size distribution and chemical composition measurements at regionally representative observatories', Atmos. Chem. Phys., 18: 2853-81.10.5194/acp-18-2853-2018 2018.

The radiative properties of clouds depend partially on the cloud droplet number concentration, which is determined by the concentration of cloud condensation nuclei (CCN) when the clouds are formed. In turn, CCN concentrations are determined by the atmospheric particle size distribution and their chemical composition. We use a novel Lagrangian modelling framework to examine the origins and history of gas and aerosol components observed at the boreal forest measurement site SMEAR II, and their potential to act as CCN. This framework combines: (a) global emission datasets, (b) backward trajectories from the FLEXible PARTicle dispersion model (FLEXPART) airmass dispersion model, (c) a detailed description of atmospheric chemistry and aerosol dynamics from the Model to Simulate the Concentration of Organic Vapours, Sulphuric Acid and Aerosol Particles (SOSAA). We apply this SOSAA-FP (FP standing for FLEXPART) framework to simulate a period from March to October 2018 with 1 h time resolution, focusing on the concentrations of CCN between 0.1 % and 1.2 % maximum supersaturation as calculated by the κ-Köhler theory (with respective dry particle diameter of activation ca. 175–35 nm). We find that the model PM₁ fraction of primary particles, sulfates and secondary organic aerosol correlate well with the observed organic aerosol and sulfate trends and explain most of the observed organic aerosol and sulfate PM₁ mass. Our results show that primary particle emissions play a considerable role in CCN concentrations even at a rural site such as SMEAR II. Changes in atmospheric cluster formation rates had a relatively weak impact on the CCN concentrations in the sensitivity runs. Enhanced cluster formation increased (decreased) the CCN concentrations for the highest (lowest) maximum supersaturation. Without any cluster formation our modelled median CCN concentrations changed by −48 % and +23 % for supersaturations of 1.2 % and 0.1 %, respectively, whereas omitting primary particle emissions had a decreasing effect in all calculated CCN supersaturation classes (−82 % and −33 % decrease in median CCN of 1.2 % and 0.1 % supersaturation, respectively). While the enhancing effect of cluster formation to high supersaturation (i.e., small diameter) CCN concentrations is unsurprising, the weak sensitivity to cluster formation rates and the decreasing effect to lowest supersaturation CCN was unexpected, as was the strong influence of anthropogenic primary emissions. The Lagrangian model framework showed its power, as it was possible to trace down the causes behind the unexpected outcomes by comparing how the particle population evolved along the trajectories in different sensitivity tests.

The radiative properties of clouds depend partially on the cloud droplet number concentration, which is determined by the concentration of cloud condensation nuclei (CCN) when the clouds are formed. In turn, CCN concentrations are determined by the atmospheric particle size distribution and their chemical composition. We use a novel Lagrangian modelling framework to examine the origins and history of gas and aerosol components observed at the boreal forest measurement site SMEAR II, and their potential to act as CCN. This framework combines: a) global emission datasets, b) backward trajectories from the FLEXible PARTicle dispersion model (FLEXPART) airmass dispersion model, c) a detailed description of atmospheric chemistry and aerosol dynamics from the Model to Simulate the Concentration of Organic Vapours, Sulphuric Acid and Aerosol Particles (SOSAA). We apply this SOSAA-FP (FP standing for FLEXPART) framework to simulate a period from March to October 2018 with one hour time resolution, focusing on the concentrations of CCN between 0.1–1.2 % maximum supersaturation as calculated by the κ-Köhler theory (with respective dry particle diameter of activation ca. 175–35 nm). We find that the model PM₁ fraction of primary particles, sulfates and secondary organic aerosol correlate well with the observed organic aerosol and sulfate trends and explain most of the observed organic aerosol and sulfate PM₁ mass. Our results show that primary particle emissions play a considerable role in CCN concentrations even at a rural site such as SMEAR II. Changes in atmospheric cluster formation rates had a relatively weak impact on the CCN concentrations in the sensitivity runs. Enhanced cluster formation increased (decreased) the CCN concentrations for the highest (lowest) maximum supersaturation. Without any cluster formation our modelled median CCN concentrations changed by –48 % and +23 % for supersaturations of 1.2 % and 0.1 %, respectively, whereas omitting primary particle emissions had a decreasing effect in all calculated CCN supersaturation classes (–82 % and –33 % decrease in median CCN of 1.2 % and 0.1 % supersaturation, respectively). While the enhancing effect of cluster formation to high supersaturation (i.e., small diameter) CCN concentrations is unsurprising, the weak sensitivity to cluster formation rates and the decreasing effect to lowest supersaturation CCN was unexpected, as was the strong influence of anthropogenic primary emissions. The Lagrangian model framework showed its power, as it was possible to trace down the causes behind the unexpected outcomes by comparing how the particle population evolved along the trajectories in different sensitivity tests.

The radiative properties of clouds depend partially on the cloud droplet number concentration, which is determined by the concentration of cloud condensation nuclei (CCN) when the clouds are formed. In turn, CCN concentrations are determined by the atmospheric particle size distribution and their chemical composition. We use a novel Lagrangian modelling framework to examine the origins and history of gas and aerosol components observed at the boreal forest measurement site SMEAR II, and their potential to act as CCN. This framework combines: a) global emission datasets, b) backward trajectories from the FLEXible PARTicle dispersion model (FLEXPART) airmass dispersion model, c) a detailed description of atmospheric chemistry and aerosol dynamics from the Model to Simulate the Concentration of Organic Vapours, Sulphuric Acid and Aerosol Particles (SOSAA). We apply this SOSAA-FP (FP standing for FLEXPART) framework to simulate a period from March to October 2018 with one hour time resolution, focusing on the concentrations of CCN between 0.1–1.2 % maximum supersaturation as calculated by the κ-Köhler theory (with respective dry particle diameter of activation ca. 175–35 nm). We find that the model PM₁ fraction of primary particles, sulfates and secondary organic aerosol correlate well with the observed organic aerosol and sulfate trends and explain most of the observed organic aerosol and sulfate PM₁ mass. Our results show that primary particle emissions play a considerable role in CCN concentrations even at a rural site such as SMEAR II. Changes in atmospheric cluster formation rates had a relatively weak impact on the CCN concentrations in the sensitivity runs. Enhanced cluster formation increased (decreased) the CCN concentrations for the highest (lowest) maximum supersaturation. Without any cluster formation our modelled median CCN concentrations changed by –48 % and +23 % for supersaturations of 1.2 % and 0.1 %, respectively, whereas omitting primary particle emissions had a decreasing effect in all calculated CCN supersaturation classes (–82 % and –33 % decrease in median CCN of 1.2 % and 0.1 % supersaturation, respectively). While the enhancing effect of cluster formation to high supersaturation (i.e., small diameter) CCN concentrations is unsurprising, the weak sensitivity to cluster formation rates and the decreasing effect to lowest supersaturation CCN was unexpected, as was the strong influence of anthropogenic primary emissions. The Lagrangian model framework showed its power, as it was possible to trace down the causes behind the unexpected outcomes by comparing how the particle population evolved along the trajectories in different sensitivity tests.