Search

Bromine monoxide (BrO) is relevant to atmospheric oxidative capacity, affecting the lifetime of greenhouse gases (i.e., methane, dimethylsulfide) and mercury oxidation. However, measurements of BrO radical vertical profiles are rare, and BrO is highly variable. As a result, the few available aircraft observations in different regions of the atmosphere are not easily reconciled. Autonomous multi-axis differential optical absorption spectroscopy (MAX-DOAS) instruments placed at remote mountaintop observatories (MT-DOAS) present a cost-effective alternative to aircraft, with the potential to probe the climate-relevant yet understudied free troposphere more routinely. Here, we describe an innovative full-atmosphere BrO and formaldehyde (HCHO) profile retrieval algorithm using MT-DOAS measurements at Mauna Loa Observatory (MLO – 19.536° N, 155.577° W; 3401 m a.s.l.). The retrieval is based on time-dependent optimal estimation and simultaneously inverts 190+ individual BrO (and formaldehyde, HCHO) SCDs (slant column densities; SCD = dSCD + SCDRef) from solar stray light spectra measured in the zenith and off-axis geometries at high and low solar zenith angles (92° > SZA > 30°) to derive BrO concentration profiles from 1.9 to 35 km with 7.5 degrees of freedom (DoFs). Two case study days are characterized by the absence (26 April 2017, base case) and presence of a Rossby-wave-breaking double tropopause (29 April 2017, RW-DT case). Stratospheric-BrO vertical columns are nearly identical on both days (VCD = (1.5 ± 0.2) × 1013 molec. cm−2), and the stratospheric-BrO profile peaks at a lower altitude during the RW-DT (1.6–2.0 DoFs). Tropospheric-BrO VCDs increase from (0.70 ± 0.14) × 1013 molec. cm−2 (base case) to (1.00 ± 0.14) × 1013 molec. cm−2 (RW-DT) owing to a 3-fold increase in BrO in the upper troposphere (1.7–1.9 DoFs). BrO at MLO increases from (0.23 ± 0.03) pptv (base case) to (0.46 ± 0.03) pptv (RW-DT) and is characterized by an added time resolution (∼ 3.8 DoFs). Up to (0.9 ± 0.1) pptv BrO is observed above MLO in the lower free troposphere in the absence of the double tropopause. We validate the retrieval using aircraft BrO profiles and in situ HCHO measurements aboard the NSF/NCAR GV aircraft above MLO (11 January 2014) that establish BrO peaks around 2.4 pptv above 13 km in the upper troposphere–lower stratosphere (UTLS) during a similar RW-DT event (0.83 × 1013 molec. cm2 tropospheric-BrO VCD above 2 km). The tropospheric-BrO profile measured using MT-DOAS (RW-DT case) and using the aircraft agree well (after averaging-kernel smoothing). Furthermore, these tropospheric-BrO profiles over the central Pacific Ocean are found to closely resemble those over the eastern Pacific Ocean (2–14 km) and are in contrast to those over the western Pacific Ocean, where a C-shaped tropospheric-BrO profile shape has been observed.

The air–sea exchange of ozone (O3) is controlled by chemistry involving halogens, dissolved organic carbon, and sulfur in the sea surface microlayer. Calculations also indicate faster ozone photolysis at aqueous surfaces, but the role of clouds as an ozone sink is currently not well established. Fast-response ozone sensors offer opportunities to measure eddy covariance (EC) ozone fluxes in the marine boundary layer. However, intercomparisons of fast airborne O3 sensors and EC O3 fluxes measured on aircraft have not been conducted before. In April 2022, the Technological Innovation Into Iodine and GV Environmental Research (TI3GER) field campaign deployed three fast ozone sensors (gas chemiluminescence and a combination of UV absorption with coumarin chemiluminescence detection, CID) together with a fast water vapor sensor and anemometer to study iodine chemistry in the troposphere and stratosphere over Colorado and over the Pacific Ocean near Hawaii and Alaska. Here, we present an instrument comparison between the NCAR Fast O3 instrument (FO3, gas-phase CID) and two KIT Fast AIRborne Ozone instruments (FAIRO, UV absorption and coumarin CID). The sensors have comparable precision < 0.4 % Hz−0.5 (0.15 ppbv Hz−0.5), and ozone volume mixing ratios (VMRs) generally agreed within 2 % over a wide range of environmental conditions: 10 < O3 < 1000 ppbv, below detection < NOx < 7 ppbv, and 2 ppmv < H2O < 4 % VMR. Both instrument designs are demonstrated to be suitable for EC flux measurements and were able to detect O3 fluxes with exchange velocities (defined as positive for upward) as slow as −0.010 ± 0.004 cm s−1, which is in the lower range of previously reported measurements. Additionally, we present two case studies. In one, the direction of ozone and water vapor fluxes was reversed (vO3 = +0.134 ± 0.005 cm s−1), suggesting that overhead evaporating clouds could be a strong ozone sink. Further work is needed to better understand the role of clouds as a possibly widespread sink of ozone in the remote marine boundary layer. In the second case study, vO3 values are negative (varying by a factor of 6–10 from −0.036 ± 0.006 to −0.003 ± 0.004 cm s−1), while the water vapor fluxes are consistently positive due to evaporation from the ocean surface and spatially homogeneous. This case study demonstrates that the processes governing ozone and water vapor fluxes can become decoupled and illustrates the need to elucidate possible drivers (physical, chemical, or biological) of the variability in ozone exchange velocities on fine spatial scales (∼ 20 km) over remote oceans.

Abstract During summer, ammonia emissions in Southeast Asia influence air pollution and cloud formation. Convective transport by the South Asian monsoon carries these pollutant air masses into the upper troposphere and lower stratosphere (UTLS), where they accumulate under anticyclonic flow conditions. This air mass accumulation is thought to contribute to particle formation and the development of the Asian Tropopause Aerosol Layer (ATAL). Despite the known influence of ammonia and particulate ammonium on air pollution, a comprehensive understanding of the ATAL is lacking. In this modelling study, the influence of ammonia on particle formation is assessed with emphasis on the ATAL. We use the EMAC chemistry-climate model, incorporating new particle formation parameterisations derived from experiments at the CERN CLOUD chamber. Our diurnal cycle analysis confirms that new particle formation mainly occurs during daylight, with a 10-fold enhancement in rate. This increase is prominent in the South Asian monsoon UTLS, where deep convection introduces high ammonia levels from the boundary layer, compared to a baseline scenario without ammonia. Our model simulations reveal that this ammonia-driven particle formation and growth contributes to an increase of up to 80% in cloud condensation nuclei (CCN) concentrations at cloud-forming heights in the South Asian monsoon region. We find that ammonia profoundly influences the aerosol mass and composition in the ATAL through particle growth, as indicated by an order of magnitude increase in nitrate levels linked to ammonia emissions. However, the effect of ammonia-driven new particle formation on aerosol mass in the ATAL is relatively small. Ammonia emissions enhance the regional aerosol optical depth (AOD) for shortwave solar radiation by up to 70%. We conclude that ammonia has a pronounced effect on the ATAL development, composition, the regional AOD, and CCN concentrations.

Mercury (Hg) is a global atmospheric pollutant. In its oxidized form (HgII), it can readily deposit to ecosystems, where it may bioaccumulate and cause severe health effects. High HgII concentrations are reported in the free troposphere, but spatiotemporal data coverage is limited. Underestimation of HgII by commercially available measurement systems hinders quantification of Hg cycling and fate. During spring–summer 2021 and 2022, we measured elemental (Hg0) and oxidized Hg using a calibrated dual-channel system alongside trace gases, aerosol properties, and meteorology at the high-elevation Storm Peak Laboratory (SPL) above Steamboat Springs, Colorado. Oxidized Hg concentrations displayed diel and episodic behavior similar to previous work at SPL but were approximately 3 times higher in magnitude due to improved measurement accuracy. We identified 18 multi-day events of elevated HgII (mean enhancement of 36 pg m−3) that occurred in dry air (mean ± SD of relative humidity = 32 ± 16 %). Lagrangian particle dispersion model (HYSPLIT–STILT, Hybrid Single-Particle Lagrangian Integrated Trajectory–Stochastic Time-Inverted Lagrangian Transport) 10 d back trajectories showed that the majority of transport prior to events occurred in the low to middle free troposphere. Oxidized Hg was anticorrelated with Hg0 during events, with an average (± SD) slope of −0.39 ± 0.14. We posit that event HgII resulted from upwind oxidation followed by deposition or cloud uptake during transport. Meanwhile, sulfur dioxide measurements verified that three upwind coal-fired power plants did not influence ambient Hg at SPL. Principal component analysis showed HgII consistently inversely related to Hg0 and generally not associated with combustion tracers, confirming oxidation in the clean, dry free troposphere as its primary origin.

Abstract The application of machine learning (ML) to solve real-world problems does not only bear great potential but also high risk. One fundamental challenge in risk mitigation is to ensure the reliability of the ML predictions, i.e., the model error should be minimized, and the prediction uncertainty should be estimated. Especially for medical applications, the importance of reliable predictions can not be understated. Here, we address this challenge for anti-cancer drug sensitivity prediction and prioritization. To this end, we present a novel drug sensitivity prediction and prioritization approach guaranteeing user-specified certainty levels. The developed conformal prediction approach is applicable to classification, regression, and simultaneous regression and classification. Additionally, we propose a novel drug sensitivity measure that is based on clinically relevant drug concentrations and enables a straightforward prioritization of drugs for a given cancer sample.

Abstract Previous efforts to measure atmospheric iodine have focused on marine and coastal regions. We report the first ground‐based tropospheric iodine monoxide (IO) radical observations over the central continental United States. Throughout April 2022, IO columns above Storm Peak Laboratory, Colorado (3,220 m.a.s.l.) ranged from 0.7 ± 0.5 to 3.6 ± 0.5 × 1012 (average: 1.9 × 1012 molec cm−2). IO was consistently elevated in air masses transported from over the Pacific Ocean. The observed IO columns were up to three times higher and the range was larger than predicted by a global model, which warrants further investigation into iodine sources, sinks, ozone loss, and particle formation. IO mixing ratios increased with altitude. At the observed levels, iodine may be competitive with bromine as an oxidant of elemental mercury at cold temperatures typical of the free troposphere. Iodine‐induced mercury oxidation is missing in atmospheric models, understudied, and helps explain model underestimation of oxidized mercury measurements.

Aircraft-based measurements enable large-scale characterization of gas-phase atmospheric composition, but these measurements are complicated by the challenges of sampling from high-speed flow. Under such sampling conditions, the sample flow will likely experience turbulence, accelerating and mixing of potential contamination of the gas-phase from the condensed-phase components on walls, and reduced vapor transmission due to losses to the inner walls of the sampling line. While a significant amount of research has gone into understanding aerosol sampling efficiency for aircraft inlets, a similar research investment has not been made for gas sampling. Here, we analyze the performance of a forward-facing laminar flow gas inlet to establish its performance as a function of operating conditions, including ambient pressure, freestream velocities, and sampling conditions. Using computational fluid dynamics (CFD) modeling we simulate flow inside and outside the inlet to determine the extent of freestream turbulent interaction with the sample flow and its implication for gas sample transport. The CFD results of flow features in the inlet are compared against measurements of air speed and turbulent intensity from full-sized high-speed wind tunnel experiments. These comparisons suggest that the Reynolds-averaged Navier–Stokes (RANS) CFD simulations using the shear stress transport (SST) modeling approach provide the most reasonable prediction of the turbulence characteristics of the inlet.

The cell painting (CP) assay has emerged as a potent imaging-based high-throughput phenotypic profiling (HTPP) tool that provides comprehensive input data for in silico prediction of compound activities and potential hazards in drug discovery and toxicology. CP enables the rapid, multiplexed investigation of various molecular mechanisms for thousands of compounds at the single-cell level. The resulting large volumes of image data provide great opportunities but also pose challenges to image and data analysis routines as well as property prediction models. This review addresses the integration of CP-based phenotypic data together with or in substitute of structural information from compounds into machine (ML) and deep learning (DL) models to predict compound activities for various human-relevant disease endpoints and to identify the underlying modes-of-action (MoA) while avoiding unnecessary animal testing. The successful application of CP in combination with powerful ML/DL models promises further advances in understanding compound responses of cells guiding therapeutic development and risk assessment. Therefore, this review highlights the importance of unlocking the potential of CP assays when combined with molecular fingerprints for compound evaluation and discusses the current challenges that are associated with this approach.

One aspirational goal of computational chemistry is to predict potent and drug-like binders for any protein, such that only those that bind are synthesized. In this Roadmap, we describe the launch of Critical Assessment of Computational Hit-finding Experiments (CACHE), a public benchmarking project to compare and improve small molecule hit-finding algorithms through cycles of prediction and experimental testing. Participants will predict small molecule binders for new and biologically relevant protein targets representing different prediction scenarios. Predicted compounds will be tested rigorously in an experimental hub, and all predicted binders as well as all experimental screening data, including the chemical structures of experimentally tested compounds, will be made publicly available, and not subject to any intellectual property restrictions. The ability of a range of computational approaches to find novel binders will be evaluated, compared, and openly published. CACHE will launch 3 new benchmarking exercises every year. The outcomes will be better prediction methods, new small molecule binders for target proteins of importance for fundamental biology or drug discovery, and a major technological step towards achieving the goal of Target 2035, a global initiative to identify pharmacological probes for all human proteins.

Since 2017 the Tropospheric Monitoring Instrument (TROPOMI) on board ESA's Copernicus Sentinel-5 satellite (S5-P) has provided the operational carbon monoxide (CO) data product with daily global coverage on a spatial resolution of 5.5×7 km2. The European Centre for Medium-Range Weather Forecasts (ECMWF) plans to assimilate the retrieved total columns and the corresponding vertical sensitivities in the Copernicus Atmosphere Monitoring Service Integrated Forecasting System (CAMS-IFS) to improve forecasts of the atmospheric chemical composition. The TROPOMI data will primarily constrain the vertical integrated CO field of CAMS-IFS but to a lesser extent also its vertical CO distribution. For clear-sky conditions, the vertical sensitivity of the TROPOMI CO data product is useful throughout the atmosphere, but for cloudy scenes it varies due to cloud shielding and light scattering. To assess the profile information, we deploy an a posteriori profile retrieval that combines an ensemble of TROPOMI CO column retrievals with different vertical sensitivities to obtain a vertical CO profile that is then a representative average for the chosen spatial and temporal domain. We demonstrate the approach on three CO pollution cases. For the so-called “Rabbit Foot Fire” in Idaho on 12 August 2018, we estimate a CO profile showing the pollution at an altitude of about 5 km in good agreement with airborne in situ measurements of the Biomass Burning Flux Measurements of Trace Gases and Aerosol (BB-FLUX) field campaign. The distinct CO enhancement in a plume aloft (length =212 km, width =34 km), decoupled from the ground, is sensed by TROPOMI but is not present in the CAMS-IFS model. For a large-scale event, we analyzed the CO pollution from Siberian wildfires that took place from 14 to 17 August 2018. The TROPOMI data estimate the height of the pollution plume over Canada at 7 km in agreement with CAMS-IFS. However, CAMS-IFS underestimates the enhanced CO vertical column densities sensed by TROPOMI within the plume by more than 100 ppb. Finally, we study the seasonal biomass burning in the Amazon. During the burning season the CO profile retrieved from the TROPOMI measurements (1–15 August 2019) agrees well with the one of CAMS-IFS with a similar vertical shape between ground and 14 km altitude. Hence, our results indicate that assimilating TROPOMI CO retrieval with different vertical sensitivities (e.g., under clear-sky and cloudy conditions) provides information about the vertical distribution of CO.

Abstract The interaction between nitrogen monoxide (NO) and organic peroxy radicals (RO2) greatly impacts the formation of highly oxygenated organic molecules (HOM), the key precursors of secondary organic aerosols. It has been thought that HOM production can be significantly suppressed by NO even at low concentrations. Here, we perform dedicated experiments focusing on HOM formation from monoterpenes at low NO concentrations (0 – 82 pptv). We demonstrate that such low NO can enhance HOM production by modulating the RO2 loss and favoring the formation of alkoxy radicals that can continue to autoxidize through isomerization. These insights suggest that HOM yields from typical boreal forest emissions can vary between 2.5%-6.5%, and HOM formation will not be completely inhibited even at high NO concentrations. Our findings challenge the notion that NO monotonically reduces HOM yields by extending the knowledge of RO2-NO interactions to the low-NO regime. This represents a major advance towards an accurate assessment of HOM budgets, especially in low-NO environments, which prevails in the pre-industrial atmosphere, pristine areas, and the upper boundary layer.

Academic and pharmaceutical industry research are both key for progresses in the field of molecular machine learning. Despite common open research questions and long-term goals, the nature and scope of investigations typically differ between academia and industry. Herein, we highlight the opportunities that machine learning models offer to accelerate and improve compound selection. All parts of the model life cycle are discussed, including data preparation, model building, validation, and deployment. Main challenges in molecular machine learning as well as differences between academia and industry are highlighted. Furthermore, application aspects in the design-make-test-analyze cycle are discussed. We close with strategies that could improve collaboration between academic and industrial institutions and will advance the field even further.