Search

A large part of the uncertainty in climate projections comes from uncertain aerosol properties and aerosol-cloud interactions as well as the difficulty in remotely sensing them. The southeastern Atlantic functions as a natural laboratory to study biomass-burning smoke and to constrain this uncertainty. We address these gaps by comparing the Weather Research and Forecasting with Chemistry Community Atmosphere Model (WRF-CAM5) to the multi-campaign observations ORACLES (ObseRvations of Aerosols above CLouds and their intEractionS), CLARIFY (CLoud-Aerosol-Radiation Interaction and Forcing), and LASIC (Layered Atlantic Smoke Interactions with Clouds) in the southeastern Atlantic in August 2017 to evaluate a large range of the model's aerosol chemical properties, size distributions, processes, and transport, as well as aerosol-cloud interactions. Overall, while WRF-CAM5 is able to represent smoke properties and transport, some key discrepancies highlight the need for further analysis. Observations of smoke composition show an overall decrease in aerosol mean diameter as smoke ages over 4-12 d, while the model lacks this trend. A decrease in the mass ratio of organic aerosol (OA) to black carbon (BC), OA:BC, and the OA mass to carbon monoxide (CO) mixing ratio, OA:CO, suggests that the model is missing processes that selectively remove OA from the particle phase, such as photolysis and heterogeneous aerosol chemistry. A large (factor of ∼2.5) enhancement in sulfate from the free troposphere (FT) to the boundary layer (BL) in observations is not present in the model, pointing to the importance of properly representing secondary sulfate aerosol formation from marine dimethyl sulfide and gaseous SO2 smoke emissions. The model shows a persistent overprediction of aerosols in the marine boundary layer (MBL), especially for clean conditions, which multiple pieces of evidence link to weaker aerosol removal in the modeled MBL than reality. This evidence includes several model features, such as not representing observed shifts towards smaller aerosol diameters, inaccurate concentration ratios of carbon monoxide and black carbon, underprediction of heavy rain events, and little evidence of persistent biases in modeled entrainment. The average below-cloud aerosol activation fraction (NCLD/NAER) remains relatively constant in WRF-CAM5 between field campaigns (∼0.65), while it decreases substantially in observations from ORACLES (∼0.78) to CLARIFY (∼0.5), which could be due to the model misrepresentation of clean aerosol conditions. WRF-CAM5 also overshoots an observed upper limit on liquid cloud droplet concentration around NCLDCombining double low line 400-500 cm-3 and overpredicts the spread in NCLD. This could be related to the model often drastically overestimating the strength of boundary layer vertical turbulence by up to a factor of 10. We expect these results to motivate similar evaluations of other modeling systems and promote model development to reduce critical uncertainties in climate simulations. Copyright:

NASA's Deriving Information on Surface Conditions from Column and Vertically Resolved Observations Relevant to Air Quality (DISCOVER-AQ, conducted in 2011-2014) campaign in the United States and the joint NASA and National Institute of Environmental Research (NIER) Korea-United States Air Quality Study (KORUS-AQ, conducted in 2016) in South Korea were two field study programs that provided comprehensive, integrated datasets of airborne and surface observations of atmospheric constituents, including nitrogen dioxide (NO2), with the goal of improving the interpretation of spaceborne remote sensing data. Various types of NO2 measurements were made, including in situ concentrations and column amounts of NO2 using ground- and aircraft-based instruments, while NO2 column amounts were being derived from the Ozone Monitoring Instrument (OMI) on the Aura satellite. This study takes advantage of these unique datasets by first evaluating in situ data taken from two different instruments on the same aircraft platform, comparing coincidently sampled profile-integrated columns from aircraft spirals with remotely sensed column observations from ground-based Pandora spectrometers, intercomparing column observations from the ground (Pandora), aircraft (in situ vertical spirals), and space (OMI), and evaluating NO2 simulations from coarse Global Modeling Initiative (GMI) and high-resolution regional models. We then use these data to interpret observed discrepancies due to differences in sampling and deficiencies in the data reduction process. Finally, we assess satellite retrieval sensitivity to observed and modeled a priori NO2 profiles. Contemporaneous measurements from two aircraft instruments that likely sample similar air masses generally agree very well but are also found to differ in integrated columns by up to 31.9 %. These show even larger differences with Pandora, reaching up to 53.9 %, potentially due to a combination of strong gradients in NO2 fields that could be missed by aircraft spirals and errors in the Pandora retrievals. OMI NO2 values are about a factor of 2 lower in these highly polluted environments due in part to inaccurate retrieval assumptions (e.g., a priori profiles) but mostly to OMI's large footprint (> 312 km2).

Increasing impacts of wildfires on Western US air quality highlights the need for forecasts of smoke emissions based on dynamic modeled wildfires. This work utilizes knowledge of weather, fuels, topography, and firefighting, combined with machine learning and other statistical methods, to generate 1‐ and 2‐day forecasts of fire radiative energy (FRE). The models are trained on data covering 2019 and 2021 and evaluated on data for 2020. For the 1‐day (2‐day) forecasts, the random forest model shows the most skill, explaining 48% (25%) of the variance in observed daily FRE when trained on all available predictors compared to the 2% (<0%) of variance explained by persistence for the extreme fire year of 2020. The random forest model also shows improved skill in forecasting day‐to‐day increases and decreases in FRE, with 28% (39%) of observed increase (decrease) days predicted, and increase (decrease) days are identified with 62% (60%) accuracy. Error in the random forest increases with FRE, and the random forest tends toward persistence under severe fire weather. Sensitivity analysis shows that near‐surface weather and the latest observed FRE contribute the most to the skill of the model. When the random forest model was trained on subsets of the training data produced by agencies (e.g., the Canadian or US Forest Services), comparable if not better performance was achieved (1‐day R2 = 0.39–0.48, 2‐day R2 = 0.13–0.34). FRE is used to compute emissions, so these results demonstrate potential for improved fire emissions forecasts for air quality models. Plain Language Summary: Increasing wildfire smoke is undoing decades of air quality progress, yet air quality forecasts often miss the most intense smoke events. This is because forecasted smoke is released at constant rates whereas the rate of smoke release from real wildfires varies in time. In this work we teach a machine learning algorithm to predict the daily change in fire heat output, a quantity that is used to calculate wildfire emissions. The machine learning algorithm uses information regarding weather, fuel moisture and amount, and firefighting efforts to make its predictions. We also tried to predict the daily change in fire heat output using only weather information but found the machine learning method to be more successful. Many federal agencies have their own ways of tracking fire weather and fuel moisture, and in this paper, we show that we can apply machine learning to the data from any of several agencies and get the same level of forecasting skill. Key Points: Random forest models trained on weather, fuel, and firefighting data surpass persistence and weather‐based methods to predict fire energyRandom forest models beat persistence across states, for most days in the 2020 fire season, and across levels of fire severityFire weather and latest fire energy predictors add most skill to the random forest; models using agency weather data perform similarly [ABSTRACT FROM AUTHOR]

Concentrations of ambient particulate matter (such as PM2.5 and PM10) have come to represent a serious environmental problem worldwide, causing many deaths and economic losses. Because of the detrimental effects of PM2.5 on human health, many countries and international organizations have developed and operated regional and global short-term PM2.5 prediction systems. The short-term predictability of PM2.5 (and PM10) is determined by two main factors: the performance of the air quality model and the precision of the initial states. While specifically focusing on the latter factor, this study attempts to demonstrate how information from classical ground observation networks, a state-of-the-art geostationary (GEO) satellite sensor, and an advanced air quality modeling system can be synergistically combined to improve short-term PM2.5 predictability over South Korea. Such a synergistic combination of information can effectively overcome the major obstacle of scarcity of information, which frequently occurs in PM2.5 prediction systems using low Earth orbit (LEO) satellite-borne observations. This study first presents that the scarcity of information is mainly associated with cloud masking, sun-glint effect, and ill-location of satellite-borne data, and it then demonstrates that an advanced air quality modeling system equipped with synergistically-combined information can achieve substantially improved performances, producing enhancements of approximately 10%, 19%, 29%, and 10% in the predictability of PM2.5 over South Korea in terms of index of agreement (IOA), correlation coefficient (R), mean biases (MB), and hit rate (HR), respectively, compared to PM2.5 prediction systems using only LEO satellite-derived observations. [ABSTRACT FROM AUTHOR]

Accurately capturing cloud condensation nuclei (CCN) concentrations is key to understanding the aerosol-cloud interactions that continue to feature the highest uncertainty amongst numerous climate forcings. In situ CCN observations are sparse and most non-polarimetric passive remote sensing techniques are limited to providing column-effective CCN proxies such as total aerosol optical depth (AOD). Lidar measurements, on the other hand, resolve profiles of aerosol extinction and/or backscatter coefficients that are better suited for constraining vertically-resolved aerosol optical and microphysical properties. Here we present relationships between aerosol backscatter and extinction coefficients measured by the airborne High Spectral Resolution Lidar 2 (HSRL-2) and in situ measurements of CCN concentrations. The data were obtained during three deployments in the NASA ObseRvations of Aerosols above Clouds and their intEractionS (ORACLES) project, which took place over the Southeast Atlantic (SEA) during September 2016, August 2017, and September-October 2018. Our analysis of spatiotemporally collocated in situ CCN concentrations and HSRL-2 measurements indicates strong linear relationships between both data sets. The correlation is strongest for supersaturations greater than 0.25% and dry ambient conditions above the stratocumulus deck, where relative humidity (RH) is less than 50%. We find CCN - HSRL-2 Pearson correlation coefficients between 0.95-0.97 for different parts of the seasonal burning cycle that suggest fundamental similarities in biomass burning aerosol (BBA) microphysical properties. We find that ORACLES campaign-average values of in situ CCN and in situ extinction coefficients are qualitatively similar to those from other regions and aerosol types, demonstrating overall representativeness of our data set. We compute CCN - backscatter and CCN - extinction regressions that can be used to resolve vertical CCN concentrations across entire above-cloud lidar curtains. These lidar-derived CCN concentrations can be used to evaluate model performance, which we illustrate using an example CCN concentration curtain from WRF-CAM5. These results demonstrate the utility of deriving vertically-resolved CCN concentrations from lidar observations to expand the spatiotemporal coverage of limited or unavailable in situ observations. [ABSTRACT FROM AUTHOR]

The southeast Atlantic is home to an expansive smoke aerosol plume overlying a large cloud deck for approximately a third of the year. The aerosol plume is mainly attributed to the extensive biomass burning activity that occurs in southern Africa. Current Earth system models (ESMs) reveal significant differences in their estimates of regional aerosol radiative effects over this region. Such large differences partially stem from uncertainties in the vertical distribution of aerosols in the troposphere. These uncertainties translate into different aerosol optical depths (AOD) in the planetary boundary layer (PBL) and the free troposphere (FT). This study examines differences of AOD fraction in the FT and AOD differences among ESMs (WRF-CAM5, WRF-FINN, GEOS-Chem, EAM-E3SM, ALADIN, GEOS-FP, and MERRA-2) and aircraft-based measurements from the NASA ObseRvations of Aerosols above CLouds and their intEractionS (ORACLES) field campaign. Models frequently define the PBL as the well-mixed surface-based layer, but this definition misses the upper parts of decoupled PBLs, in which most low-level clouds occur. To account for the presence of decoupled boundary layers in the models, the height of maximum vertical gradient of specific humidity profiles from each model is used to define PBL heights. Results indicate that the monthly mean contribution of AOD in the FT to the total-column AOD ranges from 44% to 74% in September 2016 and from 54% to 71% in August 2017 within the region bounded by 25°S - 0° and 15°W - 15°E (excluding land) among the ESMs. Using the second-generation High Spectral Resolution Lidar (HSRL-2) to derive an aircraft-based constraint on the AOD and the fractional AOD, we found that WRF-CAM5 produces 40% less AOD than those from the HSRL-2 measurements, but it performs well at separating AOD fraction between the FT and the PBL. AOD fractions in the FT for GEOS-Chem and EAM-E3SM are, respectively, 10% and 15% lower than the AOD fractions from the HSRL-2 and their similarities in the mean AODs are the result of cancellation of high and low AOD biases. GEOS-FP, MERRA-2, and ALADIN produce 24% - 36% less AOD and tend to misplace more aerosols in the PBL compared to aircraft-based observations. The models generally underestimate AODs for measured AODs that are above 0.8, indicating their limitations at reproducing high AODs. The differences in the absolute AOD, FT AOD, and the vertical apportioning of AOD in different models highlight the need to continue improving the accuracy of modeled AOD distributions. These differences affect the sign and magnitude of the net aerosol radiative forcing, especially when aerosols are in contact with clouds. [ABSTRACT FROM AUTHOR]

Biomass burning smoke is advected over the southeast Atlantic Ocean between July and October of each year. This smoke plume overlies and mixes into a region of persistent low marine clouds. Model calculations of climate forcing by this plume vary significantly, in both magnitude and sign. The NASA EVS-2 (Earth Venture Suborbital-2) ORACLES (ObseRvations of Aerosols above CLouds and their intEractionS) project deployed for field campaigns off the west coast of Africa in three consecutive years (Sept., 2016; Aug., 2017; and Oct., 2018) with the goal of better characterizing this plume as a function of the monthly evolution, by measuring the parameters necessary to calculate the direct aerosol radiative effect. Here, this dataset and satellite retrievals of cloud properties are used to test the representation of the smoke plume and the underlying cloud layer in two regional models (WRF-CAM5 and CNRM-ALADIN) and two global models (GEOS and UM- UKCA). The focus is on comparisons of those aerosol and cloud properties that are the primary determinants of the direct aerosol radiative effect, and on the vertical distribution of the plume and its properties. The representativeness of the observations to monthly averages are tested for each field campaign, with the sampled mean aerosol light extinction generally found to be within 20% of the monthly mean at plume altitudes. When compared to the observations, in all models the simulated plume is too vertically diffuse, has smaller vertical gradients, and, in two of the models (GEOS and UM- UKCA), the plume core is displaced lower than in the observations. Plume carbon monoxide, black carbon, and organic aerosol masses indicate under-estimates in modeled plume concentrations, leading in general to under-estimates in mid- visible aerosol extinction and optical depth. Biases in mid-visible single scatter albedo are both positive and negative across the models. Observed vertical gradients in single scatter albedo are not captured by the models, but the models do capture the coarse temporal evolution, correctly simulating higher values in October (2018) than in August (2018) and September (2017). Uncertainties in the measured absorption Angstrom exponent were large but propagate into a negligible (<4%) uncertainty in integrated solar absorption by the aerosol and therefore in the aerosol direct radiative effect. Model biases in cloud fraction, and therefore the scene albedo below the plume, vary significantly across the four models. The optical thickness of clouds is, on average, well simulated in the WRF-CAM5 and ALADIN models in the stratocumulus region and is under-estimated in the GEOS model; UM-UKCA simulates significantly too-high cloud optical thickness. Overall, the study demonstrates the utility of repeated, semi-random sampling across multiple years that can give insights into model biases and how these biases affect modeled climate forcing. The combined impact of these aerosol and cloud biases on the direct aerosol radiative effect (DARE) is estimated using a first-order approximation for a sub-set of five comparison gridboxes. A significant finding is that the observed gridbox-average aerosol and cloud properties yield a positive (warming) aerosol direct radiative effect for all five gridboxes, whereas DARE using the gridbox-averaged modeled properties ranges from much larger positive values to small, negative values. It is shown quantitatively how model biases can offset each other, so that model improvements that reduce biases in only one property (e.g., single scatter albedo, but not cloud fraction) would lead to even greater biases in DARE. Across the models, biases in aerosol extinction and in cloud fraction and optical depth contribute the largest biases in DARE, with aerosol single scatter albedo also making a significant contribution. [ABSTRACT FROM AUTHOR]