Search

Changing precipitation has the potential to alter nitrous oxide (N

Improved understanding of the optical properties of secondary organic aerosol (SOA) particles is needed to better predict their climate impacts. Here, SOA was produced by reacting 1-methylnaphthale...

Agricultural nitrous oxide (N2O) emission accounts for a non-trivial fraction of global greenhouse gas (GHG) budget. To date, estimating N2O fluxes from cropland remains a challenging task because the related microbial processes (e.g., nitrification and denitrification) are controlled by complex interactions among climate, soil, plant and human activities. Existing approaches such as process-based (PB) models have well-known limitations due to insufficient representations of the processes or uncertainties of model parameters, and due to leverage recent advances in machine learning (ML) a new method is needed to unlock the “black box” to overcome its limitations such as low interpretability, out-of-sample failure and massive data demand. In this study, we developed a first-of-its-kind knowledge-guided machine learning model for agroecosystems (KGML-ag) by incorporating biogeophysical and chemical domain knowledge from an advanced PB model, ecosys, and tested it by comparing simulating daily N2O fluxes with real observed data from mesocosm experiments. The gated recurrent unit (GRU) was used as the basis to build the model structure. To optimize the model performance, we have investigated a range of ideas, including (1) using initial values of intermediate variables (IMVs) instead of time series as model input to reduce data demand; (2) building hierarchical structures to explicitly estimate IMVs for further N2O prediction; (3) using multi-task learning to balance the simultaneous training on multiple variables; and (4) pre-training with millions of synthetic data generated from ecosys and fine-tuning with mesocosm observations. Six other pure ML models were developed using the same mesocosm data to serve as the benchmark for the KGML-ag model. Results show that KGML-ag did an excellent job in reproducing the mesocosm N2O fluxes (overall r2=0.81, and RMSE=3.6 mgNm-2d-1 from cross validation). Importantly, KGML-ag always outperforms the PB model and ML models in predicting N2O fluxes, especially for complex temporal dynamics and emission peaks. Besides, KGML-ag goes beyond the pure ML models by providing more interpretable predictions as well as pinpointing desired new knowledge and data to further empower the current KGML-ag. We believe the KGML-ag development in this study will stimulate a new body of research on interpretable ML for biogeochemistry and other related geoscience processes.

The ability of brown carbon (BrC) in aerosols to absorb solar radiation is an important but highly uncertain factor in climate forcing. The uncertainties are partially due to incomplete characterization of BrC chromophores and lack of authentic standards to confirm light absorption. Organonitrogen species are crucial components in atmospheric aerosols, but their light-absorbing properties remain to be fully characterized. To facilitate the molecular characterization of BrC chromophores, time-dependent density functional theory (TD-DFT) based computational chemistry approaches were used in this study to predict the light absorption spectra of 16 organonitrogen species, including nitroaromatics, nitro-heterocyclic compounds, organonitrates, and Maillard-type reaction products in BrC. Effects of basis sets, functionals, solvation, and pH on light absorption properties of these compounds were evaluated. Predicted absorption spectra were compared with experimental measurements. Overall, the PBE0 and B3LYP func...

The Salton Sea Basin in California suffers from poor air quality, and an expanding dry lakebed (playa) presents a new potential dust source. In 2017-18, depositing dust was collected approximately monthly at five sites in the Salton Sea Basin and analyzed for total elemental and soluble anion content. These data were analyzed with Positive Matrix Factorization (PMF). The PMF method resolved seven dust sources with distinct compositional markers: Playa (Mg, SO42-, Na, Ca, Sr), Colorado Alluvium (U, Ca), Local Alluvium (Al, Fe, Ti), Agricultural Burning (K, PO43-), Sea Spray (Na, Cl-, Se), Anthropogenic Trace Metals (Sb, As, Zn, Cd, Pb, Na), and Anthropogenic Copper (Cu). All sources except Local Alluvium are influenced or caused by current or historic anthropogenic activities. PMF attributed 55 to 80% of the measured dust flux to these six sources. The dust fluxes at the site where the playa source was dominant (89 g m-2 yr-1) were less than, but approaching the scale of, those observed at Owens Lake playas in the late 20th century. Playa emissions in the Salton Sea region were most intense during the late spring to early summer and contain high concentrations of evaporite mineral tracers, particularly Mg, Ca, and SO42-.

Dimethyl selenide (DMSe) is one of the major volatile organoselenium compounds released into the atmosphere through plant metabolism and microbial methylation. DMSe has been recently revealed as a precursor of secondary organic aerosol (SOA), and its resultant SOA possesses strong oxidizing capability toward thiol groups that can perturb several major biological pathways in human airway epithelial cells and is linked to genotoxicity, DNA damage, and p53-mediated stress responses. Mounting evidence has suggested that long noncoding RNAs (lncRNAs) are involved in stress responses to internal and environmental stimuli. However, the underlying molecular interactions remain to be elucidated. In this study, we performed integrative analyses of lncRNA-mRNA coexpression in the transformed human bronchial epithelial BEAS-2B cell line exposed to DMSe-derived SOA. We identified a total of 971 differentially expressed lncRNAs in BEAS-2B cells exposed to SOA derived from O3 and OH oxidation of DMSe. Gene ontology (GO) network analysis of cis-targeted genes showed significant enrichment of DNA damage, apoptosis, and p53-mediated stress response pathways. trans-Acting lncRNAs, including PINCR, PICART1, DLGAP1-AS2, and LINC01629, known to be associated with human carcinogenesis, also showed altered expression in cell treated with DMSe-SOA. Overall, this study highlights the regulatory role of lncRNAs in altered gene expression induced by DMSe-SOA exposure.

Accurate and precise retrievals of aerosol complex refractive indices (m) are essential to constraining the direct radiative effect of atmospheric aerosols. Despite this, there is no generally accepted method for constraining the uncertainty in full-distribution aerosol complex refractive index retrievals. This is in part due to condition-dependent and solution-dependent uncertainties which propagate through retrievals. Here, the Refractive Index Confidence Explorer (RICE), a program written in WaveMetrics Igor Pro, is presented. RICE applies a Monte Carlo-like method to propagate uncertainties through a full size distribution inverse Mie method (FD-IMM) for m retrievals. The m retrieval and RICE uncertainty analysis use absorption coefficients, scattering coefficients, aerosol size distributions, and measurement uncertainties as inputs. RICE iteratively tests a series of m values for their ability to produce the retrieved m under perturbed conditions. Perturbations account for uncertainties in optical, particle size, and particle number concentration measurements. RICE then uses these data to calculate semi-empirical probability distributions which are used to provide confidence intervals for the real (n) and imaginary (k) components of m. RICE provides measurement by measurement uncertainty estimations enabling estimation of uncertainty even when conditions are highly dynamic, like those associated with field measurements. When RICE is applied to idealized test cases and external data, uncertainty is shown to be dynamic in relation to the value of the retrieved m (solution) and the nature of the particle size distribution (measurement condition). Within these cases, m uncertainties were shown to be large for the upper end of n and k values explored here (i.e., n = 1.8 and k = 0.5, at 375 nm) under uncertainty conditions typical of modern particle and optical measurement technologies, suggesting FD-IMM’s usefulness may be limited by instrumental uncertainties under some measurement conditions. However, FD-IMM retrievals may still provide reasonable estimates of m when n k < 0.1. Copyright © 2021 American Association for Aerosol Research

The composition of ambient particulate matter (PM) and its sources were investigated at the Salton Sea, a shrinking saline lake in California. To investigate the influence of playa exposure on PM composition, PM samples were collected during two seasons and at two sites around the Salton Sea. To characterize source composition, soil samples were collected from local playa and desert surfaces. PM and soil samples were analyzed for 15 elements using mass spectrometry and X-ray diffraction. The contribution of sources to PM mass and composition was investigated using Al-referenced enrichment factors (EFs) and source factors resolved from positive matrix factorization (PMF). Playa soils were found to be significantly enriched in Ca, Na, and Se relative to desert soils. PMF analysis resolved the PM10 data with four source factors, identified as Playa-like, Desert-like, Ca-rich, and Se. Playa-like and desert-like sources were estimated to contribute to a daily average of 8.9% and 45% of PM10 mass, respectively....

Refractive index and optical properties of biogenic and anthropogenic secondary organic aerosol (SOA) particles were investigated. Aerosol precursors, namely longifolene, α-pinene, 1-methylnaphthalene, phenol, and toluene were oxidized in a Teflon chamber to produce SOA particles under different initial hydrocarbon concentrations and hydroxyl radical sources, reflecting exposures to different levels of nitrogen oxides (NOx). The real and imaginary components (n and k, respectively) of the refractive index at 375 nm and 632 nm were determined by Mie theory calculations through an iterative process, using the χ2 function to evaluate the fitness of the predicted optical parameters with the measured scattering, absorption, and extinction coefficients from a Photoacoustic Extinctiometer and Cavity Attenuated Phase Shift Spectrometer. Single scattering albedo (SSA) and bulk mass absorption coefficient (MAC) at 375 nm were calculated. SSA values of SOA particles from biogenic precursors (longifolene and α-pinene) were ∼0.98–0.99 (∼6.3% uncertainty), reflecting purely scattering aerosols regardless of the NOx regime. However, SOA particles from aromatic precursors were more absorbing and displayed NOx-dependent SSA values. For 1-methylnaphthalene SOA particles, SSA values of 0.92–0.95 and ∼0.75–0.90 (∼6.1% uncertainty) were observed under intermediate- and high-NOx conditions, respectively, reflecting the absorbing effects of SOA particles and NOx chemistry for this aromatic system. In mixtures of longifolene and phenol or longifolene and toluene SOA under intermediate- and high-NOx conditions, k values of the aromatic-related component of the SOA mixture were higher than that of 1-methylnaphthalene SOA particles. With the increase in OH exposure, kphenol decreased from 0.10 to 0.02 and 0.22 to 0.05 for intermediate- and high-NOx conditions, respectively. A simple relative radiative forcing calculation for urban environments at λ = 375 nm suggests the influence of absorbing SOA particles on relative radiative forcing at this wavelength is most significant for aerosol sizes greater than 0.4 µm. Copyright © 2019 American Association for Aerosol Research