Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “validation dataset”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 397 records · Page 22
All sourcesNASA NTRSDOE OSTI
DOE OSTI2026

Cybersecurity Center for Offshore Wind Energy (Final Project Report)

This project establishes a Cybersecurity Center for Offshore Wind Energy with the objective of designing and operating a cyber-physical testbed for wind energy farms (WEFs) that enables comprehensive cybersecurity research. The testbed incorporates a Supervisory Control and Data Acquisition (SCADA) system connected to turbine models via industrial-grade programmable logic controllers (PLCs) and remote terminal units (RTUs). It supports side-channel data acquisition, implementation and analysis of various cyberattack scenarios, and development of attack detection, mitigation, and best-practice guidance tailored to wind energy systems. During the project, the team expanded the number and fidelity of mathematical turbine models (MTMs), integrated these models with SCADA infrastructure, and deployed a scaled physical turbine and associated sensors. High-resolution operational and side-channel data streams were collected and used to refine machine-learning (ML)-based attack detection systems and to extend the WindCRAFT framework to multi-turbine threat scenarios. The project demonstrated a realistic, scalable environment for evaluating cyber threats, validated attack detection approaches using enriched datasets, and identified new multi-turbine and inter-turbine communication attack vectors. The resulting testbed, models, and security mechanisms provide a foundation for ongoing R&D and deployment of cyber-resilient offshore wind energy systems.

17 WIND ENERGY↗
DOE OSTI2024

Enhancing Docking Accuracy with PECAN2, a 3D Atomic Neural Network Trained without Co-Complex Crystal Structures

Decades of drug development research have explored a vast chemical space for highly active compounds. The exponential growth of virtual libraries enables easy access to billions of synthesizable molecules. Computational modeling, particularly molecular docking, utilizes physics-based calculations to prioritize molecules for synthesis and testing. Nevertheless, the molecular docking process often yields docking poses with favorable scores that prove to be inaccurate with experimental testing. To address these issues, several approaches using machine learning (ML) have been proposed to filter incorrect poses based on the crystal structures. However, most of the methods are limited by the availability of structure data. Here, we propose a new pose classification approach, PECAN2 (Pose Classification with 3D Atomic Network 2), without the need for crystal structures, based on a 3D atomic neural network with Point Cloud Network (PCN). The new approach uses the correlation between docking scores and experimental data to assign labels, instead of relying on the crystal structures. We validate the proposed classifier on multiple datasets including human mu, delta, and kappa opioid receptors and SARS-CoV-2 Mpro. Our results demonstrate that leveraging the correlation between docking scores and experimental data alone enhances molecular docking performance by filtering out false positives and false negatives.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2022

Graph-based featurization methods for classifying small molecule compounds

For over a decade, drug-induced liver injury (DILI) has posed significant drawbacks in the synthesis and development of drugs and remains a consequential concern. With finite success within the existing preclinical models, DILI is one of the main causes of drug withdrawal or termination from the market. Particularly, this withdrawal occurs during the late stages of drug development (Kullak-Ublick, 2017). Since DILI is difficult to diagnose and treat, it has become an obstacle in the drug production market that in turn affects clinicians, pharmaceutical companies, and consumers. We propose a method for learning features of DILI-positive drugs based on the graphical relationships and patterns they possess within a network of biological databases. We also train various statistical and machine learning models on these learned features in order to classify the drugs as DILI-positive or negative. Our methods include Random Forest, Neural networks, and logistic regression classification. We utilize labeled DILI-positive and DILI-negative datasets, which were developed by the FDA and the National center for toxicological research, as well as additional literature datasets (Thakkar, 2020) in order to validate our results and assess our featurization and model accuracy.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2023

Panta Rhei benchmark dataset: socio-hydrological data of paired events of floods and droughts

As the adverse impacts of hydrological extremes increase in many regions of the world, a better understanding of the drivers of changes in risk and impacts is essential for effective flood and drought risk management and climate adaptation. However, there is currently a lack of comprehensive, empirical data about the processes, interactions, and feedbacks in complex human–water systems leading to flood and drought impacts. Here we present a benchmark dataset containing socio-hydrological data of paired events, i.e. two floods or two droughts that occurred in the same area. The 45 paired events occurred in 42 different study areas and cover a wide range of socio-economic and hydro-climatic conditions. The dataset is unique in covering both floods and droughts, in the number of cases assessed and in the quantity of socio-hydrological data. The benchmark dataset comprises (1) detailed review-style reports about the events and key processes between the two events of a pair; (2) the key data table containing variables that assess the indicators which characterize management shortcomings, hazard, exposure, vulnerability, and impacts of all events; and (3) a table of the indicators of change that indicate the differences between the first and second event of a pair. The advantages of the dataset are that it enables comparative analyses across all the paired events based on the indicators of change and allows for detailed context- and location-specific assessments based on the extensive data and reports of the individual study areas. The dataset can be used by the scientific community for exploratory data analyses, e.g. focused on causal links between risk management; changes in hazard, exposure and vulnerability; and flood or drought impacts. The data can also be used for the development, calibration, and validation of socio-hydrological models. The dataset is available to the public through the GFZ Data Services (Kreibich et al., 2023, https://doi.org/10.5880/GFZ.4.4.2023.001).

54 ENVIRONMENTAL SCIENCES↗
NASA NTRS1991

Characterization and classification of sea ice in polarimetric SAR data

A more detailed characterization of the apparent thin ice types in the multifrequency, multipolarization synthetic aperture radar (SAR) dataset acquired during a campaign for validation of the DMSP SSM/I radiometer ice products is given. The emphasis is on providing a more detailed characterization of the signatures of apparent thin ice types observed in this data and the utility of these signatures for ice type classification purposes. The statistical characteristics of these signatures and their dependence on system calibration are summarized. Implications of these observations for sea ice scattering models are briefly discussed.

Kwok, R.↗
NASA NTRS2016

Validation of Cloud Properties From Multiple Satellites Using CALIOP Data

The NASA Langley Satellite ClOud and Radiative Property retrieval System (SatCORPS) is routinely applied to multispectral imagery from several geostationary and polar-orbiting imagers to retrieve cloud properties for weather and climate applications. Validation of the retrievals with independent datasets is continuously ongoing in order to understand differences caused by calibration, spatial resolution, viewing geometry, and other factors. The CALIOP instrument provides a decade of detailed cloud observations which can be used to evaluate passive imager retrievals of cloud boundaries, thermodynamic phase, cloud optical depth, and water path on a global scale. This paper focuses on comparisons of CALIOP retrievals to retrievals from MODIS, VIIRS, AVHRR, GOES, SEVIRI, and MTSAT. CALIOP is particularly skilled at detecting weakly-scattering cirrus clouds with optical depths less than approx. 0.5. These clouds are often undetected by passive imagers and the effect this has on the property retrievals is discussed.

Yost, Christopher R.↗
NASA NTRS2018

Blunt-Body Paradox and Improved Application of Transient-Growth Framework

The “Reshotko-Tumin transition criterion" based on optimal transient growth successfully correlates laboratory measurements of roughness induced transition over blunt body configurations. Even though transient growth has not been conclusively linked to the measured onset of transition, the above correlation denotes the only available physics-based model for subcritical transition in blunt body flows, since the latter do not support any modal instabilities at typical experimental conditions. Unlike other established models based on empirical curve fits that are valid for a specific subclass of datasets, the optimal-growth-based transition criterion appears to provide a reasonable correlation with measurements in various wind tunnel and ballistic range facilities and for a broad range of surface temperature ratios. This paper is focused on optimal growth calculations that improve upon significant shortcomings of the computations underlying the Reshotko-Tumin correlation. The improved framework is applied to leeward transition over a spherical section forebody that was tested in the Mach 6 Adjustable Contour Expansion wind tunnel at Texas A&M University. The computed results highlight the significance of nonparallel basic state evolution, curvature terms, and an optimization procedure that varies both inflow and outflow locations of the transient growth interval. More important, the results indicate that the modified correlation is very close to its original form, and hence, that the accuracy of the transient-growth-based transition criterion is not compromised by using a more thorough theoretical framework. Yet the results also show that the optimal energy gain up to the predicted transition onset location can be rather small, highlighting the need to further investigate the optimal growth criterion for additional experimental configurations and to also uncover the in-depth physics underlying blunt body transition.

Freestream conditions↗
NASA NTRSDate not supplied

TPSAS-NF1676L-30044-DND

The Clouds and the Earth's Radiant Energy System (CERES) project now has over 17 years accurately observed top-of-the-atmosphere (TOA) flux record for climate monitoring and diagnostic studies. The CERES FluxByCldTyp dataset, which contains cloud properties and radiative fluxes for 42 cloud types sorted by cloud top pressure and cloud optical depth, is used to investigate the clouds and their associated TOA (top-of-the-atmosphere) fluxes changes over the tropical area during ENSO events during the observed period. Unlike past studies, this study shows the impact of ENSO on cloud properties like optical depth, cloud top effective pressure and temperature and TOA LW and SW fluxes for each sub cloud type. The study reveals the detailed contributions from different cloud types for radiative characteristics during different regimes of ENSOs. This is especially important for very small net TOA radiative balance due to the cancellation of the fluxes from different cloud types. The dataset serves as a more stringent validation of climate models for cloud properties and radiative fluxes.

Sun, Moguo↗
NASA NTRS2020

Neural Network Reflectance Prediction Model for Both Open Ocean and Coastal Waters

Remote sensing of global ocean color is a valuable tool for understanding the ecology and biogeochemistry of the worlds oceans, and provides critical input to our knowledge of the global carbon cycle and the impacts of climate change. Ocean polarized reflectance contains information about the constituents of the upper ocean euphotic zone, such as colored dissolved organic matter (CDOM), sediments, phytoplankton, and pollutants. In order to retrieve the information on these constituents, remote sensing algorithms typically rely on radiative transfer models to interpret water color or remote-sensing reflectance; however, this can be resource-prohibitive for operational use due to the extensive CPU time involved in radiative transfer solutions. In this work, we report a fast model based on machine learning techniques, called Neural Network Reflectance Prediction Model (NNRPM), which can be used to predict ocean bidirectional polarized reflectance given inherent optical properties of ocean waters. This supervised model is trained using a large volume of data derived from radiative transfer simulations for coupled atmosphere and ocean systems using the successive order of scattering technique (SOS-CAOS). The performance of the model is validated against another large independent test dataset generated from SOS-CAOS. The model is able to predict both polarized and unpolarized reflectances with an absolute error (AE) less than 0.004 for 99% of test cases. We have also shown that the degree of linear polarization (DoLP) for unpolarized incident light can be predicted with an AE less than 0.002 for 99% of test cases. In general, the simulation time of SOS-CAOS depends on optical depth, and required accuracy. When comparing the average speeds of the NNRPM against the SOS-CAOS model for the same parameters, we see that the NNRPM is able to predict the Ocean BRDF 6000 times faster than SOS-CAOS. Both ultraviolet and visible wavelengths are included in the model to help differentiate between dissolved organic material and chlorophyll in the study of the open ocean and the coastal zone. The incorporation of this model into the retrieval algorithm will make the retrieval process more efficient, and thus applicable for operational use with global satellite observations.

radiative transfer↗
NASA NTRSDate not supplied

pyQuARC: Open Source Library for Earth Observation Metadata Quality Assessment

Metadata quality is essential to effective data discovery and has become increasingly vital as more Earth Science data sets become available. The Common Metadata Repository (CMR) hosts metadata describing NASA’s Earth Observation data products, which are archived across 12 Distributed Active Archive Centers (DAACs). The Analysis and Review of CMR (ARC) Team, located at Marshall Space Flight Center, conducts metadata quality assessments to ensure that these data products are discoverable, accessible, and usable. To achieve these goals, the ARC team has developed a metadata quality assessment framework to evaluate metadata completeness, correctness, and consistency. ARC uses a combination of manual and automated methods to assess these three components and identify areas of improvement; the team then collaborates with the DAACs to resolve any findings. To streamline this process, ARC is currently developing a host of scripts, known as pyQuARC, to automate metadata quality assessments as much as possible. pyQuARC is an open source library for Earth Observation Metadata Quality Assessment, and the tool utilizes ARC’s metadata quality assessment framework to make basic validation checks, pinpoint inconsistencies between dataset-level (i.e. collection) and file-level (i.e. granule) metadata, and identify opportunities for more descriptive and robust information. Since pyQuARC is also customizable, other users can make modifications as needed, and future metadata standards can also be implemented. Once pyQuARC is fully developed, it will support multiple schema types to serve the broader EOSDIS metadata community. This presentation will provide an overview of pyQuARC and its process of development while showcasing the tool’s valuable features and uses.

Jenny Wood↗
NASA NTRSDate not supplied

Summary of Shock Wave Turbulent Boundary Layer Interaction Experiments In a Circular Test Section

A series of experiments were performed at Mach 2.5 in a 17 cm diameter circular test section to characterize an impinging/reflected shock wave turbulent boundary layer interaction generated by a cone-cylinder centerbody. The cone-cylinder centerbody generates a conical shock wave that interacts with the naturally occurring boundary layer developing on the test section wall. Three different cone angles were used in the experiment to study unseparated, incipiently separated, and separated interactions. When the cone-cylinder centerbody is positioned on the centerline, a flowfield which is two-dimensional in the mean is generated. Three dimensional interactions were also created by offsetting the cone-cylinder centerbody from the test section centerline. The results are intended to provide benchmark quality datasets for computational fluid dynamics (CFD) validation without the pitfalls inherent in rectangular configurations where corner effects prohibit a truly two-dimensional flow in the mean. The experimental measurements included surface flow visualization, wall static pressure, flowfield Pitot tube pressure, constant-voltage anemometry (CVA) normal hot-wire, and particle image velocimetry (PIV) measurements. The hot-wire measurements were used to calculate mean mass flux and total temperature profiles, mass flux and total temperature turbulence intensities, and the mass flux-total temperature correlation. The PIV measurements provide three-dimensional mean velocity measurements. Agreement between the pressure, hot-wire, and PIV measurements is established in the undisturbed upstream flowfield.

Supersonic↗
NASA NTRSDate not supplied

Summary of Shock Wave Turbulent Boundary Layer Interaction Experiments in a Circular Test Section

A series of experiments were performed at Mach 2.5 in a 17 cm diameter circular test section to characterize an impinging/reflected shock wave turbulent boundary layer interaction generated by a cone-cylinder centerbody. The cone-cylinder centerbody generates a conical shock wave that interacts with the naturally occurring boundary layer developing on the test section wall. Three different cone angles were used in the experiment to study unseparated, incipiently separated, and separated interactions. When the cone-cylinder centerbody is positioned on the centerline, a flowfield which is two-dimensional in the mean is generated. Three dimensional interactions were also created by offsetting the cone-cylinder centerbody from the test section centerline. The results are intended to provide benchmark quality datasets for computational fluid dynamics (CFD) validation without the pitfalls inherent in rectangular configurations where corner effects prohibit a truly two-dimensional flow in the mean. The experimental measurements included surface flow visualization, wall static pressure, flowfield Pitot tube pressure, constant-voltage anemometry (CVA) normal hot-wire, and particle image velocimetry (PIV) measurements. The hot-wire measurements were used to calculate mean mass flux and total temperature profiles, mass flux and total temperature turbulence intensities, and the mass flux-total temperature correlation. The PIV measurements provide three-dimensional mean velocity measurements. Agreement between the pressure, hot-wire, and PIV measurements is established in the undisturbed upstream flowfield.

compressible flow↗
NASA NTRSDate not supplied

Digitizing Named Entities Found Within Letters of Agreement

Letters of Agreement (LOAs) are text-based air traffic control documents that contain procedures and actions agreed upon by the different parties, typically two or more FAA facilities, that are subject to an agreement. The documents contain among other things generic constraints, which are explicit and implicit combinations of procedures that limit a flight’s trajectory and affects pilot actions. For example, a controller may be required, to assign a specific altitude to an aircraft crossing the boundary between two airspaces. Although LOA generic constraints directly impact the trajectory of an aircraft, they are not currently available in a digital form that can be used for (or directly ingested into automated) flight planning. Instead, the constraints are manually input into an onboard or ground based system. LOA documents are primarily stored at a controlling facility and the generic constraints are implemented by experienced air traffic controllers and pilots primarily using voice instructions. This increases the workload of the controllers, likelihood of error (e.g., due to noisy communication) and makes it impractical for implementation with unmanned aircraft. Therefore, steps must be taken to make existing constraints machine interpretable to enable e.g., automated handoffs which in turn would reduce controller workload. With recent advances in natural language processing, especially the rise in digitization of text documents (e.g., medical documents) and automated extraction of information therein, it is now possible to extract flight specific constraints from LOAs. The goal of this work is to digitize named entities through a combination of natural language processing tasks: named entity disambiguation, toponym resolution, and numeric parsing to extract general constraint components contained within LOAs, herein referred to as Entity Enhancement (EE). Starting with a small list of named entities (e.g., ARTCC, Tower, Altitude and Speed), EE can extract the named entities while simultaneously converting the string-based output into a digital format using an ensemble of processes like rule-based gazetteers and syntactic-lexical patterns. The digital format contains a diverse set of information based on the entity label in question, ranging from standardized facility names to units of measure (e.g., feet) and other numeric information. Upon validating our approach using a truth dataset, we show an overall F1-Score of 0.71 for the extraction process. Looking beyond entity enhancement, we are also working towards the goal of completely digitizing the general constraints by performing EE and fitting them into a standardized exchange model (XM) such as the Aeronautical Information Exchange Model (AIXM). This will allow for easy distribution and dissemination of LOA constraints to air users, better searchability within documents, and enable ingestion into automated flight planning. Finally, we show a preliminary version of the proposed XM architecture and demonstrate how the model can be populated from the EE output.

Stephen S. B. Clarke↗
NASA NTRSDate not supplied

ENSO Tropical Cloud and TOA radiative signatures from the CERES observation

The Clouds and the Earth's Radiant Energy System (CERES) project now has over 15 years accurately observed top-of-the-atmosphere (TOA) flux record for climate monitoring and diagnostic studies. The CERES flux-by-cloud-type dataset, which contains cloud properties and radiative fluxes for 42 cloud types sorted by cloud top pressure and cloud optical depth, is used to investigate the clouds and their associated TOA (top-of-the-atmosphere) fluxes changes over the tropical area during ENSO events during the observed period. Unlike past studies, this study shows the impact of ENSO on cloud properties like optical depth, cloud top effective pressure and temperature and TOA LW and SW fluxes for each sub cloud type. The study reveals the detailed contributions from different cloud types for radiative characteristics during different phases of ENSO. This is especially important for very small net TOA radiative balance due to the cancellation of the fluxes from different cloud types. To further demonstrate the usefulness of this dataset, NCAR Community Atmosphere Model (CAM) is used to simulate the cloud and radiative changes during the period. The model cloud properties are converted to MODIS like cloud properties using modified MODIS simulator. The dataset serves as a more stringent validation of the model for cloud properties and radiative fluxes.

Moguo Sun↗
DOE OSTI2026

Probing the matter-dominated expansion with multi-redshift Lyman-$α$ BAO from DESI DR2

We present a multi-redshift Baryon Acoustic Oscillations (BAO) analysis of the DESI Data Release 2 (DR2) Lyman-$α$ (Ly$α$) forest, splitting the forest auto-correlation and its cross-correlation with quasars into three redshift bins. We obtain BAO measurements at effective redshifts $z_{\rm eff} = 2.13$, $2.40$, and $2.81$ with $\sim2.0$--$2.5\%$ precision per bin in the radial and transverse directions, corresponding to $\sim1.1$--$1.2\%$ precision for the isotropic BAO measurement. Using the same data products and modeling framework as the DESI DR2 Ly$α$ BAO analysis, we validate the pipeline on $400$ synthetic datasets and find unbiased BAO recovery with well-calibrated uncertainties. The measurements show an increase in the isotropic dilation parameter $D_V/r_d$ from $30.26\pm0.39$ to $32.22\pm0.47$ and in the Alcock-Paczyński parameter $D_M/D_H$ from $3.96\pm0.15$ to $5.63^{+0.22}_{-0.24}$. The Hubble distance $D_H/r_d$ decreases from $9.40\pm0.20$ to $7.22\pm0.17$, providing a direct measurement of the expansion history consistent with $Λ$CDM and the expected matter-dominated scaling, with $H(z)\propto(1+z)^n$ giving $n=1.34\pm0.16$. The redshift split also provides a self-consistent measurement of clustering evolution: the Ly$α$ forest bias evolves as $(1+z)^γ$ with $γ_α=3.05\pm0.16$, the RSD parameter has a redshift evolution described by $γ_β=-0.97\pm0.26$, and the quasar bias evolves with $γ_Q=1.56\pm0.23$, consistent with independent quasar clustering measurements. Combining these three-bin BAO measurements with DESI DR2 galaxy and quasar BAO measurements yields cosmological constraints consistent with the single-bin Ly$α$ BAO analysis in flat $Λ$CDM and $w_0w_a$CDM and improves curvature constraints by $\sim12\%$ in $Λ$CDM$+Ω_\mathrm{K}$.

Herrera-Alcantar, Hiram K. [Paris, Inst. Astrophys↗
NASA NTRS2017

Land Surface Temperature Product Validation Best Practice Protocol Version 1.0 - October, 2017

The Global Climate Observing System (GCOS) has specified the need to systematically generate andvalidate Land Surface Temperature (LST) products. This document provides recommendations on goodpractices for the validation of LST products. Internationally accepted definitions of LST, emissivity andassociated quantities are provided to ensure the compatibility across products and reference data sets. Asurvey of current validation capabilities indicates that progress is being made in terms of up-scaling and insitu measurement methods, but there is insufficient standardization with respect to performing andreporting statistically robust comparisons.Four LST validation approaches are identified: (1) Ground-based validation, which involvescomparisons with LST obtained from ground-based radiance measurements; (2) Scene-based intercomparisonof current satellite LST products with a heritage LST products; (3) Radiance-based validation,which is based on radiative transfer calculations for known atmospheric profiles and land surface emissivity;(4) Time series comparisons, which are particularly useful for detecting problems that can occur during aninstrument's life, e.g. calibration drift or unrealistic outliers due to undetected clouds. Finally, the need foran open access facility for performing LST product validation as well as accessing reference LST datasets isidentified.

best practice↗
NASA NTRS1995

Spectrometric Estimation of Total Nitrogen Concentration in Douglas-Fir Foliage

Spectral measurements of fresh and dehydrated Douglas-fir foliage, from trees cultivated under three fertilization treatments, were acquired with a laboratory spectrophotometer. The slope (first-derivative) of the fresh- and dry-leaf absorbance spectra at locations near known protein absorption features was strongly correlated with total nitrogen (TN) concentration of the foliage samples. Particularly strong correlation was observed between the first-derivative spectra in the 2150-2170 nm region and TN, reaching a local maximum in the fresh-leaf spectra of -0.84 at 2 160 nm. Stepwise regression was used to generate calibration equations relating first derivative spectra from fresh, dry/intact, and dry/ground samples to TN concentration. Standard errors of calibration were 1.52 mg g-1 (fresh), 1.33 (dry/intact), and 1.20 (dry/ground), with goodness-of-fit 0.94 and greater. Cross-validation was performed with the fresh-leaf dataset to examine the predictive capability of the regression method; standard errors of prediction ranged from 1.47 - 2.37 mg g(exp -1) across seven different validation sets, prediction goodness of fit ranged from .85-.94, and wavelength selection was fairly insensitive to the membership of the calibration set. All regressions in this study tended to select wavelengths in the 2100-2350 nm region, with the primary selection in the 2142-2172 nm region. The study provides positive evidence concerning the feasibility of assessing TN status of fresh-leaf samples by spectrometric means. We assert that the ability to extract biochemical information from fresh-leaf spectra is a necessary but insufficient condition regarding the use of remote sensing for canopy-level biochemical estimation.

Johnson, Lee F.↗
NASA NTRSDate not supplied

Using Machine Learning to Estimate Surface-Level SO2 Concentrations from Satellite-Based Measurements

Sulfur dioxide (SO2) is a criteria air pollutant due to its contributions to aerosol formation, rainfall acidification, and harm to human health. The placement of air quality monitoring sites is typically biased towards urban areas, leaving large areas with very limited monitoring data. The Ozone Monitoring Instrument (OMI) has been used to provide estimates of SO2 vertical column densities (VCDs) globally at spatial resolution of 10s of kms once per day. OMI SO2 VCDs have been previously used to estimate surface SO2 concentrations using chemical transport model (CTM) simulations. The CTMs use estimated emissions and assimilated meteorological data, and simulate the chemical and physical processes that determine the vertical profile of SO2, which can be used to derive a ratio between the surface concentrations and VCDs. These models are complex, computationally expensive, and have large uncertainties in the simulated surface-to-VCD ratio due to biases in emissions and relatively coarse resolution. Machine learning techniques are comparatively easier to use, much less computationally expensive to use after training, and can produce more accurate estimations of surface concentrations than the CTM-based method. The interpretation of machine learning models often poses challenges, and in some cases, non-physical variables unrelated to SO2 are used as predictors. In this work, we create an artificial neural network (ANN) to relate OMI retrievals and archived GEOS-FP boundary layer heights to surface SO2 concentrations from the ChinaHighAirPollutants ChinaHighSO2 dataset (CHAP; Wei et al., 2023) on a seasonal average timescale from 2013-2018. Our model only utilizes five variables that are directly relevant to the satellite retrieval, lifetime, and spatial distribution of SO2. The model was trained on 16 seasons (four of each) with independent validation (one of each season) and testing datasets (one of each season) to avoid overfitting. Our ANN generates surface SO2 concentrations that are sensitive (slope = 0.51) and consistent (r = 0.74) with the CHAP data, but are underpredicted by an average of 1.2 ppbv with a mean absolute error of 2.2 ppbv. These results are better than recent studies utilizing the CTM method. To our knowledge, this is the best performing machine learning model that only uses physical variables to predict surface SO2. Our work demonstrates that a carefully constructed, simple ML model can accurately estimate surface-based SO2 concentrations from satellite VCD measurements, and this technique has future promise to expend to newer, higher resolution satellites and other air pollutants.

SO2, air quality, OMI, machine learning↗