Search NASA⌕ Search

SEARCH · Search NASA

Results for “FORECAST MODELS”

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 289 records · Page 16

Short-Term Precipitation Forecast Based on the PERSIANN System and LSTM Recurrent Neural Networks

Short-term Quantitative Precipitation Forecasting is important for flood forecasting, early flood warning, and natural hazard management. This study proposes a precipitation forecast model by extrapolating Cloud-Top Brightness Temperature (CTBT) using advanced Deep Neural Networks, and applying the forecasted CTBT into an effective rainfall retrieval algorithm to obtain the Short-term Quantitative Precipitation Forecasting (0–6 hr). To achieve such tasks, we propose a Long Short-Term Memory (LSTM) and the Precipitation Estimation from Remotely Sensed Information using Artificial Neural Networks (PERSIANN), respectively. The precipitation forecasts obtained from our proposed framework, (i.e., LSTM combined with PERSIANN) are compared with a Recurrent Neural Network (RNN), Persistency method, and Farneback optical flow each combined with PERSIANN algorithm and the numerical model results from the first version of Rapid Refresh (RAPv1.0) over three regions in the United States, including the states of Oregon, Oklahoma, and Florida. We report our experiments indicate better statistics, such as correlation coefficient and root-mean-square error, for the CTBT forecasts from the proposed LSTM compared to the RNN, Persistency, and the Farneback method. The precipitation forecasts from the proposed LSTM and PERSIANN framework has demonstrated better statistics compared to the RAPv1.0 numerical forecasts and PERSIANN estimations from RNN, Persistency, and Farneback projections in terms of Probability of Detection, False Alarm Ratio, Critical Success Index, correlation coefficient, and root-mean-square error, especially in predicting the convective rainfalls. The proposed method shows superior capabilities in short-term forecasting over compared methods, and has the potential to be implemented globally as an alternative short-term forecast product.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Improving energy efficiency while reducing anthropogenic heat from buildings: how retrofits influence the building stock and urban microclimate in Los Angeles

Anthropogenic heat (AH) from buildings contributes to urban overheating, especially during heat waves, yet building retrofit studies usually evaluate energy savings without assessing impacts on AH. This study quantifies how common building retrofit measures affect both building energy use and AH emissions across the City of Los Angeles. Using a bottom-up urban building energy modeling framework coupled with high-resolution local weather from the Weather Research and Forecasting model with Building Effect Parameterization (WRF-BEP), we evaluate eleven retrofit measures and two multi-measure retrofit packages. HVAC and LED lighting retrofits provide the largest city-wide annual site energy savings, while roof coating is most effective for reducing AH. A package optimized for energy savings reduces summer site energy use by about 32% (2.3 TWh), while a package incorporating AH-focused measures reduces the total AH by over 50% (137 PJ) with minimal difference in energy savings. The AH-aware package produces substantially greater urban cooling, reducing mean near-surface air temperature by up to 0.62 ℃ and peak temperature by up to 3.79 ℃. These results show that retrofit strategies selected only for energy savings may overlook major opportunities for urban heat mitigation. The study provides a framework for integrating AH into building retrofit planning and urban heat resilience policy.

Anthropogenic heat↗

Integration of a physics-based direct normal irradiance ($\mathrm{DNI}$) model to enhance the National Solar Radiation Database ($\mathrm{NSRDB}$)

The National Solar Radiation Database (NSRDB) is an extensively used dataset that furnishes satellite-retrieved solar resource data across the United States and an expanding list of other countries. Although the NSRDB uses a physical model to compute global horizontal irradiance (GHI), it currently employs an empirical approach based on surface observations to estimate cloudy-sky direct normal irradiance (DNI). Recently, a new physics-based approach, known as the Fast All-sky Radiation Model for Solar applications with DNI (FARMS-DNI), was developed to improve DNI forecasting. FARMS-DNI integrates direct and scattered solar radiances within the circumsolar region, resulting in improved day-ahead forecasting of DNI by incorporating it into the Weather Research and Forecasting model with Solar extensions (WRF-Solar). This study incorporates FARMS-DNI into the NSRDB algorithm to produce high-spatiotemporal-resolution DNI data from satellite data. The accuracy of the NSRDB based on FARMS-DNI is analyzed using surface observations from 19 sites situated within the National Oceanic and Atmospheric Administration (NOAA) Surface Radiation Budget (SURFRAD) and Solar Radiation (SOLRAD) networks, the University of Oregon (UO) network, the U.S. Department of Energy (DOE) Atmospheric Radiation Measurement (ARM) network, and at the National Renewable Energy Laboratory (NREL). The results demonstrate that FARMS-DNI reduces the significant overestimation of DNI in the conventional NSRDB at all surface sites, particularly in cloud overcast conditions classified using both satellite retrievals and surface observations. Consequently, this new model can effectively improve the overall accuracy of the NSRDB. The results also suggest that further improvement of DNI estimates at individual time steps, however, requires advanced satellite techniques and precise identification of clouds and retrieval of cloud properties.

14 SOLAR ENERGY↗

Simulated Polarimetric Fields of Ice Vapor Growth Using the Adaptive Habit Model. Part I: Large-Eddy Simulations

Abstract The bulk adaptive habit model (AHM) explicitly predicts ice particle aspect ratio, improving the representation of microphysical processes and properties, including ice–liquid-phase partitioning. With the unique ability to predict ice particle shape and density, the AHM is combined with an offline forward operator to produce fields of simulated polarimetric variables. An evaluation of AHM-forward-simulated dual-polarization radar signatures in an idealized Arctic mixed-phase cloud is presented. Interpretations of those signatures are provided through microphysical model output using the large-eddy simulation mode of the Weather Research and Forecasting Model. Vapor-grown ice properties are associated with distinct observable signatures in polarimetric radar variables, with clear sensitivities to the simulated ice particle properties, including ice number, size, and distribution shape. In contrast, the liquid droplet number has little influence on both polarimetric and microphysical variables in the case presented herein. Polarimetric quantities are sensitive to the dominating crystal habit type in a volume, with enhancements for aspect ratios much lower or higher than unity. This synthesis of a microphysical model and a polarimetric forward simulator is a first step in the evaluation of detailed AHM microphysics.

54 ENVIRONMENTAL SCIENCES↗

Surface Energy Balance Responses to Radiative Forcing in the Central Arctic From MOSAiC and Models

The Arctic surface energy budget (SEB) couples the atmosphere with the sea ice, making it useful for both studying surface processes as well as evaluating models. Improved understanding of atmosphere-ice interactions is required to improve models, requiring year-round observations to address seasonally dependent biases. This work uses novel observations from the MOSAiC expedition to quantify the responses of surface fluxes to radiative forcing over sea ice throughout a complete annual cycle. We identify two primary regimes of flux response: an ice growth regime in winter and an ice melt regime in summer. In the growth regime, changes in radiative forcing impact upwelling longwave, sensible heat, and subsurface heat fluxes, whereas in the melt regime changes in radiative forcing primarily alter the amount of melt and subsurface transmission because the surface temperature is fixed. These observed responses of surface fluxes to radiative forcing are used to evaluate seven weather forecast models during the ice growth regime. In most models, the responses of surface fluxes to radiative forcing do not match observations. Many models also have biased downwelling longwave. One model (the Coupled Arctic Forecast System; CAFS) adequately captures both the mean radiative forcing and the flux responses in winter. CAFS is further evaluated against observations spanning the full MOSAiC year, demonstrating sufficient agreement to provide a more generalized understanding of these SEB process relationships across the Arctic.

54 ENVIRONMENTAL SCIENCES↗

3D cloud detection and tracking system for solar forecast using multiple sky imagers

We propose a system for forecasting short-term solar irradiance based on multiple total sky imagers (TSIs). The system utilizes a novel method of identifying and tracking clouds in three-dimensional space and an innovative pipeline for forecasting surface solar irradiance based on the image features of clouds. First, we develop a supervised classifier to detect clouds at the pixel level and output cloud mask. In the next step, we design intelligent algorithms to estimate the block-wise base height and motion of each cloud layer based on images from multiple TSIs. Thus, this information is then applied to stitch images together into larger views, which are then used for solar forecasting. We examine the system’s ability to track clouds under various cloud conditions and investigate different irradiance forecast models at various sites. We confirm that this system can 1) robustly detect clouds and track layers, and 2) extract the significant global and local features for obtaining stable irradiance forecasts with short forecast horizons from the obtained images. Finally, we vet our forecasting system at the 32-megawatt Long Island Solar Farm (LISF). Compared with the persistent model, our system achieves at least a 26% improvement for all irradiance forecasts between one and fifteen minutes.

97 MATHEMATICS AND COMPUTING↗

Modeling Organic Aerosols in a Megacity: Comparison of Simple and Complex Representations of the Volatility Basis Set Approach

The Weather Research and Forecasting model coupled with chemistry (WRF-Chem) is modified to include a volatility basis set (VBS) treatment of secondary organic aerosol formation. The VBS approach, coupled with SAPRC-99 gas-phase chemistry mechanism, is used to model gas-particle partitioning and multiple generations of gas-phase oxidation of organic vapors. In addition to the detailed 9-species VBS, a simplified mechanism using 2 volatility species (2-species VBS) is developed and tested for similarity to the 9-species VBS in terms of both mass and oxygen-to-carbon ratios of organic aerosols in the atmosphere. WRF-Chem results are evaluated against field measurements of organic aerosols collected during the MILAGRO 2006 campaign in the vicinity of Mexico City. The simplified 2-species mechanism reduces the computational cost by a factor of 2 as compared to 9-species VBS. Both ground site and aircraft measurements suggest that the 9-species and 2-species VBS predictions of total organic aerosol mass as well as individual organic aerosol components including primary, secondary, and biomass burning are comparable in magnitude. In addition, oxygen-to-carbon ratio predictions from both approaches agree within 25%, providing evidence that the 2-species VBS is well suited to represent the complex evolution of organic aerosols. Model sensitivity to amount of anthropogenic semi-volatile and intermediate volatility (S/IVOC) precursor emissions is also examined by doubling the default emissions. Both the emission cases significantly under-predict primary organic aerosols in the city center and along aircraft flight transects. Secondary organic aerosols are predicted reasonably well along flight tracks surrounding the city, but are consistently over-predicted downwind of the city. Also, oxygen-to-carbon ratio predictions are significantly improved compared to prior studies by adding 15% oxygen mass per generation of oxidation; however, all modeling cases still under-predict these ratios downwind as compared to measurements, suggesting a need to further improve chemistry parameterizations of secondary organic aerosol formation.

09 BIOMASS FUELS↗

A copula-based Bayesian method for probabilistic solar power forecasting

With increased penetration of solar energy sources, solar power forecasting has become more crucial and challenging. This paper proposes a copula-based Bayesian approach to improve probabilistic solar power forecasting by capturing the joint distribution between solar power and ambient temperature. A prior forecast distribution is first obtained using different underlying point forecasting models. Parametric and empirical copulas of solar power and temperature are then developed to update the prior distribution to the posterior forecast distribution. A public solar power database is used to demonstrate effectiveness of the proposed method. Numerical results show that the copula-based Bayesian method outperforms the forecasting method that directly uses temperature as a feature. The Bayesian method is also compared with persistent models and show improved performance. Furthermore, this article includes supplementary material (data and code) for reproducibility.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Integration of microseismic monitoring data into coupled flow and geomechanical models with ensemble Kalman filter

Abstract Hydraulic stimulation of low‐permeability rocks in enhanced geothermal systems, shale resources, and CO 2 storage aquifers can trigger microseismic events, also known as microearthquakes (MEQs). The distribution of microseismic source locations in the reservoir may reveal important information about the distribution of hydraulic and geomechanical rock properties. In this paper, we present a framework for conditioning heterogeneous rock permeability and geomechanical property distributions on microseismic data. To simulate the multiphysics processes in these systems, we combine a fully coupled flow and geomechanical model with the Mohr‐Coulomb type rock failure criterion. The resulting multiphysics simulation constitutes the forecast model that relates microseismic source locations to reservoir rock properties. We adopt this forward model in an ensemble Kalman filter (EnKF) data assimilation framework to jointly estimate reservoir permeability and geomechanical property distributions from injection‐induced microseismic response measurements. We show that integration of a large number of spatially correlated microseismic data with practical ensemble sizes can lead to severe underestimation of ensemble spread, and eventually ensemble collapse. To mitigate the variance underestimation issue, two low‐rank data representation schemes are presented and discussed. In the first approach, microseismic data are projected onto a low‐dimensional subspace defined by the left singular vectors of the perturbed observation matrix. The second method uses a coarser grid for representing the microseismic data. A series of numerical experiments is presented to evaluate the performance of the proposed methods and to illustrate their applicability for assimilating microseismic data into coupled flow and geomechanical forward models to estimate multiphysics rock properties.

Tarrahi, Mohammadali↗

Sensitivity of U.S. summer precipitation to model resolution and convective parameterizations across gray zone resolutions: Sensitivity Across Gray Zone Resolutions

Simulating summer precipitation is a significant challenge for climate models that rely on cumulus parameterizations to represent moist convection processes. Motivated by recent advances in computing that support very high-resolution modeling, this study aims to systematically evaluate the effects of model resolution and convective parameterizations across the gray zone resolutions. Simulations usingthe Weather Research and Forecasting model were conducted at grid spacings of 36 km, 12 km, and 4 km fortwo summers over the conterminous U.S. The convection-permitting simulations at 4 km grid spacing are most skillful in reproducing the observed precipitation spatial distributions and diurnal variability. Notable differences are found between simulations with the traditional Kain-Fritsch (KF) and the scale-aware Grell-Freitas (GF) convection schemes, with the latter more skillful in capturing the nocturnal timing in the Great Plains and North American monsoon regions. The GF scheme also simulates a smoother transition from convective to large-scale precipitation as resolution increases, resulting in reduced sensitivity to model resolution compared to the KF scheme. Nonhydrostatic dynamics has a positive impact on precipitation over complex terrain even at 12 km and 36 km grid spacings. With nudging of the winds toward observations, we show that the conspicuous warm biases in the Southern Great Plains are related to precipitation biases induced by large-scale circulation biases, which are insensitive to model resolution. Overall, notable improvements in simulating summer rainfall and its diurnal variability through convection-permitting modeling and scale-aware parameterizations suggest promising venues for improving climate simulations of water cycle processes.

54 ENVIRONMENTAL SCIENCES↗

ARM MJO Investigation Experiment on Gan Island (AMIE-Gan) Science Plan

The overarching campaign, which includes the ARM Mobile Facility 2 (AMF2) deployment in conjunction with the Dynamics of the Madden-Julian Oscillation (DYNAMO) and the Cooperative Indian Ocean experiment on intraseasonal variability in the Year 2011 (CINDY2011) campaigns, is designed to test several current hypotheses regarding the mechanisms responsible for Madden-Julian Oscillation (MJO) initiation and propagation in the Indian Ocean area. The synergy between the proposed AMF2 deployment with DYNAMO/CINDY2011, and the corresponding funded experiment on Manus, combine for an overarching ARM MJO Investigation Experiment (AMIE) with two components: AMF2 on Gan Island in the Indian Ocean (AMIE-Gan), where the MJO initiates and starts its eastward propagation; and the ARM Manus site (AMIE-Manus), which is in the general area where the MJO usually starts to weaken in climate models. AMIE-Gan will provide measurements of particular interest to Atmospheric System Research (ASR) researchers relevant to improving the representation of MJO initiation in climate models. The framework of DYNAMO/CINDY2011 includes two proposed island-based sites and two ship-based locations forming a square pattern with sonde profiles and scanning precipitation and cloud radars at both island and ship sites. These data will be used to produce a Variational Analysis data set coinciding with the one produced for AMIE-Manus. The synergy between AMIE-Manus and AMIE-Gan will allow studies of the initiation, propagation, and evolution of the convective cloud population within the framework of the MJO. As with AMIE-Manus, AMIE-Gan/DYNAMO also includes a significant modeling component geared toward improving the representation of MJO initiation and propagation in climate and forecast models. This campaign involves the deployment of the second, marine-capable, AMF; all of the included measurement systems; and especially the scanning and vertically pointing radars. The campaign will include sonde launches at a rate of eight per day for the duration of the deployment. The increased sonde launches for the entire period matches that of the AMIE-Manus campaign and makes possible a far more robust Variational Analysis forcing data set product for the entire campaign, and thus better capabilities for modeling studies and synergistic research using the data from both AMIE sites.

54 ENVIRONMENTAL SCIENCES↗

Evaluating Large-Eddy Simulations of Boundary-Layer Cloud Response to Climate Change using MAGIC Observations

Over the cool waters off California, a persistent and extensive deck of low-lying clouds forms atop a shallow, turbulent layer of moist air extending 500-2000 meters above the ocean, above which the air is much warmer and drier. Its west edge varies with weather and season and is typically half way to Hawaii. This cloud deck and similar cloud decks elsewhere reflect sunlight back to space, cooling the underlying ocean and indeed the entire planet. They must be accurately represented by a climate model to correctly simulate our current climate or future global warming. Most climate models suggest these cloud decks will thin and break up in a warmer CO 2 -enriched climate, so extra sunlight would reach the ocean surface and amplify warming worldwide. But these models have difficulty accurately simulating the cloud thickness and extent because the turbulent cloud-forming air currents are much too small to be simulated by their coarse computational grid. In 2012-2013, the DOE Atmospheric Radiation Measurement MAGIC program instrumented a container ship transiting between California and Hawaii to observe clouds and rain, small particles (aerosols), and weather conditions. Data were obtained from 20 4-6 day cruises in each direction. This study tested whether a sophisticated ‘large-eddy simulation’ (LES) of the turbulent air motions that create the clouds in this region accurately predicts the variations in cloud properties seen across the different cruises. If so, that model is more believable for predicting cloud changes in this region in a future warmer climate. Our LES used a numerical mesh over a 6.4x6.4 km region with 50 m horizontal and 5 m vertical grid spacing to simulate the cloud-forming eddies and the sharp overlying transition between moist and dry air. To better compare with MAGIC observations, we configured our model to simulate a region that moved along with the ship. We used a weather forecast model to specify effects of the large-scale air flow. We compared our results with balloon ascents, a radar and other instruments pointed upward at the clouds, measurements of downward sunlight and infrared radiation, aerosols, and weather variables for all suitable cruise periods. Our LES was remarkably skillful in predicting the cloudiness variations measured by the ship. MAGIC also sampled mesoscale cells, regions of cloud thickening and thinning over distances of 10-50 km. An LES run over a larger region 100 km on a side shows these cells form spontaneously, enhanced by strong radiative heat loss at the cloud tops.

54 ENVIRONMENTAL SCIENCES↗

Improving the National Solar Radiation Database (NSRDB) Using a Physics-Based Direct Normal Irradiance (DNI) Model

The National Solar Radiation Database (NSRDB) is a widely used resource providing satellite-derived solar data across the United States and globally. While the NSRDB employs a physical model for computing global horizontal irradiance (GHI), its current method for estimating cloudy-sky direct normal irradiance (DNI) relies on surface observations and empirical models. Recently, a novel physics-based approach, the Fast All-Sky Radiation Model for Solar applications with DNI (FARMS-DNI), was developed to enhance the DNI forecasting. FARMS-DNI incorporates both direct and scattered solar radiation within the circumsolar region, resulting in improved day-ahead DNI predictions when integrated into the Weather Research and Forecasting model with Solar extensions (WRF-Solar). This study integrates FARMS-DNI into the NSRDB algorithm to generate high-resolution DNI data from satellite resources. Our findings reveal that FARMS-DNI effectively mitigates the substantial DNI overestimation present in the conventional NSRDB across surface sites, particularly in conditions categorized as cloudy overcast. Consequently, this innovative model substantially enhances the overall accuracy of the NSRDB.

Xie, Yu↗

Improving the National Solar Radiation Database (NSRDB) Using a Physics-Based Direct Normal Irradiance (DNI) Model: Preprint

The National Solar Radiation Database (NSRDB) is a widely used resource providing satellite-derived solar data across the United States and globally. While the NSRDB employs a physical model for computing global horizontal irradiance (GHI), its current method for estimating cloudy-sky direct normal irradiance (DNI) relies on surface observations and empirical models. Recently, a novel physics-based approach, the Fast All-Sky Radiation Model for Solar applications with DNI (FARMS-DNI), was developed to enhance the DNI forecasting. FARMS-DNI incorporates both direct and scattered solar radiation within the circumsolar region, resulting in improved day-ahead DNI predictions when integrated into the Weather Research and Forecasting model with Solar extensions (WRF-Solar). This study integrates FARMS-DNI into the NSRDB algorithm to generate high-resolution DNI data from satellite resources. Our findings reveal that FARMS-DNI effectively mitigates the substantial DNI overestimation present in the conventional NSRDB across surface sites, particularly in conditions categorized as cloudy overcast. Consequently, this innovative model substantially enhances the overall accuracy of the NSRDB.

DNI↗

Conditioning multi-model ensembles for disease forecasting

In this study we investigate how an ensemble of disease models can be conditioned to observational data, in a bid to improve its predictive skill. We use the ensemble of influenza forecasting models gathered by the US Centers for Disease Control and Prevention (CDC) as the exemplar. This ensemble is used every year to forecast the annual influenza outbreak in the United States. The models constituting this ensemble draw on very different modeling assumptions and approximations and are a diverse collection of methods to approximate epidemiological dynamics. Currently, each models' predictions are accorded the same importance, or weight, when compiling the ensemble's forecast. We consider this equally-weighted ensemble as the baseline case which has to be improved upon. In this study, we explore whether an ensemble forecast can be improved by "conditioning" the ensemble to whatever observational data is available from the ongoing outbreak. "Conditioning" can imply according the ensemble's members different weights which evolve over time, or simply perform the forecast using the top k (equally-weighted) models. In the latter case, the composition of the "top-k-see of models evolves over time. This is called "model averaging" in statistics. We explore four methods to perform model-averaging, three of which are new. We find that the CDC ensemble responds best to the "top-k-models" approach to model-averaging. All the new MA methods perform better than the baseline equally-weighted ensemble. The four model-averaging methods treat the models as black-boxes and simply use their forecasts as inputs i.e., one does not need access to the models at all, but rather only their forecasts. The model-averaging approaches reviewed in this report thus form a general framework for model-averaging any model ensemble.

59 BASIC BIOLOGICAL SCIENCES↗

Power and Wind Shear Implications of Large Wind Turbine Scenarios in the US Central Plains

Continued growth of wind turbine physical dimensions is examined in terms of the implications for wind speed, power and shear across the rotor plane. High-resolution simulations with the Weather Research and Forecasting model are used to generate statistics of wind speed profiles for scenarios of current and future wind turbines. The nine-month simulations, focused on the eastern Central Plains, show that the power scales broadly as expected with the increase in rotor diameter (D) and wind speeds at hub-height (H). Increasing wind turbine dimensions from current values (approximately H = 100 m, D = 100 m) to those of the new International Energy Agency reference wind turbine (H = 150 m, D = 240 m), the power across the rotor plane increases 7.1 times. The mean domain-wide wind shear exponent (α) decreases from 0.21 (H = 100 m, D = 100 m) to 0.19 for the largest wind turbine scenario considered (H = 168 m, D = 248 m) and the frequency of extreme positive shear (α > 0.2) declines from 48% to 38% of 10-min periods. Thus, deployment of larger wind turbines potentially yields considerable net benefits for both the wind resource and reductions in fatigue loading related to vertical shear.

17 WIND ENERGY↗

Forecasting induced seismicity in Oklahoma using machine learning methods

Oklahoma earthquakes in the past decade have been mostly associated with wastewater injection. Here we use a machine learning technique—the Random Forest to forecast induced seismicity rate in Oklahoma based on injection-related parameters. We split the data into training (2011.01–2015.05) and test (2015.06–2020.12) periods. The model forecasts seismicity rate during the test period based on input features, including operational parameters (injection rate and pressure), geological information (depth to basement), and modeled pore pressure and poroelastic stress. The results show overall good match with observed seismicity rate (adjusted R 2 of 0.75). The model shows that pore pressure rate and poroelastic stressing rates are the two most important features in forecasting. The absolute values of pore pressure and poroelastic stress, and the injection rate itself, are less important than the stressing rates. These findings further emphasize that temporal changes of stressing rates would lead to significant changes in seismicity rates.

58 GEOSCIENCES↗

Simulations of Sulfate-Nitrate-Ammonium (SNA) aerosols during the extreme haze events over Northern China in 2014

Extreme haze events have occurred frequently over China in recent years. Although many studies have investigated the formation mechanisms associated with PM2.5 for heavily polluted regions in China based on observational data, adequately predicting peak PM2.5 concentrations is still challenging for regional air quality models. In this study, we evaluate the performance of one configuration of the Weather Research and Forecasting model coupled with chemistry (WRF-Chem) and use the model to investigate the sensitivity of heterogeneous reactions on simulated peak sulfate, nitrate, and ammonium concentrations in the vicinity of Beijing during four extreme haze episodes in October 2014 over the North China Plain. The highest observed PM2.5 concentration of 469 μg m-3 occurred in Beijing. Comparisons with observations show that the model reproduced the temporal variability in PM2.5 with the highest PM2.5 values on polluted days (defined as days in which observed PM2.5 is greater than 75 μg m-3), but predictions of sulfate, nitrate, and ammonium were too low on days with the highest observed concentrations. Observational data indicate that the sulfur/nitric oxidation rates are strongly correlated with relative humidity during periods of peak PM2.5; however, the model failed to reproduce the highest PM2.5 concentrations due to missing heterogeneous reactions. As the parameterizations of those reactions is not well established yet, estimates of SO2-to-H2SO4 and NO2/NO3-to-HNO3 reaction rates that depend on relative humidity were applied which improved the simulation of sulfate, nitrate, and ammonium enhancement on polluted days in terms of both concentrations and partitioning among those species. Sensitivity simulations showed that the extremely high heterogeneous reaction rates and also higher emission rates than those reported in the emission inventory

Chen, Dan↗