High-Resolution Rapid Refresh Research Articles

Frost prediction using machine learning and deep neural network models.

This study describes accurate, computationally efficient models that can be implemented for practical use in predicting frost events for point-scale agricultural applications. Frost damage in agriculture is a costly burden to farmers and global food security alike. Timely prediction of frost events is important to reduce the cost of agricultural frost damage and traditional numerical weather forecasts are often inaccurate at the field-scale in complex terrain. In this paper, we developed machine learning (ML) algorithms for the prediction of such frost events near Alcalde, NM at the point-scale. ML algorithms investigated include deep neural network, convolution neural networks, and random forest models at lead-times of 6-48 h. Our results show promising accuracy (6-h prediction RMSE = 1.53-1.72°C) for use in frost and minimum temperature prediction applications. Seasonal differences in model predictions resulted in a slight negative bias during Spring and Summer months and a positive bias in Fall and Winter months. Additionally, we tested the model transferability by continuing training and testing using data from sensors at a nearby farm. We calculated the feature importance of the random forest models and were able to determine which parameters provided the models with the most useful information for predictions. We determined that soil temperature is a key parameter in longer term predictions (>24 h), while other temperature related parameters provide the majority of information for shorter term predictions. The model error compared favorable to previous ML based frost studies and outperformed the physically based High Resolution Rapid Refresh forecasting system making our ML-models attractive for deployment toward real-time monitoring of frost events and damage at commercial farming operations.

Operational water forecast ability of the HRRR-iSnobal combination: an evaluation to adapt into production environments

Abstract. Operational water-resource forecasters, such as the Colorado Basin River Forecast Center (CBRFC) in the Western United States, currently rely on historical records to calibrate the temperature-index models used for snowmelt runoff predictions. This data dependence is increasingly challenged, with global and regional climatological factors changing the seasonal snowpack dynamics in mountain watersheds. To evaluate and improve the CBRFC modeling options, this work ran the physically based snow energy balance iSnobal model, forced with outputs from the High-Resolution Rapid Refresh (HRRR) numerical weather prediction model across 4 years in a Colorado River Basin forecast region. Compared to in situ, remotely sensed, and the current operational CBRFC model data, the HRRR-iSnobal combination showed well-reconstructed snow depth patterns and magnitudes until peak accumulation. Once snowmelt set in, HRRR-iSnobal showed slower simulated snowmelt relative to observations, depleting snow on average up to 34 d later. The melting period is a critical component for water forecasting. Based on the results, there is a need for revised forcing data input preparation (shortwave radiation) required by iSnobal, which is a recommended future improvement to the model. Nevertheless, the presented performance and architecture make HRRR-iSnobal a promising combination for the CBRFC production needs, where there is a demonstrated change to the seasonal snow in the mountain ranges around the Colorado River Basin. The long-term goal is to introduce the HRRR-iSnobal combination in day-to-day CBRFC operations, and this work created the foundation to expand and evaluate larger CBRFC domains.

Short-Term Ensemble Streamflow Prediction Using Spatially Shifted QPF Informed by Displacement Errors

Abstract Errors associated with the location of precipitation in QPFs present challenges when used for hydrologic prediction, particularly in small watersheds. This work builds on a past study that systematically shifted QPFs prior to inputting them into a hydrologic model to generate streamflow ensembles. In the original study, which used static, predetermined shifting distances, flood detection improved, but false alarms increased due to large ensemble spread. The present research tests a more informed approach by randomly selecting shift directions and distances based on the distribution of displacement errors from a sample of QPFs. Precipitation forecasts were taken from the High-Resolution Rapid Refresh Ensemble (HRRRE), and streamflow predictions were generated using the Weather Research and Forecasting hydrological modeling system, version 5.1.1, in a National Water Model 2.0 configuration. A 63-member streamflow ensemble was generated using the 9 original HRRRE and 54 shifted HRRRE members. Two ensemble updating schemes were tested in which ensemble member weights were adjusted using precipitation location and QPF displacement present at convective initiation. The ensembles using QPF shifted based on climatological spatial errors showed higher probabilistic forecasting skill, while having comparable dichotomous forecasting skill to the original HRRRE ensemble. Other methods of selecting nine ensemble members from the full 63-member suite did not show significant improvement. Flood peak timing showed frequent errors, with average timing errors around five hours early. Larger watersheds tended to have better skill metric scores than smaller basins, with increased skill added by the shifting of QPF.

Evaluation of Model Summertime Boundary Layer Cloud Development over Complex Terrain in New York State

Abstract Weather forecasting over complex terrain with diverse land cover is challenging. Utilizing the high-resolution observations from New York State Mesonet (NYSM), we are able to evaluate the surface processes of the Weather Research Forecast (WRF) Model in a detailed, scale-dependent manner. In the study, possible impacts of land–atmosphere interaction on surface meteorology and boundary layer cloud development are investigated with different model resolutions, land surface models (LSMs), and planetary boundary layer (PBL) physical parameterizations. The High-Resolution Rapid Refresh, version 3 (HRRR), forecasting model is used as a reference for the sensitivity evaluation. Results show that over complex terrain, the high-resolution simulations (1 km × 60 vertical levels) generally perform better compared to low-resolution (3 km × 50 levels) in both surface meteorology and cloud fields. LSMs play a more important role in surface meteorology compared to PBL schemes. The NoahMP land surface model exhibits daytime warmer and drier biases compared to the Rapid Update Cycle (RUC) due to better prediction of the Bowen ratio in RUC. The PBL schemes would affect the convective strength in the boundary layer. The Shin–Hong (SH) scale-aware scheme tends to produce the strongest convective strength in the PBL, while the ACM2 PBL scheme rarely resolved convection even at 1-km resolution. By considering the radiation effect of subgrid-scale (SGS) clouds, the Mellor–Yamada–Nakanishi–Niino eddy diffusivity mass flux (MYNN-EDMF) predicted the highest cloud coverage and lowest surface solar radiation bias. The configuration of SGS clouds in MYNN-EDMF would not only significantly reduce shortwave radiation bias, but also affect the convection behaviors through land surface–cloud–radiation interaction.

Single-blind determination of methane detection limits and quantification accuracy using aircraft-based LiDAR

Methane detection limits, emission rate quantification accuracy, and potential cross-species interference are assessed for Bridger Photonics’ Gas Mapping LiDAR (GML) system utilizing data collected during laboratory testing and single-blind controlled release testing. Laboratory testing identified no significant interference in the path-integrated methane measurement from the gas species tested (ethylene, ethane, propane, n-butane, i-butane, and carbon dioxide). The controlled release study, comprised of 650 individual measurement passes, represents the largest dataset collected to date to characterize GML with respect to point-source emissions. Binomial regression is utilized to create detection curves illustrating the likelihood of detecting an emission of a given size under different wind conditions and for different flight altitudes. Wind-normalized methane detection limits (90% detection rate) of 0.25 (kg/h)/(m/s) and 0.41 (kg/h)/(m/s) are observed at a flight altitude of 500 feet and 675 feet above ground level, respectively. Quantification accuracy is also assessed for emissions ranging from 0.15 to 1,400 kg/h. When emission rate estimates were generated using wind from high-resolution rapid refresh (HRRR) model (the primary wind source that Bridger uses for their commercial operations), linear regression indicates bias of 8.1% (R2 = 0.89). For 95% of controlled releases above Bridger’s stated production-sector detection sensitivity (3 kg/h with 90% probability of detection), the accuracy of individual emission rate estimates produced using HRRR wind ranged from −64.1% to +87.0%. Across all controlled releases, 38.1% of estimates had error within ±20%, and 87.3% of measurements were within a factor of two (−50% to +100% error). At low wind speed (less than 2 m/s) and low emission rates (less than 3 kg/h), emission estimates are biased high, however when removed do not impact the regression significantly. The aggregate quantification error including all detected emission events was +8.2% using the HRRR wind source. The resulting detection curves and quantification accuracy illustrate important implications that must be considered when using measurements from GML or other remote emission measurement techniques to inform or validate inventory models or to audit reported emission levels from oil and gas systems.

Short-Term Weather Forecast Skill of Artificial Neural Networks

Abstract We evaluate the short-term weather forecast performance of three flavors of artificial neural networks (NNs): feed forward back propagation, radial basis function, and generalized regression. To prepare the application of the NNs to an operational setting, we tune NN hyperparameters using over two years of historical data. Five objective guidance products serve as predictors to the NNs: North American Mesoscale and Global Forecast System model output statistics (MOS) forecasts, the High-Resolution Rapid Refresh (HRRR) model, National Weather Service forecasts, and the National Blend of Models product. We independently test NN performance using 96 real-time forecasts of temperature, wind, and precipitation across 11 U.S. cities made during the WxChallenge, a weather forecasting competition. We demonstrate that all NNs significantly improve short-range weather forecasts relative to the traditional objective guidance aids used to train the networks. For example, 1-day maximum and minimum temperature forecast error is 20%–30% lower than MOS. However, NN improvement over multiple linear regression for short-term forecasts is not significant. We suggest this may be attributed to the small number of training samples, the operational nature of the experiment, and the short forecast lead times. Regardless, our results are consistent with previous work suggesting that applying NNs to model forecasts can have a positive impact on operational forecast skill and will become valuable tools when integrated into the forecast enterprise. Significance Statement We used approximately two years of historical weather data and objective forecasts for a number of cities to tune a series of artificial neural networks (NNs) to forecast 1-day values of maximum and minimum temperature, maximum sustained wind speed, and quantitative precipitation. We compare forecast error against common objective guidance and multiple linear regression. We found that the NNs exhibit about 25% lower error than common objective guidance for temperature forecasting and 50% lower error for wind speed. Our results suggest that NNs will be a valuable contributor to improving weather forecast skill when adopted into the existing forecast enterprise.

Application of geostationary satellite and high-resolution meteorology data in estimating hourly PM2.5 levels during the Camp Fire episode in California

BACKGROUND: Wildland fire smoke contains large amounts of PM2.5 that can traverse tens to hundreds of kilometers, resulting in significant deterioration of air quality and excess mortality and morbidity in downwind regions. Estimating PM2.5 levels while considering the impact of wildfire smoke has been challenging due to the lack of ground monitoring coverage near the smoke plumes. AIM: We aim to estimate total PM2.5 concentration during the Camp Fire episode, the deadliest wildland fire in California history. METHODS: Our random forest (RF) model combines calibrated low-cost sensor data (PurpleAir) with regulatory monitor measurements (Air Quality System, AQS) to bolster ground observations, Geostationary Operational Environmental Satellite-16 (GOES-16)’s high temporal resolution to achieve hourly predictions, and oversampling techniques (Synthetic Minority Oversampling Technique, SMOTE) to reduce model underestimation at high PM2.5 levels. In addition, meteorological fields at 3 km resolution from the High-Resolution Rapid Refresh model and land use variables were also included in the model. RESULTS: Our AQS-only model achieved an out of bag (OOB) R2 (RMSE) of 0.84 (12.00 μg/m3) and spatial and temporal cross-validation (CV) R2 (RMSE) of 0.74 (16.28 μg/m3) and 0.73 (16.58 μg/m3), respectively. Our AQS + Weighted PurpleAir Model achieved OOB R2 (RMSE) of 0.86 (9.52 μg/m3) and spatial and temporal CV R2 (RMSE) of 0.75 (14.93 μg/m3) and 0.79 (11.89 μg/m3), respectively. Our AQS + Weighted PurpleAir + SMOTE Model achieved OOB R2 (RMSE) of 0.92 (10.44 μg/m3) and spatial and temporal CV R2 (RMSE) of 0.84 (12.36 μg/m3) and 0.85 (14.88 μg/m3), respectively. CONCLUSIONS: Hourly predictions from our model may aid in epidemiological investigations of intense and acute exposure to PM2.5 during the Camp Fire episode. KEYWORDS: PM2.5, GOES16, Wildland fire, Remote sensing, AOD, SMOTE, Weighted Random Forest

Assessment of the data assimilation framework for the Rapid Refresh Forecast System v0.1 and impacts on forecasts of a convective storm case study

Abstract. The Rapid Refresh Forecast System (RRFS) is currently under development and aims to replace the National Centers for Environmental Prediction (NCEP) operational suite of regional- and convective-scale modeling systems in the next upgrade. In order to achieve skillful forecasts comparable to the current operational suite, each component of the RRFS needs to be configured through exhaustive testing and evaluation. The current data assimilation component uses the hybrid three-dimensional ensemble–variational data assimilation (3DEnVar) algorithm in the Gridpoint Statistical Interpolation (GSI) system. In this study, various data assimilation algorithms and configurations in GSI are assessed for their impacts on RRFS analyses and forecasts of a squall line over Oklahoma on 4 May 2020. A domain of 3 km horizontal grid spacing is configured, and hourly update cycles are performed using initial and lateral boundary conditions from the 3 km grid High-Resolution Rapid Refresh (HRRR). Results show that a baseline RRFS run is able to represent the observed convection, although with stronger cells and large location errors. With data assimilation, these errors are reduced, especially in the 4 and 6 h forecasts using 75 % of the ensemble background error covariance (BEC) and 25 % of the static BEC with the supersaturation removal function activated in GSI. Decreasing the vertical ensemble localization radius from 3 layers to 1 layer in the first 10 layers of the hybrid analysis results in overall less skillful forecasts. Convection is greatly improved when using planetary boundary layer pseudo-observations, especially at 4 h forecast, and the bias of the 2 h forecast of temperature is reduced below 800 hPa. Lighter hourly accumulated precipitation is predicted better when using 100 % ensemble BEC in the first 4 h forecast, but heavier hourly accumulated precipitation is better predicted with 75 % ensemble BEC. Our results provide insight into the current capabilities of the RRFS data assimilation system and identify configurations that should be considered as candidates for the first version of RRFS.

The High-Resolution Rapid Refresh (HRRR): An Hourly Updating Convection-Allowing Forecast Model. Part I: Motivation and System Description

Abstract The High-Resolution Rapid Refresh (HRRR) is a convection-allowing implementation of the Advanced Research version of the Weather Research and Forecasting (WRF-ARW) Model with hourly data assimilation that covers the conterminous United States and Alaska and runs in real time at the NOAA/National Centers for Environmental Prediction (NCEP). Implemented operationally at NOAA/NCEP in 2014, the HRRR features 3-km horizontal grid spacing and frequent forecasts (hourly for CONUS and 3-hourly for Alaska). HRRR initialization is designed for optimal short-range forecast skill with a particular focus on the evolution of precipitating systems. Key components of the initialization are radar-reflectivity data assimilation, hybrid ensemble-variational assimilation of conventional weather observations, and a cloud analysis to initialize stratiform cloud layers. From this initial state, HRRR forecasts are produced out to 18 h every hour, and out to 48 h every 6 h, with boundary conditions provided by the Rapid Refresh system. Between 2014 and 2020, HRRR development was focused on reducing model bias errors and improving forecast realism and accuracy. Improved representation of the planetary boundary layer, subgrid-scale clouds, and land surface contributed extensively to overall HRRR improvements. The final version of the HRRR (HRRRv4), implemented in late 2020, also features hybrid data assimilation using flow-dependent covariances from a 3-km, 36-member ensemble (“HRRRDAS”) with explicit convective storms. HRRRv4 also includes prediction of wildfire smoke plumes. The HRRR provides a baseline capability for evaluating NOAA’s next-generation Rapid Refresh Forecast System, now under development. Significance Statement NOAA’s operational hourly updating, convection-allowing model, the High-Resolution Rapid Refresh (HRRR), is a key tool for short-range weather forecasting and situational awareness. Improvements in assimilation of weather observations, as well as in physics parameterizations, have led to improvements in simulated radar reflectivity and quantitative precipitation forecasts since the initial implementation of HRRR in September 2014. Other targeted development has focused on improved representation of the diurnal cycle of the planetary boundary layer, resulting in improved near-surface temperature and humidity forecasts. Additional physics and data assimilation changes have led to improved treatment of the development and erosion of low-level clouds, including subgrid-scale clouds. The final version of HRRR features storm-scale ensemble data assimilation and explicit prediction of wildfire smoke plumes.

The High-Resolution Rapid Refresh (HRRR): An Hourly Updating Convection-Allowing Forecast Model. Part II: Forecast Performance

Abstract The High-Resolution Rapid Refresh (HRRR) is a convection-allowing implementation of the Advanced Research version of the Weather Research and Forecast (WRF-ARW) Model that covers the conterminous United States and Alaska and runs hourly (for CONUS; every 3 h for Alaska) in real time at the National Centers for Environmental Prediction. The high-resolution forecasts support a variety of user applications including aviation, renewable energy, and prediction of many forms of severe weather. In this second of two articles, forecast performance is documented for a wide variety of forecast variables and across HRRR versions. HRRR performance varies across geographical domain, season, and time of day depending on both prevalence of particular meteorological phenomena and the availability of both conventional and nonconventional observations. Station-based verification of surface weather forecasts (2-m temperature and dewpoint temperature, 10-m winds, visibility, and cloud ceiling) highlights the ability of the HRRR to represent daily planetary boundary layer evolution and the development of convective and stratiform cloud systems, while gridded verification of simulated composite radar reflectivity and quantitative precipitation forecasts reveals HRRR predictive skill for summer and winter precipitation systems. Significant improvements in performance for specific forecast problems are documented for the upgrade versions of the HRRR (HRRRv2, v3, and v4) implemented in 2016, 2018, and 2020, respectively. Development of the HRRR model data assimilation and physics paves the way for future progress with operational convective-scale modeling. Significance Statement NOAA’s operational hourly updating convection-allowing model, the High-Resolution Rapid Refresh (HRRR), is a key tool for short-range weather forecasting and situational awareness. Improvements in assimilation of weather observations, as well as in physics parameterizations, has led to improvements in simulated radar reflectivity and quantitative precipitation forecasts since the initial implementation of HRRR in September 2014. Other targeted development has focused on improved representation of the diurnal cycle of the planetary boundary layer, resulting in improved near-surface temperature and humidity forecasts. Additional physics and data assimilation changes have led to improved treatment of the development and erosion of low-level clouds, including subgrid-scale clouds. The final version of HRRR features storm-scale ensemble data assimilation and explicit prediction of wildfire smoke plumes.

Radar Reflectivity–Based Model Initialization Using Specified Latent Heating (Radar-LHI) within a Diabatic Digital Filter or Pre-Forecast Integration

Abstract A technique for model initialization using three-dimensional radar reflectivity data has been developed and applied within the NOAA 13-km Rapid Refresh (RAP) and 3-km High-Resolution Rapid Refresh (HRRR) regional forecast systems. This technique enabled the first assimilation of radar reflectivity data for operational NOAA forecast models, critical especially for more accurate short-range prediction of convective storms. For the RAP, the technique uses a diabatic digital filter initialization (DFI) procedure originally deployed to control initial inertial gravity wave noise. Within the forward-model integration portion of diabatic DFI, temperature tendencies obtained from the model cloud/precipitation processes are replaced by specified latent heating–based temperature tendencies derived from the three-dimensional radar reflectivity data, where available. To further refine initial conditions for the convection-allowing HRRR model, a similar procedure is used in the HRRR, but without DFI. Both of these procedures, together called the “Radar-LHI” (latent heating initialization) technique, have been essential for initialization of ongoing precipitation systems, especially convective systems, within all NOAA operational versions of the 13-km RAP and 3-km HRRR models extending through the latest implementation upgrade at NCEP in 2020. Application of the latent heat–derived temperature tendency induces a vertical circulation with low-level convergence and upper-level divergence in precipitation systems. Retrospective tests of the Radar-LHI technique show significant improvement in short-range (0–6 h) precipitation system forecasts, as revealed by reflectivity verification scores. Results presented document the impact on HRRR reflectivity forecasts of the radar reflectivity initialization technique applied to the RAP alone, HRRR alone, and both the RAP and HRRR. Significance Statement The large forecast uncertainty of convective situations, even at short lead times, coupled with the hazardous weather they produce, makes convective storm prediction one of the most significant short-range forecast challenges confronting the operational numerical weather prediction community. Prediction of heavy precipitation events also requires accurate initialization of precipitation systems. An innovative assimilation technique using radar reflectivity data to initialize NOAA operational weather prediction models is described. This technique, which uses latent heating specified from radar reflectivity (and can accommodate lightning data and other convection/precipitation indicators), was first implemented in 2009 at NOAA/NCEP and continues to be used in 2022 in the NCEP-operational RAP and HRRR models, making it a backbone of the NOAA rapidly updated numerical weather prediction capability.

Winter and Wildfire Season Optical Characterization of Black and Brown Carbon in the El Paso-Ciudad Juárez Airshed

Black (EBC) and Brown (BrC) Carbon are ubiquitous constituents of atmospheric particulate matter that affect people’s health, disrupt ecosystems, and modulate local and global climate. Tracking the local deposition and sources of these aerosol particles is essential to better understanding their multidimensional environmental impact. The main goal of the current study is to measure the absorption coefficient (Babs) of particles within the Planetary Boundary Layer (PBL) of the El Paso (US)–Ciudad Juárez (Mexico) airshed and assess the contribution of black and brown carbon particles to the optical absorption. Measurements were taken during a summer, wildfire, and winter season to evaluate the optical properties of BC and non-volatile BrC. The winter season presented a variation from the background Babs in the late evening hours (3:00 PM to midnight) due to an increase in biomass burning driven by lower temperatures. The wildfire season presents the greatest variation in the Babs from the background absorption due to EBC- and BrC-rich smoke plumes arriving at this region from the US West seasonal wildfires. It was found that the international bridges’ vehicular traffic, waiting time to cross back and forth between both cities, added to other local anthropogenic activities, such as brick kiln emissions in Ciudad Juarez, have created a background of air pollution in this region. These pollutants include carbon monoxide, sulfur dioxide, nitrogen and nitric oxides, coarse and fine particulate matter dominated by BC and BrC. The absorption coefficients due to EBC and BrC of this background constitute what we have called a baseline EBC and BrC. Aided by two photoacoustic Extinctiometers (PAX), operating at 405 nm and 870 nm wavelengths, connected to a 340 °C thermal denuder to remove volatile organics, the optical properties were documented and evaluated to identify the impact of long-range transported emissions from western wildfires. The Single Scattering Albedo and the Absorption Ångstrom exponent were calculated for the winter and summer season. The Angstrom exponent showed a decrease during the wildfire events due to the aging process. The High-Resolution Rapid Refresh Smoke model, HRRR, and the Hybrid Single-Particle Lagrangian Integrated Trajectory model, HYSPLIT, were used to estimate the sources of the particles. In addition, a Vaisala Ceilometer was employed to study the vertical profile of particulate matter within the planetary boundary layer.

Multiscale and multi event evaluation of short-range real-time flood forecasting in large metropolitan areas

Many large metropolitan areas are especially susceptible to floods generated by heavy, short-duration rainfall. The high density of people, buildings and infrastructure in these areas underline the importance of developing flood resilient cities and communities. An accurate real-time flood forecasting system can support decision making for launching preparedness and response actions in short-range (hours to days), and assist mitigating the disturbances caused by floods. This study explores the short-range predictive capability of a real-time flood forecast system by coupling the High-Resolution Rapid Refresh (HRRR) meteorological forecasted variables with a fully distributed hydrological model (WRF-Hydro). We provide a comprehensive analysis of short-range (36 h) forecasts for 19flood events generated by heavy rainfall in small urban and suburban watersheds (<200 km2) in the National Capital Region of the United States. Flood forecast performance are then assessed using different metrics based on observed data from U.S. Geological Survey stream gauges and reanalysis simulations using the StageIV quantitative precipitation estimates. Results show that despite the high variability of the HRRR cycle-to-cycle precipitation forecasts, the real-time flood forecast system can produce skillful forecasts with similar overall performance as the reanalysis simulations, achieving 65 % of flood detection rate. This variability had more impact on forecast skill in smaller subbasins, and flood forecasts were more consistent for longer duration and larger spatial extent rainfall. Hourly flood forecasts performed well even in smaller watersheds, correctly detecting flood occurrence with up to 34 h lead time and resulting in low peak flow magnitude and timing errors.

Development and evaluation of the Aerosol Forecast Member in the National Center for Environment Prediction (NCEP)'s Global Ensemble Forecast System (GEFS-Aerosols v1)

Abstract. The National Oceanic and Atmospheric Administration (NOAA)'s National Weather Service (NWS) is on its way to deploying various operational prediction applications using the Unified Forecast System (https://ufscommunity.org/, last access: 18 June 2022), a community-based coupled, comprehensive Earth modeling system. An aerosol model component developed in collaboration between the Global Systems Laboratory, Chemical Science Laboratory, Air Resources Laboratory, and Environmental Modeling Center (GSL, CSL, ARL, EMC) was coupled online with the FV3 Global Forecast System (FV3GFS) using the National Unified Operational Prediction Capability (NUOPC)-based NOAA Environmental Modeling System (NEMS) software framework. This aerosol prediction system replaced the NEMS GFS Aerosol Component version 2 (NGACv2) system in the National Center for Environment Prediction (NCEP) production suite in September 2020 as one of the ensemble members of the Global Ensemble Forecast System (GEFS), dubbed GEFS-Aerosols v1. The aerosol component of atmospheric composition in the GEFS is based on the Weather Research and Forecasting model coupled with Chemistry (WRF-Chem). GEFS-Aerosols includes bulk modules from the Goddard Chemistry Aerosol Radiation and Transport model (GOCART). Additionally, the biomass burning plume rise module from High-Resolution Rapid Refresh (HRRR)-Smoke based on WRF-Chem was implemented. The GOCART dust scheme was replaced by the FENGSHA dust scheme (developed by ARL). The Blended Global Biomass Burning Emissions Product (GBBEPx version 3) provides biomass burning emission and fire radiative power (FRP) data. The global anthropogenic emission inventories are derived from the Community Emissions Data System (CEDS). All sub-grid-scale transport and deposition are handled inside the atmospheric physics routines, which required consistent implementation of positive definite tracer transport and wet scavenging in the physics parameterizations used by the NCEP's operational FV3GFS. This paper describes the details of GEFS-Aerosols model development and evaluation of real-time and retrospective runs using different observations from in situ measurement and satellite and aircraft data. GEFS-Aerosols predictions demonstrate substantial improvements for both composition and variability of aerosol distributions over those from the former operational NGACv2 system with the fundamental updates (e.g., dust and fire emission) in the atmospheric and chemical transport model.

An Evaluation of Surface Wind and Gust Forecasts from the High-Resolution Rapid Refresh Model

Abstract We utilized high temporal resolution, near-surface observations of sustained winds and gusts from two networks, the primarily airport-based Automated Surface Observing System (ASOS) and the New York State Mesonet (NYSM), to evaluate forecasts from the operational High-Resolution Rapid Refresh (HRRR) model, versions 3 and 4. Consistent with past studies, we showed the model has a high degree of skill in reproducing the diurnal variation of network-averaged wind speed of ASOS stations, but also revealed several areas where improvements could be made. Forecasts were found to be underdispersive, deficient in both temporal and spatial variability, with significant errors occurring during local nighttime hours in all regions and in forested environments for all hours of the day. This explained why the model overpredicted the network-averaged wind in the NYSM because much of that network’s stations are in forested areas. A simple gust parameterization was shown not only to have skill in predicting gusts in both networks but also to mitigate systemic biases found in the sustained wind forecasts. Significance Statement Many users depend on forecasts from operational models and need to know their strengths, weaknesses, and limitations. We examined generally high-quality near-surface observations of sustained winds and gusts from the nationwide Automated Surface Observing System (ASOS) and the New York State Mesonet (NYSM) and used them to evaluate forecasts from the previous (version 3) and current (version 4) operational High-Resolution Rapid Refresh (HRRR) model for a selected month. Evidence indicated that the wind forecasts are excellent yet imperfect and areas for further improvement remain. In particular, we showed there is a high degree of skill in representing the diurnal variation of sustained wind at ASOS stations but insufficient spatial and temporal forecast variability and overprediction at night everywhere, in forested areas at all times of day, and at NYSM sites in particular, which are more likely to be sited in the forest. Gusts are subgrid even at the fine grid spacing of the HRRR (3 km) and thus must be parameterized. Our simple gust algorithm corrected for some of these systemic biases, resulting in very good predictions of the maximum hourly gust.

A Preliminary Examination of Warm Season Precipitation Displacement Errors in the Upper Midwest in the HRRRE and HREF Ensembles

Abstract Increased operational use of convection-allowing models and ensembles offers substantial improvements for some aspects of convective weather forecasting; however, errors in quantitative precipitation forecasts (QPFs) from these models, especially those related to incorrect placement of heavy rainfall systems, limit their usefulness as an input into hydrological models. To improve understanding of QPF location errors, this study quantifies the displacement errors for the centroids of both 0–18-h accumulated rainfall and rainfall in the first hour after initiation of precipitation systems in both the High-Resolution Rapid Refresh Ensemble (HRRRE) and the High-Resolution Ensemble Forecast (HREF) for 30 events in the 2018 warm season. Ensemble member QPFs are compared to quantitative precipitation estimates (QPEs) obtained from the North Central River Forecast Center (NCRFC). HRRRE is found to have less spread in centroid locations than HREF, and both HRRRE and HREF 0–18-h QPF accumulations have less spread than the 1-h QPF accumulation when the precipitation event initiates. Furthermore, QPF centroids are most often displaced to the west in HRRRE for both 0–18-h QPF accumulation and the 1-h QPF accumulation when the precipitation event initiates. The 0–18-h QPF accumulation displacement errors can be reduced when adjustments are made to the forecasted position based upon displacement errors present in the first hour of precipitation, but only when the adjustments are a function of the intercardinal quadrant in which the initial hour QPF centroid was displaced.

Exploring Hourly Updating Probabilistic Guidance in the 2021 Spring Forecasting Experiment with Objective and Subjective Verification

Abstract This study explored how forecasters can best use the two main forms of operational convection-allowing model guidance: the High-Resolution Ensemble Forecast (HREF) system and the hourly High-Resolution Rapid Refresh (HRRR). The former represents a wider range of possible outcomes, but the latter updates much more frequently and incorporates newer observations. HREF and time-lagged High-Resolution Rapid Refresh (HRRR-TL) probabilistic forecasts of reflectivity and updraft helicity, as well as two methods of combining HREF and HRRR into hourly updating blended guidance, were evaluated for the 2021 Spring Forecasting Experiment (SFE) period. In both objective skill and the subjective ratings of SFE participants, the 1200 UTC HREF proved difficult to outperform over this sample of events, even when incorporating HRRR initializations as late as 1800 UTC. It was usually better to use either of the experimental blending techniques than to simply discard the older HREF in favor of newer HRRR solutions. The greater model diversity and dispersion of solutions within the HREF is likely primarily responsible for this result. A possible bias in diurnal convection initiation timing and coverage in the newly upgraded HRRRv4 was also investigated, including on subdomains targeted to weakly forced diurnal initiation, and was found to have little or no systematic effect on HRRRv4’s operational utility.

Estimating the economic impacts of improved wind speed forecasts in the United States electricity sector

Each year the U.S. government makes significant investments in improving weather forecast models. In this paper, we use a multidisciplinary approach to examine how utilities can benefit from improved wind-speed forecasts to more efficiently use wind-generated electricity and subsequently increase economic activity. Specifically, we examine how improvements to the National Oceanic and Atmospheric Administration's high-resolution rapid refresh model (HRRR) wind forecasts can provide (1) cost savings for utilities and (2) increase in real household income. To do so, we compare 12-h-ahead wind forecasts with real-time observations for two HRRR model transitions (i.e., when one model is operational, the other is being tested). We compare estimates of actual and predicted wind power under the publicly available and developmental models, with reduced forecast errors allowing for better utility decision-making and lower production costs. We then translate potential cost savings into electricity price changes, which are entered as exogenous shocks to eight regional computable general equilibrium models constructed for the U.S. Overall, we find that households would have seen a potential $60 million increase in real income for our sample (13% of all contiguous U.S. land-based turbine capacity), which had the updated HRRR models been in place during the two transition periods; applying our estimated savings for the sample of turbines to the entire array of turbines shows a potential real household income increase in approximately $384 million during these time frames.

Improving station-based ensemble surface meteorological analyses using numerical weather prediction: A case study of the Oroville Dam crisis precipitation event

Abstract Surface meteorological analyses serve a wide range of research and applications, including forcing inputs for hydrological and ecological models, climate analysis, and resource and emergency management. Quantifying uncertainty in such analyses would extend their utility for probabilistic hydrologic prediction and climate risk applications. With this motivation, we enhance and evaluate an approach for generating ensemble analyses of precipitation and temperature through the fusion of station observations, terrain information, and numerical weather prediction simulations of surface climate fields. In particular, we expand a spatial regression in which static terrain attributes serve as predictors for spatially distributed 1/16th degree daily surface precipitation and temperature by including forecast outputs from the High-Resolution Rapid Refresh (HRRR) numerical weather prediction model as additional predictors. We demonstrate the approach for a case study domain of California, focusing on the meteorological conditions leading to the 2017 flood and spillway failure event at Lake Oroville. The approach extends the spatial regression capability of the Gridded Meteorological Ensemble Tool (GMET) and also adds cross-validation to the uncertainty estimation component, enabling the use of predictive rather than calibration uncertainty. In evaluation against out-of-sample station observations, the HRRR-based predictors alone are found to be skillful for the study setting, leading to overall improvements in the enhanced GMET meteorological analyses. The methodology and associated tool represent a promising method for generating meteorological surface analyses for both research-oriented and operational applications, as well as a general strategy for merging in situ and gridded observations.

Using Cloud Computing to Analyze Model Output Archived in Zarr Format

Abstract Numerical weather prediction centers rely on the Gridded Binary Second Edition (GRIB2) file format to efficiently compress and disseminate model output as two-dimensional grids. User processing time and storage requirements are high if many GRIB2 files with size O(100 MB, where B = bytes) need to be accessed routinely. We illustrate one approach to overcome such bottlenecks by reformatting GRIB2 model output from the High-Resolution Rapid Refresh (HRRR) model of the National Centers for Environmental Prediction to a cloud-optimized storage type, Zarr. Archives of the original HRRR GRIB2 files and the resulting Zarr stores on Amazon Web Services (AWS) Simple Storage Service (S3) are available publicly through the Amazon Sustainability Data Initiative. Every hour, the HRRR model produces 18- or 48-hourly GRIB2 surface forecast files of size O(100 MB). To simplify access to the grids in the surface files, we reorganize the HRRR model output for each variable and vertical level into Zarr stores of size O(1 MB), with chunks O(10 kB) containing all forecast lead times for 150 × 150 gridpoint subdomains. Open-source libraries provide efficient access to the compressed Zarr stores using cloud or local computing resources. The HRRR-Zarr approach is illustrated for common applications of sensible weather parameters, including real-time alerts for high-impact situations and retrospective access to output from hundreds to thousands of model runs. For example, time series of surface pressure forecast grids can be accessed using AWS cloud computing resources approximately 40 times as fast from the HRRR-Zarr store as from the HRRR-GRIB2 archive. Significance Statement The rapid evolution of computing power and data storage have enabled numerical weather prediction forecasts to be generated faster and with more detail than ever before. The increased temporal and spatial resolution of forecast model output can force end users with finite memory and storage capabilities to make pragmatic decisions about which data to retrieve, archive, and process for their applications. We illustrate an approach to alleviate this access bottleneck for common weather analysis and forecasting applications by using the Amazon Web Services (AWS) Simple Storage Service (S3) to store output from the High-Resolution Rapid Refresh (HRRR) model in Zarr format. Zarr is a relatively new data storage format that is flexible, compressible, and designed to be accessed with open-source software either using cloud or local computing resources. The HRRR-Zarr dataset is publicly available as part of the AWS Sustainability Data Initiative.