The HRRR is a NOAA real-time 3-km resolution, hourly updated, cloud-resolving, convection-allowing atmospheric model, initialized by 3km grids with 3km radar assimilation. Radar data is assimilated in the HRRR every 15 min over a 1-h period adding further detail to that provided by the hourly data assimilation from the 13km radar-enhanced Rapid Refresh.

The HRRR ZARR formatted data was originally generated by the University of Utah under a grant provided by NOAA. They are are continuing to publish ZARR versions of HRRR data. For information about data in the s3://hrrrzarr/ please contact atmos-mesowest@lists.utah.edu.

To address the need for regularly updated wind resource data, NREL has processed the High-Resolution Rapid Refresh (HRRR) outputs for use in grid integration modeling. The HRRR is an hourly-updated operational forecast product produced by the National Oceanic and Atmospheric Administration (NOAA) (Dowell et al., 2022). Several barriers have prevented the HRRR's widespread proliferation in the wind energy industry: missing timesteps (prior to 2019), challenging file format for wind energy analysis, limited vertical height resolution, and negative bias versus legacy WIND Toolkit data (2007-2013). NREL has applied re-gridding, interpolation, and bias-correction to the native HRRR data to overcome these limitations. This results in the now-publicly-available bias corrected and interpolated HRRR (BC-HRRR) dataset for weather years 2015 to 2023. Bias correction is necessary for wind resource consistency across weather years to be used simultaneously in planning-focused grid integration studies alongside the original WIND Toolkit data. We show that quantile mapping with the WIND Toolkit as a historical baseline is an effective method for bias correcting the interpolated HRRR data: the BC-HRRR has reduced mean bias versus comparable gridded wind resource datasets (+0.12 m/s versus Vortex) and has very low mean bias versus ground measurement stations (+0.01 m/s) (Buster et al., 2024). BC-HRRR's consistency with the legacy WIND Toolkit allows NREL to extend grid integration analysis to 15+ weather years of wind data with low-overhead extensibility to future years as they are made available by NOAA. As with historical datasets like the WIND Toolkit, BC-HRRR is intended for use in grid integration modeling (e.g., capacity expansion, production cost, and resource adequacy modeling) both independently and alongside the legacy WIND Toolkit.

The High Resolution Rapid Refresh Model archive at the University of Utah

The Rapid Refresh (RAP) numerical weather model took the place of the Rapid Update Cycle (RUC) on May 1, 2012. Run by the National Centers for Environmental Prediction (NCEP), RAP runs with two versions. The first generates weather data on a 13-km (8-mile) resolution horizontal grid and the second, the High-Resolution Rapid Refresh (HRRR), generates data down to a 3-km (2-mile) resolution grid for smaller regions of interest. RAP forecasts are generated every hour with forecast lengths going out 18 hours with a 1 hour temporal resolution. Multiple data sources go into the generation of RAP forecasts: commercial aircraft weather data, balloon data, radar data, surface observations, and satellite data. This dataset contains a 13 km horizontal resolution Lambert Conformal grid covering the Continental United States (CONUS) domain.

The Rapid Refresh (RAP) numerical weather model took the place of the Rapid Update Cycle (RUC) on May 1, 2012. Run by the National Centers for Environmental Prediction (NCEP), RAP runs with two versions. The first generates weather data on a 13-km (8-mile) resolution horizontal grid and the second, the High-Resolution Rapid Refresh (HRRR), generates data down to a 3-km (2-mile) resolution grid for smaller regions of interest. RAP forecasts are generated every hour with forecast lengths going out 18 hours with a 1 hour temporal resolution. Multiple data sources go into the generation of RAP forecasts: commercial aircraft weather data, balloon data, radar data, surface observations, and satellite data. This dataset contains a 20 km horizontal resolution Lambert Conformal grid covering the Continental United States (CONUS) domain.

The National Water Model (NWM) is a water resources model that simulates and forecasts water budget variables, including snowpack, evapotranspiration, soil moisture and streamflow, over the entire continental United States (CONUS). The model, launched in August 2016, is designed to improve the ability of NOAA to meet the needs of its stakeholders (forecasters, emergency managers, reservoir operators, first responders, recreationists, farmers, barge operators, and ecosystem and floodplain managers) by providing expanded accuracy, detail, and frequency of water information. It is operated by NOAA’s Office of Water Prediction. This bucket contains a four-week rollover of the Short Range Forecast model output and the corresponding forcing data for the model. The model is forced with meteorological data from the High Resolution Rapid Refresh (HRRR) and the Rapid Refresh (RAP) models. The Short Range Forecast configuration cycles hourly and produces hourly deterministic forecasts of streamflow and hydrologic states out to 18 hours.

This dataset collects together collated and interpolated data required to run schism and are formatted for use in the model. SCHISM requires atmopheric data files in netcdf format as described here: http://ccrm.vims.edu/w/index.php/Atmospheric_forcing. These include ordered files with _air (wind, spec. humidity, pressure etc) _rad (long and short wave radiatin) and _prec (precipitation) in the name. These are often implemented using symbolic links to the files here. The data used for Bay-Delta SCHISM vary by period. For simulations centered before 2020, we use the _air data constructed by DWR based on spatial interpolation from approximately 71 public and private stations around the Bay-Delta. The _prec and _rad data are obtained from North America Regional Reanalysis data: https://psl.noaa.gov/data/gridded/data.narr.html. For simulations centered after 2020, we use the High Resolution Rapid Refresh (https://rapidrefresh.noaa.gov/hrrr/) data for all three variables. Around 2020 both options are viable and we choose whichever provides continuity across the period of study.

This repository contains the data for the paper "Local Off-Grid Weather Forecasting with Multi-Modal Earth Observation Data".

The paper presents a novel multi-modal deep learning method that downscales gridded weather forecasts, such as ERA5 and HRRR, to provide accurate off-grid predictions. The model leverages both gridded data and local weather station observations from MADIS to make predictions that reflect both large-scale atmospheric dynamics and local weather patterns.

The model is evaluated on a surface wind prediction task and shows significant improvement over baseline methods, including ERA5 interpolation, HRRR re-analysis and HRRR forecast and a multi-layer perceptron.

Use the following citation when these data or model are used:
> Yang, Q.; Giezendanner, J.; Civitarese, D. S.; Jakubik, J.;
,Schmitt E.; Chandra, A.; Vila, J.; Hohl, D.; Hill, C.; Watson, C.; Wang, S.; Local Off-Grid Weather Forecasting with Multi-Modal Earth Observation Data. arXiv, October 2024. https://doi.org/10.48550/arXiv.2410.12938

The following data is available:
- Shapefile of the Northeastern United States (NE-US, extracted from NWS)
- Shapefile containing the location and number of observations (2019-2023) of the MADIS stations in NE-US
- Processed hourly averaged MADIS data for the NE-US (2019-2023)
- ERA5 data for the NE-US (2019-2023), gridded and interpolated

For MADIS and ERA5, the following variables are available:
- u and v component of wind vector at 10 meters above ground
- temperature at 2 meters above ground
- dewpoint at 2 meters above ground

Smoke from agricultural fires is a potentially important source of fine particulate matter (PM2.5) in the US. Sugarcane is burned in Florida to facilitate the harvesting process, with the majority of these fires occurring in the Everglades Agricultural Area (EAA), where there is only one regulatory air quality monitor. During the 2022–2023 sugarcane burning season (October–May), we used public low-cost PurpleAir sensors, regulatory monitors, and 29 PurpleAir sensors deployed for this study to quantify PM2.5 from agricultural fires. We found satellite imagery is of limited use for detecting smoke from agricultural fires in Florida due to the cloud cover, overnight smoke, and the fires being small and short-lived. For these reasons, surface measurements are critical for capturing increases in PM2.5 from smoke, and we used multiple smoke-identification criteria. During the study period, median 24-hour PM2.5 concentrations increased by 2.3–6.9 µg m-3 on smoke-impacted days compared to unimp..., To track the near-source transport of smoke from fires in Florida, we used the NOAA HYbrid Single-Particle Lagrangian Integrated Trajectory (HYSPLIT) model. This model is commonly used to compute the trajectory of air parcels. We calculated 12-hour forward trajectories initiated from all HMS fire hotspots in southern Florida (-81.5o x -80o; 26o x 27.5o) during the campaign (October 2022 - September 2023). The HMS fire hotspot product combines detections from the Geostationary Operational Environmental Satellite - R Series (GOES-R)/ Advanced Baseline Imager (ABI) Fire Detection product and the Joint Polar Satellite System (JPSS)/ Visible Infrared Imaging Radiometer Suite (VIIRS) products. We used meteorological data from the NOAA High-Resolution Rapid Refresh (HRRR) model (Dowell et al., 2022), which provides conditions every 4 hours. The HYSPLIT model produces trajectory locations for every hour; however, we interpolated between the reported locations to provide 10-minute observat..., # Model output tracking smoke from agricultural fires in south Florida from October 2022 - May 2023

Dataset DOI: 10.5061/dryad.70rxwdc9k

Description of the data and file structure

This dataset includes monthly gridded output from the National Oceanic and Atmospheric Administration's Hybrid Single-Particle Lagrangian Integrated Trajectory (NOAA HYSPLIT) model. The model was initiated from every NOAA Hazard Mapping System fire hotspot that was detected by the Geostationary Operational Environmental Satellite - R Series (GOES-R)/ Advanced Baseline Imager (ABI) Fire Detection product and the Joint Polar Satellite System (JPSS)/ Visible Infrared Imaging Radiometer Suite (VIIRS) products. We used meteorological data from the NOAA High-Resolution Rapid Refresh (HRRR) model (Dowell et al., 2022), which provides conditions every 4 hours. The HYSPLIT model produces trajectory locations for every hour; however, we interpolated between the reported locations to prov...,