Yongxin Zhang Joseph Grimm Ryan Cabell Ishita Srivastava David Gochis Andreas Prein Roy Rasmussen Kyoko Ikeda Timothy Schneider 20240529 NetCDF https://doi.org/10.5066/P9JE61P7 https://www.sciencebase.gov/catalog/item/64f77acad34ed30c20544c18 This metadata record serves as documentation for the bias adjusted version of the CONUS404 (CONtiguous United States for 40-years at 4-kilometer grid spacing) atmospheric forcing dataset; however this dataset has been revised to include 45 years of data at 1-kilometer grid spacing. This is a dataset of historical conditions (water years 1980-2024, October 1, 1979-September 30, 2024) and has sufficient temporal and spatial detail to resolve mesoscale atmospheric processes, making it appropriate for forcing hydrological models and conducting meteorological analyses. This data release includes the bias adjusted CONUS404 outputs (CONUS404 BA) and the adjustment grids that were used in the bias adjustment process. The adjustment grids are approximately 40 GB in size and contain the day of year bias value needed for the bias adjustment process. The CONUS404 BA outputs are approximately 65 TB in size and contain downscaled hourly data for U-component of wind, V-component of wind, downward longwave radiation, downward shortwave radiation, air temperature, water vapor mixing ratio, surface pressure, and accumulated precipitation variables. For definition of these variables, see the attribute table included in this metadata. The air temperature and accumulated precipitation variables were both downscaled and bias adjusted while the remaining variables were just downscaled. For information on the downscaling and bias adjustments conducted on these data, see the processing steps section in this metadata. Access to Dataset The CONUS404 BA outputs and the adjustment grids are being archived on the U.S. Geological Survey's Black Pearl tape drive system. The data can be accessed through a Globus access portal here: https://app.globus.org/file-manager?destination_id=0a0d322b-1adc-4c93-bfdb-18a6a848479e. The CONUS404 BA outputs are organized into folders by water year (for example, “WY2020”) and the day of year adjustment grids in one folder titled "adjustment_grids". In order to access the files, you will need to log in with an existing Globus account or create one and then perform a file transfer. If you are transferring the files to your local computer, you will also want to install and set up Globus Connect Personal to turn your local machine into an available Globus endpoint. There is no cost for a Globus account. Once Globus is set up, you are ready to access the raw output files. Changes from version 1.0 to 2.0: The dataset was extended by two years, adding water years 2023-2024. Additionally, a small number of missing data values were found in the U2D and V2D variables in the original version of this data release. This issue has been fixed and the missing values corrected. Small edits were also made to the metadata to provide additional detail and clarity. The original metadata file has been archived in a subdirectory in the same location as the data. CONUS404 provides a retrospective high-resolution hydro-climate dataset appropriate for forcing hydrological models and conducting meteorological analyses over the conterminous United States. The data in this release have been bias-adjusted to be used in a WRF-Hydro simulation run for use in the National Water Availability Assessment. While the CONUS404 model (Rasmussen and others, 2023) simulated atmosphere has been regarded as the “best in class” in terms of a model generated estimate of the near surface climate at high spatial and temporal resolution, important biases still exist as with any model simulation. The biases of two separate fields (2-meter air temperature and precipitation) were quantified over the CONUS404 domain and it was decided that the 1 km Daymet version 3 dataset would be used as the background observational reference. The final product of the bias-correction process was an updated 1 km, hourly meteorological forcing dataset for WRF-Hydro to be used in the National Water Availability Assessment. These downscaled datasets will facilitate capturing the response of localized storms on streamflow. 19791001 20240930 publication date Complete None planned -131.1649 -63.1184 52.8980 20.1149 ISO 19115 Topic Category inlandWaters None regional climate model weather research and forecast model high resolution climate simulation hydroclimate hydrology precipitation USGS Thesaurus atmospheric and climatic processes precipitation (atmospheric) hydrology modeling USGS Metadata Identifier USGS:64f77acad34ed30c20544c18 Geographic Names Information System United States Please see 'Distribution Info' for details. Users are advised to read the dataset's metadata thoroughly to understand appropriate use and data limitations. These data are reasonably robust for a wide variety of hydrologic models like WRF-Hydro, PRMS (Precipitation Runoff Modeling System), and others. Aside from precipitation and temperature, the other variables were not bias adjusted, and some variables are known to have notable biases. Because of the bias adjustment, some of the variables are no longer internally consistent. Use in models where a bias in these other variables would create major issues are contraindicated. Please see the CONUS404 paper (Rasmussen and others, 2023) for more information on these biases. Model Data Management Function U.S. Geological Survey, WATER Integrated Modeling and Prediction Division mailing address

12201 Sunrise Valley Dr.

Reston VA 20192 US 703-648-4000 MDMF@usgs.gov National Center for Atmospheric Research (NCAR) The original CONUS404 dataset was produced using the Advanced Research WRF modeling system v3.9.1.1. The data have been modified from their original form; users should check the process step section for further details. The modeling system was run in a Linux operating system on the U.S. Geological Survey’s McKinley supercomputer. Valérie Dulière Yongxin Zhang Eric P. Salathé 20131016 publication Journal of Climate vol. 26, issue 21 n/a American Meteorological Society ppg. 8556-8575 https://doi.org/10.1175/JCLI-D-12-00818.1 Robin W. Glover 199902 publication Journal of Climate vol. 12, issue 2 n/a American Meteorological Society ppg. 551-563 https://doi.org/10.1175/1520-0442(1999)012<0551:IOSRAT>2.0.CO;2 Christopher Daly Michael Halbleib Joseph I. Smith Wayne P. Gibson Matthew K. Doggett George H. Taylor Jan Curtis Phillip P. Pasteris 20080312 publication International Journal of Climatology vol. 28, issue 15 n/a Wiley ppg. 2031-2064 https://doi.org/10.1002/joc.1688 Christopher Daly 20060327 publication International Journal of Climatology vol. 26, issue 6 n/a Wiley ppg. 707-721 https://doi.org/10.1002/joc.1322 Brian A. Cosgrove Dag Lohmann Kenneth E. Mitchell Paul R. Houser Eric F. Wood John C. Schaake Alan Robock Curtis Marshall Justin Sheffield Qingyun Duan Lifeng Luo R. Wayne Higgins Rachel T. Pinker J. Dan Tarpley Jesse Meng 20031031 publication Journal of Geophysical Research: Atmospheres vol. 108, issue D22 n/a American Geophysical Union (AGU) https://doi.org/10.1029/2002JD003118 Jeff T Falgout Janice Gordon Matthew J Davis Brad Williams USGS Advanced Research Computing 2019 dataset https://www.sciencebase.gov U.S. Geological Survey https://doi.org/10.5066/p9psw367 Beyond the data file review associated with the USGS Data Release process, the quality of a number of CONUS404 BA variables was evaluated. Because the architecture of the WRF model used to create the original CONUS404 dataset has been designed and documented to ensure consistency of scientific concepts, computation, and precision, the evaluation process is able to focus on this subset and have high certainty that the quality of the other CONUS404 variables is consistent with that subset. In addition, the selected variables (precipitation, temperature, and downward shortwave radiation) are especially important for hydrologic modeling, which is the target use case for this dataset. Negative values of near-surface specific humidity (Q2), a diagnostic output variable, were identified over a small snow-covered lake near the northern boundary. These negative values are also present in the downscaled variable Q2D. The origin of these anomalous values is unknown. The LWDOWN variable reports a value of 0 in the first time step (00Z October 1, 1979), and it is recommended that users not use this first value. The dataset does not contain null or empty values. Geographic coverage spans the conterminous United States. No horizontal positional accuracy tests were performed. No vertical positional accuracy tests were performed. P.E. Thornton M.M. Thornton B.W. Mayer Y. Wei R. Devarakonda R.S. Vose R.B. Cook 2016 publication n/a ORNL Distributed Active Archive Center https://doi.org/10.3334/ornldaac/1328 Digital and/or Hardcopy 19800101 20171231 publication date Daymet version 3 Source of the day of year temperature and precipitation values used in the bias adjustment process. Roy M. Rasmussen Fei Chen Changhai Liu Kyoko Ikeda Andreas Prein Ju-Hye Kim Timothy Schneider Aiguo Dai David Gochis Aubrey Dugger Yongxin Zhang Abby Jaye Jimy Dudhia Cinlin He Michelle Harrold Lulin Xue Sisi Chen Andrew Newman Erin Dougherty Ronnie Abolafia-Rozenzweig Nicholas Lybarger Roland Viger Krista A Dunne Kristen Rasmussen Gonzalo Miguez-Macho 2023 dataset https://www.sciencebase.gov U.S. Geological Survey https://doi.org/10.5066/p9phpk4f Digital and/or Hardcopy 19791001 20240930 publication date original CONUS404 original CONUS404 input values that downscaling and bias adjustment process was performed on. All processing steps were completed using NCL (NCAR Command Language) version 6.6.2. Downscaling: The starting point for this dataset was the original CONUS404 data release described in Rasmussen and others (2023). Using the CONUS404 original dataset with a resolution of 4 km, a subset of six selected variables: U10, V10, GLW, Q2, PSFC, and SWDOWN were downscaled to 1 km pixel size using a lapsing method and renamed to U2D, V2D, LWDOWN, Q2D, PSFC, and SWDOWN, respectively. The original temperature (T2) variable was interpolated to the 1 km grid using bilinear interpolation while the original precipitation (PREC_ACC_NC) variable was interpolated to the 1 km grid using the "conserve" option to preserve the value of the integral of data between the source and destination grid, and the variables were renamed to T2D and RAINRATE, respectively. See Cosgrove and others 2003, Daly 2006, Daly and others 2008, Glover 1999, and Duliere and others 2013 for more information about the downscaling process. original CONUS404 20230221 Generation of Adjustment Grids: Next, two of the variables (precipitation (RAINRATE) and temperature (T2D)) were bias adjusted. Precipitation data was aggregated to daily totals, while temperature data used the mean of the daily maximum and minimum temperatures for a daily mean. Then, the bias in temperature values within each CONUS404 1-kilometer cell were adjusted based on the grid cell elevation. For grid cells less than 2000 meters in mean elevation, daily temperature was adjusted using a bias adjustment factor associated with the day of the year in which it was simulated to occur. Day-of-year averages were calculated from January 1, 1980 to December 31, 2017 using Daymet version 3 (Thornton and others, 2016). For grid cells above 2500 meters, no adjustment was done. Between 2000 and 2500 meters, values were adjusted according to a linear regression between full adjustment at 2000 meters to no adjustment by 2500 meters. Bias adjustment factors for temperature are in units of Kelvin and are stored in netCDF adjustment grid files titled "CONUS404_Daymet_1km_Tmean_bias_corr_doy_XXX_CF" where XXX denotes the day of year. CONUS404 precipitation bias was calculated as the percentage bias of the modelled daily precipitation compared to the observed daily precipitation, where the observed precipitation was derived from Daymet version 3 (Thornton and others, 2016). Bias adjustment factors for precipitation are in units of percent and are stored in netCDF adjustment grid files titled "CONUS404_precip_bias_corr_1km_day_of_year_XXX_CF" where XXX denotes the day of year. Treatment of non-leap and leap years were as follows; for years with 365 days (non-leap years), bias adjustment grids with file names ending in 1-365 were used, whereas all files (1-366) are used for adjustment in leap years. A 31-day smoothing of the precipitation and temperature bias adjustments were calculated to remove noisy day-to-day values. NaN values represent modeling domains outside of the conterminous United States boundary or modeling domains that cover regions submerged in water (“water cells”). Daymet version 3 original CONUS404 2023 Bias Adjustments: The bias adjustment values from the temperature adjustment grids are added to the modelled temperature from the downscaled CONUS404 product to produce the bias adjusted temperature in CONUS404 BA. The bias adjustment values from the precipitation adjustment grids indicate the level of adjustment needed, with negative percentages requiring a downward adjustment and positive percentages requiring an upward adjustment. To perform the adjustment, the absolute decimal value of each percent is taken and added to 1. The resultant value is then multiplied by the modelled value to produce the bias adjusted precipitation in CONUS404 BA. For example, if the bias adjustment value for the date and location is -30%, and the precipitation value at that pixel is 14 mm, then the resultant precipitation would be 9.8mm (14mm * (1 - 0.3) = 9.8mm). Another way of describing this process is to multiply the precipitation value at the pixel by the decimal percentage, and the resultant value would indicate how much to increase or decrease the pixel precipitation value by (14mm * (-0.3) = -4.2mm and therefore14mm - 4.2mm = 9.8mm). These values are stored in netCDF files titled "YYYYMMDDhhmm.LDASIN_DOMAIN1" with YYYY representing year, MM representing month, DD representing day, hh representing hour, and mm representing minute. The file name structure is typical for meteorological forcing data files containing the meteorological inputs to the WRF-Hydro hydrological model for that particular date and time. The bias adjusted temperature values are stored in the T2D variable while the bias adjusted precipitation values are stored in the RAINRATE variable. RAINRATE was divided by 3600 to convert from accumulated hourly precipitation to a rate (mm/s). The final product of the bias-correction process was an updated 1 km, hourly meteorological forcing dataset for WRF-Hydro containing estimates of U-component of wind, V-component of wind, downward longwave radiation, downward shortwave radiation, air temperature, water vapor mixing ratio, surface pressure, and precipitation rate from the CONUS404 output, where the air temperature and precipitation fields had the above-described bias-adjustment applied. No further bias adjustment was derived or applied to the other variables. 2023 Raster Pixel 3840 4608 Lambert Conformal Conic 30.0 60.0 -97.0 40.0 0.0 0.0 coordinate pair 0.000000007819145331211531 0.000000007819145331211531 meters WRF-Sphere WRF-Sphere 6370000 0.0 YYYYMMDDhhmm.LDASIN_DOMAIN1 A file in NETCDF format containing CONUS404 BA outputs for 10 variables (U2D, V2D, LWDOWN, Q2D, PSFC, T2D, RAINRATE, x, y, time and reference time) where with YYYY represents year, MM represents month, DD represents day, hh represents hour, and mm represents minute. There are no values representing "No Data." NCAR Defined U2D U-component of wind at 10 meters with respect to model grid NCAR Defined -58.101 59.495 meters per second V2D V-component of wind at 10 meters with respect to model grid NCAR Defined -55.931 56.603 meters per second LWDOWN Downward solar longwave radiation flux at ground surface NCAR Defined 76.414 528.923 watts per square meter RAINRATE Precipitation rate NCAR Defined 0 0.0955209 millimeter per second T2D Air temperature 2 meters above ground surface NCAR Defined 196.25 327.30 kelvin Q2D Water vapor mixing ratio at 2 meters above ground surface NCAR Defined -0.007219 0.080471 kilogram of water moisture per kilogram of air (dimensionless) PSFC Surface pressure NCAR Defined 55587.4 105507.0 pascal SWDOWN Incoming downward solar shortwave radiation flux at ground surface NCAR Defined 0 1775.68 watts per square meter x Values of the projected x coordinates of the model cells. NCAR Defined -2303499.25 2303500.75 meters y Values of the projected y coordinates of the model cells. NCAR Defined -1919500.375 1919499.625 meters time Model time in integer minutes since 1970-01-01 00:00:00 UTC. NCAR Defined 5126400 (1979-10-01T00:00:00.000000000) 28795620 (2024-09-30T23:00:00.000000000) minutes since 1970-01-01 00:00:00 UTC reference_time Model initialization time in integer minutes since 1970-01-01 00:00:00 UTC. This is used during the model process. NCAR Defined 5126400 (1979-10-01T00:00:00.000000000) 28752480 (2024-09-01T00:00:00.000000000) minutes since 1970-01-01 00:00:00 UTC CONUS404_precip_bias_corr_1km_day_of_year_XXX_CF.nc Precipitation adjustment grid netCDF file where XXX denotes day of year. This file contains the values needed for the precipitation variable in the bias adjustment process. This file is CF compliant. NCAR Defined conus_precip_bias_corr_day_1km The bias adjustment factor for daily precipitation. These bias adjustment percentages are converted to decimals, added to 1, and multiplied by the modelled precipitation from the downscaled CONUS404 product to produce the bias adjusted precipitation in CONUS404-BA. NCAR Defined -86.5716552734375 399.999969482421 percent west_east Values of the projected x coordinates of the model cells. NCAR Defined -2303499.25 2303500.75 meters south_north Values of the projected y coordinates of the model cells. NCAR Defined -1919500.375 1919499.625 meters time Day of year NCAR Defined 1 366 day CONUS404_Daymet_1km_Tmean_bias_corr_doy_XXX_CF.nc Temperature adjustment grid netCDF file where XXX denotes day of year. This file contains the values needed for the temperature variable in the bias adjustment process. This file is CF compliant. NCAR Defined conus404_daymet_T_bias_corr The bias adjustment factor for daily air temperature at 2 meters. These bias adjustment values are added to the modelled temperature from the downscaled CONUS404 product to produce the bias adjusted temperature in CONUS404-BA. NCAR Defined NaN NaN values represent modeling domains outside of the conterminous United States boundary or modeling domains that cover regions submerged in water (“water cells”). NCAR Defined -21.5046730041503 10.3103008270263 Kelvin west_east Values of the projected x coordinates of the model cells. NCAR Defined -2303499.25 2303500.75 meters south_north Values of the projected y coordinates of the model cells. NCAR Defined -1919500.375 1919499.625 meters time Day of year NCAR Defined 1 366 day CONUS404_BA_Spatial_Metadata.nc Additional netCDF file containing CF compliant information about the modeling domain. NCAR Defined x Value of the projected x coordinates of the model cells. NCAR Defined -2303499.25 2303500.75 meters y Values of the projected y coordinates of the model cells. NCAR Defined -1919500.375 1919499.625 meters crs Coordinate reference system of the data. NCAR Defined Lambert Conformal Conic GS ScienceBase U.S. Geological Survey mailing address

Denver Federal Center, Building 810, Mail Stop 302

Denver CO 80255 US 1-888-275-8747 sciencebase@usgs.gov Unless otherwise stated, all data, metadata and related materials are considered to satisfy the quality standards relative to the purpose for which the data were collected. Although these data and associated metadata have been reviewed for accuracy and completeness and approved for release by the U.S. Geological Survey (USGS), no warranty expressed or implied is made regarding the display or utility of the data for other purposes, nor on all computer systems, nor shall the act of distribution constitute any such warranty. The dataset is not available online. The raw model outputs contained in this dataset are stored on U.S. Geological Survey Black Pearl tape drives and can be made available upon request. 20260508 Model Data Management Function U.S. Geological Survey, WATER Integrated Modeling and Prediction Division mailing address

12201 Sunrise Valley Dr.

Reston VA 20192 US 703-648-4000 MDMF@usgs.gov FGDC Content Standard for Digital Geospatial Metadata FGDC-STD-001-1998