McShane, R.R. Goodling, P.J. Diaz, J.A. Heldmyer, A.J. Hammond, J.C. 20250919 tabular digital data and application/service https://www.sciencebase.gov U.S. Geological Survey data release https://doi.org/10.5066/P132NSWY We applied machine learning (ML) models to forecast streamflow drought from 1 to 13 weeks into the future at more than 3,000 streamgage locations across the Conterminous United States. We applied two machine learning methods (long short-term memory [LSTM] neural network; light gradient boosting [LightGBM] machine) and two benchmark model approaches (persistence; autoregressive integrated moving average [ARIMA]) to predict weekly streamflow percentiles with independent models for each forecast horizon. Both ML models were trained using all percentiles (LSTM-all, LightGBM-all) and only percentiles below 30% (LSTM<30, LightGBM<30). We evaluated model performance regionally and nationally for drought occurrence (the classification performance for a future date) and for drought onset/termination (performance identifying drought starts and ends). This data release contains two zipped archives, one for model outputs (model_outputs.zip), the other for model code (model_code.zip). The model_outputs.zip folder contains files with feature importance data (importance_data.feather); model performance data (performance_data.feather); observed and predicted streamflow percentile time series data (timeseries_data.feather); additional explanatory data (explanatory_data.feather) that were used in the related primary publication to interpret model outputs in this data release; and static watershed attribute data (conus_static_inputs_gages.csv) that were used as model inputs. The model_code.zip folder contains three folders for the different modeling approaches (LSTM, LightGBM, benchmark). The benchmark folder contains two Jupyter notebooks, one for running the persistence models, the other for running the ARIMA models. The LSTM and LightGBM folders contain Jupyter notebooks, with R and Python scripts, for running the LSTM and LightGBM models. Given the need for information on current and forecast hydrological drought conditions, we developed an experimental ML tool to forecast streamflow drought occurrence, onset, and duration for more than 3,000 gaged locations across the Conterminous United States. This work highlights the challenges and opportunities to further advance hydrological drought forecasting and supports an experimental operational streamflow drought assessment and forecast tool that provides previously unavailable streamflow drought forecasts to enable the public to better anticipate and prepare for hydrological drought impacts. This product is the first version of a tool that may be substantially improved in future iterations. 20001002 20200330 ground condition Complete None planned -124.7564516 -66.9490904 49.384487 24.5232175 ISO 19115 Topic Category inlandWaters USGS Thesaurus water resources None autoregressive integrated moving average deep learning drought gradient boosting machine hydrology long short-term memory machine learning modeling neural network streamflow time series USGS Metadata Identifier USGS:687685d8d4be020e2014bdb5 None Conterminous United States No access constraints. Please see "Distribution Information" for details. These data are marked with a Creative Common CC0 1.0 Universal License. These data are in the public domain and do not have any use constraints. Users are advised to read the dataset's metadata thoroughly to understand appropriate use and data limitations. Ryan R. McShane U.S. Geological Survey Hydrologist mailing

521 Progress Circle

Cheyenne WY 82007 USA 307-287-7434 rmcshane@usgs.gov The project producing this data release was funded by the USGS Water Resources Mission Area Drought Program, with computing resources provided by USGS Cloud Hosting Solutions. Amazon SageMaker AI, R, Python, PyTorch, Jupyter, and Git were used to develop and run the models. Model outputs and model code are contained in model_outputs.zip (<1 GB) and model_code.zip (<1 MB). Hammond, J.C. 2025 tabular digital data https://www.sciencebase.gov U.S. Geological Survey data release https://doi.org/10.5066/P1X5VH96 No formal attribute accuracy tests were conducted. No formal logical accuracy tests were conducted. Dataset is considered complete for the information presented, as described in the abstract. Users are advised to read the rest of the metadata record carefully for additional details. A formal accuracy assessment of the horizontal positional information in the dataset was not conducted. A formal accuracy assessment of the vertical positional information in the dataset was not conducted. Falcone, J. 2011 tabular digital data https://doi.org/10.5066/P96CPHOT Digital 2011 publication date Falcone, 2011 Additional data used to interpret the model output in the related primary publication Hammond, J.C. 2024 tabular digital data https://doi.org/10.5066/P9JWJPNC Digital 2024 publication date Hammond, 2024 Additional data used to interpret the model output in the related primary publication Hammond, J.C. 2025 tabular digital data https://doi.org/10.5066/P1X5VH96 Digital 2025 publication date Hammond, 2025 Input used by the models and additional data used to interpret the model output in the related primary publication Price, A.N. Jones, C.N. Hammond, J.C. Zimmer, M.A. Zipper, S.C. 2021 tabular digital data https://doi.org/10.1029/2021GL093298 Digital 2021 publication date Price et al., 2021 Additional data used to interpret the model output in the related primary publication Simeone, C.E. 2022 tabular digital data https://doi.org/10.5066/P92FAASD Digital 2022 publication date Simeone, 2022 Additional data used to interpret the model output in the related primary publication U.S. Geological Survey 2025 tabular digital data https://doi.org/10.5066/F7P55KJN Digital 2025 publication date USGS, 2025 Additional data used to interpret the model output in the related primary publication Wieczorek, M.E. Jackson, S.E. Schwarz, G.E. Sekellick, A.J. Staub, L.E. 2018 tabular digital data https://doi.org/10.5066/F7765D7V Digital 2018 publication date Wieczorek et al., 2018 Input used by the models and additional data used to interpret the model output in the related primary publication Wieczorek, M.E. Wolock, D.M. McCarthy, P.M. 2021 tabular digital data https://doi.org/10.5066/P92S9ZX6 Digital 2021 publication date Wieczorek et al., 2021 Input used by the models and additional data used to interpret the model output in the related primary publication We obtained daily streamflow data for 2000-2020 from the USGS National Water Information System (USGS, 2025). We then converted 7-day average daily observed streamflow to percentile values to identify drought via consistent thresholds across streamgages. We computed de-seasonalized streamflow percentiles with the unbiased Weibull plotting position, applying a variable threshold for each day of the year that used only the values for a 30-day window surrounding that day from all years of record (for example, percentiles for August 15 used values from August 1 to August 30 for all years of record). We prepared watershed average time series of several gridded datasets (Hammond, 2025) as well as watershed average values of static watershed properties (conus_static_inputs_gages.csv; Hammond, 2025; Wieczorek et al., 2018; and Wieczorek et al., 2021). Both tree-based (LightGBM) and neural network (LSTM) models were trained and evaluated using streamflow and explanatory variable data from the period 2000-2020 for CONUS. These ML models were compared to two benchmark models: a simple persistence (no change between most recent observation and forecast period) and an ARIMA model. Separate regression models were built to predict weekly streamflow percentiles for each lead time (1, 2, 4, 9, and 13 weeks). We trained our models on a central period from October 1, 2002, to September 30, 2018, leaving the first part of the record (October 1, 2000-September 30, 2002) and the last part of the record (October 1, 2018-March 30, 2020) for model testing. Models were evaluated for their performance in matching the correct drought intensity category (moderate drought less than 20th percentile, severe drought less than 10th percentile, extreme drought less than 5th percentile) as well as for their ability to correctly forecast the onset and termination of drought events. We also performed a first-order examination on potential controls of model performance by relating our overall performance metric to several descriptors of each gage site (explanatory_data.feather; Falcone, 2011; Hammond, 2024; Hammond, 2025; Price et al., 2021; Simeone, 2022; USGS, 2025; Wieczorek et al., 2018; and Wieczorek et al., 2021). See the related primary publication for more detail. The notebooks and READMEs (model_code.zip) contain code, comments, and instructions for using AWS (Amazon Web Services) to load data from S3 (Simple Storage Service) buckets and train models in SageMaker AI on EC2 (Elastic Compute Cloud) instances. 20250801 conus_static_inputs_gages.csv CSV file containing static watershed attribute data used as model inputs. U.S. Geological Survey StaID USGS streamgage station identification number. https://doi.org/10.5066/P1X5VH96 NA No data U.S. Geological Survey USGS National Water Information System https://doi.org/10.5066/F7P55KJN S-SqKm Drainage area. https://doi.org/10.5066/F7765D7V NA No data U.S. Geological Survey 1.09 572830.2 square kilometers S-DI_EROM Reservoir storage intensity in units of days based on reservoir storage in a contributing area normalized by the mean annual streamflow. This metric indicates the duration of storage impact upstream from each stream segment relative to the typical flow condition. In addition, this metric provides an assessment of the potential influence of a dam on average and low flows because the metric estimates the number of days of flow that can be sustained by contributing area storage alone, without additional water or groundwater input. https://doi.org/10.5066/P92S9ZX6 NA No data U.S. Geological Survey 0 49023392 days S-DI_PMC Degree of regulation of a river reach based on upstream reservoir storage relative to the 30-year average annual precipitation, as well as the upstream dam and watershed areas. This metric provides an estimate of the capacity of the contributing area to store precipitation and is oriented to understanding how peak flows may be affected by dams throughout the flow network. https://doi.org/10.5066/P92S9ZX6 NA No data U.S. Geological Survey 0 51131.8562 unitless S-TOT_AET Mean annual evapotranspiration, estimated by Senay and others. https://doi.org/10.5066/F7765D7V NA No data U.S. Geological Survey 24.3 1311.25 millimeters per year S-TOT_ARTIFICIAL Flowlines coded as artificial paths. https://doi.org/10.5066/F7765D7V NA No data U.S. Geological Survey 0 86.55 percent S-TOT_AWCAVE Average value for the range in available water capacity per NHDPlus version 2 catchment. https://doi.org/10.5066/F7765D7V NA No data U.S. Geological Survey 0.04 0.25 inches of water per inches of soil depth S-TOT_BDAVE Average value for the value of bulk density per NHDPlus version 2 catchment. https://doi.org/10.5066/F7765D7V NA No data U.S. Geological Survey 0.87 1.74 grams per cubic centimeter S-TOT_BFI Base Flow Index (BFI) is a ratio of base flow to total streamflow. Base flow is the sustained, slowly varying component of streamflow, usually attributed to ground-water discharge to a stream. https://doi.org/10.5066/F7765D7V NA No data U.S. Geological Survey 4.85 87.61 percent S-TOT_CANALDITCH Flowlines coded as canals or ditches. https://doi.org/10.5066/F7765D7V NA No data U.S. Geological Survey 0 94.44 percent S-TOT_CLAYAVE Average percent of clay in soil per NHDPlus version 2 catchment. https://doi.org/10.5066/F7765D7V NA No data U.S. Geological Survey 2.32 53.08 percent S-TOT_CONNECTOR Flowlines coded as connectors. https://doi.org/10.5066/F7765D7V NA No data U.S. Geological Survey 0 4.35 percent S-TOT_CONTACT Subsurface flow contact time index. The subsurface contact time index estimates the number of days that infiltrated water resides in the saturated subsurface zone of the basin before discharging into the stream. https://doi.org/10.5066/F7765D7V NA No data U.S. Geological Survey 1.34 13941.01 days S-TOT_CWD Average number of consecutive days with measurable precipitation. It is also sometimes referred to as average storm duration. https://doi.org/10.5066/F7765D7V NA No data U.S. Geological Survey 1.45 3.6 days S-TOT_ELEV_MAX NHDPlusV2 flowline catchment's maximum elevation. https://doi.org/10.5066/F7765D7V NA No data U.S. Geological Survey 12.22 4410.95 meters S-TOT_ELEV_MEAN NHDPlusV2 flowline catchment's mean elevation. https://doi.org/10.5066/F7765D7V NA No data U.S. Geological Survey -16.41 3602.21 meters S-TOT_ELEV_MIN NHDPlusV2 flowline catchment's minimum elevation. https://doi.org/10.5066/F7765D7V NA No data U.S. Geological Survey -69.08 3158.74 meters S-TOT_EWT Average depth to water table relative to the land surface. https://doi.org/10.5066/F7765D7V NA No data U.S. Geological Survey -143.23 -2 meters S-TOT_FRESHWATER_WD Freshwater withdrawals from 1995-2000 county-level estimates. https://doi.org/10.5066/F7765D7V NA No data U.S. Geological Survey 0.03 714.02 megaliters per square kilometer S-TOT_FSTFZ6190 Average of mean day of the year of first freeze, derived from 30 years of record (1961-1990), 2km PRISM. For example, value of 300 is the 300th day of the year (Oct 27th). https://doi.org/10.5066/F7765D7V NA No data U.S. Geological Survey 0 367 day of year S-TOT_HGA Percentage of Hydrologic Group A soil. Hydrologic group A soils have high infiltration rates. Soils are deep and well drained and, typically, have high sand and gravel content. https://doi.org/10.5066/F7765D7V NA No data U.S. Geological Survey 0 91.07 percent S-TOT_HGB Percentage of Hydrologic Group B soil. Hydrologic group B soils have moderate infiltration rates. Soils are moderately deep, moderately well drained, and moderately coarse in texture. https://doi.org/10.5066/F7765D7V NA No data U.S. Geological Survey 0 100 percent S-TOT_HGC Percentage of Hydrologic Group C soil. Hydrologic group C soils have slow soil infiltration rates. The soil profiles include layers impeding downward movement of water and, typically, have moderately fine or fine texture. https://doi.org/10.5066/F7765D7V NA No data U.S. Geological Survey 0 97.69 percent S-TOT_HGD Percentage of Hydrologic Group D soil. Hydrologic group D soils have very slow infiltration rates. Soils are clayey, have a high water table, or have a shallow impervious layer. https://doi.org/10.5066/F7765D7V NA No data U.S. Geological Survey 0 99.88 percent S-TOT_IEOF Percentage of Horton overland flow. https://doi.org/10.5066/F7765D7V NA No data U.S. Geological Survey 0 55.57 percent S-TOT_KFACT Average value for the K factor per NHDPlus version 2 catchment. https://doi.org/10.5066/F7765D7V NA No data U.S. Geological Survey 0.11 0.53 unitless S-TOT_KFACT_UP Average value for the K factor in the upper soil horizon per NHDPlus version 2 catchment. https://doi.org/10.5066/F7765D7V NA No data U.S. Geological Survey 0.08 0.48 unitless S-TOT_LSTFZ6190 Average of mean day of the year of last freeze, derived from 30 years of record (1961-1990), 2km PRISM. For example, value of 100 is the 100th day of the year (April 10th). https://doi.org/10.5066/F7765D7V NA No data U.S. Geological Survey 0 208.77 day of year S-TOT_MAXP6190 Maximum average annual precipitation. https://doi.org/10.5066/F7765D7V NA No data U.S. Geological Survey 10.77 855.44 millimeters per year S-TOT_MAXWD6190 Average of maximum monthly number of days of measurable precipitation, derived from 30 years of record (1961-1990), 2.3km PRISM. https://doi.org/10.5066/F7765D7V NA No data U.S. Geological Survey 2 25.3 days S-TOT_MINWD6190 Average of minimum monthly number of days of measurable precipitation, derived from 30 years of record (1961-1990), 2.3km PRISM. https://doi.org/10.5066/F7765D7V NA No data U.S. Geological Survey 0 11.97 days S-TOT_MIRAD_2012 Moderate Resolution Imaging Spectroradiometer Irrigated Agriculture Dataset for the United States (MIrAD-US), from 250 m MODIS, for 2012. https://doi.org/10.5066/F7765D7V NA No data U.S. Geological Survey 0 80.75 percent S-TOT_NLCD19_BARREN 2019 percentage of land-use and land-cover type Barren Land (Rock/Sand/Clay). https://doi.org/10.5066/F7765D7V NA No data U.S. Geological Survey 0 42.66 percent S-TOT_NLCD19_DEVELOPED 2019 percentage of land-use and land-cover type Developed, High Intensity, Medium Intensity, Low Intensity, and Open Space. https://doi.org/10.5066/F7765D7V NA No data U.S. Geological Survey 0 99.89 percent S-TOT_NLCD19_FOREST 2019 percentage of land-use and land-cover type Deciduous Forest, Evergreen Forest, and Mixed Forest. https://doi.org/10.5066/F7765D7V NA No data U.S. Geological Survey 0 98.98 percent S-TOT_NLCD19_HERBACEOUS 2019 percentage of land-use and land-cover type Grassland/Herbaceous. https://doi.org/10.5066/F7765D7V NA No data U.S. Geological Survey 0 93.84 percent S-TOT_NLCD19_PLANTED 2019 percentage of land-use and land-cover type Pasture/Hay and Cultivated Crops. https://doi.org/10.5066/F7765D7V NA No data U.S. Geological Survey 0 94.44 percent S-TOT_NLCD19_SHRUBLAND 2019 percentage of land-use and land-cover type Shrub/Scrub. https://doi.org/10.5066/F7765D7V NA No data U.S. Geological Survey 0 98.73 percent S-TOT_NLCD19_WATER 2019 percentage of land-use and land-cover type Open Water and Perennial Ice/Snow. https://doi.org/10.5066/F7765D7V NA No data U.S. Geological Survey 0 38.47 percent S-TOT_NLCD19_WETLAND 2019 percentage of land-use and land-cover type Woody Wetlands and Emergent Herbaceous Wetlands. https://doi.org/10.5066/F7765D7V NA No data U.S. Geological Survey 0 78.45 percent S-TOT_NO10AVE Average percent by weight of soil material less than 3 inches in size that passes through a No. 10 sieve (2 millimeters) per NHDPlus version 2 catchment. https://doi.org/10.5066/F7765D7V NA No data U.S. Geological Survey 13.87 99.99 percent S-TOT_NO200AVE Average percent by weight of soil material less than 3 inches in size that passes through a No. 200 sieve (.074 millimeters) per NHDPlus version 2 catchment. https://doi.org/10.5066/F7765D7V NA No data U.S. Geological Survey 7.02 96.79 percent S-TOT_NO4AVE Average percent by weight of soil material less than 3 inches in size that passes through a No. 4 sieve (5 millimeters) per NHDPlus version 2 catchment. https://doi.org/10.5066/F7765D7V NA No data U.S. Geological Survey 16.2 100 percent S-TOT_OM Average value for the range in organic matter content (percent by weight) per NHDPlus version 2 catchment. https://doi.org/10.5066/F7765D7V NA No data U.S. Geological Survey 0.06 23.07 percent S-TOT_PERMAVE Average value for the range in permeability per NHDPlus version 2 catchment. https://doi.org/10.5066/F7765D7V NA No data U.S. Geological Survey 0.27 15.3 inches per hour S-TOT_PIPELINE Percentage of all flowline reach lengths per NHDPlusV2 catchment that is a pipeline. Pipelines are specifically man-made structures of steel, concrete, or polymers that direct surface water flows from one area to another. https://doi.org/10.5066/F7765D7V NA No data U.S. Geological Survey 0 31.47 percent S-TOT_PRSNOW Snow percent of total precipitation estimate, mean for period 1901-2000, 1km grid. https://doi.org/10.5066/F7765D7V NA No data U.S. Geological Survey 0 88.29 percent S-TOT_RDX Number of road and stream intersections. https://doi.org/10.5066/F7765D7V NA No data U.S. Geological Survey 0 132880 number S-TOT_RECHG Mean annual natural groundwater recharge. https://doi.org/10.5066/F7765D7V NA No data U.S. Geological Survey 0.99 2027.8 millimeters per year S-TOT_RF7100 Mean annual average for the Rainfall Runoff factor (R factor of Universal Soil Loss Equation) for the period 1971-2000 per NHDPlus version 2 catchment. https://doi.org/10.5066/F7765D7V NA No data U.S. Geological Survey 6.32 640.45 hundreds of foot-ton inches per acre per hour per year S-TOT_RFACT Average Rainfall Runoff Factor from Universal Soil Loss Equation per NHDPlus version 2 catchment. https://doi.org/10.5066/F7765D7V NA No data U.S. Geological Survey 63.19 6404.48 hundreds of foot-ton inches per acre per hour S-TOT_RH Average relative humidity, from 2km PRISM, derived from 30 years of record (1961-1990). https://doi.org/10.5066/F7765D7V NA No data U.S. Geological Survey 33 86.36 percent S-TOT_ROCKDEP Average value for the range in the total soil thickness examined per NHDPlus version 2 catchment. https://doi.org/10.5066/F7765D7V NA No data U.S. Geological Survey 0 69.16 inches S-TOT_SANDAVE Average percent of sand in soil per NHDPlus version 2 catchment. https://doi.org/10.5066/F7765D7V NA No data U.S. Geological Survey 3.06 92.9 percent S-TOT_SATOF Percentage of Dunne overland flow. https://doi.org/10.5066/F7765D7V NA No data U.S. Geological Survey 0 16.39 percent S-TOT_SILTAVE Average percent of silt in soil per NHDPlus version 2 catchment. https://doi.org/10.5066/F7765D7V NA No data U.S. Geological Survey 4.18 73.12 percent S-TOT_SRL25AG Estimated percent presence of the soil restrictive layer in the upper 25 centimeters of agricultural land. https://doi.org/10.5066/F7765D7V NA No data U.S. Geological Survey 0 100 percent S-TOT_STREAM_LENGTH NHDPlus version 2 flowline's length in kilometers taken directly from NHDPlusv2's NHDflowline shapefile's item, LENGTHKM. https://doi.org/10.5066/F7765D7V NA No data U.S. Geological Survey 1.2 404446.7 kilometers S-TOT_STREAM_SLOPE NHDPlus version 2 flowline's average slope. https://doi.org/10.5066/F7765D7V NA No data U.S. Geological Survey 0 0.149399 percent S-TOT_STREAMRIVER Flowlines coded as streams or rivers. https://doi.org/10.5066/F7765D7V NA No data U.S. Geological Survey 5.56 100 percent S-TOT_STRM_DENS Stream density of catchment reach. https://doi.org/10.5066/F7765D7V NA No data U.S. Geological Survey 0.03 2.43 kilometers per square kilometer S-TOT_TOTAL_ROAD_DENS Density of all road types per NHDPlusV2 catchment. Density is defined as the length of road divided by the catchment area. https://doi.org/10.5066/F7765D7V NA No data U.S. Geological Survey 0 15.75 kilometers per square kilometer S-TOT_TWI Topographic wetness index, ln(a/S); where ln is the natural log, a is the upslope area per unit contour length and S is the slope at that point. https://doi.org/10.5066/F7765D7V NA No data U.S. Geological Survey 7.36 16.46 natural log of meters S-TOT_WB5100_ANN Average annual runoff from McCabe and Wolock's Runoff Model 1951-2000 per NHDPlus version 2 catchment. https://doi.org/10.5066/F7765D7V NA No data U.S. Geological Survey 1.32 4766.98 millimeters per year S-TOT_WDANN NHDPlusV2 catchment value for the 30 year annual average (1961-1990) number of days of measurable precipitation. https://doi.org/10.5066/F7765D7V NA No data U.S. Geological Survey 11.75 228.36 days S-TOT_WTDEP Average value for the range in depth to the seasonally high water table per NHDPlus version 2 catchment. https://doi.org/10.5066/F7765D7V NA No data U.S. Geological Survey 0.44 6 feet explanatory_data.feather Feather file containing additional explanatory data used in the related primary publication to interpret model outputs. U.S. Geological Survey StaID USGS streamgage station identification number. https://doi.org/10.5066/P1X5VH96 NA No data U.S. Geological Survey USGS National Water Information System https://doi.org/10.5066/F7P55KJN dec_lat_va Latitude (NAD83). https://doi.org/10.5066/F7765D7V NA No data U.S. Geological Survey 26.7217336 48.9059574 decimal degrees dec_long_v Longitude (NAD83). https://doi.org/10.5066/F7765D7V NA No data U.S. Geological Survey -124.3157227 -67.7688889 decimal degrees drain_area Drainage area. https://doi.org/10.5066/F7765D7V NA No data U.S. Geological Survey 0 246700 square kilometers HUC02 Hydrologic Unit Code, 2-digit. https://doi.org/10.5066/P96CPHOT NA No data U.S. Geological Survey USGS Watershed Boundary Dataset https://www.usgs.gov/national-hydrography/watershed-boundary-dataset peakSWEdivP Annual maximum snow water equivalent divided by total annual precipitation. https://doi.org/10.5066/P9JWJPNC NA No data U.S. Geological Survey 0 0.7872691 unitless days_dam_impact Number of days of daily mean streamflow that can be stored by reservoirs in the watershed. https://doi.org/10.5066/P92S9ZX6 NA No data U.S. Geological Survey 0 49023392 days winter_ice_frac Fraction of days in winter (December, January, February) where streamflow is estimated because of the presence of ice. https://doi.org/10.5066/F7P55KJN NA No data U.S. Geological Survey 0 1 fraction non_perennial 1 if a site had an average of five or more days with no flow each year from 1979 to 2018, otherwise 0. https://doi.org/10.1029/2021GL093298 NA No data U.S. Geological Survey 0 If a site did not have an average of five or more days with no flow each year from 1979 to 2018. https://doi.org/10.1029/2021GL093298 1 If a site had an average of five or more days with no flow each year from 1979 to 2018. https://doi.org/10.1029/2021GL093298 dam_impacted 1 if degree of regulation (total upstream reservoir storage divided by total annual streamflow) is greater than 1, otherwise 0. https://doi.org/10.5066/P92S9ZX6 NA No data U.S. Geological Survey 0 If degree of regulation (total upstream reservoir storage divided by total annual streamflow) is not greater than 1. https://doi.org/10.5066/P92S9ZX6 1 If degree of regulation (total upstream reservoir storage divided by total annual streamflow) is greater than 1. https://doi.org/10.5066/P92S9ZX6 snow_dominated 1 if mean annual peak snow water equivalent divided by mean annual precipitation is greater than 0.25, otherwise 0. https://doi.org/10.5066/P9JWJPNC NA No data U.S. Geological Survey 0 If mean annual peak snow water equivalent divided by mean annual precipitation is not greater than 0.25. https://doi.org/10.5066/P9JWJPNC 1 If mean annual peak snow water equivalent divided by mean annual precipitation is greater than 0.25. https://doi.org/10.5066/P9JWJPNC ice_impacted 1 if 'winter_ice_frac' is greater than 0.25, otherwise 0. https://doi.org/10.5066/F7P55KJN NA No data U.S. Geological Survey 0 If 'winter_ice_frac' is not greater than 0.25. https://doi.org/10.5066/F7P55KJN 1 If 'winter_ice_frac' is greater than 0.25. https://doi.org/10.5066/F7P55KJN number_of_droughts_2 Number of drought events below the 2nd percentile identified at the site over the period. https://doi.org/10.5066/P92FAASD NA No data U.S. Geological Survey 1 39 number number_of_droughts_5 Number of drought events below the 5th percentile identified at the site over the period. https://doi.org/10.5066/P92FAASD NA No data U.S. Geological Survey 1 64 number number_of_droughts_10 Number of drought events below the 10th percentile identified at the site over the period. https://doi.org/10.5066/P92FAASD NA No data U.S. Geological Survey 1 117 number number_of_droughts_20 Number of drought events below the 20th percentile identified at the site over the period. https://doi.org/10.5066/P92FAASD NA No data U.S. Geological Survey 3 221 number number_of_droughts_30 Number of drought events below the 30th percentile identified at the site over the period. https://doi.org/10.5066/P92FAASD NA No data U.S. Geological Survey 5 314 number ave_duration_2 Mean duration of each drought event below the 2nd percentile. https://doi.org/10.5066/P92FAASD NA No data U.S. Geological Survey 7.0384615 359 days ave_duration_5 Mean duration of each drought event below the 5th percentile. https://doi.org/10.5066/P92FAASD NA No data U.S. Geological Survey 8.8888889 730 days ave_duration_10 Mean duration of each drought event below the 10th percentile. https://doi.org/10.5066/P92FAASD NA No data U.S. Geological Survey 10.0940171 1461 days ave_duration_20 Mean duration of each drought event below the 20th percentile. https://doi.org/10.5066/P92FAASD NA No data U.S. Geological Survey 11.2850679 971 days ave_duration_30 Mean duration of each drought event below the 30th percentile. https://doi.org/10.5066/P92FAASD NA No data U.S. Geological Survey 12.7887789 859.8 days max_duration_2 Longest drought duration below the 2nd percentile at each site. https://doi.org/10.5066/P92FAASD NA No data U.S. Geological Survey 11 359 days max_duration_5 Longest drought duration below the 5th percentile at each site. https://doi.org/10.5066/P92FAASD NA No data U.S. Geological Survey 21 730 days max_duration_10 Longest drought duration below the 10th percentile at each site. https://doi.org/10.5066/P92FAASD NA No data U.S. Geological Survey 37 1461 days max_duration_20 Longest drought duration below the 20th percentile at each site. https://doi.org/10.5066/P92FAASD NA No data U.S. Geological Survey 42 2205 days max_duration_30 Longest drought duration below the 30th percentile at each site. https://doi.org/10.5066/P92FAASD NA No data U.S. Geological Survey 44 2568 days total_duration_2 Total drought duration below the 2nd percentile at each site. https://doi.org/10.5066/P92FAASD NA No data U.S. Geological Survey 36 780 days total_duration_5 Total drought duration below the 5th percentile at each site. https://doi.org/10.5066/P92FAASD NA No data U.S. Geological Survey 417 1656 days total_duration_10 Total drought duration below the 10th percentile at each site. https://doi.org/10.5066/P92FAASD NA No data U.S. Geological Survey 1059 2144 days total_duration_20 Total drought duration below the 20th percentile at each site. https://doi.org/10.5066/P92FAASD NA No data U.S. Geological Survey 2371 4365 days total_duration_30 Total drought duration below the 30th percentile at each site. https://doi.org/10.5066/P92FAASD NA No data U.S. Geological Survey 3655 5161 days mean_intensity_2 Mean drought intensity below the 2nd percentile at each site. https://doi.org/10.5066/P92FAASD NA No data U.S. Geological Survey 0.107081 1.5434268 percentile mean_intensity_5 Mean drought intensity below the 5th percentile at each site. https://doi.org/10.5066/P92FAASD NA No data U.S. Geological Survey 2.0879706 3.9889442 percentile mean_intensity_10 Mean drought intensity below the 10th percentile at each site. https://doi.org/10.5066/P92FAASD NA No data U.S. Geological Survey 3.7853229 8.2916604 percentile mean_intensity_20 Mean drought intensity below the 20th percentile at each site. https://doi.org/10.5066/P92FAASD NA No data U.S. Geological Survey 9.6246516 16.4842641 percentile mean_intensity_30 Mean drought intensity below the 30th percentile at each site. https://doi.org/10.5066/P92FAASD NA No data U.S. Geological Survey 14.4909465 25.8454538 percentile max_intensity_2 Maximum drought intensity below the 2nd percentile at each site. https://doi.org/10.5066/P92FAASD NA No data U.S. Geological Survey 0.0805802 0.0875657 percentile max_intensity_5 Maximum drought intensity below the 5th percentile at each site. https://doi.org/10.5066/P92FAASD NA No data U.S. Geological Survey 0.0805802 0.087108 percentile max_intensity_10 Maximum drought intensity below the 10th percentile at each site. https://doi.org/10.5066/P92FAASD NA No data U.S. Geological Survey 0.0805802 0.087108 percentile max_intensity_20 Maximum drought intensity below the 20th percentile at each site. https://doi.org/10.5066/P92FAASD NA No data U.S. Geological Survey 0.0805802 0.087108 percentile max_intensity_30 Maximum drought intensity below the 30th percentile at each site. https://doi.org/10.5066/P92FAASD NA No data U.S. Geological Survey 0.0805802 0.087108 percentile max_severity_2 Maximum drought severity below the 2nd percentile at each site. Drought severity is measured as the integration of the duration and intensity of each event. https://doi.org/10.5066/P92FAASD NA No data U.S. Geological Survey 13.8001239 275.0046883 percentile days max_severity_5 Maximum drought severity below the 5th percentile at each site. Drought severity is measured as the integration of the duration and intensity of each event. https://doi.org/10.5066/P92FAASD NA No data U.S. Geological Survey 60.7695407 1744.5397065 percentile days max_severity_10 Maximum drought severity below the 10th percentile at each site. Drought severity is measured as the integration of the duration and intensity of each event. https://doi.org/10.5066/P92FAASD NA No data U.S. Geological Survey 240.8622079 7171.499767 percentile days max_severity_20 Maximum drought severity below the 20th percentile at each site. Drought severity is measured as the integration of the duration and intensity of each event. https://doi.org/10.5066/P92FAASD NA No data U.S. Geological Survey 701.2570508 26201.1086408 percentile days max_severity_30 Maximum drought severity below the 30th percentile at each site. Drought severity is measured as the integration of the duration and intensity of each event. https://doi.org/10.5066/P92FAASD NA No data U.S. Geological Survey 1171.5713135 50909.5312342 percentile days mean_severity_2 Mean drought severity below the 2nd percentile at each site. Drought severity is measured as the integration of the duration and intensity of each event. https://doi.org/10.5066/P92FAASD NA No data U.S. Geological Survey 7.2412562 234.6351231 percentile days mean_severity_5 Mean drought severity below the 5th percentile at each site. Drought severity is measured as the integration of the duration and intensity of each event. https://doi.org/10.5066/P92FAASD NA No data U.S. Geological Survey 22.5909077 1744.5397065 percentile days mean_severity_10 Mean drought severity below the 10th percentile at each site. Drought severity is measured as the integration of the duration and intensity of each event. https://doi.org/10.5066/P92FAASD NA No data U.S. Geological Survey 53.9934018 7171.499767 percentile days mean_severity_20 Mean drought severity below the 20th percentile at each site. Drought severity is measured as the integration of the duration and intensity of each event. https://doi.org/10.5066/P92FAASD NA No data U.S. Geological Survey 119.0765259 9583.6478848 percentile days mean_severity_30 Mean drought severity below the 30th percentile at each site. Drought severity is measured as the integration of the duration and intensity of each event. https://doi.org/10.5066/P92FAASD NA No data U.S. Geological Survey 199.8609964 12974.4650922 percentile days max_flow_deficit_2 Maximum streamflow deficit below the 2nd percentile at each site. Flow deficit is measured as the integration of the amount of flow below the drought threshold times the duration of the drought. https://doi.org/10.5066/P92FAASD NA No data U.S. Geological Survey -3669.78 90279.6 cubic feet per second days max_flow_deficit_5 Maximum streamflow deficit below the 5th percentile at each site. Flow deficit is measured as the integration of the amount of flow below the drought threshold times the duration of the drought. https://doi.org/10.5066/P92FAASD NA No data U.S. Geological Survey -1688.2 225000.5 cubic feet per second days max_flow_deficit_10 Maximum streamflow deficit below the 10th percentile at each site. Flow deficit is measured as the integration of the amount of flow below the drought threshold times the duration of the drought. https://doi.org/10.5066/P92FAASD NA No data U.S. Geological Survey -1428.1 459935 cubic feet per second days max_flow_deficit_20 Maximum streamflow deficit below the 20th percentile at each site. Flow deficit is measured as the integration of the amount of flow below the drought threshold times the duration of the drought. https://doi.org/10.5066/P92FAASD NA No data U.S. Geological Survey -143.02 1450923 cubic feet per second days max_flow_deficit_30 Maximum streamflow deficit below the 30th percentile at each site. Flow deficit is measured as the integration of the amount of flow below the drought threshold times the duration of the drought. https://doi.org/10.5066/P92FAASD NA No data U.S. Geological Survey 0 2807930 cubic feet per second days mean_flow_deficit_2 Mean streamflow deficit below the 2nd percentile at each site. Flow deficit is measured as the integration of the amount of flow below the drought threshold times the duration of the drought. https://doi.org/10.5066/P92FAASD NA No data U.S. Geological Survey -7659.8881818 12445.2888889 cubic feet per second days mean_flow_deficit_5 Mean streamflow deficit below the 5th percentile at each site. Flow deficit is measured as the integration of the amount of flow below the drought threshold times the duration of the drought. https://doi.org/10.5066/P92FAASD NA No data U.S. Geological Survey -8030.0204082 28469.3076923 cubic feet per second days mean_flow_deficit_10 Mean streamflow deficit below the 10th percentile at each site. Flow deficit is measured as the integration of the amount of flow below the drought threshold times the duration of the drought. https://doi.org/10.5066/P92FAASD NA No data U.S. Geological Survey -11164.0894737 56301.8490566 cubic feet per second days mean_flow_deficit_20 Mean streamflow deficit below the 20th percentile at each site. Flow deficit is measured as the integration of the amount of flow below the drought threshold times the duration of the drought. https://doi.org/10.5066/P92FAASD NA No data U.S. Geological Survey -4558.0890625 94399.3529412 cubic feet per second days mean_flow_deficit_30 Mean streamflow deficit below the 30th percentile at each site. Flow deficit is measured as the integration of the amount of flow below the drought threshold times the duration of the drought. https://doi.org/10.5066/P92FAASD NA No data U.S. Geological Survey -547.7848921 160562.3181818 cubic feet per second days importance_data.feather Feather file containing feature importance data. U.S. Geological Survey feature Feature used in training model. U.S. Geological Survey NA No data U.S. Geological Survey Unique values described in the data dictionary at https://doi.org/10.5066/P1X5VH96 and in the 'conus_static_inputs_gages.csv' file in this data release. importance Importance of 'feature' used in training model; gain in model accuracy by using feature. U.S. Geological Survey NA No data U.S. Geological Survey -0.9109273 325437.8579445 unitless nday_forecast Number of days ahead of model forecast. U.S. Geological Survey NA No data U.S. Geological Survey 7 91 days quantile Quantile of response variable estimated by model. U.S. Geological Survey NA No data U.S. Geological Survey 0.05 0.95 unitless model_id Model identifier. U.S. Geological Survey NA No data U.S. Geological Survey LightGBM<30 Model identifier for gradient boosting machine models trained with only streamflow observations below the 30th percentile. U.S. Geological Survey LightGBM-all Model identifier for gradient boosting machine models trained with all observations. U.S. Geological Survey LSTM<30 Model identifier for long short-term memory neural network models trained with only streamflow observations below the 30th percentile. U.S. Geological Survey performance_data.feather Feather file containing model performance data. U.S. Geological Survey site_id Site identifier. USGS streamgage station identification number. U.S. Geological Survey NA No data U.S. Geological Survey USGS National Water Information System https://doi.org/10.5066/F7P55KJN model_id Model identifier. U.S. Geological Survey NA No data U.S. Geological Survey ARIMA Model identifier for autoregressive integrated moving average models. U.S. Geological Survey LSTM-all Model identifier for long short-term memory neural network models trained with all observations. U.S. Geological Survey LSTM<30 Model identifier for long short-term memory neural network models trained with only streamflow observations below the 30th percentile. U.S. Geological Survey LightGBM-all Model identifier for gradient boosting machine models trained with all observations. U.S. Geological Survey LightGBM<30 Model identifier for gradient boosting machine models trained with only streamflow observations below the 30th percentile. U.S. Geological Survey persistence Model identifier for modeling no change between most recent observation and forecast date. U.S. Geological Survey nday_forecast Number of days ahead of model forecast. U.S. Geological Survey NA No data U.S. Geological Survey 7 91 days kap_10 Cohen's kappa of streamflow predictions classified as drought below the 10th percentile. Higher values are better. U.S. Geological Survey NA No data U.S. Geological Survey -0.2777778 1 unitless kap_05 Cohen's kappa of streamflow predictions classified as drought below the 5th percentile. Higher values are better. U.S. Geological Survey NA No data U.S. Geological Survey -0.2252422 1 unitless kap_20 Cohen's kappa of streamflow predictions classified as drought below the 20th percentile. Higher values are better. U.S. Geological Survey NA No data U.S. Geological Survey -0.4104679 1 unitless bal_acc_10 Balanced accuracy of streamflow predictions classified as drought below the 10th percentile. Higher values are better. U.S. Geological Survey NA No data U.S. Geological Survey 0.3614865 1 unitless bal_acc_05 Balanced accuracy of streamflow predictions classified as drought below the 5th percentile. Higher values are better. U.S. Geological Survey NA No data U.S. Geological Survey 0.3807692 1 unitless bal_acc_20 Balanced accuracy of streamflow predictions classified as drought below the 20th percentile. Higher values are better. U.S. Geological Survey NA No data U.S. Geological Survey 0.0980392 1 unitless n_true_positive_10 Number of true positives of streamflow predictions classified as drought below the 10th percentile. U.S. Geological Survey NA No data U.S. Geological Survey 0 131 number n_true_positive_05 Number of true positives of streamflow predictions classified as drought below the 5th percentile. U.S. Geological Survey NA No data U.S. Geological Survey 0 91 number n_true_positive_20 Number of true positives of streamflow predictions classified as drought below the 20th percentile. U.S. Geological Survey NA No data U.S. Geological Survey 0 131 number n_true_negative_10 Number of true negatives of streamflow predictions not classified as drought below the 10th percentile. U.S. Geological Survey NA No data U.S. Geological Survey 24 181 number n_true_negative_05 Number of true negatives of streamflow predictions not classified as drought below the 5th percentile. U.S. Geological Survey NA No data U.S. Geological Survey 48 181 number n_true_negative_20 Number of true negatives of streamflow predictions not classified as drought below the 20th percentile. U.S. Geological Survey NA No data U.S. Geological Survey 0 181 number n_pred_drought_10 Number of streamflow predictions classified as drought below the 10th percentile. U.S. Geological Survey NA No data U.S. Geological Survey 0 133 number n_pred_drought_05 Number of streamflow predictions classified as drought below the 5th percentile. U.S. Geological Survey NA No data U.S. Geological Survey 0 92 number n_pred_drought_20 Number of streamflow predictions classified as drought below the 20th percentile. U.S. Geological Survey NA No data U.S. Geological Survey 0 175 number n_pred_nondrought_10 Number of streamflow predictions not classified as drought below the 10th percentile. U.S. Geological Survey NA No data U.S. Geological Survey 24 181 number n_pred_nondrought_05 Number of streamflow predictions not classified as drought below the 5th percentile. U.S. Geological Survey NA No data U.S. Geological Survey 48 181 number n_pred_nondrought_20 Number of streamflow predictions not classified as drought below the 20th percentile. U.S. Geological Survey NA No data U.S. Geological Survey 0 181 number n_obs_drought_10 Number of streamflow observations classified as drought below the 10th percentile. U.S. Geological Survey NA No data U.S. Geological Survey 0 131 number n_obs_drought_05 Number of streamflow observations classified as drought below the 5th percentile. U.S. Geological Survey NA No data U.S. Geological Survey 0 91 number n_obs_drought_20 Number of streamflow observations classified as drought below the 20th percentile. U.S. Geological Survey NA No data U.S. Geological Survey 0 131 number pct_obs_drought_10 Percentage of streamflow observations classified as drought below the 10th percentile. U.S. Geological Survey NA No data U.S. Geological Survey 0 0.7237569 percent pct_obs_drought_05 Percentage of streamflow observations classified as drought below the 5th percentile. U.S. Geological Survey NA No data U.S. Geological Survey 0 0.5031847 percent pct_obs_drought_20 Percentage of streamflow observations classified as drought below the 20th percentile. U.S. Geological Survey NA No data U.S. Geological Survey 0 0.7707006 percent n_obs_nondrought_10 Number of streamflow observations not classified as drought below the 10th percentile. U.S. Geological Survey NA No data U.S. Geological Survey 50 181 number n_obs_nondrought_05 Number of streamflow observations not classified as drought below the 5th percentile. U.S. Geological Survey NA No data U.S. Geological Survey 66 181 number n_obs_nondrought_20 Number of streamflow observations not classified as drought below the 20th percentile. U.S. Geological Survey NA No data U.S. Geological Survey 36 181 number rsq Coefficient of determination of streamflow predictions for models trained with all streamflow observations. Higher values are better. U.S. Geological Survey NA No data U.S. Geological Survey 3.6526404e-13 0.9954833 unitless rsq_below_30 Coefficient of determination of streamflow predictions for models trained with only streamflow observations below the 30th percentile. Higher values are better. U.S. Geological Survey NA No data U.S. Geological Survey 4.14245e-11 1 unitless rmse Root mean square error of streamflow predictions for models trained with all streamflow observations. Lower values are better. U.S. Geological Survey NA No data U.S. Geological Survey 2.4750472 62.658318 percentile rmse_below_30 Root mean square error of streamflow predictions for models trained with only streamflow observations below the 30th percentile. Lower values are better. U.S. Geological Survey NA No data U.S. Geological Survey 0.187983 76.5385696 percentile kge Kling-Gupta Efficiency of streamflow predictions for models trained with all streamflow observations. Higher values are better. U.S. Geological Survey NA No data U.S. Geological Survey -1.0160147 0.9883621 unitless kge_below_30 Kling-Gupta Efficiency of streamflow predictions for models trained with only streamflow observations below the 30th percentile. Higher values are better. U.S. Geological Survey NA No data U.S. Geological Survey -48.248948 0.9944032 unitless timeseries_data.feather Feather file containing observed and predicted streamflow percentile time series data. U.S. Geological Survey site_id Site identifier. USGS streamgage station identification number. U.S. Geological Survey NA No data U.S. Geological Survey USGS National Water Information System https://doi.org/10.5066/F7P55KJN reference_datetime Issue date (YYYY-MM-DD) of model forecast. U.S. Geological Survey NA No data U.S. Geological Survey 2000-10-02 2020-03-23 date datetime Valid date (YYYY-MM-DD) of model forecast. U.S. Geological Survey NA No data U.S. Geological Survey 2000-10-09 2020-03-30 date model_id Model identifier. U.S. Geological Survey NA No data U.S. Geological Survey LightGBM<30 Model identifier for gradient boosting machine models trained with only streamflow observations below the 30th percentile. U.S. Geological Survey LightGBM-all Model identifier for gradient boosting machine models trained with all observations. U.S. Geological Survey LSTM<30 Model identifier for long short-term memory neural network models trained with only streamflow observations below the 30th percentile. U.S. Geological Survey LSTM-all Model identifier for long short-term memory neural network models trained with all observations. U.S. Geological Survey ARIMA Model identifier for autoregressive integrated moving average models. U.S. Geological Survey persistence Model identifier for modeling no change between most recent observation and forecast date. U.S. Geological Survey observation Observed streamflow percentile. U.S. Geological Survey NA No data U.S. Geological Survey 0.0805802 99.9194198 percentile prediction Predicted streamflow percentile. U.S. Geological Survey NA No data U.S. Geological Survey 0 100 percentile variable Variable (target) used in model forecast. U.S. Geological Survey NA No data U.S. Geological Survey var Variable (target) used in LightGBM model forecast for the percentile of the 7-day rolling mean streamflow using the variable threshold method with a 30-day window of Julian dates. U.S. Geological Survey weibull_jd_30d_wndw_7d Variable (target) used in LSTM model forecast for the percentile of the 7-day rolling mean streamflow using the variable threshold method with a 30-day window of Julian dates. U.S. Geological Survey split Split of time series used in model forecast. U.S. Geological Survey NA No data U.S. Geological Survey test Split of time series used in model forecast for testing dates (from 2000-10-02 to 2020-03-30). U.S. Geological Survey duration Duration of model forecast. U.S. Geological Survey NA No data U.S. Geological Survey P1D Duration of model forecast for daily forecast. U.S. Geological Survey P1W Duration of model forecast for weekly forecast. U.S. Geological Survey family Family of probability distribution described by 'parameter'. U.S. Geological Survey NA No data U.S. Geological Survey summary Family of probability distribution described by 'parameter' for summary statistics. U.S. Geological Survey ensemble Family of probability distribution described by 'parameter' for ensemble of models. U.S. Geological Survey parameter Parameter of distribution specified in 'family'. U.S. Geological Survey NA No data U.S. Geological Survey pred_interv_05.0 Parameter of distribution specified in 'family' for 5th percentile of distribution. U.S. Geological Survey median Parameter of distribution specified in 'family' for 50th percentile of distribution. U.S. Geological Survey pred_interv_95.0 Parameter of distribution specified in 'family' for 95th percentile of distribution. U.S. Geological Survey 1 Parameter of distribution specified in 'family' for singular realization of ensemble. U.S. Geological Survey nday_forecast Number of days ahead of model forecast. U.S. Geological Survey NA No data U.S. Geological Survey 7 91 days U.S. Geological Survey - ScienceBase mailing

Denver Federal Center, Building 810, Mail Stop 302

Denver CO 80225 USA 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. Any use of trade, firm, or product names is for descriptive purposes only and does not imply endorsement by the U.S. Government. Digital Data https://doi.org/10.5066/P132NSWY None 20250919 Ryan R. McShane U.S. Geological Survey Hydrologist mailing

521 Progress Circle

Cheyenne WY 82007 USA 307-287-7434 rmcshane@usgs.gov FGDC Content Standard for Digital Geospatial Metadata FGDC-STD-001-1998