Delaware River Basin Probabilistic Daily Streamflow Reconstruction Ensemble and Water Systems Model Data 1945-2023 (Amestoy and Reed, In Review, EMS)

Overview This data repository contains reconstructed historical streamflow ensembles and water management simulation results from: Amestoy, T.J., & P.M. Reed. (2025). Integrated River Basin Assessment Framework Combing Probabilistic Streamflow Reconstruction, Bayesian Bias Correction, and Drought Storyline Analysis. (Submitted for Review) Environmental Modeling and Software. The repository contains dataset used to evaluate water supply vulnerability and drought risk in the Delaware River Basin (DRB) using a bias-corrected probabilistic streamflow reconstruction of daily streamflow from 1945-2023 and the Pywr-DRB water management model (Hamilton, Amestoy and Reed, 2024). Relevant Links: Historic Streamflow Reconstruction Repository Pywr-DRB GitHub repository and documentation] Data Details Temporal Range: 1945-2023 Spatial Range: Delaware River Basin (DRB) Units: Million Gallons per Day (MGD) unless otherwise specified Bias-Corrected Historic Streamflow Reconstruction Ensemble The drb_historic_streamflow_ensemble_data.zip file contains two HDF5 files, each containing 1,000 realizations of daily full natural streamflow at 33 Pywr-DRB node locations across the basin. gage_flow_obs_pub_nhmv10_BC_ObsScaled_ensemble.hdf5: Primary bias-corrected probabilistic QPPQ streamflow reconstruction ensemble used in Amestoy and Reed (in preparation). This dataset combines probabilistic non-exceedance probability sampling with Bayesian Additive Regression Trees (BART) bias correction of NHM-derived flow duration curves. gage_flow_obs_pub_nhmv10_ObsScaled_ensemble.hdf5: Alternative streamflow reconstruction ensemble generated using non-bias-corrected National Hydrologic Model (NHM v1.0) flow duration curves in ungauged catchments. This dataset is provided for comparison purposes. Pywr-DRB Data The pywrdrb_data.zip folder contains all input data needed to run the Pywr-DRB model using the bias-corrected streamflow ensemble, as well as the Pywr-DRB simulation results based on the bias-corrected probabilistic reconstruction input dataset. Pywr-DRB Input Data The pywrdrb_inputs/ directory contains all required data to run the Pywr-DRB model using the bias-corrected streamflow ensemble: historic_ensembles/ catchment_inflow_obs_pub_nhmv10_BC_ObsScaled_ensemble.hdf5: Marginal catchment inflow at each Pywr-DRB node, calculate by iteratively subtracting the upstream catchment inflows from gage_flow_obs_pub_nhmv10_BC_ObsScaled_ensemble.hdf5 historic_NYC/ Pep_Can_Nev_releases_daily_2000-2021.xlsx: Reservoir release timeseries for the NYC reservoirs, provided by ODRM. Pep_Can_Nev_diversions_daily_2000-2021.xlsx: NYC transbasin diversion timeseries, provided by ODRM. NYC_storage_daily_2000-2021.csv: Reservoir storage timeseries for the NYC reservoirs, provided by ODRM. sw_avg_wateruse_Pywr-DRB_Catchments.csv: Average water withdrawal and consumptive use rates for each Pywr-DRB nodes, derived and post-processed from DRBC (2020) water use data. deliveryNJ_DRCanal_extrapolated.csv deliveryNYC_ODRM_extrapolated.csv predicted_inflows_diversions_obs_pub_nhmv10_BC_ObsScaled_ensemble.hdf5 Pywr-DRB Outputs The pywrdrb_outputs/ directory contains the results of Pywr-DRB simulations driven by the bias-corrected streamflow reconstruction ensemble: pywrdrb_results_export_obs_pub_nhmv10_BC_ObsScaled_ensemble.hdf5: Simulation results organized by result type, model run ID, and ensemble member The simulation outputs can be loaded into a pywrdrb.Data object (see example code below) with the following hierarchical key structure: Results set key (attribute) Model run ID (str) Realization number (int, 0-999) Pywr-DRB output variables (pd.DataFrame) The simulation outputs are organized into the following "results set" categories with the following keys: major_flow: Daily streamflow at key locations of interest (nodes) in the basin res_storage: Daily storage volumes for all 17 reservoirs in the model reservoir_downstream_gage: Daily streamflow at gauges downstream of major reservoirs lower_basin_mrf_contributions: Daily releases from lower basin reservoirs to maintain the Trenton minimum flow target ibt_diversions: Daily transbasin diversions to New York City and New Jersey catchment_consumption: Daily consumptive water use within each node catchment ffmp_level_boundaries: Time-varying storage boundaries defining operational zones for the Flexible Flow Management Program inflow: Daily inflow to each model node nyc_release_components: Disaggregated daily releases from NYC reservoirs by type (minimum daily, directed Montague, Trenton equivalent flow, flood, and spill) res_release: Daily total releases from all reservoirs all_mrf: Daily minimum flow requirements at all relevant locations