Data

Tools

Indexes

Applications

Experiments

Open Sustainable Technology

Recent Releases of Pywr-DRB

Pywr-DRB - Pywr-DRB v2.0.1

Overview

Pywr-DRB v2.0.1 adds package dependency versions to avoid installation issues, and contains minor bug fixes with the pywrdrb.Data() class functionality.

Package Dependencies

Recent updates to the pywr and scipy packages cause issues during installation.

We have updated the pywrdrb dependency list to include:

  • "pywr==1.27.4"
  • "scipy==1.15.3"

We specify versions across the dependency list, however the only necessary version requirements at the moment are for the two packages above.

Minor Bug Fixes and Updates

pywrdrb.Data() Updates

  • Modified the AbstractDataLoader class to support ensemble flow files in `AbstractDataLoader.get_base_results()
    • Previously, this only supported loading data from gage_flow_mgd.csv files
    • Now, it supports ensemble files (gage_flow_mgd.hdf5) for a list of realization_ids
  • Moved the flowtye_opts list creation into the __init__() of the HydrologicModelFlow() class. This is necessary to allow for custom inflow types to be recognized as a valid flowtype option when using:
  • Added "all" to the hydrologic_model_results_set_opts
    • This allows uses to load flows from all nodes (instead of just major_nodes) when running pywrdrb.Data().load_hydrologic_model_flow()
  • Modified the pywrdrb.Data().export() and pywrdrb.Data().load_from_export() to support custom data.
    • Previously, these functions only allowed for export/load of existing results_set options.
    • Now, users can add new/custom data to the pywrdrb.Data object, then export and re-load that data later.
    • See the example

Example: Modifying, exporting and re-loading data objects

Below is an example of how custom/new data can be added to the pywrdrb.Data class, and how the full data class can be exported to a new file and later reloaded.

First, load the Pywr-DRB output using the existing pywrdrb.Data functionality. See Tutorial 03 - Loading and Interpreting Output for more detail on this existing functionality.

## Existing functionality
import pywrdrb
data = pywrdrb.Data()
data.load_output(
    results_sets=['major_flow'],
    output_filenames=['./pywrdrb_output.hdf5']
)
df_flow = data.major_flow["pywrdrb_output"][0]

We can calculate new data using the loaded output data. In this example below, I'll do a simple 7-day mean flow calculation.

Then, we can store the newly calculated rolling_mean_major_flow inside the data object. Importantly, this addition must match the hierarchical data dictionary formatting.

After the data is stored in the data object, we can use the export() function.

The export function will save all attributes/contents of the data object to an HDF5 file with the given name.

## New functionality
# Calculate new metrics based on the original output data
df_flow_rolling = df_flow.rolling(window=7).mean()

# Store the new df in the data object
# Important: Format of dictionaries must match
data.rolling_mean_major_flow = {}
data.rolling_mean_major_flow["pywrdrb_output"] = {}
data.rolling_mean_major_flow["pywrdrb_output"][0] = df_flow_rolling

# Save the full data object as an export
# This will include all metrics, including the added rolling mean flow
data.export("./pywrdrb_output_with_postprocessing.hdf5")

Then, later we can re-load the modified data object using:

data = pywrdrb.Data()
data.load_from_export("./pywrdrb_output_with_postprocessing.hdf5")

Example: Loading custom hydrologic data using pywrdrb.Data()

The load_hydrologic_model_flow() function is designed to load <flowtype>/gage_flow_mgd.csv files. Thus, the resulting data is reflective of the full natural flow as modeled by the model/dataset being loaded. This flow is not routed through Pywr-DRB.

Previously, this funciton was unable to be used for custom flowtypes arguments, and was only able to load the datasets that came pre-packaged with pywrdrb installation.

The 2.0.1 patch fixes this, as shown below.

Importantly, this assumes that the custom data csv file has the same formatting as the existing pywrdrb datasets (node names as column names and datetime index) which is generally a pre-requisit for using custom data in the pywrdrb modeling workflow.

# Register custom dataset with pathnavigator
pn_config = pywrdrb.get_pn_config()
pn_config["flows/custom_data"] = "custom_data"
pywrdrb.load_pn_config(pn_config)

# List of result types to load
# these include all valid options given the input data
results_sets=['all', 'major_flow', 'reservoir_downstream_gage']

data = pywrdrb.Data()

# Previously, using flowtypes=['custom_data'] would give error
data.load_hydrologic_model_flow(flowtypes=["custom_data"],
                                results_sets=results_sets)

# The "all" results set is new, and returns gage_flow (full natural) at all nodes
data.all['custom_data'][0].head()

Hydrosphere - Freshwater and Hydrology - Python
Published by TrevorJA 10 months ago

Pywr-DRB - Pywr-DRB v2.0.0

Overview

Pywr-DRB v2.0.0 includes a major re-structure and modularization of the Pywr-DRB v1.0.2 (Hamilton et al., 2024), which emphasizes on improving the user interfaces for flexibility and ease of use. This version also greatly improves runtime efficiency and model fidelity by adding water temperature prediction and salt front location prediction capabilities while maintaining much of the core reservoir, diversion, regulatory and management functionality as the prior versions.

Please cite the release as

Lin, C.Y., Amestoy, T., Smith, M., Hamilton, A., & Reed, P. (2025). Pywr‑DRB v2.0.0 [Software]. Zenodo. https://doi.org/10.5281/zenodo.15659955

Links and References

Major Updates

1. Modularization & Packaging

  • Modularize the code and introduce new pywrdrb API, relative to that used in Hamilton et al. (2024) seeks to:

    • Improve standardization (e.g., file naming conventions, workflows, docstrings)
    • Add flexibility for running simulations with different scenarios, settings and options
    • Improve runtime efficiency
    • Improve code reliability (e.g., unit testing and more detailed package dependency specifications)
    • Introduce object-oriented interface:
    # v2.0.0 approach - object-oriented interface

mb = pywrdrb.ModelBuilder(
    inflow_type='nhmv10_withObsScaled',
    start_date="2020-01-01",
    end_date="2020-12-31"
)
mb.make_model()
mb.write_model("model.json")

# Load model
model = pywrdrb.Model.load("model.json")
recorder = pywrdrb.OutputRecorder(
    model=model,
    output_filename=output_filename,
    parameters=[p for p in model.parameters if p.name]
    )

# Execute the simulation
stats = model.run()

  • Adopts pyproject.toml to handle dependencies installation & package specifications that allows for simple pip install pywrdrb

  • All of the data is now stored in a standardized format within Pywr-DRB/src/pywrdrb/data/

Key modules & classes

alt text

  • DataPreprocessor classes:

    • Create necessary model inputs from the raw datasets (e.g., hydrological data).
    • Retrieve latest observed data (e.g., gauge flows and reservoir storages).
    • This module enable sustainable and customizable use of the model to run with latest information and user-provided inputs.

  • pywrdrb.path_manager:

    • Adopt pathnavigator for better path management that enable linking external custom datasets to pywrdrb.

  • pywrdrb.ModelBuilder:

    • Modularize the model building process for customizing simulation settings.
    • Write a model to JSON model file

  • pywrdrb.OutputRecorder:

    • Enable subset output variables to reduce memory burden.
    • Improve runtime speed by avoiding frequent IO and utilizing internal memory.
    • This enable future large-scale experiments.

  • pywrdrb.Data:

    • Convert internal variable names into user-friendly names.
    • Allow loading multiple output files
    • Enable loading subset of outputs to enhance loading speed.
    • This enable smooth comparison across simulation runs.

2. Improved policy representation with respect to Trenton flow target

  • Pywr-DRB v1.0.2 error incorrectly required NYC reservoirs to maintain the Trenton Equivalent Flow Objective at all times, which is inaccurate relative to the actual basin policy, which only requires NYC to maintain the Montague Flow Objective.
  • Pywr-DRB v2.0.0 correction: Implements the constrained Interim Excess Release Quantity (IERQ) Trenton flow bank as described by the Flexible Flow Management Program; NYC reservoirs have an annual allocation of 6.09 billion gallons for the Trenton Equivalent Flow Objective.
    • NYC reservoirs are required to maintain only the Montague flow target, as specified by the 1954 US Supreme Court Decree.
    • IERQRelease parameter (pywrdrb/parameters/banks.py) tracks bank usage and enforces constraints on NYC releases used for the Trenton Flow Objective. Bank resets each June 1st as specified in FFMP regulations
      • In the latest version, we limit the annual NYC releases for Trenton to 6.09BG
      • NOTE: The results presented in Amestoy and Reed (In Review) include the corrected IERQ operations for the Trenton Equivalent Flow Objective; NYC reservoirs are limited to a annual Trenton contribution of 6.09BG in those simulation results.

3. Enabled water temperature & salt front location prediction capabilities

  • Added custom parameters to couple LSTM models with PywrDRB-ML plug-in to predict 1) daily maximum water temperature at Lordville and 2) 7-day averaged salt front location in river mile.
  • The PywrDRB-ML plug-in is currently private pending publication, but may be available upon request.

Designed LSTMs' capability

Model From To
TempLSTM 1/1/1979 12/31/2023
SalinityLSTM 10/1/1963 12/31/2023

Temperature LSTM model

  • The temperature model is developed based on Zwart et al. (2023). In order to fit to the control purpose, we construct TempLSTM1 to predict the Cannonsville downstream gauge temperature (T_C) and TempLSTM2 to predict the East Branch flow and the natural flow to Lordville (T_i).
  • The final water temperature at Lordville (T_L) is calculated by mapping the average temperature (Tavg) to the maximum temperature (T_L) using a random forest model.
  • Introduce API to allow user input thermal control policy

Salinity LSTM model

  • The Salinity LSTM model is developed based on Gorski et al. (2024). We rebuild the model using the LSTM and BMI sturcture derived from Zwart et al. (2023) to predict 7-day averaged salt front location in river mile at each timestep.
  • Note that the salt front location has not yet be integrated into Montague and Trenton flow target operations. It is scheduled in the next release.

4. Expanded number of pre-packaged streamflow scenarios

  • The original code used in Hamilton et al. (2024) only supported Pywr-DRB simulations using 4 hydrologic inflow datasets:
    • NHM version 1.0
    • NHM version 1.0 with scaled observational inflows at some reservoirs
    • NWM version 2.1
    • NWM version 2.1 with scaled observational inflows at some reservoirs
  • The latest version adds the following hydrologic inflow scenarios:
    • WRF-Hydro simulated flow during the AORC period
    • WRF-Hydro simulated flow during the AORC period with scaled observational inflows at some reservoirs
    • WRF-Hydro simulated flow during the 1960s drought
    • WRF-Hydro simulated flow during the 1960s drought, under a +2C climate scenario
  • All of these inflow scenarios come pre-packaged with the pywrdrb source code, and are ready for simulation. The table below summarizes the scenarios, lists their scenario keys and provides their simulation periods.
Scenario Key Description Simulation Period Source New to Pywr-DRB v2.0?
"nhmv10" National Hydrologic Model v1.0 1983-10-01 to 2016-12-31 Hay & LaFontain (2020) No
"nhmv10_withObsScaled" NHM v1.0 with scaled observations 1983-10-01 to 2016-12-31 Hamilton et al. (2024) No
"nwmv21" National Water Model v2.1 1983-10-01 to 2016-12-31 Blodgett (2022) No
"nwmv21_withObsScaled" NWM v2.1 with scaled observations 1983-10-01 to 2016-12-31 Hamilton et al. (2024) No
"wrfaorc_calib_nlcd2016" WRF-Hydro AORC calibrated 1979-10-01 to 2021-12-31 NCAR^ Yes
"wrfaorc_withObsScaled" WRF-Hydro AORC calibrated with scaled Observations 1979-10-01 to 2021-12-31 NCAR^ Yes
"wrf1960s_calib_nlcd2016" WRF-Hydro 1960s drought 1959-10-01 to 1969-12-31 NCAR^ Yes
"wrf2050s_calib_nlcd2016" WRF-Hydro +2°C climate scenario 1959-10-01 to 1969-12-31* NCAR^ Yes
"pub_nhmv10_BC_withObsScaled" Reconstructed streamflow (median) 1945-01-01 to 2023-12-31 Amestoy & Reed (In Review) Yes
  • ^ National Center for Atmospheric Research. Publication unavailable.
  • * The "wrf2050s_calib_nlcd2016" scenario is meant to represent the drought of record (1960s) during a +2C warming scenario (2050). In the pywrdrb code, we adopt the 1959-1969 datetime index such that it is more easily compared to other simulation timeseries during the historic period.

5 Added support for customized datasets

  • The new code is designed to support custom streamflow datasets provided by the user.
  • This is done using the pywrdrb.pre module.
  • Users must have a single CSV file with full natural flow estimates at all model nodes, and have them labeled using the node labels.
  • Then, we provide the PredictedInflowPreprocessor to generate supporting input data required for the simulation.
  • See advanced tutorials for the detailed custom dataset workflows.

References

Hamilton, A. L., Amestoy, T. J., & Reed, P. M. (2024). Pywr-DRB: An open-source Python model for water availability and drought risk assessment in the Delaware River Basin. Environmental Modelling & Software, 181, 106185. https://doi.org/10.1016/j.envsoft.2024.106185

Amestoy, T. J. & Reed, P. M., (In Review) Integrated River Basin Assessment Framework Combining Probabilistic Streamflow Reconstruction, Bayesian Bias Correction, and Drought Storyline Analysis. Available at SSRN: https://ssrn.com/abstract=5240633 or http://dx.doi.org/10.2139/ssrn.5240633

Hay, L.E., and LaFontaine, J.H., 2020, Application of the National Hydrologic Model Infrastructure with the Precipitation-Runoff Modeling System (NHM-PRMS),1980-2016, Daymet Version 3 calibration: U.S. Geological Survey data release, https://doi.org/10.5066/P9PGZE0S.

Blodgett, D.L., 2022, National Water Model V2.1 retrospective for selected NWIS gage locations, (1979-2020): U.S. Geological Survey data release, https://doi.org/10.5066/P9K5BEJG.

Gorski, G., Cook, S., Snyder, A., Appling, A. P., Thompson, T., Smith, J. D.,
Warner, J. C., & Topp, S. N. (2024). Deep learning of estuary salinity dynamics is physically accurate at a fraction of hydrodynamic model computational cost. Limnology and Oceanography, 69(5), 1070–1085. https://doi.org/10.1002/lno.12549

Zwart, J. A., Oliver, S. K., Watkins, W. D., Sadler, J. M., Appling, A. P., Corson‐Dosch, H. R., ... & Read, J. S. (2023). Near‐term forecasts of stream temperature using deep learning and data assimilation in support of management decisions. JAWRA Journal of the American Water Resources Association, 59(2), 317-337.

Hydrosphere - Freshwater and Hydrology - Python
Published by philip928lin 12 months ago

Pywr-DRB - Pywr-DRB v2.0.0-beta

Overview

Pywr-DRB v2.0.0-beta is a pre-release version for a broader review before full v2.0.0 release.

Pywr-DRB v2.0.0-beta includes a major re-structure and modularization of the Pywr-DRB v1.0.2 (Hamilton et al., 2024), which emphasizes on improving the user interfaces for flexibility and ease of use. This version also greatly improves runtime efficiency and model fidelity by adding water temperature prediction and salt front location prediction capabilities while maintaining much of the core reservoir, diversion, regulatory and management functionality as the prior versions.

Links and References

Major Updates

1. Modularization & Packaging

  • Modularize the code and introduce new pywrdrb API, relative to that used in Hamilton et al. (2024) seeks to:

    • Improve standardization (e.g., file naming conventions, workflows, docstrings)
    • Add flexibility for running simulations with different scenarios, settings and options
    • Improve runtime efficiency
    • Improve code reliability (e.g., unit testing and more detailed package dependency specifications)
    • Introduce object-oriented interface:
    # v2.0.0 approach - object-oriented interface

mb = pywrdrb.ModelBuilder(
    inflow_type='nhmv10_withObsScaled',
    start_date="2020-01-01",
    end_date="2020-12-31"
)
mb.make_model()
mb.write_model("model.json")

# Load model
model = pywrdrb.Model.load("model.json")
recorder = pywrdrb.OutputRecorder(
    model=model,
    output_filename=output_filename,
    parameters=[p for p in model.parameters if p.name]
    )

# Execute the simulation
stats = model.run()

  • Adopts pyproject.toml to handle dependencies installation & package specifications that allows for simple pip install pywrdrb

  • All of the data is now stored in a standardized format within Pywr-DRB/src/pywrdrb/data/

Key modules & classes

alt text

  • DataPreprocessor classes:

    • Create necessary model inputs from the raw datasets (e.g., hydrological data).
    • Retrieve latest observed data (e.g., gauge flows and reservoir storages).
    • This module enable sustainable and customizable use of the model to run with latest information and user-provided inputs.

  • pywrdrb.path_manager:

    • Adopt pathnavigator for better path management that enable linking external custom datasets to pywrdrb.

  • pywrdrb.ModelBuilder:

    • Modularize the model building process for customizing simulation settings.
    • Write a model to JSON model file

  • pywrdrb.OutputRecorder:

    • Enable subset output variables to reduce memory burden.
    • Improve runtime speed by avoiding frequent IO and utilizing internal memory.
    • This enable future large-scale experiments.

  • pywrdrb.Data:

    • Convert internal variable names into user-friendly names.
    • Allow loading multiple output files
    • Enable loading subset of outputs to enhance loading speed.
    • This enable smooth comparison across simulation runs.

2. Improved policy representation with respect to Trenton flow target

  • Pywr-DRB v1.0.2 error incorrectly required NYC reservoirs to maintain the Trenton Equivalent Flow Objective at all times, which is inaccurate relative to the actual basin policy, which only requires NYC to maintain the Montague Flow Objective.
  • Pywr-DRB v2.0.0 correction: Implements the constrained Interim Excess Release Quantity (IERQ) Trenton flow bank as described by the Flexible Flow Management Program; NYC reservoirs have an annual allocation of 6.09 billion gallons for the Trenton Equivalent Flow Objective.
    • NYC reservoirs are required to maintain only the Montague flow target, as specified by the 1954 US Supreme Court Decree.
    • IERQRelease parameter (pywrdrb/parameters/banks.py) tracks bank usage and enforces constraints on NYC releases used for the Trenton Flow Objective. Bank resets each June 1st as specified in FFMP regulations
      • In the latest version, we limit the annual NYC releases for Trenton to 6.02BG
      • NOTE: The results presented in Amestoy and Reed (In Review) include the corrected IERQ operations for the Trenton Equivalent Flow Objective; NYC reservoirs are limited to a annual Trenton contribution of 6.09BG in those simulation results.

3. Enabled water temperature & salt front location prediction capabilities

  • Added custom parameters to couple LSTM models with PywrDRB-ML plug-in to predict 1) daily maximum water temperature at Lordville and 2) 7-day averaged salt front location in river mile.
  • The PywrDRB-ML plug-in is currently private pending publication, but may be available upon request.

Designed LSTMs' capability

Model From To
TempLSTM 1/1/1979 12/31/2023
SalinityLSTM 10/1/1963 12/31/2023

Temperature LSTM model

  • The temperature model is developed based on Zwart et al. (2023). In order to fit to the control purpose, we construct TempLSTM1 to predict the Cannonsville downstream gauge temperature (T_C) and TempLSTM2 to predict the East Branch flow and the natural flow to Lordville (T_i).
  • The final water temperature at Lordville (T_L) is calculated by mapping the average temperature (Tavg) to the maximum temperature (T_L) using a random forest model.

Salinity LSTM model

  • The Salinity LSTM model is developed based on Gorski et al. (2024). We rebuild the model using the LSTM and BMI sturcture derived from Zwart et al. (2023) to predict 7-day averaged salt front location in river mile at each timestep.
  • Note that the salt front location has not yet be integrated into Montague and Trenton flow target operations.

4. Expanded number of pre-packaged streamflow scenarios

  • The original code used in Hamilton et al. (2024) only supported Pywr-DRB simulations using 4 hydrologic inflow datasets:
    • NHM version 1.0
    • NHM version 1.0 with scaled observational inflows at some reservoirs
    • NWM version 2.1
    • NWM version 2.1 with scaled observational inflows at some reservoirs
  • The latest version adds the following hydrologic inflow scenarios:
    • WRF-Hydro simulated flow during the AORC period
    • WRF-Hydro simulated flow during the AORC period with scaled observational inflows at some reservoirs
    • WRF-Hydro simulated flow during the 1960s drought
    • WRF-Hydro simulated flow during the 1960s drought, under a +2C climate scenario
  • All of these inflow scenarios come pre-packaged with the pywrdrb source code, and are ready for simulation. The table below summarizes the scenarios, lists their scenario keys and provides their simulation periods.
Scenario Key Description Simulation Period Source New to Pywr-DRB v2.0?
"nhmv10" National Hydrologic Model v1.0 1983-10-01 to 2016-12-31 Hay & LaFontain (2020) No
"nhmv10_withObsScaled" NHM v1.0 with scaled observations 1983-10-01 to 2016-12-31 Hamilton et al. (2024) No
"nwmv21" National Water Model v2.1 1983-10-01 to 2016-12-31 Blodgett (2022) No
"nwmv21_withObsScaled" NWM v2.1 with scaled observations 1983-10-01 to 2016-12-31 Hamilton et al. (2024) No
"wrfaorc_calib_nlcd2016" WRF-Hydro AORC calibrated 1979-10-01 to 2021-12-31 NCAR^ Yes
"wrfaorc_withObsScaled" WRF-Hydro AORC calibrated with scaled Observations 1979-10-01 to 2021-12-31 NCAR^ Yes
"wrf1960s_calib_nlcd2016" WRF-Hydro 1960s drought 1959-10-01 to 1969-12-31 NCAR^ Yes
"wrf2050s_calib_nlcd2016" WRF-Hydro +2°C climate scenario 1959-10-01 to 1969-12-31* NCAR^ Yes
"pub_nhmv10_BC_withObsScaled" Reconstructed streamflow (median) 1945-01-01 to 2023-12-31 Amestoy & Reed (In Review) Yes
  • ^ National Center for Atmospheric Research. Publication unavailable.
  • * The "wrf2050s_calib_nlcd2016" scenario is meant to represent the drought of record (1960s) during a +2C warming scenario (2050). In the pywrdrb code, we adopt the 1959-1969 datetime index such that it is more easily compared to other simulation timeseries during the historic period.

5 Added support for customized datasets

  • The new code is designed to support custom streamflow datasets provided by the user.
  • This is done using the pywrdrb.pre module.
  • Users must have a single CSV file with full natural flow estimates at all model nodes, and have them labeled using the node labels.
  • Then, we provide the PredictedInflowPreprocessor to generate supporting input data required for the simulation.
  • See advanced tutorials for the detailed custom dataset workflows.

References

Hamilton, A. L., Amestoy, T. J., & Reed, P. M. (2024). Pywr-DRB: An open-source Python model for water availability and drought risk assessment in the Delaware River Basin. Environmental Modelling & Software, 181, 106185. https://doi.org/10.1016/j.envsoft.2024.106185

Amestoy, T. J. & Reed, P. M., (In Review) Integrated River Basin Assessment Framework Combining Probabilistic Streamflow Reconstruction, Bayesian Bias Correction, and Drought Storyline Analysis. Available at SSRN: https://ssrn.com/abstract=5240633 or http://dx.doi.org/10.2139/ssrn.5240633

Hay, L.E., and LaFontaine, J.H., 2020, Application of the National Hydrologic Model Infrastructure with the Precipitation-Runoff Modeling System (NHM-PRMS),1980-2016, Daymet Version 3 calibration: U.S. Geological Survey data release, https://doi.org/10.5066/P9PGZE0S.

Blodgett, D.L., 2022, National Water Model V2.1 retrospective for selected NWIS gage locations, (1979-2020): U.S. Geological Survey data release, https://doi.org/10.5066/P9K5BEJG.

Gorski, G., Cook, S., Snyder, A., Appling, A. P., Thompson, T., Smith, J. D.,
Warner, J. C., & Topp, S. N. (2024). Deep learning of estuary salinity dynamics is physically accurate at a fraction of hydrodynamic model computational cost. Limnology and Oceanography, 69(5), 1070–1085. https://doi.org/10.1002/lno.12549

Zwart, J. A., Oliver, S. K., Watkins, W. D., Sadler, J. M., Appling, A. P., Corson‐Dosch, H. R., ... & Read, J. S. (2023). Near‐term forecasts of stream temperature using deep learning and data assimilation in support of management decisions. JAWRA Journal of the American Water Resources Association, 59(2), 317-337.

Hydrosphere - Freshwater and Hydrology - Python
Published by philip928lin about 1 year ago

Pywr-DRB - Code and Data for Pywr-DRB paper (Hamilton et al, 2024, EMS)

This release contains all code needed to create the analysis and figures in the following paper:

Hamilton, A.L., Amestoy, T.J., & P.M. Reed. (2024). Pywr-DRB: An open-source Python model for water availability and drought risk assessment in the Delaware River Basin. (In Review) Environmental Modeling and Software.

Hydrosphere - Freshwater and Hydrology - Python
Published by ahamilton144 almost 2 years ago

Pywr-DRB - Version for initial journal submission

This release contains all code needed to create the analysis and figures in the following paper:

Hamilton, A.L., Amestoy, T.J., & P.M. Reed. (2024). Pywr-DRB: An open-source Python model for water availability and drought risk assessment in the Delaware River Basin. (In Review).

Hydrosphere - Freshwater and Hydrology - Python
Published by ahamilton144 over 2 years ago