SWAT+ Simulation

pySWATPlus executes SWAT+ simulations using the TxtInOut folder generated during project creation with QSWAT+.

To run a SWAT+ simulation, the TxtInOut folder must include all required input files, which are created during simulation setup in the SWAT+ Editor via QSWAT+.

Additionally, the TxtInOut folder must contain the SWAT+ executable. If you need help locating it, open the Run SWAT+ tab in the SWAT+ Editor to find its path. Once located, copy the executable file into the TxtInOut folder. If the executable is missing, the simulation will fail.

Once the TxtInOut folder is properly configured with the necessary input files and the SWAT+ executable, you can initialize the TxtinoutReader class to interact with the SWAT+ model:

import pySWATPlus

# Replace this with the path to your project's TxtInOut folder
txtinout_folder = r"C:\Users\Username\project\Scenarios\Default\TxtInOut"

txtinout_reader = pySWATPlus.TxtinoutReader(
    path=txtinout_folder
)

pySWATPlus offers multiple options for executing SWAT+ simulations, providing flexibility for diverse workflows and automation needs.

The following sections demonstrate different approaches, including modifying input parameters, running a single simulation, and performing batch processing.

Calibration Warning

The run_swat and run_swat_in_other_dir functions currently disable the use of the calibration.cal file.
Calibration via the calibration.cal file is still under development.

A Trial Simulation

This simple example runs a SWAT+ simulation without changing any input parameters, ensuring that the setup works correctly:

txtinout_reader.run_swat()

Simulations in a Separate Directory (Recommended)

To keep your original TxtInOut folder unchanged, it is recommended to run SWAT+ simulations in a separate directory. This approach allows you to safely modify parameters and input files without affecting your main project. The following steps demonstrate how to perform a simulation in a dedicated directory:

• Copy required files to the target directory:

  # Replance this with your desired simulation folder path
  target_dir = r"C:\Users\Username\simulation_folder_1" 
  
  # Ensure the required files are copied
  txtinout_reader.copy_required_files(
      target_dir=target_dir
  )
  

• Initialize TxtinoutReader for the target directory

  target_reader = pySWATPlus.TxtinoutReader(
      path=target_dir
  )
  

• Modify simulation timeline:

  # Update timeline in `time.sim` file
  target_reader.set_begin_and_end_year(
      begin=2012,
      end=2016
  )
  

• Set warm-up years:

  # Set warm-up years in `print.prt` file
  target_reader.set_warmup_year(
      warmup=1
  )
  

• Enable required simulation files:

  # Ensure simulation outputs for `channel_sd` object in `print.prt` file  
  target_reader.enable_object_in_print_prt(
      obj='channel_sd',
      daily=False,
      monthly=True,
      yearly=True,
      avann=True,
  )
  

• Modify esco parameter:

  # Register the `hydrology.hyd` file
  hyd_register = target_reader.register_file(
      filename='hydrology.hyd',
      has_units=False
  )
  
  # Get the DataFrame and modify the `esco` value
  hyd_df = hyd_register.df
  hyd_df['esco'] = 0.6
  
  # Overwrite the `hydrology.hyd` file
  hyd_register.overwrite_file()
  

• Run the SWAT+ simulation:

  # Another way to modify `esco` parameter and perform simulation
  params = {
      'hydrology.hyd': {
          'has_units': False,
          'esco': {'value': 0.6}
      }
  }
  target_reader.run_swat(
      params=params  # optional
  )
  

Run Configurable Simulations in a Single Function

Instead of performing each step separately as explained above, you can run a SWAT+ simulation in a separate directory by configuring all options at once using a single function:

# Configure input parameters
params = {
    'hydrology.hyd': {
        'has_units': False,
        'esco': {'value': 0.6}
    }
}

# Run SWAT+ simulation from the original `TxtInOut` folder
txtinout_reader.run_swat_in_other_dir(
    target_dir=r"C:\Users\Username\simulation_folder_2",  # mandatory
    params=params,  # optional
    begin_and_end_year=(2012, 2016),  # optional
    warmup=1,  # optional
    print_prt_control={'channel_sd': {'daily': False}}  # optional
)

Parallel SWAT+ Simulations

To run multiple SWAT+ simulations in parallel with modified parameters, the following example provides a quick guide to get started:

import multiprocessing
import tempfile

# Define multiple parameter sets
param_sets = [
    {
        'hydrology.hyd': {
            'has_units': False,
            'esco': [
                {'value': 0.6, 'change_type': 'absval'}
            ],
        }
    },
    {
        'plants.plt': {
            'has_units': False,
            'bm_e': [
                {'value': 30, 'change_type': 'absval', 'filter_by': 'name == "agrl"'},
                {'value': 15, 'change_type': 'absval', 'filter_by': 'name == "almd"'},
            ],
        }
    },
]

# Define the function
def run_simulation(params):

    with tempfile.TemporaryDirectory() as tmp_dir:
        reader = pySWATPlus.TxtinoutReader(
            path=txtinout_path
        )
        output = reader.run_swat_in_other_dir(
            target_dir=tmp_dir,
            params=params
        )

    return output


if __name__ == '__main__':
    # Run simulations in parallel
    with multiprocessing.Pool(processes=2) as pool:
        results = pool.map(run_simulation, param_sets)
    # Print output directories
    for path in results:
        print("Simulation directory:", path)