How chris-hockey/hysplit-simulations-for-cegb-power-plants works
This repository is research code for generating annual HYSPLIT transport kernels for CEGB power-plant-years, with a calibration exercise used to choose plume-rise heat values. The main workflow prepares plant data and ERA5 meteorology, runs a seven-member calibration ensemble, evaluates its exposures against pollution observations, then runs production simulations. The README documents the broad stages; sampled source confirms key runner, simulation, kernel, and analysis responsibilities. ERA5 conversion and weather-analysis details are represented at their documented directory level because their implementations were not sampled.
Main entry points: sim_run.py Production runner · ensemble_run.py Ensemble runner · kernel.py Annual kernel · 1_ensemble_treatment.py Ensemble exposure
Components 19
Input preparation
- Clean plant panel
code/data_processing/power_plants/1_pp_cleaning.py - Impute stack heights
code/data_processing/power_plants/2_stack_pred.py - Download ERA5
code/data_processing/era5/1_api_calls.py - ERA5 to ARL
code/data_processing/era5
Plume calibration
- Ensemble runner
code/simulation/ensemble_run.py - Ensemble exposure
code/analysis/ensemble_runs/1_ensemble_treatment.py - Calibration assessment
code/analysis/ensemble_runs/2_ensemble_assessment.py - Weather representativeness
code/analysis/average_weather
Simulation runtime
- Parallel batch runner
code/simulation/hysplit/batch.py - Plant-year model
code/simulation/hysplit/plant.py - HYSPLIT run
code/simulation/hysplit/run.py - Annual kernel
code/simulation/hysplit/kernel.py
Production kernels
- Production runner
code/simulation/sim_run.py - Annual kernels
code/simulation/hysplit
Other components
- Researcher
- CEGB panel data
- Copernicus ERA5
- HYSPLIT engine
- Pollution observations
Connections 25
- Researcher → Clean plant panel: prepares data
README.md - CEGB panel data → Clean plant panel: cleans panel
README.md - Clean plant panel → Impute stack heights: builds panel
README.md - Researcher → Download ERA5: requests weather
README.md - Copernicus ERA5 → Download ERA5: provides fields
README.md - Download ERA5 → ERA5 to ARL: converts fields
README.md - Researcher → Ensemble runner: starts calibration
README.md - Ensemble runner → Parallel batch runner: dispatches jobs
code/simulation/ensemble_run.py - Ensemble runner → Plant-year model: constructs source
code/simulation/ensemble_run.py - Ensemble runner → Annual kernel: executes members
code/simulation/ensemble_run.py - Annual kernel → HYSPLIT run: wraps simulation
code/simulation/hysplit/kernel.py - HYSPLIT run → HYSPLIT engine: runs and converts
code/simulation/hysplit/run.py - Annual kernel → Annual kernels: writes kernels
code/simulation/hysplit/kernel.py - Ensemble exposure → Annual kernels: reads ensemble output
code/analysis/ensemble_runs/1_ensemble_treatment.py - Pollution observations → Ensemble exposure: supplies observations
README.md - Ensemble exposure → Calibration assessment: provides station panel
README.md - Calibration assessment → Researcher: selects heat values
README.md - Researcher → Weather representativeness: checks calibration year
README.md - Impute stack heights → Production runner: supplies plant panel
README.md - Researcher → Production runner: starts production
README.md - Production runner → Parallel batch runner: dispatches jobs
code/simulation/sim_run.py - Production runner → Plant-year model: constructs source
code/simulation/sim_run.py - Production runner → Annual kernel: executes plant-years
code/simulation/sim_run.py - Production runner → Annual kernels: produces annual kernels
README.md - ERA5 to ARL → HYSPLIT run: supplies ARL meteorology
README.md