Python for Water Forecasting Services
summary
Bureau systems run operational streamflow forecasts that ingest observations, fetch climate forecasts, run runoff models and publish results: a seasonal service covering ~150 gauges (plan 350) using MSDM & WAFARi with HDF5 via PyTables; a daily 7‑day service covering 62 catchments (+40 planned) wrapping a C++ model with CFFI into STAR & HRS style workflows; a 200+ gauge historical reference dataset and a 25‑catchment water‑quality prototype. Python glues Fortran/C++ models, uses NumPy/Pandas/Matplotlib/IPython and f2py for arrays, enables rapid prototyping and reproducible verification (200-site cross‑validation currently takes ~2 weeks on shared HPC). Python + NumPy/Pandas/Matplotlib/PyTables/IPython
AI-written summary · about
chapters (16)
- 0:00Introduction
- 0:40Overview
- 2:35Background
- 5:37Technical Solution
- 7:28Why Python
- 8:30Seasonal Streamflow Forecast
- 11:12Data Exchange
- 13:20Scientific Engineering
- 13:35Graphic Design
- 15:05Streamflow
- 15:55Swift
- 18:45Reference Station
- 19:45hrs Toolkit
- 20:47etfs
- 21:50Future work
- 24:15Summary
related talks
31mUsing Python to drive Australian water availability forecasting
28mDiving into the Deep End: A Python Journey in Water Management
21mAquawatch Data Systems a cloud solution for water quality monitoring and forecasting
22mPython for bridging between researchers and service operators: from CFFI to Jupyterhub
49mMeteorology, Climate and Python: desperately trying to forget technical details
26mFrom Data to Decision: Monitoring and Improving Water Networks using AI/ML
17mGoing with the flow? Apache Airflow for operational-quality scientific workflows
29mModernizing Legacy: Wrapping a 25+ Year Computational Fluid Dynamics Codebase