Meteorology, Climate and Python: desperately trying to forget technical details
summary
A Python-based framework replaces a legacy Fortran/C++ verification stack and exposes a user-facing directive language that validates keys, inherits defaults, and expands templates into multiple plots. The system glues C/C++ science libraries, decodes meteorological formats (GRIB/NetCDF/HDF), loads data from databases, applies bias corrections and ensemble handling, and runs a compute engine to produce scores. Custom metrics require subclassing two small methods. Output uses PL templates to generate consistent multi-domain plots; the framework is operational and accepted by users.
AI-written summary · about
chapters (16)
- 0:00Intro
- 0:26Background
- 2:58History
- 4:27Thoughts (2)
- 10:28Decisions (1)
- 11:51Choices (2)
- 12:32Purpose of the talk
- 12:53The language
- 13:34The directive
- 19:22Argument checking
- 22:20Directive specialisation
- 23:20Directive behaviour
- 26:34Plotting
- 28:07Document - Plot-Curve Model
- 40:30Gallery
- 40:54Conclusion
related talks
25mEmpowering weather-satellite data users with FOSS: the Pytroll case
33mPython for Water Forecasting Services
28mPython for Planet Earth: Climate Modeling and Sustainability.. with Drishti Jain
22mFaster, Smarter Climate Mitigation Using Python 🌳
29mModernizing Legacy: Wrapping a 25+ Year Computational Fluid Dynamics Codebase
43mTowards a full stack python monitoring+analysis framework
22mPython for bridging between researchers and service operators: from CFFI to Jupyterhub
01h 27mJupyter Notebooks for geospatial data analysis