Community Infrastructure for Building and Coupling Models

The Earth System Modeling Framework (ESMF) is high-performance, flexible software infrastructure for building and coupling weather, climate, and related Earth science applications. The ESMF defines an architecture for composing complex, coupled modeling systems and includes data structures and utilities for developing individual models.

The basic idea behind ESMF is that complicated applications should be broken up into coherent pieces, or components, with standard calling interfaces. In ESMF, a component may be a physical domain, or a function such as a coupler or I/O system. ESMF also includes toolkits for building components and applications, such as regridding software, calendar management, logging and error handling, and parallel communications.

• Download ESMF: "Software -> Download/Releases"
• Reference Manual: "Users -> User Docs-> Reference Manual"
• User Guide: "Users -> User Docs -> User Guide"
• Release Schedules: Software -> Release Schedules

The National Unified Operational Prediction Capability (NUOPC) and its research partners are working toward a common model architecture - a standard way of building models. The NUOPC Layer defines conventions and templates for using ESMF, and makes it simpler to adopt the framework. 

ESMF Component Lists

• View the ESMF components that follow the NUOPC Layer conventions - a collection called the Earth System Prediction Suite (ESPS)
• View all ESMF components

Grid Remapping

RegridWeightGen: ESMF has the option to build a standalone application that can generate interpolation weights in parallel. It's fast and can handle a wide variety of grids and grid options. It's not necessary to use the rest of ESMF to use it ... just provide grid files in a standard (netCDF-based) format.

ESMF's regridding capability is growing with every release. See the latest status.

ESMPy, the Python Interface to ESMF Grid Remapping

ESMPy is an easy way to use powerful ESMF grid remapping.  The software currently allows remapping on logically rectangular grids and meshes, and will soon support observational data streams.