Skip to content

FACTS Output Types

Different steps of a FACTS experiment generate different types of projections. The key element of the FACTS framework is that any module can be run in a given step so long as it ingests and returns the expected objects. For example, a module run in the climate step must produce an Xarray xr.Dataset object with specific coordinates, dimensions and data varaibles, described below. On this page, you'll find a description of each type and references to provided functions to read and write these objects that you can adopt in your module.

Note

This page is in progress! Check back soon for more updates.

Climate Projections

Global mean sea-level (GMSL) Projections

Relative sea-level (RSL) Projections

Extreme sea-level (ESL) Projections