Welcome to stitches!#
stitches is a python-based framework that provides users with a way to easily interact with Pangeo-hosted CMIP6 data to emulate the output variables of a target Earth System Model (ESM). Figure 1 demonstrates the stitches workflow.
To use stitches, there are a number of decisions users have to make, perhaps the most important being:
Which ESM will stitches emulate?
What scenario will be the target of the emulation?
Which available CMIP6 experiments, from the ESM to be emulated, will constitute the archive, i.e., the building blocks that stitches will use to construct the target scenario?
stitches works by connecting together short segments of existing ESM simulations, relying on the fact that most variables are not path dependent, i.e., have short memory. It works by matching existing pieces (9-year windows in our specific implementation) of global average temperature trajectories from different experiments available, to the target scenario’s global temperature trajectory, which has been divided up into pieces of the same length. Once the sequence of time windows-experiments in the archive is identified that matches the target, any variable in the ESM output available for those time windows-experiments whose evolution is not path dependent can be extracted and stitched together according to the same sequence.
Why do we need stitches?#
Impact research often requires many output variables from ESMs, including but not limited to global gridded temperature, precipitation, sea level pressure, relative humidity, and more. Impact research also often requires these values on a monthly or even daily time scale.
ESMs are expensive to run, often limiting the scenarios that can be explored and the number of ensemble members that can be generated for each scenario. stitches intelligently recombines the existing model runs available from ESMs in CMIP6 into global, gridded, multivariate outputs for novel scenarios (e.g., using a simple model to translate new forcing pathways into a global temperature trajectory that becomes stitches target). on monthly or daily timescales. stitches can also be used to enrich the ensemble sizes of existing scenarios. The resulting generated gridded multivariate outputs preserve the ensemble statistics of the ESM’s data. If the ESM has saved it, stitches can emulate the ESM output at any frequency: monthly, daily or even higher.
References#
@Article{esd-13-1557-2022,
AUTHOR = {Tebaldi, C. and Snyder, A. and Dorheim, K.},
TITLE = {STITCHES: creating new scenarios of climate model output by stitching together pieces of existing simulations},
JOURNAL = {Earth System Dynamics},
VOLUME = {13},
YEAR = {2022},
NUMBER = {4},
PAGES = {1557--1609},
URL = {https://esd.copernicus.org/articles/13/1557/2022/},
DOI = {10.5194/esd-13-1557-2022}
}
Documentation#
Getting Started
Python API
Contributing