A Dependent Multi-model Approach to Climate Prediction with Gaussian Processes

Authors: Marten Thompson, Dr. Amy Braverman, Dr.Snigdhansu Chatterjee

Manuscript, data, and code associated with submission to Climate Informatics 2022.

Code

cmip6_netcdf_simplified.R creates the de-seasonalized time series. analysis.py estimates and examines our model to the observed data and CMIP6 simulations. analysis_loo.py performs the Leave-One-Out analysis described in the manuscript. These two scripts call methods defined in cmip6_eb_funcs.py. gpre_vis.R creates figures and calculates summary statistics. Running any of this code will overwrite existing contents.

Data

This work would not be possible without free and ready access granted by many authors to their data. Please see the University of East Anglia page for license and references pertaining to HadCRUT5 observed data and seasonal corrects, in particular the citations below:

• Jones, Phil D. et al.Surface air temperature and its changes over the past 150 years. https://crudata.uea.ac.uk/cru/data/temperature/abs_glnhsh.txt. Accessed 2021/11/29. 1999.

• Morice, Colin P. et al. “An Updated Assessment of Near-Surface Temperature Change From 1850: The HadCRUT5 DataSet”. Journal of Geophysical Research: Atmospheres 126.3 (2021), e2019JD032361.

We are similarly thankful for access to several CMIP6 simulations. Please see the manuscript for a full set of citations as well as

• Li, L. (2019). Cas fgoals-g3 model output prepared for cmip6 scenariomip ssp585.
• Rong, X. (2019). Cams cams-csm1.0 model output prepared for cmip6 scenariomip470ssp585.
• Semmler, T., Danilov, S., Rackow, T., Sidorenko, D., Barbi, D., Hegewald, J., Pradhan,472H. K., Sein, D., Wang, Q., and Jung, T. (2019). Awi awi-cm1.1mr model outputprepared for cmip6 scenariomip ssp585.