Open chrisroadmap opened 1 year ago
Training data is from CMIP6 models, where the lowest model has a TCR of 1.33C (https://www.ipcc.ch/report/ar6/wg1/downloads/report/IPCC_AR6_WGI_FGD_Chapter07_SM.pdf)
IPCC assesses 5th percentile of TCR to be 1.2C. We need to extend the prior at the low end to sample more low TCR values.
5th percentile of reweighted posterior is 1.28C. Don't know if this is worth the extra effort to nudge down to 1.2C
this is a tough one - kick can down road
Training data is from CMIP6 models, where the lowest model has a TCR of 1.33C (https://www.ipcc.ch/report/ar6/wg1/downloads/report/IPCC_AR6_WGI_FGD_Chapter07_SM.pdf)
IPCC assesses 5th percentile of TCR to be 1.2C. We need to extend the prior at the low end to sample more low TCR values.