CMIP6-CCAM assessment for ocean temperatures

Climate Projections and Services Team in Queensland Treasury provides dynamically downscaled climate simulations across Queensland. Prior to analysing future projections, it is critical to assess the performance of historical simulations and ensure that downscaled simulations produce realistic climate data. This repository contains scripts to process model outputs and assess the performance of dynamically downscaled historical ocean temperature simulations. In this project, the sixth phase of Coupled Model Intercomparison Project (CMIP6) global climate models (GCMs) were used as host models. Conformal Cubic Atmospheric Model (CCAM) was used to dynamically downscale CMIP6 GCMs. Analyses directory contains scripts to process model outputs (regrid, mask) and compute error evaluation and heat stress metrics, as well as a summary trend for ocean temperatures. Functions directory contains all python scripts required to execute Analysis scripts. The workflow below is iterated for CCAM, host CMIP6, bias-corrected CCAM, and bias-corrected CMIP6 simulations. Workflow : Pre-process and QC CCAM output (completed separately) Regrid Mask Model assessment PDF computation Perkins Skill Score calculation Heat stress assessment Degree heating week (DHW) calculation Summary Ocean temperature trend DHW trend Statistics and plotting (R)