Pre critical instability
Description
Pre-critical instability is emerging in the global precipitation
Files
Steps to reproduce
Data generation workflow Monthly precipitation data from ten CMIP6 models were collected for the historical period and four SSP scenarios. Observational precipitation data from GPCP were used for comparison. All datasets were regridded to a common 5° × 5° grid and analyzed using overlapping 30-year moving windows. Within each window, the time series were detrended and normalized. Early-warning indicators (variance and lag-1 autocorrelation) were computed for each grid cell. Time-evolving precipitation networks were constructed using statistically significant correlations, from which network metrics were derived. Dynamical and structural indicators were combined into a composite instability index for subsequent analysis.