Fitted Parameters of Power Transformation for rainfall in Brazil

Power Transformation: Correcting bias in MIROC6 rainfall

The Climate System Monitoring and Modeling Laboratory (LAMMOC), part of Fluminense Federal University (UFF) has tried forecasting rainfall over Brazil using one of the most known process-based models: MIROC6.

Unlike other regions of Earch, MIROC6 resulted in unacceptable bias, leading the team to explore bias correction methods that best suit this specific use case.

Power Transformation (PT) was suggested as it was used in different research papers.

The first netCDF file (PT_all_grid.nc) hosts parameters of PT after fitting it to rainfall data for the period 1961 - 2014.

Those parameters were used to correct MIROC6 forcasted rainfall over the period 2015-2100 for each of the main Shared Socio-economic Pathways (SSP).

Both raw and corrected rainfall data can be extracted via ssp_corr_245.nc, ssp_corr_370.nc and ssp_corr_585.nc  for each of ssp2-4.5, ssp3-7.0 and ssp5-8.5 respectively.

Observed Rainfall data was obtained from https://doi.org/10.1002/joc.7731.

Raw MIROC6 data was generated within the LAMMOC lab.

The associated GitHub repository hosts a usage guideline as well as relevant information.