The ISI-MIP approach presents a trend-preserving bias correction method for climate model data used in impact simulations. This method ensures that the warming signal is preserved, particularly for temperature trends and relative changes in precipitation and other variables. The approach modifies the transfer function method used in Water-MIP, adjusting both monthly means and daily variability. The method preserves the long-term trend in simulated data, which is crucial for maintaining consistency between global temperature changes and land-based impact models. It addresses limitations in adjusting variability, especially for small-scale features and extremes, while ensuring statistical consistency between climate variables. The method uses additive or multiplicative correction factors to adjust mean values and variability, preserving the relative change in precipitation with respect to temperature. The approach is applied to five CMIP5 models and uses observational data from the WATCH Forcing Data to bias-correct climate variables. The method includes steps for correcting monthly means and daily variability, with adjustments for dry days and precipitation intensity. The bias-corrected data is used to synthesize impact projections across multiple sectors under different levels of global warming. The method ensures that the long-term trend is preserved, which is essential for accurate climate impact assessments. The approach has been validated against observational data, showing improvements in matching statistical properties of simulated data. The method is applied to the HadGEM2-ES model, demonstrating its effectiveness in preserving the trend and adjusting variability. The approach is designed to maintain consistency between simulated and observed data, ensuring reliable climate impact projections.The ISI-MIP approach presents a trend-preserving bias correction method for climate model data used in impact simulations. This method ensures that the warming signal is preserved, particularly for temperature trends and relative changes in precipitation and other variables. The approach modifies the transfer function method used in Water-MIP, adjusting both monthly means and daily variability. The method preserves the long-term trend in simulated data, which is crucial for maintaining consistency between global temperature changes and land-based impact models. It addresses limitations in adjusting variability, especially for small-scale features and extremes, while ensuring statistical consistency between climate variables. The method uses additive or multiplicative correction factors to adjust mean values and variability, preserving the relative change in precipitation with respect to temperature. The approach is applied to five CMIP5 models and uses observational data from the WATCH Forcing Data to bias-correct climate variables. The method includes steps for correcting monthly means and daily variability, with adjustments for dry days and precipitation intensity. The bias-corrected data is used to synthesize impact projections across multiple sectors under different levels of global warming. The method ensures that the long-term trend is preserved, which is essential for accurate climate impact assessments. The approach has been validated against observational data, showing improvements in matching statistical properties of simulated data. The method is applied to the HadGEM2-ES model, demonstrating its effectiveness in preserving the trend and adjusting variability. The approach is designed to maintain consistency between simulated and observed data, ensuring reliable climate impact projections.