I might be missing something but do you know what data is used for the initial conditions in these models? Is it linked back to certain date? And if it's linked to a certain date, how often are the simulations updated? I hope this question makes sense.
I evaluated different downscaling approaches and reviewed current papers regarding downscaling CMIP data. Initially, I used quantile delta mapping (QDM) to preserve the distributional shape. However, this approach impacted the climate change signal. Hence I reverted to a simple linear downscaling using the ERA5-Land reanalysis as a reference. This simple form of downscaling in combination with BIAS correction ensures that average temperatures or precipitation sums are corrected without effecting the climate change signal.
Hi Patrick,
I might be missing something but do you know what data is used for the initial conditions in these models? Is it linked back to certain date? And if it's linked to a certain date, how often are the simulations updated? I hope this question makes sense.
Cheers,
Jacob
Any reference for the downscaling procedure?
Thanks, Alessandro
I evaluated different downscaling approaches and reviewed current papers regarding downscaling CMIP data. Initially, I used quantile delta mapping (QDM) to preserve the distributional shape. However, this approach impacted the climate change signal. Hence I reverted to a simple linear downscaling using the ERA5-Land reanalysis as a reference. This simple form of downscaling in combination with BIAS correction ensures that average temperatures or precipitation sums are corrected without effecting the climate change signal.