Reduce the uncertainty in model projections of precipitation change
There are large uncertainties in model simulations of the current climate and projections of future climate. Bias correction methods are often used to reduce intermodel spread based on observations. Regarding precipitation over land, we find that only model simulations of the current climate is improved while uncertainties in future projections remain large even after bias correction.
- This figure shows the cross‐model spatial correlation of the climatology (x axis) versus the cross‐model spatial correlation of the precipitation change (y axis). More details can be found in Zhang and Soden (2019)
