Cookies on this website
We use cookies to ensure that we give you the best experience on our website. If you click 'Continue' we'll assume that you are happy to receive all cookies and you won't see this message again. Click 'Find out more' for information on how to change your cookie settings.

© 2014. American Geophysical Union. All rights reserved. There is much evidence that precipitation responses to global warming involve wet regions becoming wetter and dry regions drier. This presents challenges for the interpretation of projections from general circulation models (GCMs) which have substantial biases in the location of precipitation features. While improving GCM simulated precipitation is the most desirable solution, adaptation and mitigation decisions must be made with the models already available. Many techniques have been developed to correct biases in grid point precipitation intensities, but few have been introduced to correct for location biases. Here, we describe a new technique for correcting the spatial and seasonal location of climatological precipitation features. We design this technique to respect the geometry of the problem (spherical spatial dimensions, with cyclic seasons), while conserving either precipitation intensities, or integrated precipitation amount. We discuss the mathematical basis of the technique and investigate its behaviour in different regimes. We find that the resulting warps depend smoothly on the most influential parameter, which determines the balance between smoothness and closeness of fit. We show that the technique is capable of removing more than half the RMS error in a model’s climatology, obtaining consistently better results when conserving integrated precipitation. To demonstrate the ability of the new technique to improve simulated precipitation changes, we apply our transformations to historical anomalies and show that RMS error is reduced relative to GPCP’s anomalies by approximately 10% for both types of warp. This verifies that errors in precipitation changes can be reduced by correcting underlying location errors in a GCM’s climatology.

Original publication

DOI

10.1002/2014JD022357

Type

Journal article

Journal

Journal of Geophysical Research

Publication Date

16/12/2014

Volume

119

Pages

13 - 369