The potential to narrow uncertainty in regional climate predictions

Ed Hawkins and Rowan Sutton, NCAS-Climate & Walker Institute, University of Reading

| Home | FAQ |

Frequently Asked Questions (FAQ)

1. Can I read more details?
Yes, both papers and figures for use in talks are available (BAMS, Climate Dynamics).

2. Who can I contact for more information or to make comments?
Please email Ed Hawkins (e.hawkins AT

3. Are these images copyright?
The images on these pages may be used as long as they are accompanied by the appropriate citation:
Hawkins & Sutton, 2009, BAMS
Hawkins & Sutton, 2010, Clim. Dyn.

4. What types of plot are available?
As described in the temperature paper we have constructed various ways of visualizing the sources of uncertainty:

(a) On the 'Regions' pages there are:
Fractional variance - the fractions of the total variance contained in each source of uncertainty
Relative error - shows the uncertainty divided by the mean prediction
Total error - shows the absolute values of the sources of uncertainty
Total variance - shows the absolute variances of the sources of uncertainty, stacked.
Signal to noise - shows the mean prediction divided by the total uncertainty
The slider at the top changes the period of time meaning applied to the temperature predictions. The region can be chosen on the left hand side, and this will be highlighted pink in the map.

(b) On the 'Maps' pages you can choose the source of uncertainty in annual, pentadal or decadal mean surface temperatures and show:
Fraction of total variance
Absolute value of uncertainty
Total signal to noise
The lead time can be chosen using the slider at the top. For precipitation, different seasons can be selected, but only decadal means are available.

5. Has this analysis been performed on other variables?
Similar analyses to quantify the uncertainties have been performed on Amazonian dieback projections and ozone projections.