On page 5 of the WG1 SPM the IPCC authors talk of "Widespread melting of snow" and claim that "Mountain glaciers and snow cover have declined on average in both hemispheres".

On the next page there is the following graphic for Northern Hemisphere. Note that this shows that snow cover has in fact increased since 1990. Data from Rutgers shows no reduction since the late 1980s.

More seriously, if you look at the caption carefully you can see that the IPCC have chosen to show snow cover for March-April.  No prizes for guessing why they chose this particular period.  If you look at the data for the winter period, there has been no decline in snow cover at all:



In the IPCC review process, (comments on the second-order draft, SPM-628) the Chinese Government queried this: "It is unclear why snow cover area in April is given here. Please state the reason why April snow cover is used". The response was not helpful: "Reasons for choice are left to the chapter. Now shows March-April average".  The US Government also queried this same point (comments SPM-641): "Why is April the only month mentioned? Recommend including a generalization of trends at other times of the year lest there be accusations of cherry-picking". This was also rejected. 

The IPCC do not show Southern hemisphere snow cover. In fact their claim regarding both hemispheres is false, see this research that shows no significant trend over 28 years.

The WG1 SPM claim is in fact inconsistent with the statements later in WG1. The summary of Chapter 4, page 339, says "In the Southern Hemisphere, the few long records or proxies mostly show either decreases or no changes in the past 40 years or more". Note how this has been distorted for the SPM statement. But further on in Chapter 4, it becomes clear that this statement itself is a distortion, since there is "A long-term increasing trend in the number of snow days" on the East side of the Andes and "no trend over the 1930 to 1985 period" in New Zealand (p 346).

Roger Pielke Sr discusses this error, pointing out that snow cover in the Northern Hemisphere has increased slightly since the 1980s.

Note that Northern hemisphere snow cover in January 2008 was the largest since records began in 1966, according to  NCDC data. This latest result invalidates the IPCC claims (although of course this data came out after AR4).

In summary, the IPCC have made a false claim to mislead policymakers, by cherrypicking data from only the Northern Hemisphere and only from the March-April period.