Sorry Sawzaw,
Hate to be pedantic but your graph shows the ice is 2 standard deviations away from the average not 2%. That means that there is a 95% probability that the ice cover is below the average. See here for an explanation of the stats:
68-95-99.7 rule

You hear people using phrases like 'the change is statistically significant' and they are usually implying that the new data is more than 2 deviations away from the mean.

It looks like we might be in a situation at the moment where sea temperature is rising but the air temperature doesn't seem to be. That's a pain for the experimentalists as trying to get good average sea temperature data is really hard by all accounts!