They shouldn't have been surprised. Below is a plot of actual monthly temperatures and the trends from the Hadley global data set (HADCRUT3v) and University of Alabama satellite derived lower tropospheric temperatures covering the same period as the robots measured ocean heat content. Like the robots they show a downtrend (cooling).
This during a time when atmospheric CO2 has been rising. Below is a plot of global temperatures from February 1998 to February 2008 from satellite (UAH MSU lower troposphere) (blue) and land and ocean variance adjusted surface (Hadley CRU T3v) (rose) plotted with Scripps monthly CO2 from Mauna Loa in Hawaii(green).
As you can see, the temperature has fallen since 1998 while CO2 has been steadily rising.
Dr. Roger Pielke, Senior Research Scientist at the Cooperative Institute for Research in Environmental Sciences (CIRES) at the University of Colorado, has some comments on his blog about the NPR story:
This is denial of the obvious. The observed absence of heat accumulation (of Joules) in the upper ocean (and in the troposphere) for the last four years means that there has been NO global warming in these climate metrics during this time period. It is unknown whether this is a short term aberration but, regardless, it is clear that the IPCC models have failed to skillfully predict this absence of warming. That should have been the conclusion stated at the end of the NPR story.If this is the case, why are we rushing into economy-crushing measures to stop man-made global warming that may or may not even exist?