Contributions to sea level rise from 1961 to 2003. In principle, it should be possible to add up each of the individual components of sea level rise—melting continental ice sheets in Antarctica and Greenland, retreating glaciers, the thermal expansion of near-surface water, thermal expansion of the deep ocean, and changes in water storage on land—to calculate the total rise over time. Unfortunately, early attempts to balance the sea level budget never added up. Each line on this graph shows how many millimeters each process added to or subtracted from total sea level since the early 1960s. (Graph adapted from Domingues 2008.) http://earthobservatory.nasa.gov/Features/OceanCooling/page5.php
The same flaws in the XBT data that affected Willis’ ocean heat maps showed up in the long-term historical trend (light blue). After applying a correction, the historical record shows a relatively steady increase in line with what’s shown by climate models. The remaining short-term variability is as likely to be natural variation, such as El Niño, as noise in the data. (Graph by Robert Simmon, based on data from GODAR.) http://earthobservatory.nasa.gov/Features/OceanCooling/page4.php
Willis’ map of ocean temperature change from 2004 to 2006 originally showed drops of over 1.5° Celsius in the Atlantic Ocean. The apparent large drop in temperature was due to bad data from the Argo floats and XBTs, and it disappeared when errors in these data sets were corrected. (The remaining large swings in temperature visible in these maps are due to shifting positions of ocean currents.) (Maps by Robert Simmon, based on data from Josh Willis and John Lyman.) http://earthobservatory.nasa.gov/Features/OceanCooling/page3.php
Graph of ocean heat storage compared to energy flux at the top of the atmosphere from 1994 through 2006. From 1993 to 2003, measurements of heat storage in the oceans agreed with satellite observations of net flux. After 2003, however, surface observations suggested that the ocean was losing heat, while satellite measurements of net flux showed the Earth was still slowly gaining energy. This mismatch was a hint that there might be a problem with one of the data sets. (Graph courtesy Takmeng Wong, NASA Langley Research Center.) http://earthobservatory.nasa.gov/Features/OceanCooling/page2.php
Maps of change in heat content in the oceans from 1993 through 2003, and 2003 through 2005. Josh Willis and his colleagues concluded that the world’s oceans gained heat in the decade from 1993 to 2003 (top). However, a follow-up study for the years 2003 to 2005 showed a surprisingly large decrease in heat content—about 5 times as large as the previous decade’s warming (bottom). Areas that warmed are red, while areas that cooled are blue. Note that the scale of each map is different: the 1993–2003 map ranges from -12 to +12 watts per square meter, while the 2003–2005 map ranges from -60 to 60. http://earthobservatory.nasa.gov/Features/OceanCooling/page1.php
Global climate anomalies during Heinrich events and between Dansgaard-Oeschger events. Compilation by Overpeck and Cole (2006).