Since mid last century, the average global temperature has risen by 0.6°C (1.1°F), with more than 90% of this additional heat absorbed and stored in the oceans. While global warming is the main driver of overall rising sea levels, warming does not occur uniformly across the Earth, and this leads to differences in regional effects. In the Arctic, for example, temperatures have risen twice as fast as they have in the mid-latitudes, accelerating the melting of sea ice.

Sea ice is bright and reflective and radiates solar energy back into space, whereas the darker areas of the ocean absorb it. As sea ice disappears, less energy is reflected. This leads to an expansion of the energy-absorbing darker areas of the ocean which subsequently accelerates the melting of sea ice.

The ramifications of this cycle of increased energy absorption and sea ice loss are disturbing, with recent models predicting a significant decline in the total amount sea ice in the Arctic the coming decades. Models presented here also suggest there will be an increase in both El Niño and La Niña events in the coming decades. Specifically, this is linked to the predicted expansion of the Pacific and Indian ocean warmpool, an area of the ocean with temperatures exceeding 28°C (82.4°F).

Tropical cyclones are fueled by the energy accumulated in the upper layer of the ocean. At tropical latitudes, the abundance of convection and thunderstorms helps slow the rate of ocean warming by transporting heat away from the sea surface. As the amount of energy stored in the upper layer of the ocean increases, the frequency of tropical cyclones, and the energy dissipated by them, will be effected in ways that are still difficult to predict.

Sea ice is an important component of the climate system. A sea ice cover on the ocean, and insulates the ocean from heat loss. Regional climate changes affect the sea ice characteristics and these changes can in turn feed back on the climate system, both regionally and globally. Sea ice is also a major component of polar ecosystems; plants and animals at all levels of the food chain find a habitat in, or are associated with, sea ice.

Arctic sea ice cover varies seasonally, between 6×10⁶ km² (2.3 million square miles) in the summer and 15×10⁶ km² (5.8 million square miles) in the winter (Comiso and Nishio, 2008; Cavalieri and Parkinson, 2012; Meier et al., 2012). The summer ice cover is mainly confined to the Arctic Ocean basin and the Canadian Arctic Archipelago, while winter sea ice reaches as far south as 44°N, into the peripheral seas. At the end of summer, the Arctic sea ice cover primarily consists of the previously thick, old and ridged ice types that survived the melt period. Inter-annual variability is largely determined by the extent of the ice cover in the peripheral seas in winter and by the ice cover that survives the summer melt in the Arctic Basin.

The entire extent of sea ice has been monitored using over 34 years of satellite observation.

The figure on right shows Arctic sea ice area estimate from satellite observation (as provided by NSDIC, the National Snow and Ice Data Center). The black dots show the sea ice area in September (click in the figure legend, on “Monthly (NSDIC)” to display the complete annual cycles). The blue line shows the perennial sea ice extent (the minimum is observed in September, at the end of summer). This has a decreasing trend, estimated to be about 11.5% per decade.

The orange time series shows a possible evolution, under IPCC scenario RCP 8.5. The projection was computed by averaging all CMIP5 models, and considering sea ice for a non-null thickness. The projection indicates a continuous decreasing of the perennial sea ice in the period 2006-2100, with a possible loss of about 75% of perennial sea ice. However, considering the uncertainties attached to the models, such an estimate can not be considered as accurate.

The Indo-Pacific Warm Pool (IPWP), or Pacific and Indian ocean warm pool, is the world’s largest body of warm water, and with climate change it is growing larger. Stretching through large areas of the Indian and Pacific oceans, the pool is defined by average water temperatures over 28°C (81°F), the threshold for atmospheric deep convection. Above this temperature, the influence of the water body on atmospheric circulation is significant. Indeed, because of its size and warmth, the IPWP is able to influence the whole global climate, and serves as a major source of both heat and water vapor. There considerable evidence suggesting the pool is a large contributor to the El Niño phenomenon.

The size of the Indo-Pacific warm pool varies with the season, but since the 1980s it has seen continuous growth. The graphic on the left shows an estimate of the future size of the warm pool (as a proportion of the current area), under IPCC scenario RCP 8.5. According to this scenario, the total surface of the IPWP is expected to steadily grow – more than doubling in size by the end of the 2050s (this simulation suggests a total area in 2059 about 2.6 times larger than that of 2010).

The animation underneath shows the geographical projections for the pool from 2010 to 2059. The time evolution shows expansion along the equator East to the Americas and West to Africa. The projection also suggests that the Pacific equatorial regions could show comparable high temperature water. This simple approach needs to be confirmed with future research.

The ensemble mean of the models projections show a consistent increase of the global ocean temperature. The indicators hereafter show the sea surface temperature above the climatological reference (Reynolds), accumulated along an observation time of four months. Once this accumulated temperature exceed a threshold of 2°C, a threat flag is raised (a so-called "threat level 2" flag) indicating that a significant risk threatens living species.

The following illustration shows the evolution of thermal risk according to IPCC scenario 8.5. The risk is expressed in term of a frequency of occurrence per decade: from 1 times in a decade up to a risk returning each years of the decade (with a frequency of 10 years per decade).