• Average temperature mean - 12 months

• Year-so-far

12-month average temperature mean

Using a running 12-month average temperature mean keeps data can be kept up-to-date during the year.

The graphic explained

Overview

It shows 12-month running mean of global average temperatures from three datasets, plus data from the Niño region.

What the graphic shows

Met Office-CRU (blue line),  NCDC (black line),  GISS (orange line)

The grey shaded area shows the approximate 95% uncertainty range for the data. There is uncertainty because we don't have data from every region of the world and because of inaccuracies in some measurements. The true global average is expected to lie outside this range around 5% of the time. The uncertainty range has become less in more recent years as measurements have improved. It is calculated from the Met Office-CRU dataset.

The red line shows the time series of scaled sea-surface temperature anomalies in the Niño 3.4 region (170-120°W 5°S-5°N). The global temperature response to El Niño typically lags the Niño 3.4 series by a few months.

What is being observed - running mean

World getting warmer

Average global temperatures are now some 0.75 °C warmer than they were 100 years ago. Since the mid-1970s, average global temperatures have increased at a rate of more than 0.15 °C per decade. However, is clear that since 1980 the increase in global temperatures has not been steady. 

The limited surface observations we have, backed up by more comprehensive evidence from satellites, show that warming has been greatest in the Arctic during this period. The rapid retreat of sea-ice during this period is further evidence of this warming.

Still some uncertainty

Differences in data - caused by various factors, mean any estimate of warming based on surface observations is uncertain.

Differences in data

• GISS analysis has warmed somewhat more in the past decade than the Met Office-CRU and NCDC analyses.

• The warming in the Arctic as not reflected in the Met Office-CRU analysis as surface observations are sparse. It is reflected to a greater or lesser extent in the other analyses which fill in some of the gaps, thereby giving more weight to this warming.

Some causes of differences in data

Ice

Sea-surface temperatures are used to estimate air temperature over the oceans. Where there is sea-ice (such as in the Arctic) the temperature of the sea-surface can be very different from the temperature of the air above. If the sea-ice in a particular area melts, then the air temperature above the sea and ice might change much more than the temperature of the sea-surface itself.

El Niño and La Niña

The prominent 'spike' in temperatures around 1998 is associated with the strong El Niño that developed in 1997.

Volcanoes

Large tropical volcanic eruptions are important factors affecting global temperatures on short timescales of up to a few years, The drop in global temperature from around 1991-1995 is associated with the eruption of Mount Pinatubo.

Year so far

Global average temperature, year so far Enlarge

The same three datasets are used for the comparison of the calendar year. The graphic shows the global average temperatures for the year so far and for comparable periods in earlier years.