The Washington Post picked up on the latest update to the 2005 temperature anomaly analysis from NASA GISS. The 2005 Jan-Sep land data (which is adjusted for urban biases) is higher than the previously warmest year (0.76°C compared to the 1998 anomaly of 0.75°C for the same months, and a 0.71°C anomaly for the whole year) , while the land-ocean temperature index (which includes sea surface temperature data) is trailing slightly behind (0.58°C compared to 0.60°C Jan-Sep, 0.56°C for the whole of 1998). The GISS team (of which I am not a part) had predicted that it was likely the 2005 would exceed the 1998 record (when there was a very large El Niño at the beginning of that year) based on the long term trends in surface temperature and the estimated continuing large imbalance in the Earth’s radiation budget.
In 1998 the last three months of the year were relatively cool as the El Niño pattern had faded. For the 2005 global land-ocean index to exceed the annual 1998 record, the mean anomaly needs to stay above 0.51°C for the next three months. Since there was no El Niño this year, and the mean so far is significantly above that, this seems likely.
Will a new record by a few hundredths of a degree really mean much? The important climate trends aren’t based on individual years, but on the underlying trends which have been solidly positive for decades. We still don’t expect each year to be warmer than the last due to the intrinsic variability (‘weather’) in global mean temperature (around 0.1 to 0.2°C), but at the current rate of global warming (~0.17°C/decade), new records can be expected relatively frequently. Stay tuned for further stories on this…
Update: The CRU/Met Office numbers are slightly different from the GISS analysis, but one should be careful to compare like with like. The 2005 Jan-Aug land anomaly from CRU is 0.81°C compared to 0.84°C for the same period in 1998. Their Sep update is due on the 26th, and so comparisons should become easier then.