Few studies have considered the detection of surface temperature trends at smaller regional scales or at the scale of individual model grid boxes. This is because detection of anthropogenic climate change is a signal-to-noise problem and the noise associated with natural variations of surface temperatures at regional scales is greater than at larger continental or global scales.
A brief review will be presented of detection and attribution studies of temperature changes at continental, sub-continental and regional scales. It will be shown that significant warming trends can be identified in the majority of regions across the globe and that these are consistent with the response to anthropogenic forcing, outside the range of natural internal climate variations and not consistent with the response to natural external forcing variations, such as changes in volcanic aerosols and solar irradiance. Recent observed trends in the major modes of variability of the large scale atmospheric circulation, including the NAM, COWL and SAM, do not affect this conclusion.
An example will be presenting using Central England temperature, which has the longest observed instrumental temperature record for any region. There is good agreement between the simulated variability of Central England Temperature (CET) by the HadCM3 climate model and its observed variability over the period 1700-1900. The observed warming in annual-mean CET of about 1.0°C since 1950 cannot be explained by natural climate variations and is consistent with the model response to increasing greenhouse gases and aerosols, demonstrating a significant human influence on this warming.
The IPCC conclusion in 2001 on the scale of detectable surface temperature changes is no longer correct.