Longer sets of data exist that are not global in scale. The longest is the Central England Temperature set, going back to 1659. https://en.wikipedia.org/wiki/C

Longer sets of data exist that are not global in scale. The longest is the Central England Temperature set, going back to 1659. https://en.wikipedia.org/wiki/Central_England_temperature https://en.wikipedia.org/wiki/Central_England_temperature Beyond that, people have attempted reconstruction from other evidence. For instance I found an attempt at a 2000 year reconstruction using tree ring data here: http://www.ncdc.noaa.gov/paleo/pubs/moberg2005/moberg2005.html http://www.ncdc.noaa.gov/paleo/pubs/moberg2005/moberg2005.ht... Assuming this reconstruction and others is sound, it is apparent that there definitely is a fair bit of natural temperature variability. But the recent temperature rise in the last 50 years certainly seem like aberration, in terms of rate of change.