Our methodology was inspired by the need for a more thorough data set that addressed many of the major concerns raised by global warming skeptics.
A detailed, published review of our complete methodology can be found here.
From 2010-2012, Berkeley Earth systematically addressed the five major concerns that global warming skeptics had identified, and did so in a scientific and objective manner.
The first four of these concerns were potential biases from data selection, data adjustment, poor station quality, and the urban heat island effect. Berkeley Earth’s analysis showed that these issues did not unduly bias the record.
The fifth concern related to the over-reliance on large and complex global climate models by the Intergovernmental Panel on Climate Change (IPCC) in the attribution of the recent temperature increase to anthropogenic forcings.
We obtained a long and accurate record, spanning 250 years, demonstrating that it could be well-fit with a simple model that included a volcanic term and, as an anthropogenic proxy, CO2 concentration. Through our rigorous analysis, we were able to conclude that the record could be reproduced by just these two contributions, and that inclusion of direct variations in solar intensity did not contribute to the fit.
The Berkeley Earth group has developed a new mathematical framework for producing maps and large-scale averages of temperature changes from weather station data for the purposes of climate analysis. The details of the mathematical framework used are described in the Methods paper available here.
The Berkeley Earth mathematical framework allows one to include short and discontinuous temperature records, so that nearly all temperature data can be used. The framework contains a weighting process that assesses the quality and consistency of a spatial network of temperature stations as an integral part of the averaging process. This permits data with varying levels of quality to be used without compromising the accuracy of the resulting reconstructions. The Berkeley Earth averaging process presented is extensible to spatial networks of arbitrary density (or locally varying density) while maintaining the expected spatial relationships.