Study: UN Global Warming Forecast Violates Accepted Principles

NCPA Study Reveals Forecasting Flaws That Make Conclusions Unreliable


DALLAS (February 1, 2008) - Predictions of melting ice caps, catastrophic sea level rise and severe floods and droughts are the result of a United Nation's report that violates nearly half of accepted forecasting principles, according to a new study published by the National Center for Policy Analysis (NCPA).  Consequently, the UN report is an unreliable tool for determining future public policy.   

"These dire predictions are not the result of scientific forecasting," said J. Scott Armstrong, an internationally known expert in forecasting methods from the Wharton School at the University of Pennsylvania who co-authored the NCPA study.  "Rather, they are opinions derived from a political process."

The most accepted forecasting methods were determined by internationally-known experts and expert reviewers and are available in the Principles of Forecasting handbook.  These principles were designed to be applicable to making forecasts about diverse physical, social and economic phenomena.  The NCPA study applied these forecasting principles to audit 2007'sFourth Assessment Report of the UN Intergovernmental Panel on Climate Change (IPCC), which predicted big increases in average world temperature by 2100. The audit found that:

  • Out of 140 forecasting principles, 127 are relevant to the procedures used to arrive at the climate projections in the IPCC report;
  • Of these, the IPCC report clearly violated almost half (60);
  • An additional 12 forecasting principles appear to be violated, and there is insufficient information in the report to assess the use of 38 others; therefore
  • Only 17 out of 127 applicable forecasting principles can be shown to have been followed by the IPCC.

"How many jobs are there where you can do only 13 percent of what is required and still remain employed?" asked Armstrong.

The NCPA study notes that complex forecasting methods are only accurate when there is little uncertainty about the data and the situation.  These conditions do not apply to climate forecasting where temperature data is highly variable over time and space and can often depend on when the data sample was chosen.  The authors also note that the IPCC forecasters themselves are part of the unreliability problem, as political considerations influenced all stages of the IPCC process, including writing the final version expressly to reflect the language negotiated by the political appointees to the IPCC. 

"As a result of these violations of forecasting principles, the forecasts in the IPCC report are invalid," says Armstrong.  "There is no scientific forecast supporting the widespread belief in dangerous human-caused ‘global warming.'  In fact, it has yet to be demonstrated that long-term forecasting of climate is possible at all."