Tuesday, 17 June 2008

The myth of global warming

As an early component of my studies towards a Masters degree in Statistics with the University of Newcastle I did an analysis of the Varve data. Varve data is the glacial temperature data. This information covers a 11,000 year period.

“Varve analysis is the process of counting varves or annually laminated sediments to determine the rates of change in climate and various ecosystems. Varves form when glacial advances come in contact with bodies of water such as lakes. When this process occurs, layers of sediment form on the floor of the body of water. This technique provides an opportunity to acquire detailed chronological information about the composition, displacement, and climate of that region, at that time. It was first developed by the Swedish scientist Baron de Geer in 1878.”

Climate change is a well-known natural phenomenon. It has been occurring for over 10,000 years. We have data that demonstrates this. Further, as far as I know there where NO cars 10,000 years ago.

Unlike the touted end of existence in 2007 by pseudo-scientific mystics (being nice to the quacks), 2007 didn't turn out to be the warmest ever. In fact, 2007's global temperature was essentially the same as that in 2006 - and 2005, and 2004, and every year back to 2001. The record set in 1998 has not been surpassed. In fact, the earth has become colder.

For nearly a decade now, there has been no global warming. Even though atmospheric carbon dioxide continues to accumulate - it's up about 4 percent since 1998 - the global mean temperature has remained flat. That raises some obvious questions about the theory that CO2 is the cause of climate change.

In the US, a record 44.5 inches of snow fell in New Hampshire in Dec 2008. This broke the previous record of 43 inches that was set in 1876. The Canadian government is forecast the coldest winter in 15 years, and it was close. In South America the start of winter last year was one of the coldest ever observed. In Buenos Aires in 2007 it snowed for the first time in 89 years. In August 2008, Chile reported the "the toughest winter we have seen in the past 50 years". It resulted in an estimated loss of at least $200 million in destroyed crops and livestock.

June 2007 was the coldest winter on recorded history in Australia. New Zealand's vineyards lost much of their 2007 harvest when spring temperatures dropped to record lows.
All four agencies that track Earth's temperature - the Hadley Climate Research Unit in Britain, the NASA Goddard Institute for Space Studies in New York, the Christy group at the University of Alabama, and Remote Sensing Systems Inc in California ALL report a 0.7C cooling in 2007 - a reversal of the warming that has taken place over the 20th Century.

It is possible that by 2020, the world will not have warmed for over 20 years!

The 0.7ºC of temperature change from 1950 to 2006 disappeared, we are actually over 0.5ºC cooler than the earth was in the period from 1900 to 1920.

A warm Middle Ages saw vineyards in England. Greenland got its name due to the relatively lush coastal regions encountered by contemporary exploring Vikings. These villages lasted until around the 17th Century. At this point a cooling climate reduced the snow-free land available to the settlers and indigenous people alike. This cooling period left Greenland as we know it today.

Science beats myth, lets do the former.

2 comments:

Joseph said...

For nearly a decade now, there has been no global warming. Even though atmospheric carbon dioxide continues to accumulate - it's up about 4 percent since 1998 - the global mean temperature has remained flat. That raises some obvious questions about the theory that CO2 is the cause of climate change.

Not if there's no statistical significance to this fluctuation. I've done an analysis correlating residuals of modeled time trends, and the effect is clear. This cannot be explained by coincidence or data collection errors, evidently.

Craig S Wright said...

Except that you are basing data on a small series with few of the explanatory variables. Longer series and alternative data sets exist.

Most importantly you are taking a correlation effect and calling it causal.

On top of this there is no inclusion of other variables. Effects from solar activity, volcanic action etc all need to be included. Perception based selection is a way of getting evidence to “prove” the point you want to make, not to actually come to a scientific determination of a cause.

The evidence is not clear and nor is statistically significant.