Friday, February 15, 2013

Chart of the Day: Odds of Death


The Economist writes,
ON FEBRUARY 15th DA14, an asteroid 45 metres across, will sail past the Earth at 7.8km a second (4.9 miles a second). At just 27,700km away, it is well within the range of communication satellites. It will be the closest encounter on record with an asteroid this big. In 1908 an asteroid estimated to be around 100 metres in diameter destroyed 2,000 km² of forest in Siberia. Thankfully, such events are rare. NASA has identified 9,600 "near-Earth objects" since 1995, but just 861 with a diameter of 1km or more. The greatest threat to Earth is the 140-metre wide AG5; but it has just a 1-in-625 chance of hitting Earth, and not until February 5th 2040. More prosaic things are far more dangerous. According to data from America's National Safety Council, 27 people died in 2008 in America from contact with dogs (a one in 11m chance of death). The chart below compares the odds of dying in any given year from choking, cycling, being struck by lightning or stung by a bee.
While the odds of dying from a heart disease (467:1) may seem greater than dying from an asteroid impact (74,817,414:1) may be true, the basic problem with extrapolating statistics is that we really can’t determine or we simply DON’T know when that big ONE will arrive in whatever form to claim us.

Thus we can be lulled into complacency by the use of statistics from the risks of false negative errors, or specifically, result/s that appears negative when it should not

This reminds me of what author Nassim Taleb calls as the Black Swan Theory: the extreme impact of certain kinds of rare and unpredictable events (outliers) and humans' tendency to find simplistic explanations for these events retrospectively.

For instance, the NASA suspects that DA14’s whizzing past earth may trigger tremors on some parts of the world. Has the impact from such event been incorporated in the computation of the above probabilities? 

A meteor blast in Russia caused 400 injuries a day before DA14. The meteor incident was denied by officials as having been connected with DA14.

If the authorities are right, then this simply reveals how the Russian meteor incident can be construed as an “unpredictable-outlier event” but with limited impact. In short, no one saw this coming.  Yet the incident does not satisfy the ‘extreme impact’ conditions of the black swan model.

On the other hand, if authorities could have been wrong, then the statistical odds of death may have been underestimated, since the likely methodology in arriving at such probability may have been seen only from a direct impact from a meteor/asteroid collision and not from the ancillary events, such as Russia's meteor shower (as defined by CNN), which could have been the advance party of DA14, or from other potential ramifications from meteor or asteroid flybys.

The bottom line is that people tend to overestimate their knowledge of the world and or of the universe to the point that environmental jeremiahs or ecological-phobes use scare tactics to impose social controls on us in order to shape the world according to their false ideals. 

As discussed before the world is much larger than us, where various forms of potential black swans abound: 
While we have been made aware by media of these apocalyptic scenarios through a variety of science fiction movies that could or may occur; such as huge asteroid/s crashing on earth, super volcano eruptions, alien invasion, robot uprising and many more, there are other factors such as the black hole, gamma rays from an imploding star or the unleashing of a mighty wave of solar flares from our sun, that could send our world into oblivion, unpredictably and instantaneously.
Speaking of black holes, a science or Astrophysical journal recently asserted that black holes have been growing faster than expected and have grown beyond the sphere of traditional assumptions where black holes require “galaxy collisions”. 

Our knowledge of the environment has been incomplete and keeps changing.

No comments: