In October 2000, there was a high speed derailment just south of Hatfield, some 27 km north of London. It happened around 12:30. Tragically, four people were killed.

I heard about this within half an hour of it happening - in my hotel in Washington DC.

By 12:30 that day, four people had died on Britain's roads. I can only say only this because, on average, 10 people die each day and statistically more do so after noon than before.

The train crash led to major chaos on Britain's railways. It led to the collapse of Railtrack. It also resulted - as all railway accidents do - in two major enquiries and a published report with many recommendations based on lessons learnt. At all stages it led to near world-wide near saturation coverage in all media.

The fatalities on the roads, by contrast, just led to an inquest. One or two may have hit the local papers. No enquiries, no published report, no lessons learnt - even though, because railway accidents have been painstakingly investigated since the 1840s and all low-hanging fruit has long gone, it would be easier to learn from road accidents.

I deplore this bias.

One thing it does do is give people an unbalanced view of the world. It is quite possible that, as a result of media coverage, people conclude that it's unsafe to ride trains and go by car instead - witha significantly higher risk of accident.

I also deplore the practice - especially in North America - of reporting something under the headline, "Train crash" when actually it's an idiot in a car thinking he can outpace a train approaching a level crossing.

How about some balance in reporting?

Sadly, unlikely. Railway accidents are news because they rarely happen: car crashes aren't because they happen much, much too often.

In 1990 there was an analysis of front-page reporting in the New York Times. This showed that there were 1.7 murder stories for every 1000 homicides, 2.3 AIDS stories for every AIDS death, 0.02 cancer stories for every 1000 cancer deaths and 138.2 plane crash stories for every 1000 aircraft deaths.

Balance in reporting?

In the January 2009 issue of the "Journal of Air Transport Management" there is a useful article on the possible impact of the EU's Emissions Trading Scheme (ETS).

It looks at the options for (in particular) long-haul airlines based outside the EU reducing their ETS costs by making a stop on the way.

For example, if Singapore Airlines flies direct from Singapore to Frankfurt, it will effectively pay an emissions charge for the entire flight. If however it flies via Zurich or Istanbul, it will only have to pay the charge for the shorter legs - between Zurich and Frankfurt or Istanbul and Frankfurt. So the ETS cost/passenger will be about 15 - 18 Euros less for those routings.

The article also shows the relationship between flight distance and the distance the aircraft actually flies at cruise altitude - its most efficient altitude.

For flights between 188 km and 312km, the aircraft is only at cruise altitude for 21% of the time. The rest of the time it is either climbing or descending. Between 313 km and 437km, it's at cruise altitude for 30% of the journey: this increases to 40% for flights between 563km and 687km. At 876km, it is 45%.

So only on relatively long journeys - where time makes flying competitive with rail - does an aircraft fly at cruise altitude for more than half of the time it's in the air.