Monday, 24 August 2015

Weather, biology and nonlocality

Weather forecasting is getting more accurate. In Western Europe the 5-day forecast is more reliable than was the 3-day equivalent 25 years ago. 

Philip Ball in his article ‘No hurricane tonight’ (Prospect, February 2013) argued that although eventually 10 day forecasts may be possible there will always be a limit to the achievable accuracy of the computer models used by meteorologists to predict the weather. (Ball's article was based on a new book 'Invisible in the Storm' by Roulstone and Norbury, which is an account of the mathematics of weather prediction.)

These models work by breaking down atmospheric systems into elements or pixels which obey certain laws of physics

To simulate a game of billiards on a computer you program in Newton’s three equations of motion and a representation of each ball, then compute the action of a cue hitting the ball at a certain angle with a certain force or of a moving ball hitting another ball. This then leads to an easily predictable sequence of movements.

 To simulate a racing car’s movement on a computer you need to enter a model of the race track, the car with its steering wheel, accelerator and brake  and a set of equations describing how the car responds to the controls as it meets the camber, twists and bends of the virtual track, which is also programmed into the computer. 

Meteorological models are something like this in principle except that the systems being simulated are immeasurably more complex. The numerical variables of a weather system – atmospheric pressure, temperature, humidity, wind speed, wind direction, cloud cover, cloud type, precipitation intensity and precipitation type – which are used in computer models are only sample measurements taken  from weather stations at selected places and represent only a simplified view of the real situation. To accurately describe the actual weather in terms of such parameters one would need an almost infinite number of weather stations.  Moreover, such factors as cloud type or precipitation type are over simplifications at best. So the initial data fed into a weather model gives a far from accurate picture of the real situation.

The other major well-established cause of uncertainty in meteorology is that the weather is apparently chaotic in nature. In a system described by chaos theory only small departures from the initial state can lead to huge differences at some future time, depending on the kind of initial state. For instance, the weather in the Sahara desert will not be so sensitive to initial conditions as the weather over the Atlantic Ocean.


 There appears to be a consensus that for much of the world after 10 days the divergence of actual from predicted weather will be too large for the forecast to be useful even after all possible increases in computer power, modelling and sampling. 

Yet there are two other causes of uncertainty which can upset forecasts, both of daily weather and of future climate (average weather). 

 One is not quantifiable on a global basis but can have big effects on the intensity and distribution of precipitation.  This arises from the way clouds form by the condensation or freezing of water vapour onto microscopic particles in the atmosphere (the way water vapour turns to fog in the air of a bathroom, or as a misty film on a cold mirror). These particles consist not only of all kinds of dust but viruses and bacteria sucked up in storms from plants and soil as well as sea salt and methyl compounds emitted by phytoplankton in the ocean. See and a short RTB Newsflash podcast interview with Hugh Ross (highly recommended).  These processes are impossible to simulate or forecast on a large scale, especially where living organisms play a role.

The other one is controversial at this stage of history. It cannot be accepted without first accepting what has been termed  'nonlocality physics'. This shows that every particle in the universe – past, present and future – interacts instantaneously with every other particle, irrespective of distance. It is a startling concept of reality arising from the findings of quantum physics  (experimentally verified, e.g. quantum entanglement)  and includes the deduction that during biological life human consciousness cannot be separated from nature.

What is remarkable about this merging of mind and matter into one entity is that it fits in precisely with the Gaia hypothesis of James Lovelock, which looks at the unfolding of the biosphere over billions of years and comes to the conclusion that it is interconnected and self regulating. It has thrived even though the sun’s luminosity has increased by around 25% over this period and it now appears human beings are more integral to this system than was previously supposed by mainstream science.


Reach me at