Original URL: https://www.theregister.com/2013/03/07/uk_rainfall/

The supercomputers LIED: UK rainfall is rising, but won't drown our phones

Ofcom counts jam jars above ITU's models

By Bill Ray

Posted in Science, 7th March 2013 07:02 GMT

UK rainfall fell well short of the ITU's apocalyptic predictions in 2010, and isn't rising as fast as models predicted, according to Ofcom which has been measuring the levels in jam-jars.

Ofcom's numbers come from rainfall gauges - jam jars with rulers stuck to the side - and cover 1990 to 2010 (1991-2011 in Scotland and NI), which is less sexy than supercomputer modelling but arguably more accurate, and shows rainfall increasing across almost all of the UK with only a few bits of Cornwall getting drier.

Rain increasing to 2010

The numbers used by Ofcom were collected by the Met Office and the Environment Agency, with some coming from Norway too, and much of the work done by dB Spectrum Services (under contract to Ofcom) involved amalgamating the diverse sources into a single data set.

Ofcom cares about rainfall as it gets in the way of the microwave which backhauls a good deal of our connectivity, both mobile and fixed, so knowing how much rain to expect is vital for those building networks. Until now, the builders of these networks have had to rely on global modelling from the ITU-R (the ITU's radiocomms sector), and will be pleased to hear that Ofcom's numbers are much less dramatic.

That's not to say that rainfall isn't increasing, only that it's increasing slowly, and simply extrapolating the two-decade trend for another few years shows rainfall levels as they will be in 2015 - which was the point of the whole exercise.

Rainfall in 2915

Ofcom isn't interested in why rainfall is increasing, only the reliability of the 40,000 or so microwave links the UK relies upon to maintain connectivity. Microwave works in the rain, but range drops depending on the frequency, and the density of relay points is calculated based on predicted rainfall. ®