Data Centres and Weather

[jack.h...@gmail.com]

These 'factories' are vital to the modern world.  Not everyone is in love with AI, but, like it or not, it's here to stay. 

Data centres come with huge environmental costs: water for cooling, noise (apparently a hum that is present 24/7). light pollution and heat emissions.

It is this latter that researchers in Chennai have been concentrating on.  They have a working hypothesis that the local hot spots of Data centres might, in certain atmospheric conditions (CAPE), provide the vital trigger for convection and perhaps even lead to thunderstorms.  The scientists recognise that it will be difficult to isolate the development of localised storms near data centres from the general background, but their research so far suggests some definite correlation.

Hypotheses are just thoughts until predictions are made that turn out to be facts.  The Chennai team is making forecasts for the coming Indian Monsoon season.  The hypothesis might simply prove to be wrong, but only time and testing will show whether there is some value in the idea.

Jack

[jack.h...@gmail.com]

Plenty of hits, but nobody commented on the date of my post, 1st April.

Jack

[Freddie]

It went right over my head :D

[Len W]

I didn't like the curry source.

Should have been vindaloo to give a decent value for CAPE.

Len

[Keith (Southend)]

My only thought was surely this heat could be used locally for heating houses.

Keith (Southend)

[Ashley haworth-roberts]

I was half-way through reading it (on 3 not 1 April) and my thought was 'could one data centre really generate a localised thunderstorm' and I was thinking of googling that to see if it was a serious concern. If the initial post is an April 1st hoax it is one that might have a grain of truth behind it:  AI data centres can warm surrounding areas by up to 9.1°C | New Scientist

[jack.h...@gmail.com]

That was my whole point about the April Fool hoax: it looked quite plausible.

Elsewhere, I wrote a spoof:

The RSA Encryption Standard RSA (devised by Rivest, Shamir, and Adleman) encryption relies on a fundamental mathematical principle: the immense difficulty of factorizing the product of two very large prime numbers.

and how that had been cracked with the inevitable ransomware.

The next day, I read that Quantum Computing might achieve just that by 2030.

And on the wildlife forum, I posted a totally plausible hoax.

I write these to keep my brain active.  My body at age 87 is long past its 'best-before date', but I try to make up for that with crazy ideas.

Jack