What We Don't Know


How did life begin? What's the universe made of? Why do we sleep? 42* of the biggest questions in science.

Why can’t we predict the weather? 
Wired, Issue 15.02 - February 2007

A few years ago, weather forecasts 
were totally unreliable beyond a couple of days; 
today better computer models 
make them accurate as far as a week out. 

That’s fine for figuring out 
how to pack for a business trip 
or whether you need to rent 
a big tent for the wedding reception. 

The trouble starts 
when you want to build 
a computer model to predict 
the weather over decades or centuries. 

In 1961, a meteorologist 
named Edward Lorenz 
was running a computerized 
weather simulation and decided 
to round a few decimal places off 
one of the parameters. 

The tiny tweak completely changed weather patterns. 

This became known as the butterfly effect: 
A butterfly flapping its wings in Brazil 
sets off a tornado in Texas. 

Lorenz’s shortcut 
helped launch chaos theory 
and sparked an obsession 
among meteorologists 
with feeding as-perfect-as-possible data 
into their models in an attempt 
to lengthen their forecast window.

But even refining precision 
doesn’t get us to long-term prediction. 

For that, climatologists 
need to understand boundary conditions, 
like the interactions between 
the atmosphere and the oceans. 

The goal, says Louis Uccellini, 
director of the National Centers 
for Environmental Prediction, 
is to model Earth 
as a single climate system. 

Then we can figure out what’ll happen to it next.

- Lucas Graves

No hay comentarios:

Publicar un comentario

COMENTE SIN RESTRICCIONES PERO ATÉNGASE A SUS CONSECUENCIAS