LDosser wrote:

This is actually worse than it looks. Global Climate Models (GCM)
have to take many, many variables into consideration. In general the
planet will be divided into a number of distinct areas. such as a two
by two degree grid. Once the mapping has been agreed upon, each block
is encoded for surface type and may or may not have values associated
with it. A surface type may be water, sand, ice, forest, and so on.
Typical values might be albedo or, in the case of water surface,
albedo and temperature. Now, all that considered and adding the new
knowledge on the Himalayan Glaciers, one wonders just what they were
using for data in the Himalayas! Either they made up data
(Falsified), or don't consider that in their GCM. If they made up
data, one wonders just how much else was pulled from a dark place. If
they didn't take the Himalayas into consideration, just what kind of
****ty GCM are they running?!

In 1928, the Navy Hydrographic Office built the "Brass Brain," a 2500-pound
analog computer to do tide prediction. To use it, one set knobs for some
30-odd variables affecting the tides for a given location and turned a
crank. The output was a graph showing the expected tides in feet for the
specific location and times.

The machine was retired in the '60s and moved to the Smithsonian. It was
replaced by a CDC-6600 digital computer.

During its half-century of use, the machine received two engineering
upgrades: 1) The hand-crank was replaced with a washing-machine motor, and
2) The ink-pen was replaced with a felt-tip. Preventative maintenance was
performed once a year and consisted of oiling a few gear trains.

The Brass Brain could operate without power, under water, or at any
temperature. But here's the neat part that current climate models can't
duplicate:

If you wondered whether you had entered all the parameters properly, you
simply turned the crank backwards to see if you got the tides for the
preceeding month!

Ain't technology grand?