Total Pageviews

Sunday, April 19, 2020

New climate models predict more global warming than old models ... that had predicted too much warming

In the bizarre world
of junk climate science,
computer games that
over predict global
warming by 2x to 3x.
are called "climate 
models".

Earth's response 
to increasing 
greenhouse gas 
concentrations,
is unknown,
but assumed to be
mild warming, 
based on 
lab experiments
with carbon dioxide
in artificially dried air.

In reality, we've had
intermittent global 
warming since the
1690s, of at least 
+2 degrees C.

Rising CO2 levels
could not be 
the explanation 
for most of that 
global warming.

That experience 
makes it hard to 
believe that CO2
levels suddenly 
became the 
only variable
that affects the 
global average 
temperature,
after 1975,
as many people 
claim today.




A commonly used 
measure of the
effect of CO2 is
the equilibrium 
climate sensitivity 
         ( ECS ) 

ECS is the long term
global surface 
temperature response 
to an abrupt doubling 
of atmospheric carbon 
dioxide (CO2). 

Estimates of ECS have 
historically centered
on a warming of +3˚C.
for each doubling
of the CO2 level, 
with the most recent 
United Nations 
Intergovernmental 
Panel on Climate Change 
(IPCC) report deeming
an ECS between 
+1.5 C. and +4.5˚C. 
to be most likely.

That is a huge range, and 
the range has not changed
since being published in the 
1079 Charney Report.

Those of us who do not
believe in a coming climate
crisis, mock the alarmists, 
with their never changing
ECS prediction, 
and its huge +/- 50% 
margin of error.

We like to focus on actual
measurements of global 
warming, that are 1/3 to 1/2
of what was predicted
by people who believe 
in the +3.0 C. "formula".



It has been a very 
stubborn challenge 
to get the scientific 
community to revise 
their estimated
ECS, with the goal 
of making better
global average 
temperature  
predictions.

The good news 
is ECS estimates
are now getting 
revised.

The bad news is that 
many models are 
revising ECS higher,
when they have 
already been too high,
predicting too much
global warming 
since the 1970s.

Given that models, 
on average, grossly 
over predicted 
the global warming 
in past decades, 
it's very unlikely
revising models 
to predict MORE 
global warming, 
will make their 
predictions 
more accurate 
in the future.

In fact, the ECS revisions
upwards imply accurate 
predictions are NOT the 
primary goal of climate
models.

Climate models 
are very popular.

They are very complex
methods of presenting
personal opinions about
climate change, that might
not be taken seriously
if just written on a piece
of paper.

The computer games,
called climate models,
are merely props
in a fictional 
"climate change
theater play" that 
leftists believe in,
just like Catholics
believe in god.

Very few people 
know the "models"
make wrong 
predictions, 
because the 
leftist-biased
mass media
does not think 
that's important
for you to know.



In a recent study, 
Zelinka and coauthors 
compared ECS values 
derived from CO2 
quadrupling experiments 
conducted with 27 CMIP6 
models that were available 
at the end of 
November 2019 
with values from 
all 28 models 
from the previous 
(CMIP5) 
generation. 

The latest models 
have wide-ranging 
ECS sensitivities, 
spanning values 
of +1.8 to +5.6˚C. 
-- they predict 
more warming 
than their 
predecessors 
by about +0.5˚C. 

Ten of the models 
now have 
CO2 sensitivities 
exceeding +4.5˚C, 
beyond the top of
the past IPCC's 
"most likely” range.

The primary reason 
for more global warming 
claimed by new models
was shown to be clouds. 

Specifically, both the 
water content and the 
areal coverage of 
low-level clouds 
decrease more strongly 
with greenhouse warming 
in the new models. 

That causes enhanced 
absorption of sunlight 
– an amplifying feedback 
that ultimately results 
in more global warming. 

This alleged stronger 
amplifying cloud feedback 
is particularly dramatic 
over the Southern Ocean.  

Differences between old 
(CMIP5) and new (CMIP6) 
models is in how cloud 
properties respond to 
their environment – not 
in how their environments
 change – which leads to 
a more positive cloud 
feedback in CMIP6.