If Computers Can't Predict Tomorrow's Weather, How Can They Predict What It Will Be in 100 Years?

Question

Look at the little boxes on the 5 day forcasts. They must be an inside joke at the weather channel. I think I have broke their code. Our prediction is the same weather as Chicago but two days later. But they are notoriously inaccurate. This is the same people who claim to know right down to a 10th of a degree what the temp will be in a couple of hundred years. Our poor children sweating in 140 degree temperatures, swimming to the top of the statue of liberty since the water will not evaporate by then. Global warming water will not change to clouds like ordinary pre-global warming water according to the rising sea water/disappearing coastline scenario. It just sits there getting stagnant. Take some swimming classes.

Answer 1

You make a good point.  The following was written by a computer programmer:

FACT: Only Computer Illiterates believe in "Man-Made" Global Warming.

What people do not understand is that there is no proof of "Man-Made" Global Warming without using irrelevant computer models. Yes computer models have a place in engineering but are utterly useless at fortune telling, I mean "climate prediction". With engineering you can build and test in the real world to confirm the computer model's accuracy. You can do not such thing with the planet Earth and it's climate. You cannot build a planet and it's atmosphere to "test" your computer climate model.

I am a computer analyst and can make a computer model do whatever I want by "tuning it" (adjusting variables by guessing until I get the answer I want or think is right).

GIGO: Garbage in = Garbage out

Computers need exact information and the exact procedures to process that information to get accurate answers, without that you get useless results, period. There is no way around this. Everything must be 100% understood and 100% accurate.

Computing incomplete, biased or flat out wrong data (guesses and assumptions) based on poorly understood climate physics into a "model" will give you junk results. Testing a model against past climate is an advanced exercise in curve fitting, nothing more and proves absolutely nothing. What this means is you are attempting to have your model's output match the existing historical output that has been recorded. For example matching the global mean temperature curve over 100 years. Even if you match this temperature curve with your model it is meaningless. Your model could be using some irrelevant calculation that simply matches the curve but does not relate to the real world. With a computer model there are an infinite number of ways to match the temperature curve but only one way that represents the real world. It is impossible for computer models to prove which combination of climate physics correctly matches the real world. Do not be fooled this logic is irrefutable by anyone who understands computer science and computer modeling.

To make matters worse it is not computer scientists creating these models but natural scientists coding them using Fortran. These natural scientists do not even begin to have the basic understanding of computer science or proper coding practices. Sloppy and buggy code is littered inside these climate model programs yet there is next to no accountability for any of this. How do you separate a programming error from a temperature anomaly? How can you replace observational data with a complex mathematical equation? You can't.

Processing more complex data in more complex ways via guessing gives you more complex junk results. But since the models have been "tuned" (guesstimated or deliberately altered to get the results they want) they get results that "seem" likely or even convincing to the average computer illiterate, except it is all based on a complex serious of guesses and assumptions and absolutely meaningless for prediction.

Nothing is emotional about computers they are pure logical machines, 1 + 1 must = 2. Imagine trying to use random numbers to get a right answer on a calculator but you do not know if you are to add or multiply those numbers and you have no way to confirm that "right answer" except to wait 50-100 years. Sound crazy? Welcome to Global Climate Modeling.

All the computer illiterates are convinced that because something is done on a "super computer" that costs "millions of dollars" it is infallible. The more complex the model, the more "mysterious" it seems to the average person. The public gives computer climate models this mystical aura because they are largely computer illiterate about how they actually work and when they hear the term "computer" they do not want to sound or feel stupid, so they nod their heads and go along with it.

What the modelers do is they keep playing with the numbers in a much more complex way until they think they guess right. A useless exercise. These same climate model computers are used to predict your weather and you know how accurate they are. But d**n! Al Gore and Gavin Schmidt can certainly tell your what the climate will be 50-100 years from now! Give me a break. Don't be fooled that modeling climate is different than the weather or one is more accurate than the other long term. The difference is simply a matter of resolution and scale.

Why are we not turning to models to predict the future for everything? Because they can't, not even remotely. Some of them work "sort of" for the weather in very, very short term results (1-3 days) until all the data they are processing that is wrong combined with all the data they are missing and the millions of variables they are not accounting for start to kick in and grow exponentially the farther out the model runs and wham - the model is wrong. No kidding, there are simply way too many variables that they cannot account for and the computer power necessary to even start to take these variables into account does not exist.

You are expected to believe that they can "model" the climate 50-100 years in the future when they cannot even give you accurate weather 3 days out? Don't be fools, I do this for a living, Computer Models cannot predict the future with anything as complex as the Earth's climate.

Answer 2

you need to study hydrology and evaporation

http://en.wikipedia.org/wiki/Hydrology

http://en.wikipedia.org/wiki/Evaporation

long before the temperatures reach 140 degrees the torrential rains will set in and you will be very wet.

this is what the earth will be like at ten degrees warmer

http://anthropology.si.edu/humanorigins/...

http://anthropology.si.edu/humanorigins/...

as the earth warms the deserts will disappear do to the increased rain fall.

plus the increased cloud cover will stop the temperature increase.

this will cause a increase it plant growth breaking down the amount of CO2 in the air. now the earth will cool.

all this will stop the earth from ever reaching 140 degrees.

there are many inland area that are dry now that will become lakes like the great basin of the US

http://en.wikipedia.org/wiki/Great_Basin

this area and many others like it around the world will take up the extra water from the ice caps.

Answer 3

It was just reported that Lake Mead will go dry in 13 years.  Its also been shown that glaciers hundreds of years old are melting.  Does that not indicate the climate change?  What about the tempatures always breaking records every week. doesn't that indicate something is happening.

Answer 4

You have to remember that the AGW alarmists have set it up now so they can't be wrong, even if they are. They claim that low temps and high temps are caused by AGW. Snow and rain are caused by AGW. The lack of storms and the abundance of storms is caused by AGW. It's been 10 years since the self proclaimed hottest year on record ever. I'd say that's a trend. AGW is a huge liberal, media driven scare tactic.

Look at it like this, in layman's terms. Five years ago we'd hear about Britney Spears maybe once a month. Things were going smooth, she'd just be on tour etc.

Now, we get daily, literally daily, updates on what is going on with her. Why? Because she's a trainwreck and anything but smooth sailing. That's what sells. That gets a "rise" in people.

AGW alarmists are nothing but nervous nellies who think they are in the "in" crowd by alarming people around the world with their hypocritical rhetoric.

No thanks.

Answer 5

In many ways its easier to see and project long term changes to climate than it is to forecast local weather over a short time period.

Climate tells you what clothes to buy and have in your wardrobe, weather tells you what to wear each day. If you go on holiday the area’s climate tells you what clothes to pack and what weather to expect for that time of year. The climate of an area or region is the average pattern of weather conditions in a region over a long period of time. Weather is constantly changing; climate is more stable over time.

Weather is local and short-term. If it rains or is sunny in the town where you live tomorrow, that's the weather. It's what is going on in the atmosphere at a particular place and time.

But you probably have a very good idea what type of weather to expect for your town in any given January or July. That’s its climate.

The weather is the type and severity of events such rainfall, snowfall, mists, frosts, cloud, winds, sunshine, droughts and so on that happen within a climate zone.

We know enough from trends to recognise when a region’s climate zone is changing.

For example tropical savannahs are a transitional stage between an arid climate and those of a tropical rainforest.

If a savannah receives less and less rainfall then it begins to take on the characteristics of an arid area. When its rainfall, over a period of years, matches the amount found in an arid area then the vegetation and landscape will also begin to match those already found in an arid area. If it receives more rainfall than usual over a period of years then vegetation from neighbouring rainforest will take the opportunity to expand and colonise the grassland, which is now able to support forest.

In other words predicting climate is predicting the probability of events over time. We also have a historical record to compare events with.

Given we can detect changes to climate we then have to look at the likely causes to try and determine how long and severe they will be, and at how any changes will effect other climate mechanisms and what the impacts of those will be.

We can tell from the trends what the likely impacts will be and attempt to plan for them; in areas likely to suffer from increased drought for example.

--------------------------------------...

I’ll give some examples from the country I live in:

During the 20th century, the annual mean Central England temperature warmed by about 1°C.  The North Sea is now 1°C warmer than it was a decade ago.

The growing season has lengthened by about a month in central England since 1900, with the onset of spring occurring around two to three weeks earlier than just 30 years ago.

Between 1972 and 1999, British bird species extended their breeding ranges north by an average of 18.9 km in response to increasing mean annual temperatures at the northern end of their distribution. Hawthorn and hornbeam are coming into leaf earlier, and most spring-flowering plant species are flowering earlier – typically by around two weeks compared to pre-1990.

The four wettest winters in south-west England since records began have all been in the last 10 years. A larger proportion of winter precipitation falls in heavy rain showers than 50 years ago.

Sea levels on average around the UK are about 10cm higher than they were in 1900.

The UK has become twice as stormy over the last 50 years, with an increase in heavy rain showers. High pressures have increased on average by 3mb and low pressures decreased by 3mb since 1950. This results in much windier weather.

There’s an obvious trend there related to temperature. We can also compare current conditions with regions that currently have warmer conditions and see what things are like there. For much of southern England the trend projects that decade on decade the climate is becoming more like that of Mediterranean regions. That will impact on the type of crops we can grow, on water supplies and so on.

Climate tells farmers what crops they can plant and when to harvest them. Or if you can grow crops at all. And climate determines the vegetation that will grow in a region and the type of wildlife that will live there.

Answer 6

Two answers to your question in one:

1) Predicting the weather is based on realtime events which are constantly changing. For example, a frontal system may appear to be moving to the south when it suddenly shifts to the north, bring totally different weather.  

Predicting climate change is very similar in its being calculated, but the events it is stipulated on are changing much less rapidly. There is a relationship between gases in the atmosphere and the global average temperature that cannot be denied.  If we continue to tilt the scales and vent more gases, the temperature will, most likely, rise.

2.) Meteorologists regularly predict local temperature and climate data for a short period, while global warming data is calculated for the entire planet and over longer periods of time.  Climatologists will NEVER claim to be able to tell you what the exact temperature will be on September 6th, 2108 in Brooklyn.  They can give you a strong estimate of what the global average temperature will be if the current trends continue, however. This is because they have a greater sampling size for climate data.  The entire planet, over 100 years, instead of one city, count, state/province, country for 1 week.

Think of it this way.  Statistically, flipping a fair coin has a 50% chance of going either heads or tails.  That does not mean that the first 6 flips will be 3 heads and 3 tails.  However, if we increase the sample size to 2000 flips, the ratio of heads to tails will be almost 1:1, though it will admittedly be unequal in all likelihood.  The same is true of weather and climate data.  Local 5 day forecasts are essentially the same as flipping a coin only 5 times and then coming to a conclusion, while global climate data is like flipping the coin 2000 times.

Answer 7

Global warming has raised temperatures one degree over 100+ years.  The El Nino Southern Oscillation changes temperatures a degree or more from year to year, and we're in a strong La Niña cooling influence at the moment.  So we have a .01 degree upward trend per year for long term climate vs. a 1 or 2 degree downward trend for short term weather.  One outweighs the other by at least 100X in magnitude.  Can anyone really question which one will dominate for the next year or two, or in local weather variations from day to day?  Duh.  Of course the underlying global warming climate trend will be temporarily masked by major weather fluctuations.  That's why it's pointless to confuse weather with climate.

Fortunately we have 100 year records, so the long term trend is clear and we won't get fooled by short term weather fluctuations.  

Sea levels rising could eventually slam world economies, but in the much shorter term what will affect more people will be water shortages.  For example, for Americans from Wyoming to New Mexico to California, declining snowpack is likely to become a major issue very soon:

http://www.nytimes.com/2007/10/21/magazi...

Steven Chu, a Nobel laureate and the director of the Lawrence Berkeley National Laboratory, one of the United States government’s pre-eminent research facilities, remarked that diminished supplies of fresh water might prove a far more serious problem than slowly rising seas.

Chu noted that even the most optimistic climate models for the second half of this century suggest that 30 to 70 percent of the snowpack will disappear. “There’s a two-thirds chance there will be a disaster,” Chu said, “and that’s in the best scenario.”

Answer 8

they cant.  They just predict that things will increase in a linear fashion, which IS NOT SCIENCE.

They cant predict anything.  It isnt possible.  The computers are not capable of predicting certain complex factors such as amount and distribution of cloud cover.

They cant forecast temps in the arctic more than 2 days ahead.  The ten day forcast always shows that the temps will be near the average temperature 2 days from now, but nope, it is still always -20 below, and the new 10 day forecast shows it will warm up to the average, and it NEVER HAPPENS.  Sorry, its hard to explain that I guess.

Answer 9

It is so easy to be cynical, not so easy to deal with a difficult subject seriously.

Answer 10

You're confusing weather and climate.