Weather forecasting is more about interpreting the results than writing the code for the model, and, realistically there are only the British and American weather supercomputers. Everyone else is borrowing the GRIB data and then doing their own interpretations.
As for that interpretation, a surprising amount of forecasters operate much like how gamblers work.
Think of a horse racing scene, the horses and the riders have form, i.e. how they have performed in the past. Then there is the course, this may or may not suit a given horse/rider combo. Weather, time of day and other factors come into it. Not to mention the prize money and the competition.
A gambler looks at the form, the weather and works with their niche of specialist knowledge, similar with weather forecasting.
The 'bet' might be what the weather is going to be in Dudley tomorrow or in Barnstaple, or anywhere else. The goal of the forecaster is to get it right. The observations are the key to it and the GRIB forecast data is not to be considered strictly accurate. Even if it is accurate it still needs interpretation. Even with the data and an eye on the metosat images it is incredibly hard to get the likelihood of rain right.
Sometimes a step outside is needed too - with a local forecast.
The resolution of the GRIB is not that good, so if you have a town at the foot of a mountain and next to a lake, it might not have a realistic prediction. The GRIB data might be for the top of the mountain or the middle of the lake. You need an expert to actually do an educated guess as to what temperature it will be there in the morning.
Forecasters actually understand some things that are going on in the now, so a storm off Africa that goes across to the U.S seaboard as some tropical storm will come back across to the British Isles as rain. Forecasters don't tell you that there is a queue of storms heading across the Atlantic on the way to the U.S. and inevitably going to bring rain to the UK after that. But that information could inform your holiday plans for a few weeks ahead. The computer and the meteosat images might show all this but they won't interpret it for you.
Then, when you get down to it, forecasting can often be done with horse betting mentality, like a gambling addiction.
What would be cool is to have a consumer grade version of the software the weather forecasters use, where you can go forward and back in time layering up different types of data at different altitudes. But weather forecasting is like its own magic circle and it would be betrayal of sorts to share the tricks and the props with the audience.
Modeling climate change is much more complex than the simple heuristic pattern matching of the human mind. You're talking about a huge, complex, and fundamentally chaotic system; the latter property precludes the usefulness of heuristic based predictions by a 4 year old (or a human of any age).
There's a reason we run massive numerical simulations on multi billion dollar HPCs.
*Depending of course on how you define processing power. Sure, in terms of raw flops it is estimated that human brains are indeed orders of magnitude more performant than supercomputers, but for the type of operations required to simulate something like climate, humans don't come anywhere close to the accuracy or precision of results you can obtain from a gridded 3D model fed into a purpose built simulation - without which you really can't answer any but the most basic of climate change questions.
Personal anecdote: my grandmother is capable of predicting rain 1-2 days ahead of time based on smell and just "a feeling." Granted, we probably don't remember all of the times that she was wrong.
See [0]. PDEs can be solved discreetly in three dimensions on a three-dimensional grid (3D model, sort of).
EDIT: Ah, I see what you're saying. Yeah, I believe we fundamentally agree and our posts describe the same reasoning in two different ways.
not even close in processing power of a 4 year old