12 ms·

Fundamentally there's one problem that plagues global warming research, and that's that we're grasping at a system so huge and complex that we have absolutely n

Fundamentally there's one problem that plagues global warming research, and that's that we're grasping at a system so huge and complex that we have absolutely no understanding of it. Sure people believe they understand the system, but spend a month in England and watch how weather forecasts pan out. I don't see how, when some of the biggest computer systems are being used to predict weather systems and we get it wrong, how scientists expect us to believe their computer models that follow less variables and use lots less processing power. All evidence so far says global warming is real, however I can't help but feel all the computer models they use are complete bunk. It strikes me as fraud, it's a computer program and everyone here knows you can get a computer program to do whatever you want. I see it as highly susceptible of bias, which makes it very hard to ever believe their predictions. There's a lot of real science used in global warming research, like using ancient sediments and ice cores to extrapolate how much CO2 was in the atmosphere n centuries ago. However, then we get pseudoscience with computer models of systems infinitely more complex than anything we've ever dealt with; it's akin to comparing a stick figure and a full anatomical diagram of a human down to every capillary for modelling the human body. We're at the stick figure when it comes to the global environment and we're trying to predict somethings effect like we have the whole picture.

LargeWu 17y ago
I mostly agree with you, except it's dangerous to compare weather models to climate models. I think this statement by the author is a bit disingenuous: "If correlation doesn't prove causation, what does? Taken to its logical conclusion, you could observe the same result in a hundred experiments and dismiss it as merely a strong correlation." The problem here is that there are no actual experiments done on the climate - it's all models, the outputs of which are essentially unverifiable. In other words, nobody is taking a test planet Earth and a control planet Earth and filling the test version's atmosphere up with CO2 and seeing what the difference is, much less repeating this and getting the same results. That is why, despite the fact that these models are based on scientific theory, the models themselves are not, strictly speaking, "science". For the record, I believe global climate change is probably happening, and is probably caused by human activity.
emmett 17y ago
You clearly did not read the article you're critiquing, because he spends screens and screens of text discussing exactly the difference between climate and weather. In short, climate is predicting that a candle will make the room warmer. Weather is predicting the exact swirls and eddies of the flame. Further, your objections to computer modeling are incorrect. Of course you can get a computer program to say whatever you want, but you can't make a computer model say whatever you want. Computer models are a subset of computer programs constrained to attempt to model reality according to some set of assumptions. You can make equations say whatever you want too, but that doesn't make Maxwell's laws incorrect.
- azgolfer 17y ago
  There is no way to build anything close to an accurate model of the climate. It has far too many independent variables. What really makes it ridiculous is that the are always claiming to use the latest "super" computers when every (good) CS student knows that you can't solve problems like these with any size computer. Also - regarding non scientists influencing the debate - seems to me a lowly patent clerk had a pretty large influence on physics around 1905.
- electromagnetic 17y ago
  Actually assuming the climate and weather aren't connected is fundamentally ignorant. The El Nino and La Nina effects are the biggest example of this, in that it can change regional temperatures across the entire globe, it can cause algal blooms, which in turn are a key part of the climate model. A prolonged El Nino is reported to have helped provoke the French Revolution by causing famines all over Europe. Crop failures and record low temperatures/rainfall have profound effects on the environment and the local biospheres climate effects. Death of biological material (crop failures) is usually tied to CO2 and Methane release through the biological processes that result, this is not to mention that prolonged cold periods (El Nino lasted over a decade at the time of the French revolution) result in less plant growth and more CO2 production through fuel burning. Then it would have likely led to a lot more deforestation than it likely would today. My objections to computer modeling is that all they are is a model. If you put the wrong equation into the model, the answer is wrong. The one thing I've never heard factored in is the amount of thermal energy emitted by living organisms and our society. From the estimates I've read, all the biomass on earth produces roughly 3-4 days the amount of solar energy received from the sun. Essentially the Earth is receiving 369 days worth of energy, for only 365 days in a year. It's all a matter of equations, and the computer models have always put out unsatisfactory answers. I remember one being laughed at for predicting ocean rises of 1 cm per year, and I believe we've had approximately 0.01 cm per year. My point is that the climate scientists' jobs depend on the climate being in danger. If there weren't fires we wouldn't have firemen, and if we didn't have crimes we wouldn't have police, so if there isn't a climate crisis we wouldn't have climate scientists and there's no provisions to protect us from rogue scientists but there are provisions to protect us from rogue cops. None of these models are repeated, no one can confirm the results as even being valid, they're just guesses. Perhaps I just have a problem with trusting people in white coats telling me the sky is falling, but it all sounds too much like Chicken Little to me. I need to know the firemen aren't burning buildings down before I'm willing to pay them to put out fires. Equally I need to know Climate Scientists aren't hyping this up just to keep their careers alive. As I've already said, I don't have a problem with the hardcore evidence and I believe global warming is happening. My problem is with the explanations and predictions, because they all seem either too simplistic or just bunk. Perhaps it's just a problem in the translation from the scientific and non-scientific lexicons, just like the major problems with explaining evolution to an American.

LargeWu 17y ago

I mostly agree with you, except it's dangerous to compare weather models to climate models. I think this statement by the author is a bit disingenuous: "If correlation doesn't prove causation, what does? Taken to its logical conclusion, you could observe the same result in a hundred experiments and dismiss it as merely a strong correlation." The problem here is that there are no actual experiments done on the climate - it's all models, the outputs of which are essentially unverifiable. In other words, nobody is taking a test planet Earth and a control planet Earth and filling the test version's atmosphere up with CO2 and seeing what the difference is, much less repeating this and getting the same results. That is why, despite the fact that these models are based on scientific theory, the models themselves are not, strictly speaking, "science". For the record, I believe global climate change is probably happening, and is probably caused by human activity.

emmett 17y ago

You clearly did not read the article you're critiquing, because he spends screens and screens of text discussing exactly the difference between climate and weather. In short, climate is predicting that a candle will make the room warmer. Weather is predicting the exact swirls and eddies of the flame. Further, your objections to computer modeling are incorrect. Of course you can get a computer program to say whatever you want, but you can't make a computer model say whatever you want. Computer models are a subset of computer programs constrained to attempt to model reality according to some set of assumptions. You can make equations say whatever you want too, but that doesn't make Maxwell's laws incorrect.

azgolfer 17y ago
There is no way to build anything close to an accurate model of the climate. It has far too many independent variables. What really makes it ridiculous is that the are always claiming to use the latest "super" computers when every (good) CS student knows that you can't solve problems like these with any size computer. Also - regarding non scientists influencing the debate - seems to me a lowly patent clerk had a pretty large influence on physics around 1905.
electromagnetic 17y ago
Actually assuming the climate and weather aren't connected is fundamentally ignorant. The El Nino and La Nina effects are the biggest example of this, in that it can change regional temperatures across the entire globe, it can cause algal blooms, which in turn are a key part of the climate model. A prolonged El Nino is reported to have helped provoke the French Revolution by causing famines all over Europe. Crop failures and record low temperatures/rainfall have profound effects on the environment and the local biospheres climate effects. Death of biological material (crop failures) is usually tied to CO2 and Methane release through the biological processes that result, this is not to mention that prolonged cold periods (El Nino lasted over a decade at the time of the French revolution) result in less plant growth and more CO2 production through fuel burning. Then it would have likely led to a lot more deforestation than it likely would today. My objections to computer modeling is that all they are is a model. If you put the wrong equation into the model, the answer is wrong. The one thing I've never heard factored in is the amount of thermal energy emitted by living organisms and our society. From the estimates I've read, all the biomass on earth produces roughly 3-4 days the amount of solar energy received from the sun. Essentially the Earth is receiving 369 days worth of energy, for only 365 days in a year. It's all a matter of equations, and the computer models have always put out unsatisfactory answers. I remember one being laughed at for predicting ocean rises of 1 cm per year, and I believe we've had approximately 0.01 cm per year. My point is that the climate scientists' jobs depend on the climate being in danger. If there weren't fires we wouldn't have firemen, and if we didn't have crimes we wouldn't have police, so if there isn't a climate crisis we wouldn't have climate scientists and there's no provisions to protect us from rogue scientists but there are provisions to protect us from rogue cops. None of these models are repeated, no one can confirm the results as even being valid, they're just guesses. Perhaps I just have a problem with trusting people in white coats telling me the sky is falling, but it all sounds too much like Chicken Little to me. I need to know the firemen aren't burning buildings down before I'm willing to pay them to put out fires. Equally I need to know Climate Scientists aren't hyping this up just to keep their careers alive. As I've already said, I don't have a problem with the hardcore evidence and I believe global warming is happening. My problem is with the explanations and predictions, because they all seem either too simplistic or just bunk. Perhaps it's just a problem in the translation from the scientific and non-scientific lexicons, just like the major problems with explaining evolution to an American.