Welcome to the MacNN Forums.

If this is your first visit, be sure to check out the FAQ by clicking the link above. You may have to register before you can post: click the register link above to proceed. To start viewing messages, select the forum that you want to visit from the selection below.

You are here: MacNN Forums > Community > MacNN Lounge > Political/War Lounge > Weather Data Not Always Accurate—Effect on Climate Change Data?

Weather Data Not Always Accurate—Effect on Climate Change Data?
Thread Tools
ghporter
Administrator
Join Date: Apr 2001
Location: San Antonio TX USA
Status: Offline
Reply With Quote
Dec 15, 2008, 10:32 PM
 
An interesting and fairly well known science writer, Forest Mims III, who has a column published in my local paper, points out that weather observation data is not always accurate. I mentioned this in my somewhat tangential discussion with Oreo Cookie in the Climate Change and Fear thread, but I didn't have anywhere near the data that Mims presents in his above linked article.

Among issues he presents include things like 69% of 534 surveyed weather stations were physically located in spots that obviously had an impact on their temperature observations (and by the way were contrary to the NOAA guidelines for locating weather observation equipment). While NASA and NOAA were not exactly happy to learn of these problems, at least they listened and may have done something to prevent future inaccurate measurements.

I don't know if any of these observation stations was there before the parking lot/air conditioning plant/road/etc. that caused them to no longer be reliable, or if the people who installed them just didn't pay attention, or maybe they just didn't care. But I think that if we're going to be effective at reducing how much impact humans have on the environment, we should have accurate data on that impact, its degree and speed, and where it's actually occurring as opposed to natural phenomena causing whatever's happening.

Is is possible that this sort of thing, combined with new information like this regarding the fact that 2007 has been the coolest this decade may change some people's minds about how fast we're affecting the planet? Or is it more likely that too many people hold such polar and ingrained beliefs about the issue that, regardless of the data, they'll hang onto those beliefs to the end?

Glenn -----OTR/L, MOT, Tx
     
finboy
Registered User
Join Date: Mar 2000
Location: Garden of Paradise Motel, Suite 3D
Status: Offline
Reply With Quote
Dec 15, 2008, 10:37 PM
 
It's passed from the religious to the political at this point, and there are lots of folks (many indoctrinated in public schools) who will go to their graves believing that global warming is upon us. Or even better, believing that "global warming is demonstrated by the current trend toward global cooling." Wow.
     
The Crook
Dedicated MacNNer
Join Date: Sep 2008
Status: Offline
Reply With Quote
Dec 15, 2008, 11:27 PM
 
Global warming deniers are generally only a small number of hard-core conservatives.

The level of support for their cause tends to get over-represented on an internet message board, while the scientific consensus gets underemphasized. There is no disagreement among all the major scientific organizations except among a few crackpot individuals on boards like these. No amount of "respect" given to their views will change their minds.

And I tend to think there's a fair amount of internet forum psychology behind posts like ghporter's original post. At PWL, it's socially acceptable and a sign of independence to doubt global warming in whole or in part. I tend to feel no such desire to ingratiate myself with global warming deniers anywhere.

The subject of the linked article is interesting, but we have no idea what it means in any sort of scientific context. What do the major scientific organizations have to say about it? The conclusion someone might draw from that article is that global warming isn't nearly as bad as previously thought because the data is skewed towards warmer temperatures. I find that very difficult to believe given the breadth of the consensus that global warming is real, it's man-made, and it's urgent.

Crooked Member of the MacNN Atheist Clique.
     
Wiskedjak
Posting Junkie
Join Date: Jun 2002
Location: Calgary
Status: Offline
Reply With Quote
Dec 16, 2008, 09:31 AM
 
Originally Posted by finboy View Post
It's passed from the religious to the political at this point, and there are lots of folks (many indoctrinated in public schools) who will go to their graves believing that global warming is upon us. Or even better, believing that "global warming is demonstrated by the current trend toward global cooling." Wow.
Yep, and visa-versa as well. There are lots of folks who will go to their graves denying that climate change is upon us even if the ice sheets disappear, sea levels rise and "extreme weather" becomes the norm.
     
stupendousman
Addicted to MacNN
Join Date: Nov 2003
Status: Offline
Reply With Quote
Dec 16, 2008, 09:49 AM
 
Originally Posted by Wiskedjak View Post
Yep, and visa-versa as well. There are lots of folks who will go to their graves denying that climate change is upon us even if the ice sheets disappear, sea levels rise and "extreme weather" becomes the norm.
Pet peeve time....

Confusing "climate change" with "man-made climate change".

I don't think anyone denies that the climate is changing, has changed, and always has. What is in debate is how much influence man is having in how it currently has been changing, and the extend to which it has changed. Sometimes I believe that people who do this are just making a mistake in the precision of their words. Other times, I'm pretty sure it's done on purpose in order to try and distort the arguments against man-made climate change.

After eons of radical climate changing with zero evidence that there was any influence by man, some are skeptical that circumstantial evidence (which might be skewed by a lack of good controls) might show that man might have some small effect on what the weather is doing, and that even if man is having some effect that the "solutions" offered are unnecessary and won't provide any kind of reasonable cost/benefit incentive.
     
Warren Pease
Mac Enthusiast
Join Date: Jan 2007
Status: Offline
Reply With Quote
Dec 16, 2008, 09:55 AM
 
So what is the effect? He walks us through his hypothesis, that the location of stations biases towards warmer temperatures, but he doesn't actually tell us how much this bias is. And to be fair to the author, it's not his research, rather just a rehashing of old arguments.

Originally Posted by http://en.wikipedia.org/wiki/Urban_heat_island
Note that not all cities show a warming relative to their rural surroundings. For example, Hansen et al. (JGR, 2001) adjusted trends in urban stations around the world to match rural stations in their regions, in an effort to homogenise the temperature record. Of these adjustments, 42% warmed the urban trends: which is to say that in 42% of cases, the cities were getting cooler relative to their surroundings rather than warmer. One reason is that urban areas are heterogeneous, and weather stations are often sited in "cool islands" - parks, for example - within urban areas.
I think the 'citizen scientists' need to follow through with the research, rather than just coming up with an idea. They apparently have pretty unfettered access to temperature data, seeing as they blog about it before the government agencies that collect it have time to look at it. It shouldn't be too hard to compare the temperature record of each station they visited with nearby rural stations and come up with a number.

Regardless, surface temperature are only one subset of the myriad data that supports global warming.
     
stupendousman
Addicted to MacNN
Join Date: Nov 2003
Status: Offline
Reply With Quote
Dec 16, 2008, 09:58 AM
 
Originally Posted by Warren Pease View Post
Regardless, surface temperature are only one subset of the myriad data that supports global warming.
The world has warmed and cooled forever. I don't think we have any additional data to prove that.
     
Warren Pease
Mac Enthusiast
Join Date: Jan 2007
Status: Offline
Reply With Quote
Dec 16, 2008, 10:31 AM
 
Originally Posted by stupendousman View Post
The world has warmed and cooled forever. I don't think we have any additional data to prove that.
We don't have any additional data, other than surface stations, that the earth has cooled and warmed historically? Not sure I understand what you are getting at...
     
ghporter  (op)
Administrator
Join Date: Apr 2001
Location: San Antonio TX USA
Status: Offline
Reply With Quote
Dec 16, 2008, 10:42 AM
 
Originally Posted by The Crook View Post
Global warming deniers are generally only a small number of hard-core conservatives.

The level of support for their cause tends to get over-represented on an internet message board, while the scientific consensus gets underemphasized.
What if that consensus is based on flawed, incorrect data? Shouldn't real scientists take the accuracy and reliability of all of their data seriously enough to relook both the rate and destination of any change in light of knowing that a lot of data is either skewed or completely wrong?

Glenn -----OTR/L, MOT, Tx
     
OreoCookie
Moderator
Join Date: May 2001
Location: Hilbert space
Status: Offline
Reply With Quote
Dec 16, 2008, 11:02 AM
 
I'm writing a longer reply at the moment, but I can just finish this after work.
I don't suffer from insanity, I enjoy every minute of it.
     
Warren Pease
Mac Enthusiast
Join Date: Jan 2007
Status: Offline
Reply With Quote
Dec 16, 2008, 12:41 PM
 
Originally Posted by ghporter View Post
What if that consensus is based on flawed, incorrect data? Shouldn't real scientists take the accuracy and reliability of all of their data seriously enough to relook both the rate and destination of any change in light of knowing that a lot of data is either skewed or completely wrong?
Scientists should be, and are, concerned about the accuracy and precision of their data. Where again has it been shown that the data is "completely wrong?"

In the broader scope of things, I don't think it proves anything that 2007 may be the coolest of the decade with respect to global warming. Eight of the ten warmest years have occurred in the past 10 years. The solar cycle has been waning over this period which may explain the slight cooling seen within the decade, but as the solar cycle gears up again, along with other factors, I can see the temperatures rising with new record highs.
     
stupendousman
Addicted to MacNN
Join Date: Nov 2003
Status: Offline
Reply With Quote
Dec 16, 2008, 12:50 PM
 
Originally Posted by Warren Pease View Post
Not sure I understand what you are getting at...
The Earth's climate is always changing. You don't don't have to convince us. At one point, a good chunk of the world was cold and covered in ice. We get it. No one really disputes that.

What is in dispute is HOW MUCH weather is changing and WHY. I understand though it's just easier for people who want to convince those not paying attention that it's man-made, if you simply say that the dispute is based on something more easily proven like the fact that our climate changes - sometimes getting warmer, sometimes hotter.
     
OreoCookie
Moderator
Join Date: May 2001
Location: Hilbert space
Status: Offline
Reply With Quote
Dec 16, 2008, 02:27 PM
 
@stupenduousman
Global warming is concerned with the earth climate, not the weather.
I don't suffer from insanity, I enjoy every minute of it.
     
Wiskedjak
Posting Junkie
Join Date: Jun 2002
Location: Calgary
Status: Offline
Reply With Quote
Dec 16, 2008, 02:47 PM
 
Originally Posted by stupendousman View Post
Pet peeve time....

Confusing "climate change" with "man-made climate change".

I don't think anyone denies that the climate is changing, has changed, and always has. What is in debate is how much influence man is having in how it currently has been changing, and the extend to which it has changed. Sometimes I believe that people who do this are just making a mistake in the precision of their words. Other times, I'm pretty sure it's done on purpose in order to try and distort the arguments against man-made climate change.

After eons of radical climate changing with zero evidence that there was any influence by man, some are skeptical that circumstantial evidence (which might be skewed by a lack of good controls) might show that man might have some small effect on what the weather is doing, and that even if man is having some effect that the "solutions" offered are unnecessary and won't provide any kind of reasonable cost/benefit incentive.
It wasn't much more than a year ago that Conservatives, even on this board, were arguing against the existence of *any* form of climate change, man-made or not. Some conservatives have obviously not received the memo and still argue that there is *no* climate change happening. finboy is obviously arguing that there is no such thing as "global warming".

So, I wonder, if you are saying that no one denies that the climate is changing, are you arguing *against* the OP?
     
stupendousman
Addicted to MacNN
Join Date: Nov 2003
Status: Offline
Reply With Quote
Dec 16, 2008, 04:11 PM
 
Originally Posted by OreoCookie View Post
@stupenduousman
Global warming is concerned with the earth climate, not the weather.
The earth climate is exactly what I was referring to.

Climate:The meteorological conditions, including temperature, precipitation, and wind, that characteristically prevail in a particular region.

The meteorological conditions, including temperature, precipitation, and wind, that characteristically prevail on Earth has been changing from the beginning of time.
     
stupendousman
Addicted to MacNN
Join Date: Nov 2003
Status: Offline
Reply With Quote
Dec 16, 2008, 04:19 PM
 
Originally Posted by Wiskedjak View Post
It wasn't much more than a year ago that Conservatives, even on this board, were arguing against the existence of *any* form of climate change, man-made or not. Some conservatives have obviously not received the memo and still argue that there is *no* climate change happening. finboy is obviously arguing that there is no such thing as "global warming".

So, I wonder, if you are saying that no one denies that the climate is changing, are you arguing *against* the OP?
No. I believe that if you read what the OP wrote in the linked thread, he concedes that the Earth's climate may be changing. In fact, he specifically mentions possible cooling. What he appears to be arguing against is that there is conclusive evidence of catastrophic unstoppable man-made global warming.

Again, we can have climate change without having CUMMGW™. You can have one without the other.
     
OreoCookie
Moderator
Join Date: May 2001
Location: Hilbert space
Status: Offline
Reply With Quote
Dec 16, 2008, 04:33 PM
 
Originally Posted by stupendousman View Post
The earth climate is exactly what I was referring to.

Climate:The meteorological conditions, including temperature, precipitation, and wind, that characteristically prevail in a particular region.
No. Climate is weather averaged over time.
I don't suffer from insanity, I enjoy every minute of it.
     
OreoCookie
Moderator
Join Date: May 2001
Location: Hilbert space
Status: Offline
Reply With Quote
Dec 16, 2008, 04:46 PM
 
Originally Posted by ghporter View Post
I don't know if any of these observation stations was there before the parking lot/air conditioning plant/road/etc. that caused them to no longer be reliable, or if the people who installed them just didn't pay attention, or maybe they just didn't care.
It's not a matter of `reliability', it's a matter of systematic errors. All quantities measured by scientists have systematic and statistic uncertainties. So a `real' result of a measurement of a voltage looks like this:

1.2 V +/- 0.2 V +/- 0.1 V

For the lack of subscripts, I have assumed that the errors are symmetric (they needn't be) and that I wanted to separate systematic and statistic uncertainties (which is often not done, instead a combined uncertainty is used). Measurements without uncertainties are meaningless in science. Needless to say that it's standard to model a bias (if necessary), but this isn't always necessary.

Think of a scale that always shows 1 pound less than the actual weight. If you want to know how much weight you have lost (or gained), it will still be accurate, even though your wife like to keep this setting, because that's what the clothes weigh (that's the justification my mom has used when I was a kid ). Similarly, if a thermometer shows temperatures that are too 1 or 2 ˚C too warm, but consistently too warm, then I can still reliably tell whether the average temperature has changed or not.

So if they officially write that they accept a systematic error of 20 %, then a systematic error of this magnitude enters into the considerations. You may say that `this is very large,' but this is another matter. The only measurements scientists don't trust are those that do not have an error estimate associated to them.
Originally Posted by ghporter View Post
But I think that if we're going to be effective at reducing how much impact humans have on the environment, we should have accurate data on that impact, its degree and speed, and where it's actually occurring as opposed to natural phenomena causing whatever's happening.
First of all, all predictions regarding climate change (and weather patterns) are only statistical, meaning there is a probability that an event occurs -- or not. The second, equally significant characterization is the `spread' of the outcome: if the possible results are very spread out, e. g. the events that the sun will shine at 25 ˚C or snow at -5 ˚C are both possible at similar probability, the forecast will not be very useful. So let's start with weather forecasts: you don't start with `one set of values' (e. g. the measurement data of the temperatures at the weather stations at a certain point in time), you start with an ensemble of temperatures, pressures, water and moisture content of the air, etc. For each of these initial data, you simulate how the weather patterns will change according to the model at hand. After simulating the ensemble of initial data, you can analyze the distribution of final results. The distribution of final results has an `average' and a `spread' again which gives you an indication what will likely happen and how likely deviations are.

Arguably weather forecasts, while obviously not perfect, have become a very reliable and accurate tool, also in every-day life. Whether people are proponents or opponents of global warming, it's fair to say that the vast majority of them takes the weather forecast as a serious indication of the weather (e. g. when planning trips or whatnot). Just on a phenomenological level, if you consider weather forecasting as a black box, it has become a very successful and widely accepted tool.

Weather forecasts are (by their very nature of being local forecasts) arguably much more sensitive to uncertainties in the initial data, e. g. problems in local weather stations may manifest themselves much more directly (in the evening news) if they aren't properly handled. This problem is tackled in several ways: (i) One obvious way is to have multiple methods to measure a quantity (e. g. surface temperature measurements by satellite and weather stations). This gives you more accurate results or more accuracy with respect to slightly different points of view. (ii) The uncertainties of measurements enter via choosing appropriate values of mean and spread of your initial data.

If we turn to modelling the climate, i. e. `averaged weather' over a period of time, then of course the things that have been learned modelling the weather enter either directly or indirectly. Climate models are examples of very complex `multiscale systems' meaning that processes on the molecular level (absorption or resorption of radiation in the atmosphere) as well as on the global scale (currents in the oceans, formation/melting of ice on the poles) influence the weather on the same scale. On first glance, this seems like an impossible challenge, after all, if we cannot predict the weather beyond, say 14 days or so, how can we predict the climate? The big benefit here is that we are looking at averages. Just think of the stock market: if you look at the stock of a particular company, than the share price during a particular day may vary considerably. So people look at averages that behave in a much more predictable fashion, you don't see the `small spikes' anymore. Averages are `more stable', both numerically and in the sense of measurement results.

These models are also checked for consistency in several different ways, e. g. you can try and see whether your algorithms reproduce the climate patterns of the past or you can run a more complex simulation on shorter time scale to check whether the results obtained with the `simpler'* simulation give matching results. The simplest one would compare daily or weekly temperatures with the result of weather simulations.

* The simulations on longer scales needn't be simpler. In many instances, different effects may be important on different time scales. Take the precession of the earth's axis in time. This is irrelevant for today's weather forecast and probably also for the forecast of the weather in the next 100 years. But important if you want to simulate the earth's climate for 500,000 years.


So what about your initial question: are the results of climate simulations meaningless because of large systematic uncertainties in some of the measurements? Well, they are taken into account already in several ways: (i) Optimization of initial data used in weather forecasts. This is done for a long, long time and you can `immediately' check whether forecasts are accurate or not. Note that land-based measurements aren't available for all regions anyway, so the initial data for weather forecasts need to be obtained and estimated by other means. (ii) Weather and climate predictions are made using a spread of initial conditions and not just one set of initial temperatures, pressures, etc. (iii) Cross-checking of initial data using independent ways of measurement whenever possible.

Strictly speaking, this is not a fourth point, but it is connected to this question: some people here allude that scientists `tweak' the initial data to get a result that is more to their liking. These simulations are not simple evaluations of a function at a point. They are the result of many, many cpu hours of numerics and you can't possibly know how to fine-tune your initial data to get a result of their liking.
Originally Posted by ghporter View Post
Is is possible that this sort of thing, combined with new information like this regarding the fact that 2007 has been the coolest this decade may change some people's minds about how fast we're affecting the planet? Or is it more likely that too many people hold such polar and ingrained beliefs about the issue that, regardless of the data, they'll hang onto those beliefs to the end?
Think of a biased coin that gives you tails with a probability of 70 % and heads with only 30 %. (This means if you throw the coin very often, the relative frequency of tails will approach 70 % while that of heads will approach 30 %.) A single (or very few) `measurements' will not tell you anything about a trend. The fact that humans influence the global climate is not the question anymore, it's a question of how much. This is even true if you don't believe in an anthropogenic contribution to global warming, then the answer would be `not appreciably much.'
I don't suffer from insanity, I enjoy every minute of it.
     
stupendousman
Addicted to MacNN
Join Date: Nov 2003
Status: Offline
Reply With Quote
Dec 16, 2008, 05:10 PM
 
Originally Posted by OreoCookie View Post
No. Climate is weather averaged over time.
Take your argument up with dictionary.com

Besides, weather averaged over time has changed as well.
     
Wiskedjak
Posting Junkie
Join Date: Jun 2002
Location: Calgary
Status: Offline
Reply With Quote
Dec 16, 2008, 06:27 PM
 
Originally Posted by stupendousman View Post
Again, we can have climate change without having CUMMGW™. You can have one without the other.
No doubt. And, some of us were arguing this point before conservatives switched to it from "There is no global warming".
     
ghporter  (op)
Administrator
Join Date: Apr 2001
Location: San Antonio TX USA
Status: Offline
Reply With Quote
Dec 16, 2008, 10:47 PM
 
Originally Posted by OreoCookie View Post
It's not a matter of `reliability', it's a matter of systematic errors. ... The only measurements scientists don't trust are those that do not have an error estimate associated to them.
The issue isn't "are these measurements accompanied with error estimates" but rather "are those error estimates accurate and appropriate?" Having a standard for where and how to install a temperature sensor, and associated expected error ranges for such sensors is only part of the problem. What about UN-NOTED errors, such as seasonally biased errors like having a paved area on the north or south side of the sensor? Such siting is explicitly warned against in the NOAA standards, so if the person installing that sensor ignores that standard, what other standards are being ignored? Further, if 50 different people install the same sensor in 50 different, non-standard sites, how can this unknown variation in siting be accounted for? I'll admit I'm only a "jackleg" scientist; I don't have a lot of formal training in building large experiments. But if you have such a large source of error that is impossible to account for due to data sources being installed without regard to installation standards, how can you use that data?

Originally Posted by OreoCookie View Post
So what about your initial question: are the results of climate simulations meaningless because of large systematic uncertainties in some of the measurements?
I never even intimated "meaningless." Instead I asked if knowing that you have used inaccurate, slanted and poorly collected data should trigger a reevaluation of the conclusions based on that data.

Originally Posted by OreoCookie View Post
Think of a biased coin that gives you tails with a probability of 70 % and heads with only 30 %. (This means if you throw the coin very often, the relative frequency of tails will approach 70 % while that of heads will approach 30 %.) A single (or very few) `measurements' will not tell you anything about a trend. The fact that humans influence the global climate is not the question anymore, it's a question of how much. This is even true if you don't believe in an anthropogenic contribution to global warming, then the answer would be `not appreciably much.'
Climate and weather are far more complex than a simple coin toss, even a known to be biased coin toss. We do not yet know all of the factors involved in even medium-term climate change. Knowing that you're using poor data, especially when you can't even state "all the data is off by X amount" but rather "we don't know how far off any of the data." So building a simplified statistical model of this extremely large and enormously complex system based on this faulty data seems to be an exercise in building statistical models, not in building a truly representative model.

I'd personally like to see NOAA evaluate EVERY non-government owned sensor location, and decertify any and all that don't comply with their standards. I'd also like to see them regularly reevaluate those sites to ensure that no changes have been made that impact the known accuracy (and thus error range) of the sensors.

I'm not suggesting that climate scientists are idiots, only that they seem to be taking data that cannot be reasonably relied upon at face value.

Glenn -----OTR/L, MOT, Tx
     
The Crook
Dedicated MacNNer
Join Date: Sep 2008
Status: Offline
Reply With Quote
Dec 16, 2008, 11:03 PM
 
Originally Posted by ghporter View Post
I'm not suggesting that climate scientists are idiots, only that they seem to be taking data that cannot be reasonably relied upon at face value.
Grists's "How to Talk to a Global Warming Skeptic" feature covers this point about the temperature data being "unreliable"

Go ahead, put aside the direct surface temperature measurements -- global warming is also indicated by:
  • Satellite measurements of the upper and lower troposphere
  • Weather balloons show very similar warming
  • Borehole analysis
  • Glacial melt observations
  • Declining arctic sea ice
  • Sea level rise
  • Proxy Reconstructions
  • Rising ocean temperature

All of these completely independent analyses of widely varied aspects of the climate system lead to the same conclusion: the Earth is undergoing a rapid and substantial warming trend. Looks like the folks at NASA and CRU know what they are doing after all.

Crooked Member of the MacNN Atheist Clique.
     
OreoCookie
Moderator
Join Date: May 2001
Location: Hilbert space
Status: Offline
Reply With Quote
Dec 17, 2008, 06:24 AM
 
Originally Posted by ghporter
The issue isn't "are these measurements accompanied with error estimates" but rather "are those error estimates accurate and appropriate?"
That's always a subject of scrutiny. But in an `experiment' that large, you are bound to have erroneous data, even if the placement of weather stations is impeccable. A malfunctioning thermometer or a transmission error can enter unreliable data.

Again, this is a problem that isn't new and not a peculiarity of climate predictions. And I still don't know whether any data from weather stations actually enters climate predictions. They use grid sizes of 100+ km for their simulations, so I strongly doubt it!
Originally Posted by ghporter
What about UN-NOTED errors, such as seasonally biased errors like having a paved area on the north or south side of the sensor?
You can never quantify all possible systematic errors. That's a reality of science. Nevertheless, you can try to minimize them by using several different methods of measurement or calibration. One of the first (and to me, most boring things) was to learn to calibrate instruments before (or rather for) use: to calibrate an ultra-low temperature thermometer, you'd use other `known' thermometers and extrapolate from way mark to way mark.
Originally Posted by ghporter
Climate and weather are far more complex than a simple coin toss, even a known to be biased coin toss.
Yes, but this was just a response to `well, 2007 was a cold year, so how come we speak of global warming?'
A single data point (or even a few) cannot give you any trends. The interpretation of the results is statistical -- just as a coin toss.
Originally Posted by ghporter
We do not yet know all of the factors involved in even medium-term climate change.
No, we don't. And there are other factors why more research is necessary (e. g. large grid size in the range of 100 km), but many of them are alleviated with the increase in cpu horse power and other research. Scientists always try to include as many factors as necessary, but nothing will help the fact that a climate model is always a simplified picture of what actually happens.

A few years ago, weather simulations couldn't calculate ensembles of initial data, now we can. Ditto for climate simulations. Things improve. But that doesn't change that the current consensus is (which has been built over many, many hundred man hours of research and cpu time) that there is a significant anthropogenic contribution to global warming.

Whenever you hear someone say that `because effect has been neglected, the results can't be trusted.' The average increase in temperature that is projected to happen within this century is a consequence of many, many different, interdependent processes (climate scientists like to call them feedback loops). You cannot isolate one mechanism or another and assign responsibility to it. You simply have to try it out (e. g. run the same simulations with and without a certain mechanism). The other reason why things cannot be arbitrary is because climate simulations must obey restrictions placed upon them by weather simulations. They are the `conservation laws'. This provides a sanity check.

So if a weather simulations starting from `the same' data yields compatible results (meaning you need to average the results of the weather simulation suitably and compare them to the predictions of the climate simulation), then you have another sanity check.

So individually, you have a point: we certainly don't know enough now and lots of things need to be explored. But that doesn't change that the current state of research simply is what it is.
Originally Posted by ghporter
Knowing that you're using poor data, especially when you can't even state "all the data is off by X amount" but rather "we don't know how far off any of the data." So building a simplified statistical model of this extremely large and enormously complex system based on this faulty data seems to be an exercise in building statistical models, not in building a truly representative model.
How do you know that climate models actually use poor data from the weather stations? Grid sizes are way too large to include these small data points.
Originally Posted by ghporter
I'm not suggesting that climate scientists are idiots, only that they seem to be taking data that cannot be reasonably relied upon at face value.
I was trying to convince you that any scientists, especially those under tremendous scrutiny never take data at face value. As long as you associate reasonable uncertainties (systematic and statistic) to your data, scientists never take measurement results at face value. `They know that you can't measure the height of a penny with a yard stick.' Your impression is incorrect.
I don't suffer from insanity, I enjoy every minute of it.
     
ghporter  (op)
Administrator
Join Date: Apr 2001
Location: San Antonio TX USA
Status: Offline
Reply With Quote
Dec 17, 2008, 11:23 AM
 
This sort of science should not be so opaque. For the public to understand the problem and get behind ANY solution for it, they need to have plenty of confidence that the people looking at the problem are looking at good, accurate data in such a manner as to correctly see what effect man has on the world. While I'm sure that there are plenty of people taking this subject VERY seriously and being extremely fastidious about the data they include in their analysis, I am also absolutely certain that NOAA has included erroneous and even just plain "wrong" data in the "historical record" of weather in the U.S. because they were unaware of poorly installed sensors.

NOAA depends on a huge number of individual observation stations for hour-by-hour weather observation data. This data is integrated by NOAA and a number of other agencies (like the Air Force Weather Agency) to provide a global picture of the weather as it happens. ALL of these agencies use data from all sorts of sources, including "PIREPS," pilot reports of weather conditions at various altitudes. So Joe Blow's weather station on his back 40 is given plenty of credence by NOAA, and though they don't enforce anything, they expect that Joe is going to have installed his sensors according to their standards. All of these weather agencies use this data in near-real time and cannot afford the time or manpower to examine the data for inconsistency or bias. Since so much of climate change reporting is made in reference to historical data, this lack of rigor in data collection bothers me.

Originally Posted by OreoCookie View Post
I was trying to convince you that any scientists, especially those under tremendous scrutiny never take data at face value. As long as you associate reasonable uncertainties (systematic and statistic) to your data, scientists never take measurement results at face value. `They know that you can't measure the height of a penny with a yard stick.' Your impression is incorrect.
Here's where the opacity of climate change science comes in. I have worked closely with weather professionals, from the folks that maintain sensors to the folks that work the large-scale simulations. Their work has always depended on the sensors to be accurate and valid, without bias or systematic error beyond some very small margin. For example, the "great new wind gauge" installed throughout the Department of Defense had no moving parts to fuss with and being all solid state was a joy to work on. Until the maintainers noted that these sensors began differing from their calibration references almost immediately. These "hot wire" sensors measure wind speed and direction by noticing how much current it takes to keep a circular array of these "hot wires" at a given temperature. The problem is that birds figured this out really quickly and started "cooking their crickets" by impaling them on the hot wires! If regular checks of these weather devices weren't required, a whole lot of data would have been collected (and pilots would have had a lot of problems with winds on take off and landing) before the sensors actually failed entirely and people found out that they were acting as "bird barbies." (I am NOT making that up-I had to process a number of hardware failure reports on these devices, along with the resulting modification orders.) So I'm not just making up issues to argue. I have real experience that gives me some pretty strong reason to worry about how all of this information is used.

So I think it comes down to this: Since weather agencies are reporting this data as accurate, do climate scientists actually verify that the data was collected properly? How would they do that, since NOAA doesn't vet that information in the first place? And most importantly, why don't these climate scientists bother to tell us about how careful they are? I know more about the Large Hadron Collider than I do about domestic climate change science, and that really bothers me.

Glenn -----OTR/L, MOT, Tx
     
The Crook
Dedicated MacNNer
Join Date: Sep 2008
Status: Offline
Reply With Quote
Dec 17, 2008, 02:43 PM
 
Originally Posted by ghporter View Post
Since so much of climate change reporting is made in reference to historical data, this lack of rigor in data collection bothers me.
Read the Grist article.

And that will be my response to you until you do- in every single reply and every single global warming thread. It will help if you read the whole feature, too. Excellent reading for both global warming deniers and proponents alike.

Crooked Member of the MacNN Atheist Clique.
     
ghporter  (op)
Administrator
Join Date: Apr 2001
Location: San Antonio TX USA
Status: Offline
Reply With Quote
Dec 17, 2008, 03:56 PM
 
Originally Posted by The Crook View Post
Read the Grist article.

And that will be my response to you until you do- in every single reply and every single global warming thread. It will help if you read the whole feature, too. Excellent reading for both global warming deniers and proponents alike.
I'm not at all denying anything. I just think that climate scientists in general should do a better job of everything, especially making it plain that they have ensured they're using good data.

And while Grist's entries about "the temperature record is simply unreliable" bring up good points, I think I'm saying something in the opposite direction. I think that the data showing the direction of change is indeed compelling. I just think that scientists open themselves up to criticism by not strongly ensuring that there is no doubt about the accuracy, validity, consistency, and other qualities of their data. I also think that NOAA would do a much better job of weather (not climate) analysis if they were to require compliance with their guidelines for siting and accuracy.

(I'm NOT going to devote the rest of the week reading all the articles Grist links on that page. If this specific article wasn't what you meant, please let me know.)

Basically, this is akin to the numerous groups that wanted to either confirm or refute General Relativity through total eclipse observations. Campbell originally had what seemed to be really strong proof that Einstein was wrong, but because his tools were not state of the art, he chose to wait for another eclipse-and managed to see that his original observations had been flawed. If we use flawed equipment (whether it's flawed in how it works or where it's located), we make it easy to refute conclusions we make based on what that equipment tells us, whether or not we are able to account for error, bias and such in that data.

Glenn -----OTR/L, MOT, Tx
     
mattyb
Addicted to MacNN
Join Date: Feb 2008
Location: Standing on the shoulders of giants
Status: Offline
Reply With Quote
Dec 17, 2008, 05:27 PM
 
This book, is a VERY interesting read.
     
OreoCookie
Moderator
Join Date: May 2001
Location: Hilbert space
Status: Offline
Reply With Quote
Dec 17, 2008, 05:42 PM
 
Originally Posted by ghporter View Post
This sort of science should not be so opaque.
Neither should quantum mechanics or cardiothoracic surgery. But these topics are that complicated.

The criticism you voice concerns details (which is good), but when experts are linked to that tell you that `these factors are taken into consideration,' you seem to insist that `you don't believe the statement to that effect.'
Originally Posted by ghporter View Post
While I'm sure that there are plenty of people taking this subject VERY seriously and being extremely fastidious about the data they include in their analysis, I am also absolutely certain that NOAA has included erroneous and even just plain "wrong" data in the "historical record" of weather in the U.S. because they were unaware of poorly installed sensors.
Again, the misunderstanding is on your side: there is no `erroneous data', just data with a larger-than-optimal systematic uncertainty. Although sometimes called systematic error, it's more accurate to call it uncertainty associated to your data, because its root cause is usually not improper handling in experiments. Other times (like now), just because of sheer size of experiment, even under close-to-optimal conditions, you'd need to pre-process your data (something very common in science, as you know the big detectors at the LHC have several `sieves' for events that filter only the most relevant and accurate data). Some members have given you ways (abstract and concrete) of how to ensure quality of your data (sanity checks, cross-checks using other, independent methods of measurement, for instance).

What are you missing here?
Originally Posted by ghporter View Post
All of these weather agencies use this data in near-real time and cannot afford the time or manpower to examine the data for inconsistency or bias. Since so much of climate change reporting is made in reference to historical data, this lack of rigor in data collection bothers me.
Weather ≠ climate. Even if you are sceptical about the quality of the data, do you think the end result, the weather forecasts you see on TV or various web pages, is comparatively accurate? If there is some big, unsolved problem with the data from the weather stations, shouldn't it show up first (and much more noticeably) in weather forecasts? This is an important question.
Originally Posted by ghporter View Post
For example, the "great new wind gauge" installed throughout the Department of Defense had no moving parts to fuss with and being all solid state was a joy to work on. Until the maintainers noted that these sensors began differing from their calibration references almost immediately. [...] I have real experience that gives me some pretty strong reason to worry about how all of this information is used.
Your example is very different from the original topic at hand: you're talking about real-time data for a specific use. I'd be concerned if the US military (just like any other large airport, I presume) does not maintain its weather monitoring system.
Originally Posted by ghporter View Post
So I think it comes down to this: Since weather agencies are reporting this data as accurate, do climate scientists actually verify that the data was collected properly?
Again, same old mistake: professional weather agencies always give the margin of error associated to their data. The data is only accurate within that margin of error. I don't know how this is done concretely in practice. Warren Pease has quoted a study on whether weather stations in urban environments tend to under- or overestimate temperatures. So yes, scientists think about this.
Originally Posted by ghporter View Post
How would they do that, since NOAA doesn't vet that information in the first place? And most importantly, why don't these climate scientists bother to tell us about how careful they are? I know more about the Large Hadron Collider than I do about domestic climate change science, and that really bothers me.
There is no such thing about domestic climate change science, the climate needs to be understood and simulated globally. I've linked to a nice 12-page article in the other thread. Even if you are sceptical about portions of it, it does a good job at explaining how weather and climate predictions work, and goes into some detail.
I don't suffer from insanity, I enjoy every minute of it.
     
stupendousman
Addicted to MacNN
Join Date: Nov 2003
Status: Offline
Reply With Quote
Dec 17, 2008, 05:43 PM
 
Originally Posted by Wiskedjak View Post
No doubt. And, some of us were arguing this point before conservatives switched to it from "There is no global warming".
As it's defined by liberals and those who whose religion is based on faith in MMGW, they still do.

Again, you're confusing the normal cyclical (or due to natural causes) fluctuations in global temperature which no one really every denied, and the notion that the world is getting irrationally hotter (caused by man) with no end in sight unless we do a bunch of stuff that even the scientists who support the idea admit will do little to effect a change.
     
ghporter  (op)
Administrator
Join Date: Apr 2001
Location: San Antonio TX USA
Status: Offline
Reply With Quote
Dec 17, 2008, 08:12 PM
 
Originally Posted by OreoCookie View Post
Neither should quantum mechanics or cardiothoracic surgery. But these topics are that complicated.
Quantum mechanics does not require a huge buy-in by the public. Cardiothoracic surgery has enough transparency to allow the public to understand that there are good surgeons and bad surgeons. What climate scientists need to do is make the basis of their claims clear, even if they can't explain the data in detail to the common layman.

Originally Posted by OreoCookie View Post
The criticism you voice concerns details (which is good), but when experts are linked to that tell you that `these factors are taken into consideration,' you seem to insist that `you don't believe the statement to that effect.'
Because these scientists don't seem to bother with telling the public what they're doing, just what their conclusions are. Instead of saying "don't worry about this level of detail, we have it covered" they should be saying "we've had to deal with a huge amount of data, and some of it hasn't been very simple or easy to use, but we've managed to handle it." I haven't seen that yet.

Originally Posted by OreoCookie View Post
Again, the misunderstanding is on your side: there is no `erroneous data', just data with a larger-than-optimal systematic uncertainty. Although sometimes called systematic error, it's more accurate to call it uncertainty associated to your data, because its root cause is usually not improper handling in experiments. Other times (like now), just because of sheer size of experiment, even under close-to-optimal conditions, you'd need to pre-process your data (something very common in science, as you know the big detectors at the LHC have several `sieves' for events that filter only the most relevant and accurate data). Some members have given you ways (abstract and concrete) of how to ensure quality of your data (sanity checks, cross-checks using other, independent methods of measurement, for instance).

What are you missing here?
I don't see it as "systematic uncertainty" when there are numerous UNKNOWN uncertainties, such as day-to-day temporal biases (due to solar heating of nearby items around the sensor). Or not. We do not know whether there are absolutely clean data sources or not because there seems to be very little review and QC of this weather observation data, at least outside of DoD. Where does the independent measurement of local conditions come from? Is IT correctly collected? How does anyone know?

Originally Posted by OreoCookie View Post
Weather ≠ climate. Even if you are sceptical about the quality of the data, do you think the end result, the weather forecasts you see on TV or various web pages, is comparatively accurate? If there is some big, unsolved problem with the data from the weather stations, shouldn't it show up first (and much more noticeably) in weather forecasts? This is an important question.
Again, climate change is stated in terms of historical weather observations. If those observations are off (not exactly "wrong," just not well synchronized, or just plain off because of the influence of that big swamp cooler system two meters away...), then the "historical observations" that we're talking about are off, and the comparison becomes tainted and maybe even spurious.

Originally Posted by OreoCookie View Post
Your example is very different from the original topic at hand: you're talking about real-time data for a specific use. I'd be concerned if the US military (just like any other large airport, I presume) does not maintain its weather monitoring system.
Real-time data BECOMES the historical record of weather. While it may be fine to use this stuff with a hefty grain of salt when making predictions, it is still the "recorded high temperature for Podunk, Kentucky." That's the point; predictions are not an issue, but the historical recorded high and low temperatures are still the "record" to which climate data is compared.

Originally Posted by OreoCookie View Post
Again, same old mistake: professional weather agencies always give the margin of error associated to their data. The data is only accurate within that margin of error. I don't know how this is done concretely in practice. Warren Pease has quoted a study on whether weather stations in urban environments tend to under- or overestimate temperatures. So yes, scientists think about this.
Again, margin for error is used in predictions, but NOT in "recorded temperatures."

Originally Posted by OreoCookie View Post
There is no such thing about domestic climate change science, the climate needs to be understood and simulated globally. I've linked to a nice 12-page article in the other thread. Even if you are sceptical about portions of it, it does a good job at explaining how weather and climate predictions work, and goes into some detail.
Valid point about "domestic" climate science. However, I do have a fairly good picture of how weather information is collected in the U.S., while I do not have a clue how it is collected in Japan, or Serbia, or South Africa, so I don't know what sort of challenges scientists there face. I consider my country's scientists to have a lot of advantages here, and if they face the data collection challenges I see, then how much bigger is the problem in less affluent countries?

I still say that this whole climate change issue is a real and serious problem. What I'm concerned about is that the public at large, worldwide, need to understand what's going on, and if all the bases aren't covered, it gives naysayers ammunition to pick little holes in simplified explanations and cast doubt on the conclusions, no matter how well built those conclusions are. I am NOT saying that there is a problem with the conclusions. I AM saying that maybe climate scientists should mention something about how goofy weather observations may be used, but the data is properly interpreted so that it's not a problem.

Glenn -----OTR/L, MOT, Tx
     
Helmling
Mac Elite
Join Date: Apr 2005
Status: Offline
Reply With Quote
Dec 17, 2008, 11:20 PM
 
Ok, so maybe all the weather stations in the U.S. are in parking lots, but there seems to be another weather station that's relevant:

http://edition.cnn.com/2008/TECH/sci...f=mpstoryemail
     
OreoCookie
Moderator
Join Date: May 2001
Location: Hilbert space
Status: Offline
Reply With Quote
Dec 18, 2008, 05:56 AM
 
Originally Posted by ghporter
I don't see it as "systematic uncertainty" when there are numerous UNKNOWN uncertainties, such as day-to-day temporal biases (due to solar heating of nearby items around the sensor).
But that's what they are scientifically, uncertainties.


So let's make this simple. I've asked a few questions in my other post, but I think they got lost (mostly, because I hid them well in the middle of my post ).
(1) Do you think (near-term, say up to 3 days) weather forecasts you see on TV or on various web pages are comparatively accurate? Do you take them into consideration when you plan your weekend or go on a trip?
I don't suffer from insanity, I enjoy every minute of it.
     
ghporter  (op)
Administrator
Join Date: Apr 2001
Location: San Antonio TX USA
Status: Offline
Reply With Quote
Dec 18, 2008, 11:07 AM
 
Around here, weather forecasts are not reliably accurate or inaccurate. Local geography has a strong impact on actual weather, and air masses tend to either speed up or (more usually, especially with cold air) slow down, so one day's forecast may be correct, but not for the given timeframe. Funny, but this local geography has been this way for centuries, and you'd think that the effects of being on the transition between the inland uplands and coastal plains would be something that could be fairly easily accounted for.

But again the forecast is not the point, the record of observations is. How can you account for measurements that can be absolutely unbiased AND measurements that can be horribly biased in a number of different ways without knowing which ones are and are not biased? Yeah, that's probably a basic science question, but there it is.

Glenn -----OTR/L, MOT, Tx
     
OreoCookie
Moderator
Join Date: May 2001
Location: Hilbert space
Status: Offline
Reply With Quote
Dec 18, 2008, 11:24 AM
 
Originally Posted by ghporter View Post
But again the forecast is not the point, the record of observations is.
The two are related because weather forecasts are based on that data.

So do you think weather forecasts are accurate enough to be a useful tool in everyday life or not?
I don't suffer from insanity, I enjoy every minute of it.
     
BadKosh
Professional Poster
Join Date: Aug 2007
Location: Just west of DC.
Status: Offline
Reply With Quote
Dec 18, 2008, 12:29 PM
 
it's NOT the data collection that is flawed, but the formulas and conclusions. The inability of computer models to closely track the KNOWN REALITY suggests they don't have the right combination of elements or are using the values incorrectly. So far, the proof isn't there.
     
Helmling
Mac Elite
Join Date: Apr 2005
Status: Offline
Reply With Quote
Dec 18, 2008, 01:50 PM
 
The ice, people! Pay attention to the ice!
     
Chongo
Addicted to MacNN
Join Date: Aug 2007
Location: Phoenix, Arizona
Status: Offline
Reply With Quote
Dec 18, 2008, 06:03 PM
 
Originally Posted by mattyb View Post
This book, is a VERY interesting read.
I heard this guy on Coast to Coast AM a couple of nights ago
Ice Age Now
45/47
     
ghporter  (op)
Administrator
Join Date: Apr 2001
Location: San Antonio TX USA
Status: Offline
Reply With Quote
Dec 18, 2008, 10:58 PM
 
Originally Posted by OreoCookie View Post
The two are related because weather forecasts are based on that data.

So do you think weather forecasts are accurate enough to be a useful tool in everyday life or not?
In South Texas, no. In a lot of places, weather forecasts are only somewhat useful. They are certainly not good for much more than telling you whether or not you might need a heavy coat or umbrella. Typical forecasts predict specific temperatures and provide a "chance of precipitation" value (which is poorly understood by the public), but these forecasts, whether by "Gloria the weather girl" or the National Weather Service, are often so far off that they require hour-by-hour revision by the forecasters, especially when an air mass speeds up or slows down. Meteorologists are extremely bad at predicting how long phenomena like inversion layers will persist. With all that said, at least these predictions are generally "in the ballpark," whether you're talking about San Antonio or Seneca.

The issue I have, which is related to all of this above, is that these forecasts, with huge margins of error, are based on data that is not necessarily dependable. I do not know whether a local forecast might or might not be more accurate if every bit of local data were scrupulously obtained in the most pure and perfect form, but that's moot because local forecasts depend on national and global data to go beyond a few hours in the future.

Glenn -----OTR/L, MOT, Tx
     
OreoCookie
Moderator
Join Date: May 2001
Location: Hilbert space
Status: Offline
Reply With Quote
Dec 19, 2008, 08:30 AM
 
Originally Posted by ghporter View Post
In South Texas, no. In a lot of places, weather forecasts are only somewhat useful.
If you don't think that weather forecasts have become a reliable every-day tool, then you can't be convinced that climate simulations work.
( Last edited by OreoCookie; Dec 19, 2008 at 09:23 AM. )
I don't suffer from insanity, I enjoy every minute of it.
     
ghporter  (op)
Administrator
Join Date: Apr 2001
Location: San Antonio TX USA
Status: Offline
Reply With Quote
Dec 19, 2008, 10:35 AM
 
I didn't say they weren't "reliable." They're just not terribly precise, and I do not know if this is due to lack of ability to simulate future air mass interactions or use of sloppy data or something else. On the very short term—hours to a day or so—weather predictions tend to be close to what actually happens, though the predictions of specific temperatures are often off and again the "chance of precipitation" issue is not well presented. On the longer term, I do see how statistically these blips and bloops smooth out and produce solid trends. That's no big deal to grasp. Depending on a weather forecast to determine whether or not to irrigate crops (or your lawn) around here is a good way to wind up with a dead field. Knowing that it's not a bad idea to have that coat or umbrella with you is very useful, but it's still not precise enough information for the average person to really believe in weather prediction. Meteorologists do a poor job of demonstrating how accurate they are on the log term, and that makes it difficult for the layman to trust them. If weather forecasters produced stats for how close they were versus NWS and others' forecasts, (and even publicly competed against each other in this area), I think there would be a lot better appreciation for their craft. Just think of the color commentary!

But again, how is the normal bloke supposed to be able to appreciate the accuracy of these climate trend lines if he sees his local weather dude telling him "rain today with a high of 56" and it's dry and 60 where he is? I think he'll see that as "these guys are just guessing" and dismiss ALL weather related predictions. And when you add things to this phenomenon like reports about inaccurate and poorly collected weather data, that just makes the average bloke more sure that everything about weather is guesswork.

Glenn -----OTR/L, MOT, Tx
     
OreoCookie
Moderator
Join Date: May 2001
Location: Hilbert space
Status: Offline
Reply With Quote
Dec 19, 2008, 12:10 PM
 
Originally Posted by ghporter View Post
I didn't say they weren't "reliable."
The first reply was wishy-washy (`not reliable, not unreliable'), in the second post you write that weather forecasts are `only somewhat useful', that predictions are `often so far off' that predictions need `hour-by-hour revisions'. These are quotes from your post. Clearly I read from your posts that you do not think weather forecasts are reliable.

It's a simple thing to compare weather forecasts with measurements from some select few weather stations. So you can quantify in a very good way how precise forecasts are.

In either case, I don't agree with your comment that weather forecasts aren't trusted or widely accepted (`… makes it difficult for the layman to trust them'). Most people don't think twice about this. And they intuitively know that long-term predictions (3+ days) are only `trends.' Yet they still look at them to see whether they should plan a trip to Lake Garda for a long weekend full of mountain biking (we ended up cancelling it, it rained three out of four days*). Scientists and non-scientists alike took the weather forecast serious.
But I don't think it's useful to convince you that (near term) weather forecasts have become very reliable. I respect your opinion.

In any case, the point I was trying to make is actually very simple:

(1) Near-term weather forecasts have become very accurate as has been checked directly.
(2) All problems regarding data collection, quality assurance and uncertainties arise in the context of weather forecasts. Weather forecasts (being local and `more statistically uncertain') are much more sensitive to problematic data input.
(3) Because near-term weather forecasts are accurate, scientists have acquired lots of experience with this and data quality assurance has been subject of research. (Don't forget that weather forecasts are a business these days.)
(4) Climate simulations have weather simulations `at heart' (they are used as consistency checks, but obviously not every `consistent' model gives good predictions). The data problem has been subject to much research already (see point (2)), so all the experience from weather forecasting can be put to good use. Climate simulations (just like any simulation of a simplified representation of the earth) are not perfect, some of its limitations (lack of resolution) is due to limitations in computing power. But many people are working on it independently.
(5) Climate simulations are much, much more coarse-grained than weather simulations (~100+ km grid size vs. ~10~1 km grid size). Any bias (if any) in the (averaged) data should be much less pronounced than in weather simulations.
(6) Code of climate simulations is comprised of several hundreds of thousands lines of code. It's impossible to `tweak' a few parameters so as to selectively to get `intended results'.
(7) There is no vast conspiracy of scientists, it's safe to take off your tinfoil hat now

If you accept that (near-term) weather forecasts have become very accurate, then the rest follows.

* Weather predictions in the mountains are a lot, lot more difficult, because the weather may be very different on two sides of the mountain or different elevations.
I don't suffer from insanity, I enjoy every minute of it.
     
olePigeon
Clinically Insane
Join Date: Dec 1999
Status: Offline
Reply With Quote
Dec 19, 2008, 12:48 PM
 
Originally Posted by stupendousman View Post
After eons of radical climate changing with zero evidence that there was any influence by man, some are skeptical that circumstantial evidence (which might be skewed by a lack of good controls) might show that man might have some small effect on what the weather is doing, and that even if man is having some effect that the "solutions" offered are unnecessary and won't provide any kind of reasonable cost/benefit incentive.
Sorry, pet peeve of mine. People who talk out of their ass and have zero understanding of the process, try to make a point, obscure data, and nit pick results to better suit their opinion without providing any evidence to support their claim.

History update for you stupendousman: humans haven't been around for eons.
"…I contend that we are both atheists. I just believe in one fewer god than
you do. When you understand why you dismiss all the other possible gods,
you will understand why I dismiss yours." - Stephen F. Roberts
     
olePigeon
Clinically Insane
Join Date: Dec 1999
Status: Offline
Reply With Quote
Dec 19, 2008, 12:53 PM
 
Originally Posted by The Crook View Post
Grists's "How to Talk to a Global Warming Skeptic" feature covers this point about the temperature data being "unreliable"
That doesn't change their minds. Trust me, I've tried it over and over again. There are quite a few people on this forum that look at it then exclaim it's a conspiracy amongst scientists. Just like the moon landing was faked.
"…I contend that we are both atheists. I just believe in one fewer god than
you do. When you understand why you dismiss all the other possible gods,
you will understand why I dismiss yours." - Stephen F. Roberts
     
stupendousman
Addicted to MacNN
Join Date: Nov 2003
Status: Offline
Reply With Quote
Dec 19, 2008, 03:47 PM
 
Originally Posted by olePigeon View Post
Sorry, pet peeve of mine. People who talk out of their ass and have zero understanding of the process, try to make a point, obscure data, and nit pick results to better suit their opinion without providing any evidence to support their claim.
Is that better or worse than an ad hominem attack?

History update for you stupendousman: humans haven't been around for eons.
....and yet, the climate of the Earth has constantly changed. Thanks for helping me make my point.
     
Warren Pease
Mac Enthusiast
Join Date: Jan 2007
Status: Offline
Reply With Quote
Dec 19, 2008, 04:15 PM
 
Originally Posted by stupendousman View Post
....and yet, the climate of the Earth has constantly changed. Thanks for helping me make my point.
Historic climate change doesn't negate the present effects of humans. And previous global climate change are explained by natural cycles and non-cyclical tectonic movements, and now the human presence - not sure what to call it; terraforming implies the opposite of what we are currently doing with our planet.

To ghporters concerns about weather prediction - Firstly, no one has shown the measurements at any station are skewed or incorrect.

Secondly, weather is chaotic - that is the nature of the beast. When Texas thunderstorms pass through, it can be hailing where you are and a mile away bbq's are going on uninterupted (albeit with a nice light show in the sky) - what would you want to see on the forecast for that day? For that hour? Sunny? Rainy? 80 degrees at the bbq? or 50 where the hail is falling?

Thirdly, heavy coat? In Texas? Really?
     
Warren Pease
Mac Enthusiast
Join Date: Jan 2007
Status: Offline
Reply With Quote
Dec 19, 2008, 04:50 PM
 
Originally Posted by ghporter View Post
But again, how is the normal bloke supposed to be able to appreciate the accuracy of these climate trend lines if he sees his local weather dude telling him "rain today with a high of 56" and it's dry and 60 where he is? I think he'll see that as "these guys are just guessing" and dismiss ALL weather related predictions. And when you add things to this phenomenon like reports about inaccurate and poorly collected weather data, that just makes the average bloke more sure that everything about weather is guesswork.
This is true of any chaotic system - stock market for example (which i'll reiterate i know little about so play along with me here). Look at any given stock performance on a day and it doesn't necessarily hold true for the entire market. One goes down, another up. Just like one person's thermometer is his/her backyard may not read the same as the one the weatherman is using. Is one person wrong, and the other right? No. They both are correct. For that spot.

A volatile stock market can be seen as a sunny day for one, a stormy day for another, but each persons position adds something to the overall rise or fall of the stock market. The weather networks (not single data points) are like that - individual stations may report different things, but the overall representation is comprised of the average of each one. From this we can get a trend of what the market is like or the weather.

The layman knows that things like the weather aren't consistent or predictable. What is it they say about the weather? If you don't like it - give it a day - it'll change.
     
olePigeon
Clinically Insane
Join Date: Dec 1999
Status: Offline
Reply With Quote
Dec 19, 2008, 05:05 PM
 
Originally Posted by stupendousman View Post
Is that better or worse than an ad hominem attack?
I didn't name anyone in particular. I also support my claims with science journals and other media containing reviewed scientific data.

You just choose to ignore it or completely misinterpret it or twist it in a way that is not consistent with the findings. I doubt I'm the only one who's noticed.

Originally Posted by stupendousman View Post
....and yet, the climate of the Earth has constantly changed. Thanks for helping me make my point.
See my reply above.
"…I contend that we are both atheists. I just believe in one fewer god than
you do. When you understand why you dismiss all the other possible gods,
you will understand why I dismiss yours." - Stephen F. Roberts
     
ghporter  (op)
Administrator
Join Date: Apr 2001
Location: San Antonio TX USA
Status: Offline
Reply With Quote
Dec 19, 2008, 06:22 PM
 
Originally Posted by OreoCookie View Post
The first reply was wishy-washy (`not reliable, not unreliable'), in the second post you write that weather forecasts are `only somewhat useful', that predictions are `often so far off' that predictions need `hour-by-hour revisions'. These are quotes from your post. Clearly I read from your posts that you do not think weather forecasts are reliable.
Yes, I was wishy-washy, mainly because I didn't want to formulate a detailed response if I could get away with a couple of examples instead. Obviously I didn't make my point so here goes: I implied that weather forecasts are not "reliable" in the sense that I have yet to see any forecast that gives the expected margin of error or (more important, I think) the location at which the expected temperature and precipitation would be most likely. As in "high tomorrow 56˚and cloudy with a 30% chance of rain" rather than "at the airport, the high should be 56˚±5 and most of the area should be cloudy; chances are there will be measurable rain in about 30% of our area at some time today". Precision measurements have a given range of measurement tolerance, ±0.04mm or ±3°F, so leaving out the expected uncertainty of a prediction based on precise measurements makes those predictions look like absolutes, thus undermining the perceived accuracy of the prediction.

I understand this, and accept implicitly that, given where the basic "this location" observations are made, that the predictions are valid for that location, and further that the rain chance in any forecast applies to "some measurable precipitation" within that percentage of the given area. But I'm not just anybody, and I've studied this sort of thing, as well as having worked closely with both meteorologists and weather equipment maintainers. The average Joe (particularly here in South Texas) looks at weather forecasts as a fair guess most of the time, rather than highly accurate within very tight tolerances at the location from which the primary observations are made. The whole "our official temperature is what's recorded at the airport" thing is completely lost on most people.
Originally Posted by OreoCookie View Post
(1) Near-term weather forecasts have become very accurate as has been checked directly.
(2) All problems regarding data collection, quality assurance and uncertainties arise in the context of weather forecasts. Weather forecasts (being local and `more statistically uncertain') are much more sensitive to problematic data input.
(3) Because near-term weather forecasts are accurate, scientists have acquired lots of experience with this and data quality assurance has been subject of research. (Don't forget that weather forecasts are a business these days.)
(4) Climate simulations have weather simulations `at heart' (they are used as consistency checks, but obviously not every `consistent' model gives good predictions). The data problem has been subject to much research already (see point (2)), so all the experience from weather forecasting can be put to good use. Climate simulations (just like any simulation of a simplified representation of the earth) are not perfect, some of its limitations (lack of resolution) is due to limitations in computing power. But many people are working on it independently.
(5) Climate simulations are much, much more coarse-grained than weather simulations (~100+ km grid size vs. ~10~1 km grid size). Any bias (if any) in the (averaged) data should be much less pronounced than in weather simulations.
(6) Code of climate simulations is comprised of several hundreds of thousands lines of code. It's impossible to `tweak' a few parameters so as to selectively to get `intended results'.
(7) There is no vast conspiracy of scientists, it's safe to take off your tinfoil hat now

If you accept that (near-term) weather forecasts have become very accurate, then the rest follows.

* Weather predictions in the mountains are a lot, lot more difficult, because the weather may be very different on two sides of the mountain or different elevations.
Here's the issue, which I have managed to not really state explicitly but is what seems to be the root of what bugs me about the whole issue. I, myself understand and accept the whole list above, with the caveat that I believe #2 requires a few assumptions that may not be warranted (as in "how accurate COULD local forecasts be if all local observations were made up of exceptionally "clean" data without any biasing phenomena involved in its collection?). I have not, I think, even suggested that I, personally, don't accept the validity of either weather forecasts or climatological forecasts. What I do see and have tried to express is that the common layman with an abysmal level of science education in public schools, has NO BACKGROUND from which to view the prima facie predictions of meteorologists or climatologists with as anything but black magic. And when the press makes a buck by publishing stories about how "2007 was the coolest year this decade" and "NASA was forced to admit that their multi-million dollar climate measuring satellites were wrong" that this undermines the acceptance of climate predictions.

Maybe this is really about me thinking "gee, most people these days probably believe science is either made up or beyond human understanding." But the items I quoted in my first post are certainly very good examples of science not giving ANY attention to letting the average person know what the heck they're doing and why. People complained that the Apollo program spent billions just to thumb our collective nose at the Russians, but they ignore how much of what we just accept as normal today came from that program (Velcro as an everyday item, microprocessors becoming affordable due to manufacturing scale, biometric telemetry, portable color television cameras, and so much more), and that the program cost less than a soda a day... Climate change has the potential to change the entire planet in such a way as to change all society and how we live. If scientists don't start explaining what the heck they're doing about this, people are going to resist the changes that are needed to moderate our impact on climate until it's way, way too late.

Originally Posted by Warren Pease View Post
This is true of any chaotic system - stock market for example (which i'll reiterate i know little about so play along with me here). Look at any given stock performance on a day and it doesn't necessarily hold true for the entire market. One goes down, another up. Just like one person's thermometer is his/her backyard may not read the same as the one the weatherman is using. Is one person wrong, and the other right? No. They both are correct. For that spot.

A volatile stock market can be seen as a sunny day for one, a stormy day for another, but each persons position adds something to the overall rise or fall of the stock market. The weather networks (not single data points) are like that - individual stations may report different things, but the overall representation is comprised of the average of each one. From this we can get a trend of what the market is like or the weather.

The layman knows that things like the weather aren't consistent or predictable. What is it they say about the weather? If you don't like it - give it a day - it'll change.
The stock market is an excellent analogy here. It seems that a whole lot of people think of the market as black magic, and another whole lot of people behave as if it's all controlled by some small, shady group out to control the world. And while the basic idea is quite straightforward, being able to "get a feel for it" is difficult because the very chaotic nature of thousands of independent agents interacting is hard to grasp. You don't see Science Channel pieces on how the markets work, or for that matter how weather forecasts work. You see complete idiots out chasing tornadoes (not chasing away from them, mind you, but toward them!), and condescendingly oversimplified stuff about how electric lights work...

People who didn't get a unit on meteorology in 5th grade like I did, and who didn't study science through junior high and high school like I did, won't have a clue about how these things work unless scientists start explaining what they are doing, why it's important, and why stuff like temperature sensors placed in the middle of a tarred roof aren't really of any importance in the overall results of their work. What we need is a new Isaac Azimov, who could explain anything to anyone with smooth wording and tight logic that made sense without having to depend on elaborate diagrams or three prior articles...

I remain unsatisfied that scientists even bother to think about the common layman and what he knows about what they do, especially in this.

Glenn -----OTR/L, MOT, Tx
     
PaperNotes
Registered User
Join Date: Jan 2006
Status: Offline
Reply With Quote
Dec 21, 2008, 01:07 PM
 
Originally Posted by The Crook View Post
Global warming deniers are generally only a small number of hard-core conservatives.
This is nonsense and the typical Orwellian propaganda we have seen for a number of years now that the majority of the public just yawn at.

There are many thousands of scientists and economists who don't see any alarm in the small amount of global warming we have seen and can't find any evidence that manmade carbon dioxide emissions have a dangerous effect on the world's climate systems.

Unfortunately the media reports otherwise. The media only serves the the wealthiest now. Government and media organisations who all go with a begging bowl to the major central bankers will not dissent. The bankers see a major market for trading in carbon emissions so the media and politicians overwhelmingly state that the planet is in peril and that global warming is already destroying the world.
( Last edited by PaperNotes; Jan 9, 2018 at 05:43 AM. )
     
PaperNotes
Registered User
Join Date: Jan 2006
Status: Offline
Reply With Quote
Dec 21, 2008, 01:16 PM
 
Originally Posted by olePigeon View Post
That doesn't change their minds. Trust me, I've tried it over and over again.
Your idea of scientific facts don't help. You claimed that the temperature in the Antarctic is rising by 3.5C each decade. This is bullshit. For such an increase to occur we would have to see a four fold increase over that in the troposphere.

Yet all we have is 0.8C at ground level over the last 150 years and half that higher up in the troposphere where the most warming should be occurring if global temperatures were being forced up by carbon dioxide emissions.

The rise in temperatures we have seen is exactly what we should have expected coming out of the Little Ice Age.

Just today another alarmist (yes, you are one) said in The Independent (cough) that global temperatures are rising faster than ever. The media one week reports that temperatures are at their lowest for a decade and then the next week they say temperatures are rising faster than ever. They can't make up their minds because global warming alarmism is based on faith and prophecy instead of scientific observation of reality.

It's religion.
( Last edited by PaperNotes; Jan 9, 2018 at 05:43 AM. )
     
 
Thread Tools
 
Forum Links
Forum Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts
BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Top
Privacy Policy
All times are GMT -4. The time now is 07:41 PM.
All contents of these forums © 1995-2017 MacNN. All rights reserved.
Branding + Design: www.gesamtbild.com
vBulletin v.3.8.8 © 2000-2017, Jelsoft Enterprises Ltd.,