Visualizations : Flawed Data - Temp Over Time and CO2 Levels : Temperature increases cause increases in co2...

Don’t forget to read the comments on this one!

Hey! Apparently there is a correlation between Co2 level and temperature… :/

This view shows the correlation between CO2 and temp. That tendril at the right, is that the world spinning out of control in recent years?

Interesting indeed, the collaborative manipulation of data like this is really something! I hadn’t tried those arrangements, and now it shows something else quite interesting. Adding the year as the size makes it a little messy, but if you then scroll over that tendril, you’ll see that all the points are within the last 2K years cluster. What does this mean though that such high CO2 levels are preceeding significant changes in temperature? Something, much worse is yet to come?

It seems to me that the tendril on the right, composed of recent anthropogneic CO2 points, says that the coefficient of temperature versus CO2 for anthropogneic CO2 is roughly 1 degree C per 100 ppm of CO2, and this applies to CO2 levels above 300 ppm.

In contrast, for the older non-anthropogenic CO2 levels below 300 ppm the coefficient of temperature versus CO2 is an order of magnitude higher, being roughly 10 degrees C per 100 ppm of CO2.

This is very surprising to me. Has this difference been discussed in the climate modelling community? IN the latest UN IPCC report on global warming?

If you look here you’ll notice something interesting… As CO2 level go up there doesn’t seem to be an immeidately apparant shift in temperature. Of course, if you look at the data more closely you come to understand that the lack of increase is probably due to a presentation error which is, unfortunately, difficult to overcome. While we have many data points for the 2k year slice we have significantly fewer as we move back in time. I believe that this changes the perception of the graph significantly.

I’m not saying that global warming doesn’t exist and that CO2 rises aren’t a problem. Only that this data set might not be the right way to go about illustrating it.

Frank, Correllation does not mean causation. There are far more processes at work during warmer periods than cold periods which would release more Co2. It should be greater during warmer periods and lower in cold periods. That doesn’t mean that the CO2 caused the warming.

I know, I never said there was a causation between the two (see original post). There most certainly is a correlation though..

The problem is, the data just doesn’t go back far enough. We need the same data over a 500-1000 million year period. The fossil records of climate in the 60 – 250 million BCE period indicate a global tropical or semi-tropical climate. What we seem to be entering into in terms of global climate may have been experienced numerous times in the 4+ billion year history of the planet.

Yes, but- maybe it is all a scare. Maybe this is all part of a cycle. Maybe. It’s Pascal’s wager, only it’s our kids frackin lives at stake. I won’t roll those dice, sorry, just because it would be easier. Sometimes you suck it up and assume the worst, because if you don’t and you’re wrong – well, then you’ve just screwed a buch of people because you can’t control your appetites.

Snobby people make poor arguments.

As I look at the data, it is clear that there is a periodic cycle where the temperature rises and the CO2 level increases. But this occured at a time when man was not in the picture. In the past warming cycles, did the CO2 increase cause the warming, or did the warming cause the CO2 increase, or both? The periodicity of the warming cycles is also interesting — what caused that? The bottom line is we have to be careful not to mix-up cause and effect. Just because we see a warming cycle at this time, and we see a huge increase in CO2 concentration likely due to man’s activities, does that mean the CO2 increase is the cause? From my talks with global modeling scientists at LLNL in the 1990’s (I worked there as a scientist), the jury is still out whether CO2 at the current concentrations can account for solar energy trapping because the radiative exchange at high altitudes is so poorly understood, plus the impact of cloudiness, particulates, plus that the sun’s output is not constant over time (an area I worked in). All in all, the picture of man’s impact on global temperatures is still “cloudy”.

Plot temperature deviation on the Y-axis and CO2 concentrations on the X-axis. Notice how tight the line gets and how the slope goes nearly flat when we get to CO2 concentrations above 290ppm. Could this mean that we have reached the maximum effect CO2 will have on temperature rises and further increases in CO2 will not cause a faster rise in temperatures?

Couldn’t this also show that natural CO2 is a function of temperature, not the other way around?

For those who still have some doubts about the anthropogenic causes for global warming, and to know how far the atmosphere and natural forces are being modified by us, please read the 2007 book “we are the weather makers” by the scientist Tim Fannery.

It is clear that CO2 concentration is off the chart, and it is intuitive that this has been caused by Man’s unchecked oxidation of carbon.

There has been a lot of work since the 1990s in climate modeling, and the ice core evidence (from which these data were collected) has come out around 2002 or so. This has exploded many of the arguments from the 1990s.

Whether there are other, mitigating factors that will saturate/check the greenhouse effect remain to be seen. However, it seems undeniable that Global Climate Change is an important issue to be dealt with whether the extra Carbon leads to a monotonic increase in global average temperatures.

After watching a BBC documentary http://video.google.com/videoplay?docid=-4520665474899458831

After watching a BBC documentary:
http://video.google.com/videoplay?docid=-4520665474899458831
where they were claiming there was a 800 years lag between the earth temperature raise and the CO2 levels increase I downloaded the data from this site and run a cross correlation between the temperature and CO2 data. What I found is that the CO2 increases or decreases, on average, AFTER 350 there has been a temperature increase or decrease.
It then looks like the CO2 levels are a consequence of the temperature and not the other way around. (As claimed in the documentary)
From the data also looks like there is a 100.000 years cycle going on and we are getting close to one of its peaks.

Said so, the CO2 increase of the last 100 years is much higher that what is possible to see in the previos 400.000 and I don’t see other explanation than it’s caused by men.

I am interested in which technique was used to compute the Xcorrelation function with the unequally spaced (in time) data provided

The thing that jumps out at me in this view is that the current cycle peak is far wider than the previous peaks. It looks like after the rise out of the most recent ice age the temp started to drop quickly, as it has in previous cycles, then started to oscillate up and down within a new (higher) range. Without being able to expand the horizontal (X = time) axis, it is hard to make out the details, but it almost looks like the oscillation begin some 3-4k years ago, well before man would have had any significant influence on global CO2 levels. Or, since we have far more data regarding recent years, is the recent sharp rise disguising a peak that would have appeared as narrow as the previous cycles?
Either way, we have definitly remained at the peak of the current cycle far longer than we have previously.

This view is interesting. I see a significant shape difference between the last 200 years and the previous few hundred thousand.

It only seems to be CO2 that is outlying, temperature is lining up with the natural cycle of things.

Wow, very cool to see that this is still getting so much interest. I want to further caveat the way I created the data set from the original datasets here:
http://cdiac.esd.ornl.gov/trends/trends.htm
I had to compile several different text files that used different time scales and had disparate data points. For the times scales I converted, data with AD years to BP. There was also the problem that temperature and CO2 were done separately so they don’t have data points for all the same years. To address this I did a simple macro that when through all the years that there was data for and inserted copies of the last data point for either CO2 or temp deviations. It may be more appropriate to use some kind of averaging, especially since many point will be closer to a more recent entry, but I just always used the previous. Anyway, I would invite anyone more savvy to revisit the original text files. Regardless, there is certainly a clear correlation between these two datasets, which cannot be ignored. The fact that we have outliers should raise some questions. One question someone has was that since temp hasn’t gone up is it reasonable to assume we’ve reached the maximum impact. My opinion is that there is likely a large delay between the change of CO2 and temp, but since we have made unnatural changes in CO2, it will not follow the previous trends which followed some more natural coupling. From this data it appears to me that the two do in-fact mutually affect one another, this should be even more alarming because as temp does rise, even more CO2 may get dumped from a natural process that we aren’t fully accounting for. This could be releases from carbon sinks, like permafrost fields and the ocean and/or a killing of exchangers such as forests or plankton. Anyway, I’m glad this has raised so many good questions, it would be nice to try and pull in some REAL scientists into the discussion. The public is ready to be engaged, and mediums like Many Eyes make it possible to provide more accurate discussions, based on REAL data instead of broad abstractions.

Why So Gloomy?

There is no compelling scientific evidence that the warming trend we’ve seen will amount to anything close to catastrophe. What most commentators—and many scientists—seem to miss is that the only thing we can say with certainly about climate is that it changes. The earth is always warming or cooling by as much as a few tenths of a degree a year; periods of constant average temperatures are rare. Looking back on the earth’s climate history, it’s apparent that there’s no such thing as an optimal temperature—a climate at which everything is just right. The current alarm rests on the false assumption not only that we live in a perfect world, temperaturewise, but also that our warming forecasts for the year 2040 are somehow more reliable than the weatherman’s forecast for next week.

A warmer climate could prove to be more beneficial than the one we have now. Much of the alarm over climate change is based on ignorance of what is normal for weather and climate. There is no evidence, for instance, that extreme weather events are increasing in any systematic way, according to scientists at the U.S. National Hurricane Center, the World Meteorological Organization and the Intergovernmental Panel on Climate Change (which released the second part of this year’s report earlier this month). Indeed, meteorological theory holds that, outside the tropics, weather in a warming world should be less variable, which might be a good thing.
In many other respects, the ill effects of warming are overblown. Sea levels, for example, have been increasing since the end of the last ice age. When you look at recent centuries in perspective, ignoring short-term fluctuations, the rate of sea-level rise has been relatively uniform (less than a couple of millimeters a year). There’s even some evidence that the rate was higher in the first half of the twentieth century than in the second half. Overall, the risk of sea-level rise from global warming is less at almost any given location than that from other causes, such as tectonic motions of the earth’s surface.

Richard S. Lindzen
Newsweek International
http://www.msnbc.msn.com/id/17997788/site/newsweek/

Lindzen is the Alfred P. Sloan Professor of Meteorology at the Massachusetts Institute of Technology. His research has always been funded exclusively by the U.S. government. He receives no funding from any energy companies.

Isn’t it amazing that whomever posted this did not mention that carbon dioxide emissions FOLLOWED temperature rise and fall by 800 to 1000 years, thereby debunking the claim that “global warming” is caused by the increase in carbon dioxide emissions?

By the way, this data is probably from the Vostook icecore sample from Antarctica.

I apologize for the errors, I tried to explain my imperfect solution, but it apparently wasn’t factored into many folks considerations. There were several comments about the temp climbs preceding the CO2 climbs, despite my cautions about how I created the data pairs. I’ve since revised the script that created the pairs so the date nearest to a corresponding data point was used as opposed to simply the last data point. This is why I’ve changed this title to flawed. The new dataset is here:

http://services.alphaworks.ibm.com/manyeyes/data/SgoRsIsOtha6nEVIYgB3J2-

new visualization is here:

http://services.alphaworks.ibm.com/manyeyes/view/Sm4H4JsOtha6-tkwkPGDJ2-/S#m4H4JsOtha6zsUoiPGDJ2-

The normalized data may also be of interest:

http://services.alphaworks.ibm.com/manyeyes/view/SgoRsIsOtha6RF__h7G3J2-

“Notice how tight the line gets and how the slope goes nearly flat when we get to CO2 concentrations above 290ppm. Could this mean that we have reached the maximum effect CO2 will have on temperature rises and further increases in CO2 will not cause a faster rise in temperatures?”

Basically, no. There are diminishing returns to the effect of additional CO2, but that’s not what you’re seeing here. Over the preindustrial period, changes in CO2 and temp were very slow, so temps had time to equilibrate, thus they reflect the full effect of CO2 (and vice versa). Recently, CO2 has gone up very fast, and temperature has not yet caught up. It takes a long time to heat up the earth, especially the ocean. If you wait long enough, you’ll see the full effect of today’s high CO2 on temperatures, especially as other feedbacks kick in.

“Isn’t it amazing that whomever posted this did not mention that carbon dioxide emissions FOLLOWED temperature rise and fall by 800 to 1000 years, thereby debunking the claim that "global warming” is caused by the increase in carbon dioxide emissions?"

No. Temperature and CO2 are in a feedback loop; they mutually influence one another. The fact that the initiation of deglaciations was not caused by CO2 does not preclude CO2 from serving as an amplifier of temperature variations, just as eggs causing chickens does not preclude chickens from causing eggs.