next message in archive
next message in thread
previous message in archive
previous message in thread
Index of Subjects
Index of Subjects
Hi Again, June 11, 2008
Let me assure you that I know do something about reading graphs,
approaches that can be used to minimize noise in measurements and a
couple of other things. But it seems some printers are more equal than
others and one or both of us may be to some degree dealing with printer
artifact. My printout of methane, for whatever reason, was useless,
which is why I stuck with CO2.
My graph is 93 mm/800,000 years or 8600 years/mm and where the
variates increase rapidly (i.e. the points of interest in this case) the
lines are as thin as 0.34 mm (measured by ocular micrometer at 40X).
But the precision of measuring an average offset, between a printed
line and a drawn line, is not constrained by the width of either line
provided the imaginary dimensionless ideal line is on average in the
middle of the printed and drawn lines and provided care is taken to
measure from the middle of one line to the middle of the other.
A brief search failed to locate that graph that we discussed several
years ago but I did find this commentary to the effect that warming
preceeds CO2 by about 800 years.
http://www.realclimate.org/index.php/archives/2004/12/co2-in-ice-cores/
As noted previously this initial reversed order (first temp then
CO2) does not rule out a greenhouse effect that further increases
temperature by feedback.
Qualitatively, the model that seems to fit is a warming that is
initiated by the solar cycle, a subsequent increase in CO2 & CH4 that is
caused by this warming (and based on Ruddiman, probably even more
warming due to greenhouse effect which of course generates more of these
gasses).
Yt, DW
Christopher Majka wrote:
> Hi Dave,
>
> I think one would want to see the original data on which the graph
> was based. Using the graph itself, I think your measurements are
> below the level of its resolution. Printed, the graph measures 71.5
> mm and represents 800,000 years so each millimeter = 11,189 years. I
> measure the line width (where it is a single track and not
> fluctuating to create a wider area) as ~ 0.45 mm, in other words the
> line width itself (which is really the minimal level of resolution of
> the graph) is 5,035 years. Most of your time lags (5,200, 900, 1,700,
> & 2,600 years) are at or below this resolution level so measurements
> below this threshold would only be noise in the signal. Even the
> largest "lag" is only 1.7 times the width of the line, and even that
> would be valid only in places where the line width is at its minimal.
> As you point out the line is much wider (up to ~ 1.15 mm = 12,867
> years) at other points where there have been fluctuations over a
> shorter time interval.
>
> Cheers!
>
> Chris
>
> On 10-Jun-08, at 10:02 PM, David & Alison Webster wrote:
>
>>
>>
>> Christopher Majka wrote:
>>
>>> Hi Dave,
>>>
>>> On 10-Jun-08, at 8:02 PM, David & Alison Webster wrote:
>>>
>>>> Figure 1a shows without doubt that temperature, CO2 & CH4 are
>>>> related but--- temperature starts to increase BEFORE CO2 & CH4
>>>> start to increase. There may subsequently be positive feedback,
>>>> but at least initially, higher temperatures are driving CO2 &
>>>> CH4 increase not the converse.
>>>>
>>>> I suppose one could call this 'tceffe esuohneerg a' but it would
>>>> be awfully difficult to pronounce.
>>>
>>>
>>>
>>>
>>> You must have preternaturally acute vision to be able resolve this
>>> from Figure 1a. :-> I'm looking at it at the moment and, as you
>>> point out, does show a clear historical relationship of
>>> temperature, CO2, and CH4 but the graph, which represents 800,000
>>> years, is only 3.75" wide on my screen. At that level of
>>> resolution, these graphs, set one over the other, look absolutely
>>> co-incident to me. I tried printing this out and running a
>>> vertical rule over it, but I don't see how one could possibly
>>> discern whether temperature, carbon dioxide, or methane are
>>> increasing before or after one another at this resolution. :->
>>
>>
>> Hi Chris, June 10, 2008
>> A larger scale graph would make it more evident but-- I printed it
>> and drew fine vertical lines to join the ticks at 600, 400 and 200
>> thousand YBP. Looking at major abrupt increases in temperature and
>> the center as opposed to either edge of ink lines (some lines are
>> wide, some narrow)--
>>
>> 1) Temperature starts to rise 4 mm left of the 400 line, CO2 starts
>> 3 mm left [note 1mm~ 8700 years],
>> 2) Temperature starts to rise 6.5 mm right of the 400 line, CO2
>> starts 7.2 right [lag of ~ 6000],
>> 3) Temperature starts to rise 6mm left of the 200 line, CO2 starts 5
>> mm [lag of ~8700 yrs],
>> 4) Temperature starts to rise 3.4 left of the 200 line, CO2 starts
>> 2.8 mm [lag of 5200],
>> 5) Temperature starts to rise 1.3 right of 600, CO2 starts 1.2
>> [advance of ~900],
>> 6) Temperature starts to rise 7 right of 200, CO2 starts 7.2 [lag of
>> 1700],
>> 7) Temperature starts to rise 1.7 left of 0, CO2 starts 1.4 [lag of
>> ~2600].
>>
>> Based on these 7 points the average lag [as represented in this
>> graph] of CO2 behind temperature is about 4500 years. What really
>> matters of course is the data behind the graph and the date
>> resolution of these data. But this figure reminds me a lot of one
>> that we discussed several years ago; don't have time to locate it now.
>>
>
next message in archive
next message in thread
previous message in archive
previous message in thread
Index of Subjects