Search This Blog

Friday, September 4, 2015

Statistical Extrapolations of Climate Trends


















CO2

For some years, climate activists have claimed that we can't wait until we have all the data on AGW/CC, because it will be too late by then.  But it is 2015 now, and we do have substantial data on many of the trends climate models have been used to predict up until now.  I have been working with some of this data from the NOAA, EPA, and 2015 satellite datasets, on global temperatures, methane, and CO2 levels. Not surprisingly, there are some excellent fits to trend lines, which can be extrapolated to 2100.

A word of caution here, though:  trend lines cannot predict events that may change them, and some lines simply don't make scientific sense.  For example,
PPM CO2 Increases from 1959 - 2014












 
Clearly CO2 increases are not going to follow this 6'th order polynomial trend, although it did give the best fit of all trends I tried.  This is understandable.  CO2 increases vary considerably from year to year.  Better is to follow atmospheric CO2 levels, which I have done below:

PPM CO2 Levels 1980











 
Notice I've used a quadratic trend line here, which fits with > 99.99% correlation.  According to CO2 forcing calculations, this leads to an approximate 1.7 degree C increase in temperature above 2014, or 2.7 degrees overall since 1900. This fits well with official IPCC numbers, albeit on the low end.  It also agrees with temperatures plotted from NOAA data:









 
The increase from 2014 to 2100 is about 1.6 degrees, or 2.6 degrees by the IPCC. Very encouraging!

But not very hopeful, if many climate scientists are right.  It's generally predicted that an additional rise above one degree from current or more may lead to unacceptable consequences, most of which I'm sure you've read about.  Can we reduce them to acceptable levels?  And in a way that doesn't crash the global economy -- no, civilization itself -- leading to billions dead and the survivors back in the Dark Ages?  I think there is, and have tried to apply my thoughts to the above chart.  The result is the new chart below: 

I have used an exponential type decay on the CO2 increase model to project how emissions trends might be reduced down to virtually zero without (I hope) harming the global economy.  This new trend would not only depend on political/economic conditions, but mainly on developments in science and technology..At present the trend is about 2.3 PPM/year, and will increase to 3.2 PPM by the end of this century if nothing is done.  I calculate that this will reduce  a projected CO2 increase from 240 PPM down to an additional mere 72 PPM.  The calculated temperature increase from 2015 is then only about +0.5 degrees above 2014 conditions.

I must add, however, that even reducing carbon emissions to zero -- even removing the gas from the atmosphere -- will not mean atmospheric levels will drop quickly. About ten times as much CO2 is dissolved in the oceans than is in the air (the greatest majority is in living matter and carbonate rocks); so that, even if can reduce atmospheric levels, an equilibrium process will redistribute some of the dissolved gas back into the air.  For example, to remove a billions tons of CO2 quickly and permanently from the latter might require removing almost 10 billion tons from the ocean in order to maintain equilibrium.

On the other hand, it is also likely that anthropogenic CO2 has not fully equilibrated with the oceans  (it is supposed to have a half-life of ~100 years in the air), so that even if we added no more, it would probably absorb further, and atmospheric levels would slowly decline.  I do not know how this might affect the 2100 CO2 levels should the scheme above be applied.  http://www.nature.com/nature/journal/v488/n7409/full/nature11299.html



Methane

Of course, CO2 is not the only greenhouse gas climate scientists have been worrying about.  The other main other culprit has been methane.  Therefor, I have plotted atmospheric methane levels from 1980-2014.  Instead of plotting the gas directly, I have converted to CO2 PPM equivalents.













 
Strange, to say the least, especially if the quadratic fit is correct!  At first I didn't believe what I seeing.  Recently however, I encountered several articles,
http://wattsupwiththat.com/2015/08/19/the-arctic-methane-emergency-appears-canceled-due-to-methane-eating-bacteria/
http://fabiusmaximus.com/2015/08/20/ipcc-defeats-the-methane-monster-apocalypse-88620/, www.epa.gov/climatechange/indicators, and
http://www.climatechange2013.org/images/report/WG1AR5_Chapter02_FINAL.pdf, which shows rises in methane levels diminishing over time. This is possibly due to recently discovered sources of  methane consuming bacteria, found mainly in the arctic but probably widely distributed.  These bacteria naturally convert methane to CO2, and become even more active when temperature rises as it is doing.  If the quadratic trend line (best fit) is correct, then methane levels in 2100 will be about the same as 1980, peaking at 2040.  A linear trend line, on the other hand, shows methane increasing to almost twice that level.  Either way, it appears that methane can no longer be counted as a serious greenhouse gas with any certainty.




Satellite Data

I've also plotted NOAA (blue) versus satellite temperature change data (orange):














The differences are striking.  The satellite data indicate a full degree less warming by 2100 than the NOAA data, or only +0.7 degrees above current temperatures; however, the varriation is also much larger, so this might not be accurate.  Incidentally, for those who follow this issue, the so-called "hiatus" (of "climate-denier" fame) is based on the satellite data (plotted from 1996-2014 below):














So the pause is real, at least for the satellite data.  The fallacy, I think, is to assume that because temperatures have been falling for some eighteen years, we can declare global warming to be over.  The entire data set does not support that, and it is common for global temperatures to decrease (and increase) over short periods.  So beware.

I have puzzled over why the NOAA and satellite datasets differ as much as they do.  Since NASA has both geostationary and polar satellites, it shouldn't be a lack of global coverage, though there could be systematic errors.  There is also the often said accusation that the NOAA data have been "adjusted" (meaning fudged in this case) to make global warming appear worse than it is (and I have read about some peculiar adjustments), but I don't have no data to back this claim up. It could also be, of course, that satellites measure a higher region in the troposphere, while the NOAA data is strictly ground level.

Naturally, it could be any combination of any of these reasons.


Sea Level Rise and Global Ice















Now we have three conflicting data sets!  CISRO (30 cm -> 2100) and NASA (40 cm) agree the best, and they are both best within predicted range for the middle of the next century (~ one meter rise from 1900).  Of course, if CO2 emissions are reduced in the fashion described above, these numbers should not as high; I have no data to demonstrate this, however.


















These data show an approximately 10,000 cubic kilometer loss in Arctic ice over the last 35 years, leading to a three cm sea level rise during that period.  Unfortunately, I could not obtain the raw data for this chart, so I cannot draw a trend line through the data (the straight line came with the chart). We can, however, calculate a geometric increase from this data, by assuming rising temperatures will double each succeeding 35 year period.  This give 2015-2050 = 20,000 km^3, 2050-2085 - 40,000 km^3, and 2085-2120 = 80,000 km^3.  As we are only extrapolating to 2100, this approximately adds an another 20,000 + 40,000 + 40,000 = 100,000 km^3 melted ice, enough to raise sea levels another 30 cm over 2015.  It is in excellent agreement with the CISPRO and NASA sea level trendline above.

I haven't included Antarctic sea ice because, up to the present, it doesn't show enough shrinkage or expansion (estimates differ) to extrapolate a significant trend.

What about ice coverage?  This is important because a significant shrinkage here could reduce the Earth's albedo (solar reflectance) and strongly enhance global warming.  Again, I can only include a chart as presented from it's source without raw data   http://arctic.atmos.uiuc.edu/cryosphere/IMAGES/global.daily.ice.area.withtrend.jp:










Very little, if any global sea ice area is demonstrated; however, if melting follows the path I have outlined above, it is quite possible that these levels will at least gradually shrink, raising global warming by reducing the planet's overall albedo.


Population Issues














It is, I hope, evident that the planet's increasing human population only adds more greenhouse gases, thereby increasing global warming.  Of equal interest, however, is the per capital emission rates.  To calculate this, I took the data from chart 2, and divided the points from the chart with population to produce the blue line:












The orange line is similar to that used to show how CO2 can be reduced to near zero before the century is out.  See my comments about this modified trend above.


Conclusions

Global warming, along with the dire consequences this study supports, is very real and no hoax or scheme by some secret cabal of government and the scientific community.  Yet neither is it an unstoppable catastrophe as some have alleged -- in fact, the trends revealed here mostly confirm to the more conservative estimates (model outputs) of how serious the problem is.  That is no reason for complacency, however, for even the these models inexorably lead to serious problems.

I suggest, however, that these problems can be at least strongly ameliorated at least, if we invest in the basic science and technologies that alone can address the issue.  For myself, I am strongly convinced that we will do so; more than that, that we have already begun. 

Citation signal

From Wikipedia, the free encyclopedia https://en.wikipedia.org/wiki/Cit...