Being In It for the Species The Kolbrin Bible Complete Danjeon Breathing System Home Study System

Author Topic: Earth's Atmosphere  (Read 1200 times)


  • Global Moderator
  • Prolific Member
  • *****
  • Posts: 2785
  • Karma: +32/-0
Earth's Atmosphere
« on: May 19, 2015, 07:48:47 AM »
   Folks, this is scary.  You can see the graphs by going directly to the article/site.

   •   Today (May 18, 2015) at 11:37 AM

The Institute of Science in Society
Science Society Sustainability

This article can be found on the I-SIS website at
If you would like to be removed from our mailing list unsubscribe at
or email

ISIS Report 18/05/15
O2 Diving Towards Danger Point
Not only is O2 dropping faster than CO2 rising, it is diving towards the danger point much faster than previously thought, O2 accounting is urgently needed Dr Mae-Wan Ho
New analysis of O2 records raises alarm
In [1] O2 Dropping Faster than CO2 Rising (SiS 44), I highlighted new research showing the depletion of atmospheric oxygen accelerating since 2003, which coincided with the biofuels boom, and warned that climate policies focusing exclusively on carbon sequestration could be disastrous for all oxygen-breathing organisms including humans. I also call for O2 accounting in climate policies. The article attracted many comments not all favourable (see
Now, six years later, a detailed analysis of the atmospheric O2 records in 9 stations around the world shows that O2 is not just falling faster than CO2 is increasing, it is actually dropping more than 10 times faster than previously thought [2]. If it continues at this rate, the danger point will be reached in thousands of years or less, instead of tens of thousands of years.
One should bear in mind that although the proportion of O2 in the atmosphere is about 21 %, much, much higher than CO2 (currently ~0.4 %), the dangerous level of oxygen according to the US Occupational Safety and Administration and the US National Institute for Occupational Safety and Health is 19.5 %. In humans, failure of oxygen energy metabolism is the single most important risk factor for chronic diseases including cancer and death (see [1]).
The Scripps O2 Program
The new analysis [2] has been carried out by Valeri Livinia and colleagues at National Physical Laboratory, Teddington, and John Innes Centre, Norwich in the UK. They used datasets on atmospheric O2 levels collected by the Scripps Institution of Oceanography at La Jolla, California in 9 stations around the world (see Figure 1) [3] under the direction of Prof Ralph Keeling, who pioneered the measurement techniques in his Ph D thesis at Harvard University in 1988 [4]. Records since 1989 are available from Scripps Pier and Alert in Alaska, although these are not continuous [5]. Continuous records from 7 stations extend back to 1993, and data for the remaining two, Cold Bay in Alaska and Palmer Station in Antarctica, are available back to the mid-1990s. Oxygen levels are measured as changes in the ratio of oxygen to nitrogen O2/N2 of sampled air relative to a reference sample of air pumped in the mid-1980s and stored in the Scripps laboratory. The unit is ‘per meg’ such that a decline of 1 per meg is equal to 1 part per million of oxygen or 0.0001 %; or 1 molecule of oxygen per 4.8 molecules of all gases in the atmosphere, not counting water vapour [5].
The Scripps O2 datasets also have variable timescales, with an average time interval of two weeks. Their length varies from 391 data points (Cold Bay) to 688 (Mauna Loa). For the analysis, all series were interpolated to obtain 4096 data power points separated by equal time periods.

Figure 1   Nine Scripps stations measuring changes in atmospheric O2 around the world
The down-trend is faster than linear for all stations
The data from the nine stations are displayed in Figure 2, where the downward trend is obvious in all the datasets; some looking faster than linear.

Figure 2   Downward trend of O2 concentration in all nine datasets
Livinia and colleagues applied Bayesian inference to the O2 concentration records, and confirm that they are all dropping at least quadratically (second order polynomial) rather than linearly (first order). Bayesian inference avoids the danger of over-fitting by automatically penalising model complexity. The Bayesian Information Criterion (BIC) is a criterion for model selection, the one with the lowest BIC value being preferred.
They tested several types of polynomial models (orders from 1 to 4, including the linear Ax +B and parabolic Ax2 + Bx + C), logarithmic A + Blogx and exponential –expAx +B, using BIC, Akaike Information Criterion AIC and Akaike corrected AICs for model selection. The results plotted in Figure 3 show that both linear and exponential models have the highest Information Criterion values, and are discarded in favour of polynomial models, for which the quadratic is taken as the most feasible (and most conservative) selection.
Continue reading here
Or read other recent reports on the ISIS science website here

If you find this report useful, please support ISIS by subscribing to our magazine Science in Society, and encourage your friends to do so. Or have a look at the ISIS bookstore for other publications


Surviving the Planet X Tribulation: A Faith-Based Leadership Guide

Surviving the Planet X Tribulation: A Faith-Based Leadership Guide

This uplifting and entertaining guide is written to give you, the reader, confidence and hope through effective leadership techniques and survival community strategies designed for an extended tribulation. Learn more...


Marshall's Motto

Destiny comes to those who listen, and fate finds the rest.

So learn what you can learn, do what you can do, and never give up hope!