Higher CCl4 emissions? Eli is shocked, shocked

About a week ago, maybe a bit more, NASA, nay Goddard Space Flight Center released a breathless press release which spread near and far even unto our buddies on the other side of reality and some on our side.

Eli is here to tell you that just about everyone missed the real discoveries in the paper and underlying work, and the paper itself let alone the press release did not tell the whole story.  According to everybunny else the take home was

However, the new research shows worldwide emissions of CCl4 average 39 kilotons per year, approximately 30 percent of peak emissions prior to the international treaty going into effect.

It has been long known that there are significant fugitive emissions of just about all CFCs and ilk and that bottom up inventories of atmospheric emissions always fall short of reality.  This has consequences for our understanding.  It should be no real surprise that the emissions are higher than officially reported.  Of course, the issue is quantifying the amount of emissions and tracing them back to their source and here the Liang paper makes an important contribution.  Yes, IF the Montreal Protocols (MP) were being rigidly enforced globally we would not be seeing such emissions, and we would be seeing a faster decline, but  the best of times has not yet arrived.  OTOH, it is vitally important to know if the fugitive emissions themselves are decreasing.

Buried down at the bottom of the press release is this afterthought

In addition to unexplained sources of CCl4, the model results showed the chemical stays in the atmosphere 40 percent longer than previously thought. The research was published online in the Aug. 18 issue of Geophysical Research Letters.

Geez, even tho it is a press release they could have provided a link.  To Eli this is the most important of the results in the paper.  If one simplistically looks at the problem as a one box model, to explain the slower than expected falls in CCl4 there are two possibilities, either the destruction rate is slower than expected or the emissions rate is higher or some combination of the two.  The problem with deciding which is which, is that the two interact.  If your lifetime is too short it will look like the amount of emissions are high, and if the lifetime is too long the amount of emissions will look too low.  Liang et al have a nice way of showing this

Figure 2. CCl4 global mean trend (ppt/yr) as a function of total lifetime and emissions from the two-box model (gray contours). Purple contours indicate the emissions and τCCl4 ranges that yield IHGs within the observed 1.1–2.0 ppt range (2-σ) between 2000 and 2012, using the current best estimate EFn of 0.94. Red (Advanced Global Atmospheric Gases Experiment (AGAGE)-based) and blue (GMD-based) numbers show emissions and lifetimes derived using the observed IHG and trend for individual years (2000–2012). The dark (light) gray shading outlines the range of emissions and τCCl4 that can be reconciled with the observations for EFn of 0.94 (0.88–1.00). The black diamond symbol shows our current best estimate for τ (thick and thin red bars indicate 1-σ and 2-σ uncertainties, respectively) and the upper limit bottom-up potential emissions for 2007–2012 (thick blue bar shows 1-σ variance) with 1-σ uncertainty shown in black-hatched shading.

IHG- Inter hemispherical gradient   EFn - fraction of emissions in the Northern Hemisphere.  

The black diamond indicates the results one would get using a bottoms up (reported emissions) estimate of emissions and the previous best estimate of the Total removal lifetime of 25 years.  Eli has added the purple dot showing the estimates of Liang, et al for 2007-2012 emissions with a 35 year CCl4 total lifetime. The green dot is the result of an earlier study of Xiao, et al on emissions between 1996 and 2004 that used a lifetime which is too short.   Xiao's estimate of emission rates was 74 Gg/yr on average between 1996 and 2004 using a total lifetime of 25 + ~5 years, which is shown by the green dot far to the right.  OTOH, moving up the 0.9 ppm/year contour to the new inferred lifetime of Liang et al., 35 years, brings that estimate of emissions down to ~ 40 Gg/year.

Liang et al divide the world into two boxes, the Northern and Southern hemisphere.  Since most of the emission is in the Northern Hemisphere, there is an inter hemispherical gradient which can be used to calculate the total emissions and lifetime.

Cx is the concentration in each hemisphere, τns the time for inter-hemispheric mixing,  the αs are deposition losses in the oceans and soils and the STE represents dilution of tropospheric air with down moving stratospheric air with CCl4 diminished by photolysis.  f is a factor to convert emissions to concentrations.  The prevalence of land in the Northern Hemisphere and oceans in the Southern has to be taken into account as well as various issues about global circulation.  Liang, et al, then make a number of simplifications to yield

From which with various hocus pocus, they produce the graph above.  Surprisingly the latest evaluation of stratospheric ozone depleting molecules has some surprising reappraisals, including raising the photochemical lifetime of CCl4 from 35 to 44 years, perhaps another post.  In a straight-forward way this would explain part of the unexpectedly slow decrease in atmospheric carbon tet. Liang and other authors on Liang et al were part of that revaluation.  With reasonable values for ocean and soil deposition, the 35 year total lifetime that Liang, et al find is, well, reasonable.  FWIW soil deposition looks really slow, like about 1000 years.  Eli and maybe the Weasel remember arguments about that.

Prominent in the abstract and the press release is that the average emissions over the 2000-2012 time period were 3934⁴⁵ Gg/yr.  Somewhat less, well a lot less, prominent, you have to read the paper, is that emissions have been decreasing.  Between 2007 and 2012 they decreased to between 31 and 45 Gg/yr.  Simply taking the average of these gives an average of 35.5 Gg/yr.  Simple math tells us that the average emissions between 2000 and 2007 would then be an average of 41.5 Gg/yr which is consistent with the numbers shown in Liang's Figure 2 above, with the earlier years clustering to the right of the graph in the 40 Gg.year area and the later ones to the left.  Emissions are decreasing.  Not as fast as we would like, but they are decreasing.

Where is of course the question all bunnies want to know.  Several jumped to the conclusion that all the fugitive emissions are from China and India.  

Inverse 3D modeling is IEHO the best choice for quantifying sources total emissions, reported and unreported, and in 2010, X. Xiao and about 20 friends took a shot  in Atmos. Chem. Phys. 10, 10421, for the period 1996-2004.  As with all such things, that paper was not perfect, and with the passage of time, some of the problems with it have become clearer, but taken together with the new Liang paper there are a number of take homes.  By looking at the time history of CCl4 at stations around the globe Xiao et al was able to infer the location and average carbon tet emissions from various locations during the study period.  By comparison, if you simply want global emissions, the advantage of the Liang, et al method is that the two box model is robust at the price of resolution

For convenience, Xiao et al divided the world up into eight boxes and tried to trace emissions geographically on a finer 64 x 128 point grid

Group I:  Europe, NW Asia (Russia and the stans), and N. America

In this group emission fell rapidly (factor of 5) from the pre-Montreal Protocol estimates.  

Group II Australia/NZ (very small), S. America and Africa

Substantial percentage increases but still minor contributors in the period.  Interestingly African emissions at 6.7 Gg/yr were three times that of S. America.  The regions of major emissions in Africa were South Africa and the Mediterranean coast (think oil production and maybe petrochemicals) but not Nigeria and Angola.  In S. America, the Atlantic coast of an industrializing Brazil was the major source

Group III:  So. Asia (the Mid-East oil patch, India and Pakistan), SE Asia (China and Indo-China, Japan and S. Korea)

Major increases above the pre-MP baseline.  Not very surprising, but note that Japan and S. Korea were bright red as are areas associated with oil production.  

Bottom line is that Xiao et al find that the yearly global emissions fell from ~130 Gg/yr pre-MP to about 65 Gg/yr in 2004 with a slight downward trend.  In addition to estimating the source term, Xiao et al estimated the ocean sink loss.  Using Liang's Figure 2, and tracing up the contour to a global lifetime of 35 years, this is not out of line with Liang, et al for a similar period, ~42 Gg/yr.

So yes, CCl4 in the atmosphere is decreasing ~1% per year, slower than bunnies expected, due to nature (a longer atmospheric lifetime) and fugitive emissions (which are also decreasing).