What’s Up with the Bad Chemistry?

Watts Up With That is a very, very silly website.

Here’s what I mean: In a recent article at WUWT, chemical engineering graduate Steve Burnette tries to dismiss concerns about ocean acidification, but his claims are outright wrong when they are coherent. The centerpiece of the article is a calculation meant to estimate the change in ocean pH over the 20th century. Unfortunately, it’s been badly bungled.

First, let’s go over a few preliminaries.

Figure 1: The carbonic acid equilibrium system. Atmospheric CO2 is in equilibrium with dissolved CO2, which in turn is in equilibrium with carbonic acid. The acid has a tendency to lose its hydrogen ions, forming bicarbonate ion, which it is also in equilibrium with. The bicarbonate can also lose a hydrogen ion, forming carbonate. Keep in mind that these steps work in reverse; in particular, carbonate ions can react with hydrogen ions to form bicarbonate.

When fossil fuels are burned, about a third of the resulting carbon dioxide ends up dissolved in the oceans. There, it undergoes a series of reactions, first combining with water to form carbonic acid, then losing protons one at a time. The result is a mixture of dissolved CO2, bicarbonate, and carbonate ions, as well as an increase in hydrogen ions, which increase the acidity of the seawater.

Burnette mentions a few principles which are useful for analyzing this scenario. The equilibration between gaseous CO2 and dissolved CO2 is expressed by Henry’s law, which states that the concentration of dissolved CO2 is proportional to the partial pressure of atmospheric CO2. This means that, as CO2 concentrations increase in the air, they will increase in the oceans too. Mathematically:

p = k*c

…where p is the partial pressure, c is the concentration, and k is the Henry’s coefficient.   (Equation 1)

The Henry’s coefficient is itself a function of temperature; this reflects the fact that gasses dissolve better in cold water than in warm water. This temperature dependence is important to ocean chemistry calculations in a warming world.

k(T) = k(T₀) * exp(-c * (1/T – 1/T₀) )

… where T is temperature, T₀ is a reference temperature, and c is a constant. (Equation 2)

So far this is all straightforward chemistry, but then Steve Burnette begins to spin out, starting with faulty data:

“I decided to use the EPA’s stated 1.5C temperature increase since 1917…“

No citation is given for this statistic, and indeed, it appears to be an overstatement. The actual warming is closer to 0.8C (which is suspiciously close to 1.5F) No explanation of the start date is given.

“…and an increase from 280ppm for my concentration …”

No citation for this datum either, and again it’s wrong: 280 ppm is approximately the preindustrial level of CO2, but the Industrial Revolution started around 1750. By 1917, the concentration was closer to 302 ppm, according to the Law Dome record.

“…I used 10C as my current water temperature and 390ppm as my current CO2 concentration. Atmospheric pressure was assumed at 1atm. I also kept hearing a pre-industrial pH value of 8.2 … “

Again, none of these data are cited. 390 ppm is a bit low, with Mauna Loa Observatory reporting in the neighborhood of 395 ppm. 8.2 is a correct preindustrial ocean pH (Zeebe 2012), but again 1917 was post-industrial revolution.

With that cleared up, let’s look at the calculations themselves. The CO2 system is consists of six parameters: carbonate and bicarbonate concentration, total alkalinity, total dissolved inorganic carbon (DIC), pH, and CO2 pressure. (Zeebe 2012) Knowing any two of these parameters determines the other four. The chemistry can be tricky and involved, so to avoid a bunch of nasty calculations, I decided to download a CO2 system calculator. There are several out there; I settled on the SWCO2 package as a quick calculator, and also downloaded seacarb, an R-based seawater chemistry library, for more involved computations.

Let’s accept an initial ocean temperature of 9.2 C (10C at present minus 0.8 C of warming). Heck, let’s go ahead and accept an initial ocean pH of 8.2. And let’s use the correct pCO2 of 302 ppm. This allows us to characterize the initial state of the ocean; in particular, we can use SWCO2 to calculate TA, the total alkalinity.

T = 9.2C
pH = 8.2
pCO2 = 302 ppm
--->
TA = 2604.16 umol/kg

Okay, that’s 1917. Let’s turn the clock forward a bit. The CO2 concentrations and the temperatures both rise, but the total alkalinity stays the same. It doesn’t vary with increased CO2 alone. (Zeebe 2012) That means that we can use it as a carbonate system parameter to calculate present-day pH:

T=10C 
TA = 2604.16 umol/kg 
pCO2 = 395 ppm 
---> 
pH = 8.10

In other words, we get a change in pH of about 0.1 units. That’s in good agreement with published estimates for the change in ocean pH since preindustrial times. (Brewer 1997, Caldeira & Wickett 2003).

What did Steve Burnette get? A change of 0.001 units. That’s orders of magnitude off, on a logarithmic scale! How did he arrive at this answer? I’m afraid I’m still not entirely sure, as his calculations are a bit opaque. When they can be followed, I get different answers than him, even when starting with his faulty data. Some figures don’t have units, others appear to have incorrect units. Partial pressures are expressed in Pascals, while the Henry’s coefficients are expressed in L*atm/mol. Variable names are recycled, introducing ambiguity.  Calculations of Henry’s Law ( Equations 1 and 2) are laboriously worked through, but the more difficult calculation of pH change is skimmed over. Throughout the article, there is a sense that the author is tossing equations at the reader, but they seem to obscure as much reasoning as they reveal.

Burnette’s wonky chemistry continues. For example, he seems to believe that “as CO2 increases, Carbonate ions increase”. But this is simply not true in the oceans (Figure 2).

Figure 2. Carbonate and bicarbonate concentrations in seawater as a function of pH; note that the two graphs are on different vertical axes. As CO2 is added, it consumes carbonate and generates bicarbonate. Data calculated with seacarb and plotted with pygal.

It seems he is applying a naive understanding of le Chatlier’s principle, whereby carbonic acid loses hydrogen ions to become bicarbonate ions, which are then available to form carbonate ions. It’s intuitive, but it ignores the fact that the hydrogen ions released by carbonic acid tend to combine with existing carbonate ions, to form more bicarbonate. (Take a look back at Figure 1 if this isn’t clear)

H+ + CO3(2-) <=> HCO3-

Even stranger, he mentions this reaction (his Equation 6), but doesn’t seem to understand it. This is especially odd because it’s the key to ocean buffering, which Burnette makes a big deal of.

Let’s back up. A chemical buffer is a solution which resists changes to its pH. Seawater is a buffer system because it contains CO2 in a number of different states: carbonic acid, bicarbonate, and carbonate. If we try to raise the pH by taking away hydrogen ions, the system will respond by releasing some hydrogen ions from the carbonic acid and bicarbonate ions. The change in pH that we tried to make will be dampened. Similarly, if we try to add hydrogen ions, the carbonate and bicarbonate ions will absorb some of them, shifting the distribution of the CO2 system and again dampening the pH change. This is bad news because it’s not just the pH but the distribution of the carbonate system which is important to calcifiers, as we’ll see later. And remember, the calculated and modeled pH changes occur in spite of the ocean’s buffering; they would be even larger without it.

So we’re in the odd position that in some places Burnette is neglecting the effects of the equilibrium between bicarbonate (as in his calculation of the ocean pH change); in some places his argument centers on it (in the case of ocean buffering) and in some places he gets it completely backwards (as when he expects increased CO2 to increase carbonate concentrations)!

At other times, he has clearly not done any sort of literature review. In his summary, he claims: “There have been no experiments to demonstrate harm, only hypothesis and models.”

But this is flat-out wrong. For example, the results from (Anthony et al  2008) “indicated that high CO2 is a bleaching agent for corals and [crustose coraline algae] under high irradiance, acting synergistically with warming to lower thermal bleaching thresholds”. (DuPont et al 2008) found a “significant mortality increase in the low pH [high CO2] treatments versus controls” in echinoderm larvae. (Gazeau et al 2007) found that ocean acidification decreases shellfish calcification. (Wood et al. 2008) found that some organisms can ramp up their calcification rate to compensate, but only at the expense of muscle mass. (Albright et al 2010) found that high CO2 levels negatively impact juvenile corals as well as adults. (Comeau et al 2010) found decreased calcification in pteropods, important links in the Arctic food chain. The list goes on.

Indeed, although Burnette claims that “The experimental framework for testing carbonate organisms with increasing CO2 is easy, yet unperformed,” the framework has been developed and performed with considerable nuance (Riebesell et al 2010).

He also throws in a talking point we saw way back when we were debunking Dr. Everett’s testimony: “The organisms most susceptible to ocean acidification from CO2 evolved at a time when concentrations were 15 times higher than today.” This is true, but calcifiers evolved in times of environmental equilibrium, not perturbation states like today. (Kump et al 2008) explain: “In long-term quasi-steady-state conditions, there is also sufficient time for evolutionary innovation and adaptation by the biota to low pH conditions. Only in significant and geologically “rapid” departures from steady-state carbon cycling will both pH and saturation fall together, stressing calcifying marine organisms at a rate that may be beyond their ability to adapt and evolve.”

Furthermore, calcium carbonate saturation state depends on calcium ion concentrations, which have declined since the high-CO2 past. (Bradley Opdyke) This matters because calcifiers like corals are sensitive to a quantity called the saturation state of calcium carbonate, and this depends on both calcium and carbonate ion concentrations. When CO2 flows into the ocean and releases protons, they are buffered by the carbonate system, which results in a lower carbonate ion concentration, and hence a lower saturation state.  A higher calcium concentration (ie, higher alkalinity) can compensate for a lower carbonate concentration. In fact, a high-CO2 steady state will tend to have more alkalinity flowing to the oceans due to increased weathering. (Kump et al 2008) In short, “seawater chemistry comparisons between the Cretaceous, for instance, and the near future cannot be based on one carbonate system parameter alone.“ (Zeebe 2012)

Burnette wants us to look to the past to understand the impacts of ocean acidification on marine life: “…we can look back at history and watch how CO2 trends match with carbonate critter fossil records”. However, he doesn’t seem to have done so himself. Past analogs for the present event were unpleasant episodes (Kump 2009) and ocean acidification has recently been implicated in the Permian-Triassic mass extinction, a tremendous ecological crisis which hit ocean calcifiers especially hard. (Hinojosa et al 2012). This history is not mentioned or discussed.

In short, Burnette’s argument does not inspire confidence in his thesis. The calculations are off when they are coherent, and there is no engagement with established research on the matter. And it is just the one of the more recent posts from WUWT; there are plently of other lulz to be had. This article is unique in that it bungles geochemistry instead of geophysics or statistics.

Watts Up With That is a very silly website.

~~~

Albright R, Mason B, Miller M, & Langdon C (2010). Ocean acidification compromises recruitment success of the threatened Caribbean coral Acropora palmata. Proceedings of the National Academy of Sciences of the United States of America, 107 (47), 20400-4 PMID: 21059900

Anthony KR, Kline DI, Diaz-Pulido G, Dove S, & Hoegh-Guldberg O (2008). Ocean acidification causes bleaching and productivity loss in coral reef builders. Proceedings of the National Academy of Sciences of the United States of America, 105 (45), 17442-6 PMID: 18988740

Brewer, Peter G. (1997). Ocean chemistry of the fossil fuel CO 2 signal: The haline signal of “business as usual” Geophysical Research Letters, 24 (11) DOI: 10.1029/97GL01179

Caldeira, Ken, & Wickett, Michael (2003). Anthropogenic carbon and ocean pH Nature, 425 DOI: 10.1038/425365a

Comeau S, Jeffree R, Teyssié JL, & Gattuso JP (2010). Response of the Arctic pteropod Limacina helicina to projected future environmental conditions. PloS one, 5 (6) PMID: 20613868

Sam Dupont,, Jon Havenhand, William Thorndyke, Lloyd Peck, & Michael Thorndyke (2008). Near-future level of CO2-driven ocean acidification radically affects larval survival and development in the brittlestar Marine Ecology Progress Series, 373, 285-294 DOI: 10.3354/meps07800

Gazeau, Frédéric, Quiblier, Christophe, Jansen, Jeroen M., Gattuso, Jean-Pierre, Middelburg, Jack J., & Heip, Carlo H. R. (2007). Impact of elevated CO2 on shellfish calcification Geophysical Research Letters, 34 (7)
Hinojosa, J. L., Brown, S. T., Chen, J., DePaolo, D. J., Paytan, A., Shen, S.-z., & Payne, J. L. (2012). Evidence for end-Permian ocean acidification from calcium isotopes in biogenic apatite Geology, 40 (8), 743-746 DOI: 10.1130/G33048.1

Kump, Lee, Bralower, Timothy, & Ridgwell, Andy (2008). Ocean Acidification in Deep Time Oceanography, 22 (4), 94-107 DOI: 10.5670/oceanog.2009.100

Riebesell U., Fabry V. J., Hansson L. & Gattuso J.-P. (Eds.), 2010. Guide to best practices for ocean acidification research and data reporting, 260 p. Luxembourg: Publications Office of the European Union.

Wood, Hannah L, Spicer, John I, & Widdicombe, Stephen (2008). Ocean acidification may increase calcification rates, but at a cost. Proceedings of the Royal Society B, 275, 1767-1773 DOI: 10.1098/rspb.2008.0343