The Weekly Carboholic: Cassava, sorghum yields drop, toxicity rises with more CO2

Posted on July 1, 2009




Cassava and sorghum are tubers that form the protein base for hundreds of millions of people. But while there’s a great deal of protein in the plant, there’s also cyanide in the plant’s leaves. Whether the leaves are poisonous or not depends partly on how much protein there is – more protein means that the cyanide is less toxic and the plants are safe to eat for man and beast alike. But according to a new study reported in Reuters, higher carbon dioxide (CO2) concentrations means both less protein and more cyanide, a toxic combination.

According to the article, an Australian team grew cassava and sorghum under different CO2 concentrations that approximated the various projected climate disruption scenarios for the rest of this century. What they found was that “the amount of cyanide relative to the amount of protein increases” and that “[a]t double current CO2 levels, the level of toxin was much higher while protein levels fell.” As a result, cassava-dependent communities could be poisoned, especially when experiencing a drought.

The article pointed out a greater worry, however – at high CO2 concentrations, the crop yields fell significantly:

[Monash University researcher Ros Gleadow said] “There’s been this common assumption that plants will always grow better in a high CO2 world. And we’ve now found that these plants grew much worse and had smaller tubers.”

CO2 has been referred to as “plant food” in some circles. This study suggests that this is not necessarily the case. Other studies have discovered increased crop yields due to more CO2 in the atmosphere could actually lead to more starvation as the protein content of those crops falls dramatically.

Both studies illustrate that the “plant food” meme is false, at least as it applies to the staple crops people actually eat.


Differences between aerosol effects in models vs. observations largely explained

Aerosols like pollution, airborne dust, black carbon particles, even sulfur dioxide (SO2) have many different effects. Black carbon absorbs solar radiation and heats up the air or melts the snow and ice it settles on. Sulfur dioxide cools the planet when blasted into the stratosphere by a volcano, but may heat up the Earth and produce acid rain when located lower in the atmosphere. Airborne dust and pollution increase the rate of cloud formations – except when they decrease the rate instead. Scientists know that clouds and aerosols interact greatly, and since the effects of clouds on climate – and of climate on clouds – remains one of the few major unknowns in climate models, improving scientific understanding of aerosols is similarly critical to improving climate model predictions.

One of the recent problems with aerosols is that satellite measurement-derived estimates of aerosol radiative forcing (hereafter referred to as SDRF, for satellite derived radiative forcing) have differed by from modeled predictions of aerosol RF (MRF, modeled radiative forcing) by up to a factor of two, well outside the margins of error for both measurements and models. Even worse, there was also a statistically-significant difference between two different sets of SDRFs . Scientists haven’t been able to determine why there was such a large difference between and among the SDRFs and between SDRF and MRF until now. A new paper published in the journal Science claims to have not only explained the differences between the different aerosol RFs, but to explain differences between the two satellite-based datasets as well.

According to the paper, the discrepancy is a result of two assumptions made in the process of calculating aerosol RFs. The first assumption made in the calculation of SDRF is that “there is no radiative effect of the aerosols within cloudy sky areas.” Models, on the other hand, don’t make this assumption. The second assumption is that the there was no anthropogenic aerosols prior to 1750 (defined as the start of the industrial era), a false assumption. In addition, there is a third difference between SDRF and MRF that isn’t a difference in starting assumption – the SDRFs don’t have complete earth coverage. The MODIS satellite measurements that are the basis of most calculated SDRFs can’t take measurements over highly reflective terrain like ice and desert, and so significant swaths of the Earth’s surface can’t be observed. Again, the models don’t have this problem.

model-satelliteThere are two ways to prove that the different calculated RFs are actually statistically the same – demonstrate that SDRFs can be made equal to the MRFs, or demonstrate that the MRFs can be made equal to the SDRFs. The paper does both. First, by using model data to fill-in the places that the satellites can’t measure and then by changing the initial assumptions used in SDRF calculations, the paper illustrates that the satellite-based RFs are equal to the modeled RFs (marked in blue in the image at right). Then, by changing the model parameters to match the assumptions underlying the SDRFs, the paper illustrates that the MRFs were made to be equal to the satellite-based numbers (marked in red in the image at right). If you notice, the two MODEL lines (Int and Ext) look very similar to the MODIS (Model) line, just as the MODEL (Sat & opt obs) line looks very much the same as MODIS line.

blackcarbonThe paper also identifies what specific aerosol is mostly responsible – black carbon, aka soot. As the image shows, there has been a massive increase in the amount of black carbon present in the atmosphere since pre-industrial times (“more than a factor of six”). As a result, the reflectivity of the Earth’s atmosphere has dropped:

The global mean annual average single scattering albedo computed in the model for all aerosols at 0.55 um is 0.986 at pre-industrial conditions and 0.970 at present-day conditions. Thus the aerosol in present times is approximately twice as absorbing as that in pre-industrial conditions.

There are two caveats, however. The first is that the SDRFs are not model independent at this time – model data is used to fill in the parts of the satellite observations that the MODIS instruments can’t detect. This means that, if the models are wildly wrong, then the SDRF calculations are going to be wrong as well, although not as wrong as the models would be alone (the error would be proportional to the area filled in with model data).

The second, and more important, caveat is that the changed assumption about the pre-industrial aerosol levels may not actually be correct. Given that the new assumption also explains differences between two different SDRFs means that the assumption is likely to be correct, but further research will be necessary to test the validity of the new assumption.

All in all, this is a valuable study that will both improve climate modelling and quell some concerns about differences between models and observations of aerosol radiative forcing.

Thanks to paper author Gunnar Myhre for a review copy of his paper.


methhydrateMethane clathrates proposed for energy and carbon sequestration

An article in New Scientist suggests that countries are looking to methane clathrates (methane frozen into ice) for two purposes – a source of natural gas and a carbon sequestration opportunity.

As an energy supply, the methane held in clathrate form under the Arctic, off the coast of Japan and India, and elsewhere around the world hold significant potential. The article says that these deposits are estimated to hold trillions of cubic meters of methane that could, if questions of scale and safety can be worked out, power hundreds of millions of homes for a decade or more. But there are significant problems.

The first is that the methane held in the clathrates are difficult to extract – either the ice has to be melted or the pressure that helps keep the methane locked into the ice must be lowered. The article says that researchers tried the melting method and found it took too much energy, but that the decompression technique appeared to work well, and has been powering an industrial furnace in Siberia for decades.

Which brings us to the second problem. Extracting methane from clathrates on large scales runs the risk of destabilizing the entire deposit, and depending on where the deposit is and how large it is, that could result in underwater landslides that cause tsunamies. Any life close to the “methane burp” would probably be asphyxiated as well. And if the burp was really big, it could produce short-term climate effects around the world – methane is moderately powerful greenhouse gas as compared to CO2.

Some researchers are hoping to extract clathrate a different way, though – by replacing the methane in the ice with CO2. This has the supposed benefit of sequestering the CO2 – but only if you assume that there will never be such a thing as a “CO2 burp” out of a destabilized CO2 clathrate deposit.

I understand the interest in this, and I think additional research is warranted. But industrial scale deployment of a methane clathrate harvesting technology should not be deployed until the risks and potential safety issues have been well documented and are understood.

Image credits:
Northern Arizona University
Science Paper, “Consistency between satellite-derived and modeled estimates of teh direct aerosol effect”

Posted in: Uncategorized