Baffled by BERN, but beware of biofuels
Today’s project – well it was going to be this morning’s project – was to evaluate the two papers I recently tried to summarise one here: from Hansen et al and from Kharecha and Hansen.
The first question I asked myself was whether the conclusions of these papers are compatible with my Deep Green perspective. What is the Deep Green perspective? You may well ask. The point is that we have to attend to all parts of the Earth’s carbon cycle.
Both papers referred to something called the BERN carbon cycle model. At first I thought BERN was a place, then that it was a person (short for Bernie, perhaps). It turns out of course, thanks to Part 1 of the 4th IPCC Assessment Report (4AR, for short) – best doorstop I ever bought (tip, always ask for a discount) – that BERN is an acronym. I don’t know what BERN stands for, but I have learnt that it is an example of an EMIC – an Earth System Model of Intermediate Complexity, believe it or not.
What it seems is built in to BERN is the idea that a “pulse” of carbon emissions put into the atmosphere is gradually absorbed over time, such that a defined proportion remains in the atmosphere after t years (though it seems the model does allow this to be modified in order to simulate carbon cycle feedbacks). The equation is a series of exponential functions, given in Khurecha and Hansen. This is fine if we’re talking about a single emission of carbon – say a volcano going off in an oilfield – and its gradual re-absorption over time. But what appears to be being done is to try to apply this principal to emissions over a number of years by “integrating the results from 1850 to year t“. After much thought and fiddling with spreadsheets, I’m still baffled why anyone would want to do this.
Tied in with the BERN equation is the idea of the Airborne Fraction (AF) of fossil-fuel emissions, which has been observed to be remaining roughly constant at about 60% (AR4, p.139).
The justification behind both the BERN equation and the (dangerous) AF concept is, it seems, the idea that the processes that remove carbon dioxide from the atmosphere – ocean uptake and the fertilisation effect on land – remove a (roughly) fixed proportion each year of the difference between the current atmospheric level of CO2 and the equilibrium value.
Now, my point is that there is, in the real world, no necessary relationship between our emissions and the rate of uptake of carbon by the biosphere. Our emissions go into the atmosphere changing the level of CO2 – now over 380ppm, compared to about 280ppm before industrialisation – but the uptake processes depend on the level in the atmosphere, not the annual change in the level. It is therefore very dangerous to drift into assuming a constant AF.
To test my hypothesis, I stuck some numbers into a spreadsheet – mainly the rough CO2 levels and emissions between 1960 and the present, calculating absorption rates based on IPCC data (AR4, p.26) extrapolating into the future (spreadsheet available on request). All I’m doing is adding what carbon we’re putting in to the atmosphere, and what’s being taken out each year, to what’s already there to produce a time-series. And, lo and behold (actually I was a tad surprised), it is indeed the case that the AF is fairly constant under these assumptions at about 60%.
The trouble is, to my mind, thinking in terms of an AF and integrating annual pulses of emissions is really odd way to look at the problem.
The simple way to expose the limitations of this approach is to see what happens if you start to decrease emissions. And, sure enough, the AF drops considerably (and can even go negative) if you do this. In other words, the whole approach only works while CO2 emissions and atmospheric levels are increasing at a fairly steady rate.
But this is only a minor problem. The AF is extremely sensitive to saturation of the processes to remove CO2 from the atmosphere. And it appears that removal by the oceans is indeed saturated (AR4, p.26 & elsewhere).
If we allow for the fact that the rate of carbon uptake by the oceans is not going to increase, then (of course) we see that the AF increases, if our emissions continue to increase, as in the graph below:
The implication of all this is rather important. If we continue to increase our emissions, then we face a double whammy. A higher proportion of the higher level of emissions will remain in the atmosphere.
At this point I realised that I had constructed a rough carbon cycle model in my spreadsheet. I was able to test different emission scenarios, varying the behaviour of other parts of the carbon cycle. And I can report that we are indeed in big trouble. Here are some preliminary conclusions:
1. If we’re to keep CO2 levels below 450ppm – the absolute optimistic minimum to prevent dangerous climate change of 2C or more – global carbon emissions have to peak by 2015 and decline steeply, that is, by about 1.5% pa – my 2045 peak example led to 630ppm by 2070. I don’t believe anything faster than 1.5% is feasible – in fact, even that looks like a silly number to me, though I’m open to persuasion.
2. In a scenario where we do stabilise CO2 levels, then (of course) the AF declines dramatically, and eventually goes negative. The AF is a double-edged and misleading concept if used as part of a causal explanation. It can only be sensibly used – IMHO – to make the rhetorical point that increasing our rate of carbon emissions will make our problems much worse very quickly.
3. In my spreadsheet carbon cycle model, the outcome is sensitive not only to the emission peak and rate of decline and to the timing of the peak rate of ocean uptake (I’ve assumed 2010, which may be optimistic – AR4 implies it has already peaked). Critical, also, is the rate of land uptake of CO2. My model has the fertilisation effect dependent on the area of natural ecosystem. Now, here we have another double whammy. If we clear forest (and, worse, wetlands) to create grazing land, or to grow more crops – for example, biofuels – then we not only emit carbon as we do so, we also reduce the rate of carbon uptake by the fertilisation effect.
4. If land (and/or ocean) uptake of CO2 goes into reverse as the planet warms we are completely screwed, to put it scientifically. If this happens it will be almost impossible to keep atmospheric CO2 (not CO2 equivalent, just CO2) below 45ppm.
In fact, my conclusion is that it will be practically impossible to keep CO2 below 450ppm unless we:
1. Start reducing fossil fuel carbon emissions within the next decade; and
2. Significantly increase the area of forest and wetland over the next century.
Given point 2, it might be a good idea to suspend all incentives and quotas for biofuels.
0 thoughts on “Baffled by BERN, but beware of biofuels”
“I’m still baffled why anyone would want to do this.”
“The justification behind both the BERN equation and the (dangerous) AF concept is, it seems, the idea that the processes that remove carbon dioxide from the atmosphere – ocean uptake and the fertilisation effect on land – remove a (roughly) fixed proportion each year of the difference between the current atmospheric level of CO2 and the equilibrium value.”
“Now, my point is that there is, in the real world, no necessary relationship between our emissions and the rate of uptake of carbon by the biosphere. Our emissions go into the atmosphere changing the level of CO2 – now over 380ppm, compared to about 280ppm before industrialisation – but the uptake processes depend on the level in the atmosphere, not the annual change in the level. It is therefore very dangerous to drift into assuming a constant AF.””
I agree with all the sentiments and am similarly baffled by BERN. Do you think all of the processes depend on the level in the atmosphere dCO2(Outflow)/dt = a(CO2 – p) where a & p are constants. This would be my simple understanding.
– ie is BERN *completely* wrong?
Or are there some processes that are BERN-like e.g. equilibrium with the upper oceans so that dCO2(Outflow)/dt = a1(CO2 -p) + b(dCO2(Inflow)/dt)
I’d like it that the simple model (that bern is completely wrong) was the valid one.
Sorry, your second equation makes sense! According to my analysis, some processes of carbon exchange between the atmosphere and the oceans are indeed acting to reduce the atmospheric level over time so that:
dCO2(Outflow)/dt = a1(CO2 -p) + b(dCO2(Inflow)/dt)
The BERN-like term b(dCO2(Inflow)/dt) is dominated, if that’s any consolation, by the non-BERN-like one!
I discuss your point further and have revisited the issues discussed in this post, here.