Feedbacks and tipping points

 From  ....and Then There's Physics

There’s been some discussion on Twitter about feedbacks, runaways, and tipping points. The issue is that some seem to confuse these and sometimes imply that we could cross thresholds where we’ll undergo a runaway. I thought I would briefly try to explain these terms.

In the context of climate change, external factors that can lead to warming are typically called forcings. This would be things like changes to the solar flux, volcanic eruptions, and our release of greenhouse gases into the atmosphere. Feedbacks are then responses to this externally driven warming that either act to amplify, or suppress, the warming. Some of these are fast, such as water vapour and clouds, while others are slower, such as changes to vegetation or ice sheets. Some are also negative and quite strong (such as the Planck response). This means that even though the overall effect of these feedbacks is to amplify the externally-driven warming, it is limited (the negative feedbacks eventually balance the the effect of the change in forcing and the resulting positive feedbacks). For example, if we were to double atmospheric CO2, we’d expect to eventually warm by about 3oC.

A runaway, on the other hand, typically refers to what happened on Venus. Essentially, virtually all of the CO2 was released into the atmosphere, the warming was so substantial that any liquid water evaporated and was eventually lost to space, most atmospheric molecules lighter than CO2 were also lost to space, and the surface warmed by many 100s of oC. On the Earth, such a runaway is simply not possible, because most of the carbon, that can then form CO2, is locked up in the lithosphere. We can’t emit enough CO2, either through anthropogenic influences or naturally, to undergo a runaway.

Finally, a tipping point refers to us crossing some threshold where the climate system changes (tips), irreversibly, into a new state. There is the possibility of a global tipping point, but this is seen as very unlikely. However, it is possible that we could cross thresholds where some parts of the system undergo essentially irreversible changes. Examples would be melting of the West Antarctic Ice Sheet, the Greenland Ice Sheet, Amazon rain forest die-off, release of carbon from the permafrost, and the disappearance of summer Arctic sea ice.

If we were to cross any of these tipping threshold, then the changes would further amplify the warming (through either releasing additional CO2, or methane, or changing the albedo) and – in the case of the ice sheets – would lead to substantial sea level rise. There are a few things to bear in mind, though. The timescales are typically long; if we cross a tipping threshold it will still take a long time (centuries) for the full effect to manifest. Also, we don’t have a particularly good idea of where these thresholds might lie; we could already have crossed some, or might not do so unless we were to warm substantially. Additionally, there is still debate as to whether or not some of these are truly irreversible; if we could artificially draw down atmospheric CO2 would some, like Arctic summer sea ice, then reverse?

So, one does have to be slightly careful as to what one implies about tipping points and their significance. On that note, I was going to highlight a recent paper by James Annan that might initiate an interesting discussion about using tipping points in the public narrative. However, this has got long enough, so I will leave that to another post.

Update: Something I meant to stress, is that even though crossing some tipping point might lead to irreversible changes that would amplify our warming, it is still limited. There is only so much that these effects can change the albedo, and there is a limit to how much CO2, or methane, that they could release.

Update 2: If you want to read a more detailed description of the runaway greenhouse effect there is a very good Skeptical Science post by Chris Colose. There is a formal aspect to a runaway that I hadn’t appreciated, but was made aware of by MarkR’s comment. It involves a condensable greenhouse gas (such as water vapour) in equilibrium with a surface reservoir, accumulating in the atmosphere so that it eventually limits the outgoing longwavelength flux. If the incoming flux exceeds this, then energy will accumulate without there being any way to balance this. The surface will then undergo runaway warming until the entire surface reservoir of the condensable greenhouse gas has been depleted.

For water vapour, this longwavelength limit would occur at about 310W/m^2, which – given our albedo of 0.3 – is currently greater than the flux we’re receiving from the Sun. This essentially means that water vapour in the atmosphere is in a regime where it condenses, rather than accumulates. We, therefore, can’t undergo such a runaway in our current state, and no level of CO2 emissions will trigger it.

Note: none of the above suggests that we can just relax and not bother about cutting emissions.  Just because we won't have a Venusian runaway rise in temperatures does not mean we won't have steadily worsening climatic conditions as temperatures rise.  The ± 1 degree C we have had so far is already leading to record floods, droughts, hurricanes, heatwaves and bushfires.   2 degrees, which seems entirely possible given that most of the world is still not approaching the issue of global warming with enough urgency, will make all these climate extremes much worse.