On my trip to Queens University last week, I participated in a panel session on the role of social media in research. I pointed out that tools like twitter provide a natural extension to the kinds of conversations we usually only get to have at conferences – the casual interactions with other researchers that sometimes lead to new research questions and collaborations.

So, with a little help from Storify, here’s an example…

In which we see and example of how Twitter can enable interesting science, and understand a little about the role of existing social networks in getting science done.


  1. well, you have to be a little careful with rules of thumb. If forcing is linear, you can solve the basic ‘climate’ equation exactly (e.g where dT/dt = F – T/S ; T is temp anomaly, S is sensitivity, F is the forcing, heat capacity defined to be 1). Linear forcing in time has F ~ t for t>0, thus

    T = S*t – S^2 * (1 – exp( -t/S )), t>0

    so it’s only really linear at timescales long compared to the response time. For short times it is quadratic in t.

    Sensitivity to emissions is more complicated because CO2 concentration is a complicated function of emissions (multi-exponential), emissions themselves have been growing faster than exponential in any case, and as mentioned, forcing is logarithmic in C. It so happens that these complications smooth out and you do get a reasonably linear *equilibrium* temperature increment as a function of emission (see Allen et al, Meinshausen et al, Nature special issue in 2010(?)), but it’s not obvious that this would work out. But this is not a statement about the transient response in temperature – that depends on the pathway of the emissions.

    Issues related to the dependence of S on base climate and included feedbacks are interesting but beside the point.

  2. Hehe, great to see that my article made a splash!

    I second Gavin’s comment (Gavin Schmidt? or a different Gavin?) about the important distinction between transient response and equilibrium temperature. The linear relationship depends on slow feedbacks, which aren’t taken into account for TCR, or even most GCM measurements of S2x to my understanding. If you take out these slow carbon feedbacks the nice proportionality falls apart.

    Here is the Allen et al paper you were probably thinking of: http://www.nature.com/nature/journal/v458/n7242/full/nature08019.html

  3. @Kaitlin : Apologies – your comment ended up in the spam box; don’t know why. Anyway sounds like you’re having a great summer at UVic!
    (PS. Yes, that’s Gavin Schmidt)

Join the discussion: