Reading
- NAP report on “Abrupt impacts of climate change: anticipating suprises.”
- Nice overview of major climate tipping points. Interesting that the 2011 IPCC report put to rest concerns about a sudden transition of the thermohaline circulation within this century, despite good evidence that it is a bistable dynamic. Likewise appears to dismiss concerns of a tipping point related to the release of permafrost methane, thus putting aside the two most well-known and well-studied examples. Meanwhile no shortage of other concerns, from species extinctions and other “climate impacts” to sea-level rise, ice sheet melting, acidification, ocean oxygen depletion and other examples. (See table 4.1)
- Box 4.2 covers early warning signals, and a citation to my Nature piece (with my name misspelled) echoes our skepticism of generic warning signals and overall handles these issues rather well. (pg 153)
- Recommends a global Abrupt Change Early Warning System (ACEWS) with a rather nice discussion of some existing monitoring efforts one would build on, and highlighting the need for modeling and synthesis as well.
- Nice overview of major climate tipping points. Interesting that the 2011 IPCC report put to rest concerns about a sudden transition of the thermohaline circulation within this century, despite good evidence that it is a bistable dynamic. Likewise appears to dismiss concerns of a tipping point related to the release of permafrost methane, thus putting aside the two most well-known and well-studied examples. Meanwhile no shortage of other concerns, from species extinctions and other “climate impacts” to sea-level rise, ice sheet melting, acidification, ocean oxygen depletion and other examples. (See table 4.1)
Two very interesting papers by @jameshowison discuss issues of practices and incentives in software publication. In particular, 10.1145/1958824.1958904, pdf is an excellent survey of both some common and different challenges in high-energy physics vs. structural biology vs. bioinformatics (would have been great to see a field like ecology included in which the software used is potentially more heterogenous and less central still…) Likewise, his second paper (pdf), which focuses on the evolution and versions of BLAST, includes an interesting discussion with particular messages for funders on how research software incentives differ from the open source software movement, and what possible incentives might ameliorate these challenges.
- Piece from Matt Pennell on punctuated equilibrium and reply critiquing it. Sounds like the two agree more than they let on, with confusion in the terminology lying not so much in its creation as in its subsequent use in the literature.
- Pennell, M. W., Harmon, L. J. & Uyeda, J. C. 2014 Is there room for punctuated equilibrium in macroevolution? Trends Ecol. Evol. 29, 23-32. (doi:10.1016/j.tree.2013.07.004)
- Lieberman, B. S. & Eldredge, N. 2014 What is punctuated equilibrium? What is macroevolution? A response to Pennell et al . Trends Ecol. Evol. , 1-2. (doi:10.1016/j.tree.2014.02.005)
- somewhat less clarity in the debate on generic early warning signals in medical contexts:
- Bos, E. H. & De Jonge, P. 2014 “Critical slowing down in depression” is a great idea that still needs empirical proof. Proc. Natl. Acad. Sci. U. S. A. 111, 2014. (doi:10.1073/pnas.1323672111)
- Wichers, M., Borsboom, D., Tuerlinckx, F., Kuppens, P., Viechtbauer, W., van de Leemput, I. a., Kendler, K. S. & Scheffer, M. 2014 Reply to Bos and De Jonge: Between-subject data do provide first empirical support for critical slowing down in depression. Proc. Natl. Acad. Sci. 111, E879-E879. (doi:10.1073/pnas.1323835111)
Misc
- Melissa lab group meeting on statistical estimation from arial survey data on sea otters. Very neat stuff.
- Going over ropensci reporting materials
- Going over NSF reporting requirements (see subsequent post).
Meeting with Steve
- Go over derivation – had dropped a \(\sigma^2\) term in the denominator so wasn’t getting the expected Gamma-like integral.
Next steps:
- Consider random-walk prior / functional rather than linear forcing
- Explore implementation via Kalman filtering