Friday links: are regime shifts even a thing, rewilding English, minimal maps, and more (UPDATED)

Also this week: read ALL the Terry McGlynn things, ice ages vs. the Slutsky-Yule effect, grad student mental health, the null hypothesis for p-hacking, and more!

From Meg:

Thanks to Terry McGlynn for pointing me to this great post by Nash Turley (in the comments of my post on academics as humans). In it, Nash argues that mental health issues present the greatest barriers to success among grad students, and gives suggestions on how to acknowledge and address those problems. I agree with Nash that this can be a major barrier for students. Colleagues from many different universities have said this is an important issue they face in their graduate programs.

Speaking of Terry, he had a really important, thought-provoking post this week on the NSF Graduate Research Fellowships, and how they perpetuate privilege and end up being a barrier to diversifying STEM. This is not an issue I’d considered before, but I’m glad Terry made me start thinking of it!

Here’s a call to “rewild our language of landscape”, to add back in words related to nature. It includes this passage:

The same summer I was on Lewis, a new edition of the Oxford Junior Dictionary was published. A sharp-eyed reader noticed that there had been a culling of words concerning nature. Under pressure, Oxford University Press revealed a list of the entries it no longer felt to be relevant to a modern-day childhood. The deletions included acorn, adder, ash, beech, bluebell, buttercup, catkin, conker, cowslip, cygnet, dandelion, fern, hazel, heather, heron, ivy, kingfisher, lark, mistletoe, nectar, newt, otter, pasture and willow. The words taking their places in the new edition included attachment, block-graph, blog, broadband, bullet-point, celebrity, chatroom, committee, cut-and-paste, MP3 player and voice-mail.

As someone who loves words and nature, I loved reading this piece! I particularly liked this line:

to use language well is to use it particularly: precision of utterance as both a form of lyricism and a species of attention

Long, but beautifully written! (ht: Jacquelyn Gill)

Here’s an interesting blog post by an evolutionary biologist, John Stanton-Geddes, who has decided to leave academia for a career as a data scientist. (My only quibble: starting academic salaries are definitely negotiable at least in some cases.) ht: Titus Brown

I also enjoyed this post from BiochemBelle about changing her plans to follow an academic path (which is often seen as the default) to a non-academic path. She talks about the fears and anxieties and doubts associated with that decision, but also about the pride and gratification she gets from following her own path.

These “minimal maps” are really neat and show where the nation’s croplands, deciduous forests, water, and urban areas are. (ht: Elena Bennett)

From Jeremy:

Charley Krebs points us to two recent reviews of the surprisingly weak evidence for alternative stable states, regime shifts, and tipping points in ecology. Yup. Frankly, even in microcosms it’s hard to find decent examples, and that’s not for lack of trying.

The myth of Europe’s Little Ice Age. It’s an artifact of smoothing the data, known as the Slutsky-Yule effect. (UPDATE: an informed commenter pushes back hard against this one.)

Following up on last week’s post on evidence for widespread p-hacking, this old post from Daniel Lakens (commenting on a study I’ve linked to previously) argues that p-hacking should not be expected to increase the frequency of p-values just below 0.05. I’ll continue to follow this discussion with interest.

How come the college-level textbooks in any given field all use the same “classic” (in the sense of “really old”) examples, have the same organization, etc.? Doesn’t this have the effect of preserving the field in aspic? I’ve complained about this in the past, now Terry McGlynn takes up the call for truly new textbooks. And see the comments in both posts, where Markus Eichorn reveals that he is on it.

12 thoughts on “Friday links: are regime shifts even a thing, rewilding English, minimal maps, and more (UPDATED)

  1. RE: rewild our language–I bet you’d find the same pattern (less nature, more tech) in Ecology classrooms and journals too.

  2. Great links, many thanks!

    Jeremy: Though I have been critical of some of the regime shift literature, I would quibble with your paraphrasing of the two reviews Krebs links. In my opinion, those reviews are making almost the opposite point: far from concluding that a vast literature has measured very carefully to look for regime shifts and fail to find them, what they really show is that very little literature has actually presented the data needed to test these hypotheses in the first place — that is, papers that turn up in search engines using terms like “threshold” or “regime shift” very rarely have the time-series data measurements to really test these assertions.

    I think this has much more to do with the fact that it is immensely difficult to conduct the experiments needed that can really tie to such observations to mechanisms of regime shifts than it has to do one way or another with how common these phenomena are in the natural world. I agree with you that even many of the microcosm experiments are not truly convincing, but again I find that is due to limitations of the design and data collection. In my mind, the work of Jeff Gore’s group at MIT in experiments led by Lei Dai have really been exceptional in this regard, and take advantage of a degree of control and replication you just can’t create that easily in meso-scale or whole-lake experiments.

    There’s also an issue of terminology that seems to perpetually cloud this research. Much of the literature uses a much broader (and usually more vague) definition of the terms ‘regime shift,’ ‘threshold’ and ‘transition’ than is captured by the classic Holling / Walters saddle-node bifurcation event; and much of the literature claiming to consider ‘regime shifts’ is doing so only under a weaker definition than those reviews have in mind. Unfortunately, both these reviews and the papers they discuss continue to present their own definitions of the terms that continue to capture multiple different mathematical phenomena that have qualitatively different behaviors. (We tried making this point with a Venn diagram in Figure 1 here:, perhaps unsuccessfully. Alternative ideas to explain this would be much appreciated!).

    I guess both the issues of terminology and issue of when absence of evidence is evidence of absence are quite a bit broader than this context too (density-dependence debate, anyone?). What do you think?

    • I very much agree with Carls interpretation of the results of the two review papers. They rather demonstrate absence of evidence for regime shifts, alternative attractors etc. than the opposite, at least for real ecosystems.

      • It’s true that absence of evidence isn’t the same as evidence of absence. But if absence of evidence is mistaken for presence of evidence, that’s a problem. I take Charley to be making the case that ecologists are too often claiming regime shifts in the absence of evidence (or, equivalently, using an overbroad and vague definition of “regime shifts”).

  3. Re: new textbook organization.

    I suspect the reorganizers will discover that we’ve stuck with round wheels all this time for good reasons. :). IMO its kind of the same story as the various “innovative” attempts to rid us of the old and tired lecture method of teaching – so many great ideas! If only they actually worked….;)

    • Hmm…not so sure about that. See Markus’ comments. He was motivated because the current content and organization of ecology textbooks failed to reflect what he and other ecologists actually do all day. Obviously can’t judge a new textbook without having seen it. But I do think there’s more motivation for a new ecology textbook than just the vague sense that “new”=”better”.

      • Judge it when you see it… I can only tell you what I’m trying to do, and whether I’ve succeeded will be for others to decide. Just going over figure permissions this weekend. BTW, it’s Eichhorn with two h’s.

    • Jim, so what you’re saying is that approaches involving teaching other than lecturing don’t work? There’s not really any data to support that idea, and there are oodles of data that contradict that idea.

      • Well terry, I’m not an education researcher. And I suspect your an ecologist. But while other approaches may have some merits, there’s a reason the lecture has survived since before Socrates. Having read a small but wide sampling of education research, I’d also caution against putting too much stock in it. IMO a well prepared and delivered lecture is extremely effective.

        WRT textbook organization, I can’t argue ecology – I’m not an ecologist. I’m speaking more generally about science textbooks. Just the same, having made some modest efforts myself, I found that reorgs lead to new probs that the reorg creates – probs that are worse than the original.

  4. The so-called myth of the European Little Ice Age, and certainly the link to which you refer, is pretty dubious.
    I asked a paleoclimatologist for some comments:
    “Regarding the so-called slutsky-Yule effect: current climate syntheses do take that into account.
    The selection of the Dutch data is rather cherry-picking. That reconstruction is based on a limited number of tree-ring records. By now, everybody should know that the traditional method of just concatenating overlapping tree-ring records into a single 800-year reconstruction has the tendency to downscale the importance of long-term trends relative to interannual variation. Current methods take care of this problem, and use both high- and low-resolution proxies (for example Moberg et al. 2005), and/or combined analyses of time series on a larger spatial scale (see”
    There’s much more to say about that link, but codswallop is what describes it best.

Leave a Comment

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.