Ghost research: taking stock of work that disappears

I recently discovered a citation (in Breen, 1990) referring to a short article my father wrote for National Fisherman, in 1988, in which he reported on a new kind of lobster trap with a “catch escape panel”, aimed at reducing bycatch. My dad had a steady freelance gig at the time with National Fisherman, and the article was one among several he wrote while researching “ghost traps” – lobster traps, specifically, but really any lost fishing gear (nets, lines) that disappears underwater for reasons random, accidental, or deliberate.

Ghost Lobster Trap

A ghost trap (via the Bangor Daily News).

With lobster traps, it’s easy to imagine what happens. To retrieve traps and the lobsters in them, a fisher works her way along from floating buoy to buoy (painted in her own distinctive, garish, high-contrast colors for good reason – try tracing a series of such bobbing breadcrumbs in fine let alone foul weather). Each buoy is connected to a heavy “sink line” that is in turn fixed to a trap, which sits on the seabed (where the lobsters live). If something – a propeller from a passing boat, for example – parts the sink line, then the buoy drifts off with the current and the trap is lost.

And yet the trap keeps fishing. Hence the term “ghost fishing” – which refers to lost gear that continues to catch whatever hapless critters get stuck or entangled (and die) in it. In the 1980s there was a regulatory push in the US to make lobster traps deliberately less effective. Their redesign made it easier for more lobsters to come and go from the trap, and, constructed from coated but degradable wire, the idea was that a lost trap would eventually crumble into a little rusty pile.

There was a lot of interest, vested and curious, in how lobsters interacted with the new gear. In one of my all-time favorite data-gathering experiments, a group of researchers deployed traps mounted with video cameras to observe how lobsters behaved among the traps in the fishing grounds (rather than in a tank with bubblers and a plastic castle). They found the traps caught 6% of the lobsters that entered them. (Here is where an accomplished commercial-fishing friend of mine would add that you can’t legally keep everything in the trap when you haul it up. By my buddy’s own back-of-the-envelope numbers, the trap catches ~10% of what’s moving through it, and of that, ~10% is legal to keep. All things being equal, he’s harvesting ~1% of the lobsters hanging around any given trap.)

I went looking for literature on ghost fishing because I had an awkward analogy to satisfy.

Over the calendar months of 2015 and 2016 I sank a huge amount of time, effort, and energy into grant proposals. Two (both for large sums of money) I led outright: one was an unusual solo shot, in which I was the only investigator named; the other had a more conventional structure, involving co-investigators from two other institutions, a visiting researcher from a third, and project partners from three non-governmental organizations.

(An aside: “large sum of money” is relative. The sums at stake in my world dwarf most humanities grants by an order of magnitude, but are themselves an order of magnitude smaller than most grant awards in the biomedical sciences. The lion’s share of a “large sum” is to employ one or more persons for two or three years as a dedicated, full-time researcher, and to pay their institutional overheads. A portion of a grant is for travel and field or experimental work. Very little tends to go toward stuff – “consumables.”)

The solo bid died by faint praise in an initial cull. The other garnered six peer reviews (which required a formal response from us), went to a review panel, but ended there. I prepped another grant that ultimately failed to launch (lacked a green-light). A fourth grant – for a travel exchange, inexpensive but exciting – I co-wrote with a close colleague, but we weren’t funded. And I was aboard (not leading) three major consortium grants, each involving many, many collaborators. (Two failed, but – yes – one hit. And my experience with one of the not-funded-would-be consortia did motivate this piece for the Careers section of Nature.)

Closing the old year and opening the New Year ’tis always a season for reflection. Until last month (which saw the arrival of two publications, here and here, both long in the making), there was little to ink 2016 into my permanent academic record. The past year has been one of transience – not just in leaving one institution to begin at another, but especially in the ephemera of the funding chase. One of my colleagues is adamant about recording that ephemera: he lists all grant proposals, funded and not, in his full CV because each grant represents so much work. He says each one should count for something – at least by being written down somewhere.


Process. (And my daughter’s process.)

And that’s “ghost research”: the drafts you write and revise and circulate to collaborators and revise and recirculate; the emails and meetings and Skype chats and phone calls you have with partners and colleagues; the careful bartering you do with impossibly patient finance-office staffers – all that work you invest in proposing a project that will only exist formally if funded.

Which means the research seafloor is piled high with ghosts.

Fishers hate losing gear. Their hearts are in their job (as is true for many of us, who may be so lucky), so a part of themselves always regrets the collateral damage to the animals and environment on which they depend: to not tend one’s traps is to be neglectful, is to abuse that which feeds you. And losing gear is f*£&!^g expensive. A trap lost is a trap that needs replacing – along with its weights and hardware and line and splicing and a buoy and a paint job and the time required to deal with all that. Literally a sunk cost.

A year spent pitching proposals into the funding abyss is also expensive, directly and indirectly. Writing an honest grant requires that you believe in it, and with that belief comes emotional investment. But to submit that grant requires emotional detachment because a rejection, whenever it comes, is never personal. (Which is precisely why the rejections hurt. Oh, if only they were personal!)

How far does this analogy go? Maybe it begins to twist. Maybe, unlike a ghost trap, a ghost project that “keeps fishing” is a good thing – maybe it resurfaces some months or years later, all tangled up with other ideas you’ve been thinking about; maybe it gets dragged up from the depths quite by surprise when you meet a new colleague or start working on a proposal that seemed unrelated. It is possible to flip unsuccessful proposals into articles, which I’ve tried (here, for example). Articles form the basis of a researcher’s track record, which in turn constitutes (as an abbreviated CV) the frontispiece of most grant proposals. Haul up an empty trap; re-bait it; set it again. And even funded grants are ephemeral – all traps eventually need replacing. But, undeniably, they do help pay the grant fisher’s costs (which have many forms), and the best bait for funding is evidence of previous funding. So there’s a limit to how much ghost research a person can generate. Balance sheets can’t run in the red indefinitely.


Cut&paste mock-up for a diagram that never was.

There’s one more turn to make here that may help explain why ghost research can be so frustrating, if you are silly enough to attempt its accounting. (See “emotional investment,” above.) The reason your CV is the only place such a reckoning can exist, should it exist, is that no one else is burning to know about it. Your colleagues, however empathetic they may be, certainly aren’t; those in supervisory roles probably have a sense of your dogged determination, but they’re pondering your funding situation approximately never. They’re busy people. Ghost research is frustrating because to tell everyone what you’re doing, or to list it in an overlong, belated year-end blog post, would be to behave in a particularly annoying way.

You know it when you’re on the receiving end. During finals week, for example, it’s the difference between those people studying in the library and those other people who’ve come to the library to complain, at volume, about how much they have to study. In formal terms, it’s the difference between instrumental and expressive behavior. Turns out that no one, not even among your family and dearest friends, really wants to know and understand your laundry list as well as you do, nor do they have the energy to help you process your anxieties about the implications of its myriad items (unless they are paid to do so in a professional capacity, and even then…). And this is especially true when your job, it turns out, involves always maintaining some version of the same impossibly long list – for decades. If it’s not one grant call one year, it’s another grant call the next. Et cetera, ad infinitum, ad absurdum.

Maybe, in academic research pursuits, sunk costs are just inherently high and unavoidable. Moreover, ghost research is intrinsic to research. It just is. Not everything you do is worth keeping or should be kept – ask your college poetry teacher, or ask Picasso (lines I’ve quoted here before): “When you begin a picture, you often make some pretty discoveries. You must be on guard against these. Destroy the thing, do it over several times. In each destroying of a beautiful discovery, the artist does not really suppress it, but rather transforms it, condenses it, makes it more substantial. What comes out in the end is the result of discarded finds. Otherwise, you become your own connoisseur. I sell myself nothing.”

Unfortunately, constructing a CV is kind of like being your own connoisseur in order to convince other people to pay good money for what you’ve got. (Picasso is not interested in helping you compile your promotion portfolio.)

But maybe in ghost research there’s some functional analogue to a “catch escape panel.” Maybe that’s a space, real or imagined, in which to keep beautiful discards to then find and transform and condense and resubstantiate them. No one else needs to know exactly how much ghost research you have drifting out there – but maybe it’s important that you know where to look if you had to find it.

Haul, re-bait, set again; haul…



Upped sticks – come find us at University of Southampton (@geogsouthampton)!

img_4378An update on lab doings – I’ve moved from the School of Earth & Ocean Sciences at Cardiff to the Geography & Environment Unit at the University of Southampton, effective 12 September 2016.

I’m grateful for my time in Cardiff, and I’m excited for new opportunities in Southampton.

For the next while, I’m back to living that recurring dream in which I know I have a meeting or an appointment somewhere, but I don’t remember what it’s about and I can’t find the room I’m supposed to be in.

“You must be new here.”

Plenty of developments to post about, coming soon – there’s a small raft of material (research and teaching) making its way into daylight…



A writing exercise: manually mapping one draft onto another

Scott Armstrong & I just tried this writing exercise:

Required materials include two different drafts of the same manuscript (here, let’s call one “early” and one “final”), several sheets of flip-chart paper, scissors, tape, and a handful of coloring markers. Make sure there are line numbers in the margins of both manuscripts, and print them single-sided.

Rotate the flip-chart paper to landscape orientation. Tape the final draft on the left side, one page per flip-chart sheet.

And now comes the fun (and time-consuming) part.

With the scissors, cut the early draft into individual sentences. Then begin reading through the final draft. Wherever material from the early version appears in the final one, tape the relevant strips from the early draft onto the right side of the chart paper. You might need to cut whole sentences into clauses or fragments. Matches don’t have to be verbatim. Use the markers to draw lines between words, phrases, or concepts connecting the two versions, and to otherwise annotate and clarify how the various elements on a given sheet relate to each other.

The result might look something like this:


A few things to look for and ponder – preferably with your coauthor(s) at a big conference table where you can stand back and see the sheets as a set:

  • do differences in respective line numbers appear to exhibit a pattern (e.g., high numbers in the early draft now correspond to low numbers in the final draft), or does the right side of each sheet look like it was attacked by a random number generator? either way, what does that convey about how one draft is organized relative to the other? how does the reconstituted early draft reveal structure(s) in the final?
  • how much of the early draft do you have left over? what do those remaining scraps contain? why isn’t that content in the final draft? (and you might be surprised at how much of the early draft is still in the final, especially if they read very differently…)
  • what role does each section of the document perform? why is the final draft arranged the way it is? (or, looking again, should you take the scissors to it, too?)

The added benefit of taking a pair of scissors to a draft you’ve labored over is the reminder – the tactile reminder – that the manuscript is an object, and is best regarded with objectivity. You can cut it up because it’s just a thing – it’s not an extension of your person, or of you as a person. It is a representation of your good idea; if it’s not representing your idea well, then the document needs to change until it does what it’s supposed to do.

(My first experience of an exercise like this was in a workshop on personal essays. We’d spent 72 hours pouring our hearts into 1500 words about our grandmothers, friends, a memory beloved or sensitive. We thought we were done. Then we were handed scissors and told to hack the whole thing into its constituent parts and experiment with different ways to tape it back together. Some people are more comfortable with that initial confrontation with objectivity than others.)

If you try it, give me a shout. I’d be interested to hear about how it went – what worked, what didn’t, what surprised you.



Syllabus: The Global Ocean (@globaloceanS16)


Malta Freeport, from Pretty Bay.

Another very brief shout –

Embedded here is an abridged copy of the syllabus for this year’s version of The Global Ocean, the gateway module for Cardiff’s first-year students in Marine Geography. As with my second-year module (Marine & Coastal Resource Systems), you can dip into the readings we’ll be doing from week to week.

And like @marcoresys2016, we’ll be maintaining a Twitter feed to aggregate news and stories, breaking and old, on a broad spectrum of ocean topics. Three different students will curate the feed each week. If you’re interested, please follow along – you can find us at @globaloceanS16. We’re already well underway.

You might notice that works by Prof Phil Steinberg have a significant presence throughout the semester. I’m a fan. (Check out the topics in his Research pages.) To this reader, his writing offers the most holistic perspective – and most comprehensive articulation – of what marine geography is as a concept and sub-discipline. “Navigating toward multiple horizons: Toward a geography of ocean-space” (The Professional Geographer, 1999), is a great place to start, but I recommend his monograph The Social Construction of the Ocean (Cambridge University Press, 2001).


Syllabus: Marine & Coastal Resource Systems (@marcoresys2016)

A very brief shout –

Embedded here is an abridged copy of the syllabus for this year’s Marine & Coastal Resource Systems module, part of the core curriculum for Cardiff’s second-year students in Marine Geography. You can check out the readings we’ll be doing. Ambitious? You bet. But fun.

Also, between now and May-ish (at least) we’ll be maintaining a Twitter feed to aggregate news and stories, breaking and old, on topics broadly related to marine and coastal resource systems. Three different students will curate the feed each week. If you’re interested, please follow along – you can find us at @marcoresys2016.


Hidden scale dependency in conserving working woodlands

A bogged-out skidder road after a major woodcut.

A bogged-out skidder road after a major woodcut.

I’ve been sitting on this post a long time, in part because I haven’t known how to write it up.

A couple of years ago, I wrote this manuscript and submitted it to two different peer-review journals (in series, not in parallel) as a sort of commentary piece. The first outlet rejected it with a brief but polite review. With the second, the editor offered to publish it, pending major corrections to incorporate changes a game reviewer had suggested. I thanked the ed for his interest, then withdrew the piece. Now I’m pinning it here, to this would-be bulletin board, and starting a new category, “unpublished”, that I hope not to use too often.

A hidden scale dependency in conserving working woodlands” is the title, and the Abstract is as follows:

Because plans for large-scale landscape preservation in the US do not rely exclusively on lands held in trust, conservation programs have a vested interest in forest stewardship by private landowners. Selective harvests for commercial sale are often highlighted as a financial incentive for owners of non-industrial “family forests” to sustainably maintain the working character of their acreage rather than subdivide it or convert it for development. However, the business costs inherent in even a small-scale commercial timber harvest typically mean that forest parcels smaller than approximately 80 acres are too small to support a financial return. Statistics for private forest ownership in the U.S. suggest this minimum scale makes commercial harvest incentives effectively inaccessible to more than 90% of forest owners. Rural landscape conservation and commercial timber harvests depend on the same economies of scale to be viable. Designs for regional-scale forest conservation need to account for non-industrial but nonetheless commercial economies of scale that set an inherent limit on financial incentives intended to foster stewardship activity among family-forest landowners.

Key subcategories of forest subdivision and ownership in the US, using data compiled from Butler (2008). Of all privately owned, non-industrial forest land, 11% is in parcels smaller than 50 acres (black wedge), and therefore below the minimum size typically necessary to support a commercial harvest. This acreage category accounts for 90% of all private forest land owners.

Key subcategories of forest subdivision and ownership in the US, using data compiled from Butler (2008). Of all privately owned, non-industrial forest land, 11% is in parcels smaller than 50 acres (black wedge), and therefore below the minimum size typically necessary to support a commercial harvest. This acreage category accounts for 90% of all private forest land owners.

I wrote this piece because the subject is something I care about. I have a vested interest in it. I grew up on an unusually productive (high-grade), mixed-species, 30-acre woodlot that’s been cut selectively three times in my life (in the interest of full disclosure, I missed the first cut by a year or two), overseen each time by the same forester (who co-authored the manuscript), and twice by the same logger (a one-man, one-skidder operation). Technically, my family is among the 90% of forest owners (Fig. 1) for whom commercial harvest should, by the numbers, be inaccessible.

I also care because I’ve seen a lot of woodlots flipped for development back home. Hence the question I still can’t answer: is there any land use of financial value to a land-owner – cash-on-the-nailhead value – that can trump the financial incentive of development? Scholarship on small-scale and “family” forestry suggests there are a lot of options. While I have the utmost respect for campaigns like Wildlands & Woodlands, I’m afraid I remain skeptical about the real alternatives available to the 90%. It’s one thing if your little plot sits on a watershed of value to a conservation interest. But what if you can’t sell to a trust, and can’t afford to set up a stand-alone conservation easement? What if the timber quality’s poor or middling at best, but you want to leave it wooded? Forestry literature is full of social surveys that expound the many virtues of woodlot ownership, highlighting all the rewards that aren’t financial – the pleasure of manual labor, of environmental stewardship however humble, of working outdoors. If I’d been polled, I’d have expounded them too. But the non-financial benefits of a small woodlot can quickly scatter in the face of immediate financial need. If you’re too small to harvest for commercial sale, you’re probably too small to retain as a conservation asset. But almost never are you too small to subdivide.

I pulled the submission because I couldn’t answer two primary questions posed in peer-review: Who cares if 90% of homeowners are left out of landscape conservation agendas? (If their plots are too small, isn’t it a case of what’s done is done?) And if we did care, what kind of policy instruments should be out there to do something about it? In fact, aren’t there already a number of conservation instruments for family forests? So what’s the matter with them?

This manuscript falls into a gap between two audiences. To conservationists, the subcategory of smallholdings actually amounts to not much – 11% of non-industrial US forest land. Speaking of economy of scale, imagine a campaign to organize such a huge constituency to conserve so little (in relative terms) and such diffuse physical space. The economy-of-scale argument is obvious – these properties, however many of them there are, are too small to save. Likewise, forestry disciplines aren’t interested because economy-of-scale in commercial forestry practices isn’t news.

So, for now, the audience is me, mourning the blighting of rural landscapes. And maybe you, if you’re still reading.

Economy of scale is surely a key fulcrum in the dynamical relationship between conservation and development (and, for that matter, between other competing uses for a given resource). I’ll embed it in future work, somewhere – perhaps in an agent-based model of land-use change (which I’ve attempted before), or similar (with some inspiration from ecology).

Please call me before you subdivide.


POSTSCRIPT (14 Jan 2015) – After I posted this, friend&colleague Evan Goldstein suggested I upload this manuscript to PeerJ PrePrints. So I did. It now has a (citable) DOI, and is out there in an Internet world that’s wider than this bloggy one. Check out the re-routed link above.

Hello & goodbye – a belated lab update

Regarding comings and goings in the Environmental Dynamics Lab, two important updates – which come with my apologies, because there’s “late-breaking news” and there’s news that’s just late.


Congratulations to Dr Paul Evans ’15.

First, huge congratulations to Paul Evans – now Dr Paul Evans – for graduating with his doctorate in July’s ceremonies. Paul & I started working together in 2012, not long after I arrived in Cardiff. Perhaps his capacities as a multi-sport athlete helped him endure many, many months of editorial comments and annoying marginalia from me.

Paul aboard Cardiff's R/V Guiding Light.

Paul aboard Cardiff’s R/V Guiding Light.

His PhD thesis, Hydrodynamic characteristics of macrotidal straits and implications for tidal-stream turbine deployment, was ultimately a win. He has one peer-reviewed paper already out, and another in the review pipeline. Over the past few months, he’s been postdoc-ing in the School of Engineering here at the U, but he’s soon off for a position at Intertek, an industry-facing energy/water consultancy based here in Cardiff.

Scott at the steep front of a washover lobe at Three Cliffs Bay.

Scott at the steep front of a washover lobe at Three Cliffs Bay.

Second, a huge welcome to Scott Armstrong, who joined the lab as a PhD student back in October. (Like I said – sometimes there’s news that’s just late.) Scott worked with me on his MESci thesis project coastal overwash dynamics, which we turned into a paper for Geology earlier this year.

Scott (et alia) give the new lab space a facelift.

Scott (et alia) give the new lab space a facelift.

For now, he’s turned his attention from morphodynamic experiments to human–coastal systems and the dynamics of developed coastlines. He’s already generating great stuff, and has a manuscript in motion – exciting things to come. With a little help from his colleagues, he also redecorated the new lab clubhouse (new to us, anyway) while I was on vacation.

Nice one.

%d bloggers like this: