Friday, September 27, 2013

Parrotfish and sand production (part 2)

Stoplight Parrotfish (Sparisoma virde)
The sight is common on the reef: a parrotfish swims by and a thin stream of coral dust and sand particles is ejected from underneath.  Coral sand is made here in the belly of the beast.   

Here’s a picture I took last week in Bonaire.  You can see the stream of sand coming from below.  

(BTW - do you know how hard it was to get this shot?  I took a LOT of photos, trying to capture this magical moment.)  
Parrotfish caught in the act of creating more sand.

Our question is simple:  How much sand is created by an average parrotfish in the course of a year.  One set of sources says it’s 200 pounds, another says it’s 2000.  One of those can’t be right. 

This kind of discrepancy comes up all the time, and it’s the reason you don’t want to single-source anything, even simple facts like this.  (In an earlier SearchResearch episode wefound widely varying numbers for how far a specific bird flies.)  

So where do these numbers come from?

200 pounds / fish / year?  The source for most searcher’s “200 pounds” claim seems to be from the Wikipedia article on Parrotfish which points to the article by Thurman (see below for reference list).   The Wikipedia entry looks like this: 

Unfortunately, the link to the Thurman article specified in Wikipedia is dead (it was last checked in 2009) and the target site no longer supports this reference.  I spent some time searching for this chapter, and while I can find lots of references to it, I can’t find the book chapter itself. 

Even worse, many of the references to this chapter I found are clearly copies of the citation and link from Wikipedia.  (Even at such well-respected sites as the Encyclopedia of Life whose parrotfish page--—would be considered authoritative—merely quote the Thurman reference and give the same link to a dead page.) Sigh.  

So went to my local library’s website and inter-library loaned the darn thing.  Amazingly enough, I got the book after only 3 days!  And there, on page 310 is the assertion:  “By eating coral (and corraline red algae) a single parrot fish may produce up to 90 kg  (190 lbs) of sand per year.” 

Case closed?  Not yet.  Unfortunately, THEY don’t give any reference for that bit of data.  Where did THEY get that information from??   

2,000 pounds / fish / year?  The other data SearchResearchers found was the "1 ton / year" number.  The source for most searchers’ “2,000 pounds” seems to originate from the Ichthyology department’s page about parrotfish at the Florida Museum of Natural History (FMNH). 

On that page they say “Parrotfish may produce as much as one ton of coral sand per acre of reef each year.” 

But read that carefully:  This means “all of the parrotfish on the reef in 1 acre” are producing around 2,000 pounds of sand over the entire acre.  That’s the contribution by all of the parrotfish working (so to speak) together. 

Sadly, they don’t give a reference for their source material either.  And when I search for the exact phrase as seen on the FMNH site,

     [“Parrotfish may produce as much as one ton of coral sand per acre of reef each year”]

it’s clear that all of the references are using the FMNH page as their source.  (Typically, much of the text in the article is quoted directly, including that phrase. This kind of not-thought-out copying always makes me suspicious.  )

However, by being diligent and following various sources, I also found the 2,000 pound reference as coming from the book “Fishes of the Bahamas” by (Bšhlke & Chaplin, 1993).  The citations would look like this:  
Bšhlke, J.E., and C.G. Chaplin 1993. Fishes of the Bahamas and Adjacent Tropical Waters. University of Texas Press, Austin, 464-482. 
So I go to Google Books and find…  this citation is slightly wrong.  The first author’s surname is Böhlke, not Bšhlke.  (Some character encoding failed and people weren't careful about their cites.)

And, once again, this book falls into the “scanned on Google Books, but snippet-view only” (which doesn’t help).  It’s not scanned on or any of the other sources.  So…

… back to the library.  I ordered this book and got it along with the Thurman book.  Thing is, “Fishes of the Bahamas” is a giant tome:  771 pages of fish data.  Not much plot, but still pretty interesting to read (if you’re a diver). 

In this book I found the sentence:  “On a Bermuda study reef it was recently estimated that one ton per acre per year of this material [“calcareous material,” i.e., coral sand] passes through the intestinal tracts of reef fishes and is redeposited as fine sand, and that the scarids [ parrotfish ] are primarily responsible.” [pg 465]

And the reference given here is…   (finally, a real reference!)
Bardach, John E. "Transport of calcareous fragments by reef fishes." Science 133.3446 (1961): 98-99.

By going to Google Scholar and searching for that author + title I was able to get to the original article:

Now we’re getting somewhere.  In that paper (published originally in the well-respected journal, Science) the author writes:   It was found that browsing omnivorous reef fishes which rely, in part, on a plant diet ingested and redeposited at least 2300 kg of such material on a 1-hectare study reef near Bermuda.  

Let’s do some Google conversions to check this number. 


     [ 2000 pounds per acre in kg per ha ] 

means  2241.7 kg per ha (a hectare is 10,000 square meters) 


     [ 2300 kg per ha in pounds per acre ] 

means  2052.0 lbs per acre (an acre is pretty close to 1 football field).  

I wanted to get the full text of the paper, but that’s $20 for a “short-term use” (1-day) license.  As you can imagine, I’m not about to pay for that. 

But we still don’t know how much sand/fish/year that represents.  To figure this out we need to know How many parrotfish live in on an acre of reef?  

Hoping to figure out this sub-puzzle, I started doing searches like this, using the scientific name of the Stoplight parrotfish: 

     ["Sparisoma viride" sand OR sediment production ]

This is reasonably productive (esp. when done in Google Scholar) and leads to several papers on the topic.  The most cited paper is:
Bruggemann, J. H., et al. "Bioerosion and sediment ingestion by the Caribbean parrotfish Scarus vetula and Sparisoma viride: implications of fish size, feeding mode and habitat use." Marine ecology progress series. Oldendorf 134.1 (1996): 59-71.
I got a PDF of the paper by doing this search:

[ filetype:pdf  “Bioerosion and sediment ingestion by the Caribbean parrotfish Scarus vetula and Sparisoma viride" ]
That led me to the PDF file, which I printed out and carefully read. 

This is when I realized that this is a much more complex problem than just looking up the number of parrotfish / square meter on the reef. 

By reading this (and a few other papers) I learned that the density of parrotfish on the reef varies hugely by depth.  (Shallow water has a lot; water deeper than 50 feet has fewer.)  They also sleep at night, and feed only between 7AM and 6PM (and, btw, they mate daily, between 7AM and 9AM… but I digress).  I also found out that the quantity of coral ingested when they bite varies a good deal by the size of the fish, but it varies even MORE depending on the kind of parrotfish.

Among other things, we HAVE to be careful not to mix up our parrotfish species (e.g, Hawaiian parrotfish with Caribbean parrotfish, or even parrotfish that live side-by-side in the waters of Bonaire—as the paper points out “…the erosion rates per bite by Sparisoma viride is more than an order of magnitude higher than in Scarius vetula.”  That is, the Stoplight parrotfish (Sparisoma viride) has a bite size that’s ten times larger than that of Scarius vetula (Queen parrotfish). 

But we’re getting somewhere. 

If you read this paper carefully, you’ll eventually find Figure 5 “Scarus vetula and Sparisoma viride Erosion rates per bite (upper graphs) and per day (lower graphs) as a function of fish size.” 

Figure 5 from Bruggemann, et al. (Yellow highlighting mine)

If you look at the graph in the lower right corner, that’s a chart of “mean daily erosion rate (g ind -1 d -1)” – that is, “grams per individual fish per day.”  Although the “grams / day” vary greatly by fish size, the average is right around 300 grams / fish / day.   Basic math shows us:

300 grams * 365 days = 109,500 grams / year

Or, 109.5 kg / year (204 pounds), which is close enough (given all of our averaging) to our earlier estimate of 95 kg / year.  (And, incidentally, nothing close to 2,000 pounds / fish / year.) 

If this is correct, then to get 2000 lbs of sand contributed per acre by all of the parrotfish, we’re going to need around 10 Stoplight parrotfish / acre.  By my direct observation on the reefs of Bonaire, that seems low, but let’s see if we can’t find some numbers in the literature. 

Another paper I found from my previous query is

Choat, J. Howard, et al. "An age-based demographic analysis of the Caribbean stoplight parrotfish Sparisoma viride." Marine Ecology Progress Series 246 (2003): 265-277.

He “conservatively estimates” 4 individuals / 300 square meter plot.

     [ 1 acre in square meters ]

is 4046.86 square meters in an acre.  If we divide 4046.86 / 300, we should get an estimate of the number of Stoplight parrotfish / acre (according to Choat).  This number is 13.5, a little higher than we estimated before, but not a crazy number. 

Search Lessons:  First, this one took me a long time.  I’m guessing I spent about 10 hours researching this, and I learned a huge amount along the way.  (Not just about the sex life of parrotfishes, but also about why it’s so difficult to estimate the total number of fish on the reef.) 

But I figured out that 1 Stoplight parrotfish generates around 0.6 pounds of sand / day (on average, mid-depth of the reef). 

I did this by going back to the original papers (using Scholar and Google Books) to dig out the charts and original data. 

In some cases, it’s hard to get to the original source document.  As you saw, I had to use my local library to get to the original paper by Thurman.  But once I had it, the rest of the story unfolded easily.  (Although slowly—I had to look up a lot of words.  Acanthurids and scarids and fork length were all new terms for me). 

This is a nice example of sensemaking: Pulling together information from multiple resources and cross-checking them against each other. 

In truth, I read much more than I’ve reported on here.  I didn’t want to tell you that I actually read about 10 papers, some of which were duds.  They seemed good from the abstract, but the reality was that the paper encoded the data in an unhelpful way or that it was about parrotfish on some other continent, or the wrong kind of fish. 

AND... a beautiful 2 minute video of parrotfish feeding from the "Blue Planet" BBC series (discovered by regular reader Ramón... thanks!).   Note the part at  56 seconds in--a lovely shot of sand being delivered to the reef.

Reference List

Bruggerman JH, van Kessel AM, van Rooij JM, Breeman AM (1996) Bioerosion and
sediment ingestion by the Caribbean parrotfish Scarus vetula and Sparisoma viridae: implications of fish size, feeding mode and habitat use. Mar Ecol Prog Ser 134:59-71.

Choat, J. Howard, et al. "An age-based demographic analysis of the Caribbean stoplight parrotfish Sparisoma viride." Marine Ecology Progress Series 246 (2003): 265-277.

Gygi RA (1975) Sparisoma viridae(Bonnaterre), the stoplight parrotfish, a major
sediment producer on coral reefs of Bermuda. Ecol Geol Helv 68:327-359

Thurman, H.V; Webber, H.H. (1984). "Chapter 12, Benthos on the Continental Shelf". Marine Biology. Charles E. Merrill Publishing. pp. 303–313. 

van Rooij, Jules M., et al. "Resource and habitat sharing by the stoplightparrotfish, Sparisoma viride, a Caribbean reef herbivore." Environmental Biology of Fishes 47.1 (1996): 81-91.


For comparison purposes of parrotfish feeding and sediment production on the Great Barrier Reef:

Bellwood D.R. (1995a) Direct estimate of bioerosion by two parrotfish species,
Chlorurus gibbus and C. sordidus, on the Great Barrier Reef, Australia. Mar Biol 121:419-429
Bellwood D.R. (1995b) Carbonate transport and within reef patterns of bioerosion and
sediment release in parrotfishes (family Scaridae) on the Great Barrier Reef. Mar Ecol Prog Ser 117:127-136.
Bellwood D.R. (1996) Production and reworking of sediment by parrtofishes (family

Scaridae) on the Great Barrier Reef, Australia. Mar Biol 125:795-800.


  1. Dr Russell, Parrot fishes and your study are wonderful, and your photos of them are quite impressive. When I did the challenge saw a lot of images and none of them is like yours and specially like the one with Parrot fish eating coral.

    It is fascinating what these beautiful fishes do and the way you did the research. I don´t know if it is possible but you should add this study to Google Scholar. I'm sure many will find it just as amazing and for those in that field that could leat to make it even bigger.

    Thanks for sharing with us, Dr. Russell.

  2. I appreciate the process. This is a lesson for sure in 'sensemaking'. This depth of search is overwhelming. I would get just so far before I would start sending some emails or making phone calls to experts in the field if I could find them. What I really learned is allowing the information to flow until it makes sense and is deemed credible.

    Another observation is that we as searchers can help future searchers by correcting crowd source pages like Wikipedia. I tried to do that once but I didn't know what I was doing. I will based on your comments endeavor to learn how and to do my part.

    And my final comment is I really appreciated how you made your answer flow. Being able to read through a rather lengthy document could have been choppy and hard to follow. I tend to read through comments once then go back and check out links/references. So again I appreciated the process you have shown us.

    1. Thanks, Rosemary. As I've mentioned, I take a lot of notes as I do research, then convert them into something like this.

      The *reality* of those notes is that they're pretty long. I was going to write the long, unexpurgated version of this search (including all of the side-tracks and things I read, charts I drew, other avenues pursued).. but it got very long and difficult to follow.

      I'm really glad this makes sense. The truth is this is the coherent half of my research. The hardest part is knowing when something is NOT working out, and giving up on it. I'll try to write about that one day...

      WRT fixing up Wikipedia.. YES! I think it's part of our obligation as careful researchers to go and fix up errors as we discover them.

      In this case, I'm postponing fixing up those articles until this weekend (so my readers can go see the errors for themselves).

      But you should definitely learn how to fix Wikipedia. It's good for you immortal soul.

  3. Dr. Russell - that raised the search/sensemaking bar considerably - the methodology was very instructional as Rosemary stated -
    it was an epistemological tour de force

    Ran across this and thought it might be enjoyable, interesting details -
    Jerry Ligon/Bonaire
    best watched on Vimeo
    other Arcturus Productions from Bonaire.

    Ramón and Rosemary are becoming search/sensemaking beasts - hope that translates as a compliment - mis compañeros de búsqueda (my fellow seekers)

    I feel like a piker - or muskier, in comparison, eh.

    These are mostly for Ramón, but others may be interested - they point toward a direction that search is taking and what Google is developing, especially in terms of spoken search (as opposed to text input) - I would assume that translation would eventually be rolled into the spoken query/response…?
    The divergence of spoken ui serp results from text input is very interesting - moving closer to the Gene Roddenberry/Star Trek model of "Computer…"
    or in this specific case: "OK Glass"… (thanks Amanda)

    SEL Hummingbird
    BI write up Hummingbird
    conversational search/chrome
    Google Now/predictive search

    fwiw: I emailed Dr. Choat at James Cook University/School of Marine and Tropical Biology in Townsville - thought he might be interested in the questions you had posed (and you had cited his work in [part 2]) and hoped he might respond - will keep you posted if that happens…
    he has been busy - saw a post on a Red Sea dive in April:
    Al-Bahr Al-Ahmar
    and a small Parrotfish tidbit on the sleeping sac - sleepy time

    Google is wrestling with some interesting notions of knowledge - how to know what to know, how deep to know, how to modulate and update knowledge,
    how to interface and apply and even anticipate what knowledge to access… all very heady stuff. Am currently leaning toward the phantasmagorical -
    hoping that the mystery doesn't get wrung out with the successful query…

    btw, looking for Choat brought this - something the Googleplex might consider before embarking on the new campus construction?
    I'll leave it to you to search out the possible value…
    JCU/Smoking Ceremony

    1. Thanks Remmij for your links, I'll read it to learn more.

    2. meant to include this from current Smithsonian; Errol Morris muses about the "unknown unknowns" and how that notion relates to one of the great search questions from the last century… a good read that pertains to the nature and elusiveness of knowledge, information and what can be known… ichthyological Stoplight (& other varieties of Pfish) sand production included (and why Google isn't Wolfram|Alpha - but that is casting a larger net.) may also fall in that category.
      frame 313

      Ron Rosenbaum:
      "Is it something specific to the JFK case or is it true more generally about knowability? Morris’ new film, after all, about Donald Rumsfeld—famous for his line about “unknown unknowns”—is called The Unknown Known."

      “My favorite quote in my favorite movie review,” he says, “appeared in a 1941 review of Citizen Kane by [the great Argentine fabulist Jorge Luis] Borges and it contained Borges quoting, as he often did, [British writer G.K.] Chesterton, who said there is nothing more frightening than a labyrinth with no center. The loneliest of all labyrinths. The labyrinth to nowhere. Where there is no Minotaur at the center, there’s just winding corridors, more twists and more turns.”

      “Here’s my problem,” Morris replies. “My article of faith is that there’s a real world out there in which things happen. The real world is not indeterminate. I don’t want to hear people misinterpreting the Heisenberg Uncertainty Principle. Something happened. The problem is not about the nature of reality. We know somebody killed Kennedy and there’s an answer to the question of who and why.

      “Another thing we know is that we may never learn. And we can never know that we can never learn it. We can never know that we can’t know something. This is the detective’s nightmare. It’s the ultimate detective’s nightmare.”

      back to Stanford, Heisenberg

    3. Parrotfish sleeping sac link - was dropped in translation, apologies:
      gnathiids net