Dr. Rob Dillon, Coordinator

Monday, March 19, 2012

Toward the Scientific Ranking of Conservation Status – Part III

Thank you all for your comments on my posts of December and January, regarding my new method for assigning conservation status ranks (1). Almost all of you addressed your emails to me personally, rather than posting comments to the blog. But I do think some of the points you raised are of sufficiently general interest to warrant sharing.

Many of you commented regarding the obvious sampling biases in the relative abundances of the freshwater gastropods listed on Table 1 in my January post (2). Here’s a cute example:
Maybe "commonness" is actually "commonly seen in the field without a microscope-ness." Large things (elephants, Oprah, etc.) tend to be noticed more than tiny things (hydrobiids, lawyer's souls, etc.)
Yes, I agree that the conservation ranks I suggested back in January do indeed reflect sampling biases against small-bodied species, and against species that are difficult to identify, and against species unusual or cryptic in their habitat.

My biological intuition suggests to me, for example, that the most common freshwater gastropod species in southern Atlantic drainages may actually be either Physa acuta or Ferrissia fragilis. The Physa abundances in Table 1 are certainly biased downward by taxonomic difficulties – several species (all common) cannot be distinguished as juveniles, and hence no Physa of any species can sometimes be tallied in samples where physids most certainly do occur. The abundance of the large-bodied Campeloma decisum was probably biased upward in the casually-collected samples we obtained from museums, and the small-bodied Ferrissia fragilis biased downward in the quantitative (or semi-quantitative) macrobenthic data we obtained from natural resource agencies. Limpets are rarely recovered from kick-samples.

But such routine sampling biases, irritating though they may be, are nevertheless random with respect to the object for which these data were tabulated. There is no reason to suspect that species warranting conservation concern are different in their body size, habitat choice, or taxonomic nuisance than more common species.

Much worse, from the standpoint of our purposes here, must be conservation-biased oversampling, the sampling error in favor of rare species generated as a consequence of the misbegotten system under which we currently labor. For today we first identify our putatively endangered species by pseudoscience, and then secondarily fund directed surveys to hunt that anointed subset specifically.

In the early 1980s, for example, Hugh Porter was funded by a North Carolina Wildlife Resources Commission grant to survey Lake Waccamaw (3). When I visited the NC Museum of Natural Sciences in 2005, there were so many lots of the (apparently endemic) Lake Waccamaw Floridobia (cataloged as "Cincinnatia sp."), differing only by transect number, that I simply closed the drawers and moved on (4). The (just 4) records of Waccamaw Floridobia shown in January Table 1 result from my own (arbitrary, but admittedly subjective) culling efforts.

Footnote #5 of my January post bears further attention in this regard. To compile my Table 1, I footnoted “We have screened any date-duplicates from the databases we have obtained from secondary sources, including museums and state natural resource agencies, as well as any nearly-neighboring collections, such as those taken upstream and downstream of single bridges.” This exercise was much more complex than my footnote made it sound, but was rendered necessary by overly-intensive sampling, often brought on by narrowly-directed surveys like Porter’s.

The situation regarding the endemic hydrobiid fauna of Georgia is similar. Fred Thompson has surveyed this fauna exhaustively, lodging in the Florida Museum of Natural History (for example) 49 records of Notogillia sathon and 29 records of Spilochlamys turgida from five small counties in central Georgia (5). Our colleague Charles Watson was also awarded a USFWS grant to survey this same fauna in 1995 (6), lodging 5 records of Notogillia and 4 records of Spilochlamys in the NCSM . These records I have boiled down to the 22 and 15 shown in Table 1, respectively, almost certainly every accessible spot where every Notogillia and Spilochlamys currently in existence can be sampled.

Meanwhile, January Table 1 showed just 16 records of Fontigens nickliniana, an ecologically-similar hydrobiid common in every hardwater spring in western Virginia (ranging all the way to Michigan, if you can believe it, Note 7) and hence attracting no interest from funding agencies or conservationists.

Io fluvialis must (literally) be the poster-child for conservation-biased oversampling (witness crawling to the left in the FMCS logo above). The snails are as spectacular on the hoof as any freshwater gastropod worldwide, but populations are currently restricted to the Clinch, Powell, Holston and Nolichucky Rivers in western Virginia and East Tennessee. The raw database I received in 2005 from Brian Watson, my colleague at the Virginia Fish & Game, included 128 (mostly historical) records of Io from Virginia alone, to which 54 records were added from directed surveys. I culled these records as best I could, but reference to the map at left (detail – click for larger) suggests that the 42 records ultimately remaining in our database continue to represent a gross over-estimate of the abundance of Io relative to the other 37 freshwater gastropods inhabiting East Tennessee River drainages.

The FWGNA survey we released in August found Pleurocera clavaeformis to be the most common freshwater gastropod in East Tennessee drainages, represented by 289 records as against our 42 records of Io. But my simple search of the Global Biodiversity Information Facility this morning returned 456 “occurrences” of Io fluvialis in the museums of the world, and just 204 occurrences of Pleurocera (“Elimia” or “Goniobasis”) clavaeformis (8).

So here’s my bottom line for the month of March. Not only is the method I proposed back in January scientific, it will be more effective in ranking species for conservation priority than the current system as developed by NatureServe. What is needed, now more than ever, is a continental survey of our freshwater gastropod fauna, conducted in a manner that is objective with respect to conservation status. Welcome to the FWGNA project… 14 years old, and still toddling forward (9).


(1) Toward the Scientific Ranking of Conservation Status:
Part I - [12Dec11]
Part II – [9Jan12]

(2) Table 1 - The 57 species of freshwater gastropods inhabiting the southern Atlantic drainages of the United States, ranked by their abundances in the FWGNA database 1/2012. [PDF]

(3) For references to Hugh Porter’s research see my post:
Crisis at Lake Waccamaw [16July10]

(4) My search of the online database at the NC Museum of Natural Sciences this morning only returned nine records of “Cincinnatia sp.” from Lake Waccamaw. That’s just a small fraction of their actual holdings, if my 2005 notes are correct. [NCSM]

(5) The Florida Museum of Natural History online database can be accessed here [FLMNH].

(6) Watson, C. (2000) Results of a survey for selected species of Hydrobiidae (Gastropoda) in Georgia and Florida. In Freshwater Mollusk Symposia Proceedings, Part II, eds. Tankersley, Warmolts, Watters, Armitage, Johnson & Butler, pp. 233 - 244. Columbus: Ohio Biological Survey.

(7) Hershler, R., Holsinger, J. & Hubricht, L. (1990) A revision of the North American freshwater snail genus Fontigens (Prosobranchia: Hydrobiidae). Smithsonian Contributions to Zoology, 509, 1-49.

(8) For more about the GBIF see:Freshwater Gastropod Databases Go Global! [26May09]

(9)  I initially made Parts I and II of this series available as a pdf separate back in January.  That document has been updated to a "version 19Mar2012" here [pdf].