Response to John Morse
Date: Thu, 3 Sep 1998 13:59:19 -0400 From: "WHITE, PETER S."To: jmorse@clemson.edu (John Morse) Cc: keith_langdon@nps.gov, pick@pick.uga.edu, wfharris@utk.edu, chuck_parker@nps.gov, msharkey@southey.ca.uky.edu, jmorse@clemson.edu Subject: Re: Fwd: Inventory Plan John: A short response is: Yes. The appropriate way to regulate sample effort will vary with the taxon. At the simplest level this would just be a standardized sampling strategy for particular groups (at least for that component of the sampling which is more quantitative) that would be worked out by the TWGs. Thus Fred Coile records the number of spiders and species he collects as a function of duration of collection. How many species does he see in 1 minute, 10 minutes, 1 hour? In your message you note that different creatures require different sampling strategies--numbers of light traps vs. distances traveled by the insects. So one sample might draw insects from 100 m radius; another from 1 km radius. You suggested that the former would require more light traps to cover the same amount of space. You'd have to space the light traps closer than 100 m to each other in the first case to get the available insects in the habitat. I agree. If we want to compare apples and apples, then let us assume that you want to collect the available insects within 1 km on the sample location. In the case of long-distance insects, you'd need just 1 light trap. In the case of short-distance insects you'd need a grid of light traps all 100 m apart. Both samples are aimed at the same grain size of land--1 km--but they use different grain size of sampling units (100 m vs. 1 km). I've picked this numbers arbitrarily, but if I understand correctly, I am with you every step of the way. TWGs and investiagors must determine these parameters. I want to point out another way of looking at grain--and this is the one I was trying to get across. Regardless of the field method--that is, regardless of the intensity of the light, the density of traps, the flight distances of the insects, or other parameters--the best way to understand the number of individuals seen and the number of species recorded is to document the dimensions of the sample itself. For example, even if you had the wrong light bulb on the wrong grid spacing, the numbers of individuals and species recorded from that light trap would be a function of the duration it was on each night, the number of nights it was on, etc... One could express the number of species as a function of the number of individuals in the sample (rarifaction is a method by which such distributions of species among individuals are resampled to produce a random species-individual curve) or as a function of the sampling effort. My main point was that the number of species recorded is a function of the dimensions of the sample window. Those dimensions include grain in space and duration in time, but also how many samples there are and how these are distributed across space or through time. If grain in space, duration in time, extent in space, and extent in time are all known, the species list becomes are repeatible sample. If each also has nested subsets, you can draw a curve that represents the gain of individuals and species as the grain in space, duration in time, extent in space, and extent in time, are varied. That curve has more information in it than the number of species at any particular fixed sampling window...but at the very least (that is, even if the curve is not drawn and the observations are not in nested sets), the sampling window should be documented so at least we know what the sampling effort was. I'll try again on this in person (and with a black board)...the idea is a simple and straightforward one, though challenging to describe. I believe most taxonomists and systematists intuitively understand the principles because they often adjust sample effort to the characteristics of organisms and their distributions. As they inventory, they gain an intuitive feel for how complete their list is (it gets harder to find a new species) and how thoroughly they've covered a habitat. If their objective is not a complete inventory and if they have limited time, they are the first to call their checklist "preliminary" or a "working checklist" or even to just document their specimens without claiming to complete a checklist. My perspective is to formalize this intuitive approach, at least for a portion of the effort or as a recommended but not mandatory guideline. =========================================================== Peter S. White email: pswhite@unc.edu Department of Biology -- CB# 3280 University of North Carolina at Chapel Hill Chapel Hill, NC 27599-3280 USA Biology Phone: 919-962-6939 Biology FAX: 919-962-1625 NCBG Phone: 919-962-0522 NCBG FAX: 919-962-3531 Home Phone: 919-967-4926 Web information: www.unc.edu/depts/biology/white.html ===========================================================
Discover Life in America | Science | Inventory Design | Peter White - 3 September, 1998 |