The Eastern Bering Sea Shelf The Eastern Bering Sea Shelf : oceanography and resources / edited by Donald W. Hood and John A. Calder easternberingsea00hood Year: 1981 Intertidal scouring by sea ice 1115 Pielou (1975) argues that when the total number of species present is being estimated by the data and the community is small, as in the present study, one cannot statistically test the fit of the observed data to the theoretical distribution. However, we can test the difference between the empirical distributions with tests of the Smirnov type (Pielou 1975, Conover 1971). Visual comparison of
The Eastern Bering Sea Shelf The Eastern Bering Sea Shelf : oceanography and resources / edited by Donald W. Hood and John A. Calder easternberingsea00hood Year: 1981 Intertidal scouring by sea ice 1115 Pielou (1975) argues that when the total number of species present is being estimated by the data and the community is small, as in the present study, one cannot statistically test the fit of the observed data to the theoretical distribution. However, we can test the difference between the empirical distributions with tests of the Smirnov type (Pielou 1975, Conover 1971). Visual comparison of the species-importance curves for Amak and Akun islands and the Pribilof Islands showed that biomass of mollusks was most evenly distributed among the species at Akun Island (Fig. 64-4). The curve for Akun Island appeared closest to the lognormal distribution. The species-biomass distributions become less and less uniform (, toward a greater concentration of biomass among a few dominant species) at Amak, Otter, and St. George islands, in that order (). Eighty percent of the wet weight was concentrated among one or two dominant species at Amak, Otter, and St. George islands, whereas five species shared 80 percent of the wet weight at Akun Island. The curve for St. George Island most closely approached the geometric series distribution. Using the Smirnov test, I tested only the species- importance curves for St. George Island and for Akun Island, because the Smirnov test and others like it which can detect differences in the form of empirical distribution functions are valid for situations involv- ing more than two samples only if sample sizes are equal (Conover 1971). Because n (sample size: the number of points which determine the curve) for the species-abundance curve is the number of species in the collection and not the number of samples col- lected, sample size cannot be controlled. Values of n were unequal among the four sites. The Smirnov test of the empirical
Size: 1787px × 1119px
Photo credit: © Bookworm / Alamy / Afripics
License: Licensed
Model Released: No
Keywords: archive, book, drawing, historical, history, illustration, image, page, picture, print, reference, vintage
