Wyniki wyszukiwania - BazTech

Ograniczanie wyników

Znaleziono wyników: 2

Liczba wyników na stronie

Wyniki wyszukiwania

Wyszukiwano:
w słowach kluczowych: species-area relation

Sortuj według:

Ogranicz wyniki do:

Species area relations generated by theoretical relative abundance distributions : parameter values, model fit and relation to species saturation studies

Ulrich W.

Polish Journal of Ecology

2001

Vol. 49, nr 1

67-77

Using model assemblages and random samplings the relations between 8 relative abundance distributions (RADs)(broken stick,log-series, power fraction, random fraction, Sugihara fraction, and two types of Zipf-Mandelbrot models) and resulting species-area relationship (SPARs) were studied. It is shown that the model fit of the power function and the exponential SPAR model depends mainly on the number of species per unit of area, the fraction of singletons in the sample, and the total species number in the assemblage. Sugihara and power fraction RADs did not necessarily led to power function SPARs but are characterized byrelatively high slope values in comparison to other distributions. Random placement and sampling of individuals of Zipf-Mandelbrot and log-series distributions resulted in curvilinear local vs. regional plots and the slope value z of the power function SPAR was not necessarily constant but could be forced to become constant by introducing a correction factor into the power function SPAR. The implications of these findings for detecting local species saturation are discussed.

On the scale dependence of evenness

Ulrich W.

Polish Journal of Ecology

2001

Vol. 49, nr 1

91-94

A recently reported (Wilson et al. 1999) effect of spatial scale on evenness is studied and it is shown that such a pattern is not necessarily an effect of changes in community structure at different scales but may simply result as a byproduct from constrains introduced by maximum and minimum allowed densities due to the sampling procedure. Evenness is found to be constant only if the species area relationship of the community under study has exactly the parameter values that are given by the parameter values of the relative abundance distribution of the community. Because such a situation will seldom occur under natural circumstances scale dependence of the evenness (and of related descriptors of structure) is expected to be a general feature.