ヒョウシキカン オ アシ ニ ハメラレタ ヤチョウ ノ シュルイ ト コタイスウ ノ カンケイ
Relation Between the Number of Bird Species and the Number of Individuals in Ringing Work
d0030009n007.pdf 912 KB
Trials were made to fit Fisher's logarithmic series and Preston's truncated discrete lognormal distribution to bird ringing records obtained in Baltic coast. Poland by Busse and Gromadzki. Four series of species abundance data were tallied from the records by spring, autumn and both seasons 1967, and by the eight-year totals 1960-67. Every series of the data were well described by both mathematical models. The ringing records in each series were within the limits of the sampling error of the numbers expected according to both distributions, but the probabilities for X^2 were a little greater for the lognormal distribution than for the logarithmic series excepting the record obtained in autumn 1960, in which they are almost equal. If we add the X^2 values for the four series, ∑X^2 = 65.977, n = 82 for the logarithmic series (P ≒ 0.69) and ∑X^2 = 57.753, n = 78 for the discrete lognormal distribution (P ≒ 0.85). It could be said that the lognormal distribution provided a somewhat better description of the ringing records than the logarithmic series.
One of parameters characterizing the former model, R^^∧ values, which depend upon the intensity of the ringing work, were consistent over both spring and autumn seasons X^2 = 2.926, n = 1, P ≒ 0.087, and a weighted mean of R^^- = 0.9992±0.0001. A corresponding test for the agreement of the <α^^∧>'s, the index of diversity, showed a significant discrepancy, X^2 = 46.930, n = 1, P<0.001, indicating the different patterns in migration between two seasons.
For the lognormal model, the relevant statistics are the estimated mean μ^^∧ and the standard deviation σ^^∧ of the whole population. Comparing seasons the standard deviation proved a little more uniform X^2 = 0.101, n = 1, P ≒ 0.751, giving a combined estimate of <σ^^∧>_c = 1.2583 ± 0.1450, whereas for μ^^∧,X^2 = 0.412, n = 1, P ≒ 0.521, and their combined estimate of <μ^^∧>_c = 0.7356 ± 0.2251.
Based on the result of fitting the two mathematical models to the data, we could conclude that the bird ringing work were quite properly carried out in every seasons and every years. If we could have several season's or year's ringing records and if there was any discrepancy of parameters among season's or year's species abundance distributions, we might consider the discrepancy as the results of environmental disturbances on the land of starting of migration or on the way of migration. Natural forces, such as volcanic eruptions, earthquakes, fire and unusual climatic conditions and weathers, and human activities, such as the production of forest monocultures and agricultural land, urbanization, industrialization, wide use of pesticides, chemical contamination, pollution of various kinds are considered as the sources of environmental disturvances.
Shimane University, Faculty of Agriculture