Premium
Populations of Sitobion avenae and Aphidius ervi on spring wheat in the northwestern United States Spatial distribution and sequential sampling plans based on numerical and binomial counts
Author(s) -
Feng M. C.,
Nowierski R. M.,
Zeng Z.
Publication year - 1993
Publication title -
entomologia experimentalis et applicata
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.765
H-Index - 83
eISSN - 1570-7458
pISSN - 0013-8703
DOI - 10.1111/j.1570-7458.1993.tb01658.x
Subject(s) - sitobion avenae , aphididae , biology , aphid , homoptera , tiller (botany) , braconidae , parasitoid , negative binomial distribution , linear regression , agronomy , statistics , botany , pest analysis , mathematics , hymenoptera , poisson distribution
During a three‐year field survey of the English grain aphid, Sitobion avenae (F.) (Homoptera: Aphididae) and the parasitoid, Aphidius ervi Haliday (Hymenoptera: Aphidiidae; represented by aphid mummies) on spring wheat, data sets were generated which consisted of 47 estimates of mean density ( m aphids per tiller), variance ( s 2 ), and the proportion of empty tillers ( P 0 ) for S. avenae and 22 estimates of the same statistics for A. ervi , respectively. Each estimate of the aphid and parasitoid populations was based on counts per individual tiller on each sampling occasion. Taylor's power law was found to fit the data sets better than Iwao's m * ‐ m regression model. Taylor's slope for S. avenae (1.3076) was significantly higher than that for A. ervi (1.1519), indicating that S. avenae was more aggregated than A. ervi. Based on temporal changes in spatial aggregation represented by an index, 1/ k [where k was estimated as m 2 /( s 2 ‐ m)], S. avenae was most aggregated at low densities during the early infestation period and tended to be less aggregated as density increased, whereas A. ervi was more likely to be randomly distributed. A common k was detected for neither S. avenae nor A. ervi because the slopes from the linear regression of k on m significantly exceeded 0. Sequential sampling plans, based on numerical (complete) and binomial (presence or absence) counts, for the aphid and parasitoid species were developed using Taylor's parameters and those estimated from the linear regression of ln( m ) on ln[‐ ln[ P 0 )], respectively. Suggestions are made concerning the use of the sampling plans and the levels of precision that may be attained.