Ácaros na cultura de soja: genótipos, danos e tamanho de amostra

• Ano de defesa: 2014
• Autor(a) principal: Fiorin, Rubens Alex
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Santa Maria; BR; Agronomia; UFSM; Programa de Pós-Graduação em Agronomia
• Palavras-chave em Português: Precisão experimental; Soja; Reamostragem; Danos; Dimensionamento de amostra; Experimental precision; Soybean; Resampling; Damages; Sample size dimensioning; CNPQ::CIENCIAS AGRARIAS::AGRONOMIA
• Link de acesso: http://repositorio.ufsm.br/handle/1/3239

The study aimed to evaluate the influence of soybeans genotypes on spider mites populations, quantify the occurring damage from spider mite attack and determinate leaflet number collected from different genotypes to estimate the spider mite population. Two studies were carried, in São Sepé (20 genotypes) and in Santa Maria (25 genotypes). The experiments were carried in randomized block design with four replications in 4,5 and 5,0 x 25 m experimental units. Weekly samplings were carried collecting 25 leaflets from the medium stratum and 25 leaflets from the superior soybean plant stratum in each genotype and evaluated an area of 20 cm2 of each leaflet. To determinate the sample size was used the data from evaluations which at least one genotype presented average population superior to one spider mite cm-2. To estimate spider mite number was considered the number of immature + adults spider mites, averages were compared with t bootstrap test. Sample size was estimated for an amplitude of 2 and 4 spider mites 20cm-2 and the optimal sample size calculus. To quantify spider mite damage in each genotype was maintained infested plots and not infested plots by pulverizations of acaricide. The predominant specie was Mononychellus planki. Population of spider mites vary in different genotypes and concentrates on the plant superior stratum. The necessary sample size is crescent as population grows, at the beginning of the infestation, 50 leaflets are enough with CIA95% (confidence interval amplitude with 1-p=0,95) maximum equal to 2 spider mites 20cm-2. To quantify higher populations 150 leaflets is necessary with CIA95% maximum equal to 4 spider mites 20cm-2. Yield variation as response to spider mite populations attack depend on the studied genotype and to all genotypes there is difference between the infested and not-infested plots. Average damage on Santa Maria experiment was 493 kg ha-1 and São Sepé 427 kg ha-1 and average gain of 33,4%.