About
  PDF
Full Text
(52 K)

Potential usefulness of Newly Available Statistical Methods for Modeling Meloidogyne incognita, Thielaviopsis basicola and Influential Soil Factors on Yield

Scott Monfort, Craig Rothrock, Andronikos Mauromoustakos, and Terry Kirkpatrick

ABSTRACT

Meloidogyne incognita and Thielaviopsis basicolapopulation densities and soil factors that influence their densities and distribution were evaluated in 2001 and 2002 in a commercial cotton field in southeastern Arkansas to evaluate their effects on yield. The 15-acre field was subdivided into 512 grid plots (30.5 meters x 3.9 meters) and sampled for M. incognita in April, May, July, and October. The April and October samples were also assayed for T. basicola. Soil fertility at planting and soil texture were also determined from each grid plot. The fumigant nematicide, 1,3-dichloropropene (Telone II) was applied in strips through the field at 14.2, 28.4, and 42.6 liters/ha. to create zones within the field with differential population densities of the nematode. Grid plots were aggregated and averaged based on four soil textural ranges (0-30 % sand, 31-45 % sand, 46-60 % sand, and > 60 % sand) and on the nematicide treated zones within each textural class. Statistical analyses were conducted in JMP SAS software utilizing three different methods: multiple regression, neural networks, and partitions. Data variables that were important in explaining cotton yield differences were percent sand, nematicide application, M. incognita (April 2002 density), and T. basicola (October 2001 density). Using these variables, all three models explained at least 89 % of the variability of the 2002 yield.





[Main TOC] | [TOC] | [TOC by Section] | [Search] | [Help]
Previous Page [Previous] [Next] Next Page

Document last modified 04/27/04