Quantification of the effect of host patch configuration on the abundance of Bemisia tabaci in central Argentina: a multimodel inference approach.

ABSTRACT

Bemisia tabaci is a complex of species, which is considered the most common and important pest of a wide range of crops belonging to many different botanical families. In Argentina, this species is recognized as a vector of geminiviruses, and Middle East-Asia Minor 1, Mediterranean, New World and New World 2 have been found to coexist in the same area. Landscape elements, like habitat patch area and isolation, define the habitat configuration and have a direct effect on insect populations between and within host patches. In this paper, we analyse the effect of potato patch configuration on the distribution and abundance of B. tabaci. Potato patches were identified using Landsat TM5 and TM7 images, and a supervised classification was performed to quantify the spatial distribution of the patches in the whole study area. Potato patch metrics were estimated using Fragstats 4.4. Generalized linear mixed models were employed to analyse the relationship between whiteflies and landscape configuration, through a multimodel inference approach, finding that B. tabaci abundance and landscape metrics were very variable. After a multimodel selection process, we found that perimeter-to-area ratio and Euclidean distance between patches were the variables that best explained whitefly abundance in potato patches. Implications of these findings are discussed.