Methodological studies on the precision of variety trials with sugar beet – effect of sample size and head rows

Abstract

In Germany, the guidelines for variety trials with sugar beet require plots with 80-100 beets for a representative estimation of root yield. If possible, one or two head rows shall be planted perpendicular to the plots to avoid border effects at open alleys. The optimum sample size was determined in the 1970s and the effect of head rows has not yet been systematically studied. The aim of the present study was to quantify the effects of sample size and head rows on the precision of yield and quality data of sugar beet. Two series of field trials were carried out in Germany in 2016-2017. In the series "sample size", conducted at 9 environments, samples of 30, 60, 90 and 120 beets were compared. With increasing number of beets, coefficients of variation and least significant differences for yield and quality parameters decreased, especially when the sample size was increased from 30 to 60 beets. Compared to the current standard of 90 beets, the precision of samples with 60 beets was similar for sugar content but lower for root yield. It is concluded that the current sample size should not be reduced. In the series "head rows", ten variety trials were conducted without and with head rows. The precision of trials with and without head rows did not differ in general. The effect of head rows on least significant differences for root yield and sugar content was not consistent among environments. With head rows, absolute root yield (tonnes per ha) was lower and sugar content was higher than without, but the rank of varieties in relative white sugar yield remained unchanged. Sugar beet varieties can thus be tested in trials with and without head rows without losing precision, even within the same trial series.