Simulation of design-unbiased point-to-particle sampling compared to alternatives on plantation rows.
Abstract
This study compares the design-unbiased point-to-particle sampling on plantation rows to five alternative methods using computer simulation. The results indicate that the G-MR and the NG-RM had the best root mean square error percent (RMSE%).On the G row, the G-RM and the NG-MR were biased significantly. The fixed-length plot is design-unbiased and so did not show bias in the simulations. It performed well in RMSE% but not as well as some of the other methods. Ducey's method showed slightly higher RMSE% than some of the alternative methods but is design-unbiased and so unbiased for any spatial distribution including the G and NG simulation rows used here.