Investigation of the optimum plot size for accurate yield evaluations in agriculture by Dr. Marcus Jones

Su Mo Tu We Th Fr Sa
26 27 28 29 30 31 1
2 3 4 5 6 7 8
9 10 11 12 13 14 15
16 17 18 19 20 21 22
23 24 25 26 27 28 29
30 1 2 3 4 5 6
Date/Time:Thursday, 20 Sep 2018 from 4:00 pm to 5:00 pm
Location:2050 Agronomy Hall
Channel:Agronomy Department
Actions:Download iCal/vCal | Email Reminder
Field trials investigating yield benefits of agronomic inputs utilize randomization, replication, and blocking as fundamentals of statistical design. Plot size is an overlooked design fundamental that could influence the outcome of trials, as fields possess a number of characteristics that tend to be nonhomogeneous and unpredictable.

A series of uniformity trials were conducted across the upper Midwest during the 2015-2016 cropping seasons to better understand the effect of plot length, width, shape, and size on data variability measured as maize yield. An area roughly 0.3 hectare was bulk planted with a single maize hybrid and divided into 144 experimental units, 4.6 m in length with 0.76 m spacing. The middle two rows were harvested and yield calculated. Statistical variance was calculated for 31 different plot combinations ranging in size from 27.9 m2 to 1,003 m2. The results demonstrate that variance decreases as plot size increases. Variance was reduced over 70% with plots larger than 83 m2 using a 27.9 m2size plot as comparison. When plot size was held constant, the shape of the plot did not consistently lower variance. The data indicate a tradeoff between plot size and replicates and fewer replicates are needed as plot size increases. While increasing replicates can reduce variance, the benefit decreases with each additional replicate and fails to address edge and alley effects. Plot size can be used to manage field variability and should be considered along with replication and
randomization when planning a yield trial.