This research project was conducted by Vasilis Liakos, a Ph.D. student at the Panepistimio Thessalias in Greece during his internship at the Univeristy of Georgia between April and December, 2012.
Cotton production is very important in United States and all over the world. USA is the second larger producer of cotton second to China. Cotton is very sensitive to management practices. For this reason, farmers are very careful with the application of the inputs as they have to apply them at the correct time and at the correct rates. Within-field variability is high even in small fields while an uneven distribution of plant biomass in fields is present in most fields. The aim of Precision Agriculture (PA) practices is to manage the field variability. This can be achieved with the division of the fields into small sections called management zones (MZs). Using MZs, farmers can use different treatments to meet the real needs of the plants.
A uniform application of fertilizers is inefficient as it does not consider field variability. During the last decade, variable rate application (VRA) has become increasingly popular. At both study fields, the experimental design consisted of two variable rate application strategies (VRA1 and VRA2) and a control. The fields were divided into nine 6-row strips allowing 3 replicates of the VRA treatments and 3 replicates of the control. The design was not randomized. Rather the strips were assigned to a repeating pattern (VRA1, Control, VRA2) to allow for direct comparison of adjacent VRA and Control treatments. This was done primarily because the fields were terraced and not all the strips were immediately adjacent to each other and we wished to avoid excessively complicating the comparisons with distance-induced variability.
The two VRA treatments (VRA1 and VRA2) used in this study differed primarily in the amount of N applied to the areas of the field with the highest vigor (high NDVI). The results show that applying more N to the high vigor areas (VRA2) resulted in higher overall yields and higher levels of profitability compared to the control. The VRA1 treatment resulted in lower overall yields, much lower use of N, but the same level of profitability as VRA2. The same pattern was observed during 2011 in a field immediately adjacent to Field 1. Profitability is not always associated with the highest yield and cotton producers should focus on this factor more than on yield.