Corn yield variability on the Des Moines Lobe of Iowa: Assessment of extent and soil-related causes

摘要

Precision agriculture techniques are an essential component to modern row crop agriculture in Iowa and can be used to create crop yield variability maps via geographic information systems. The first objective of this thesis was to explore the methodology that could be used to locate significant long-term corn yield variability on the Iowa Des Moines Lobe. A 158-ha site, consisting mainly of Clarion, Nicollet and Webster soil map units and containing multiple years of geo-referenced corn (Zea mays) yield data was selected. A cluster analysis tool was performed to locate spatially consistent corn yield variability (high, low and mean yield clusters). The mean yield of the entire site from 2005 to 2011 was 12052 kg ha-1 compared to the high and low clusters which were 13747 kg ha-1 and 10420 kg ha-1, respectively. It was observed that 12% of the site was consistently variable, of which 82%, occurred within the Clarion. This identification approach could be used as a basis for variable rate management which could increase yield, net profitability and sustainability. The second objective was to explore the relationships of specific soil characteristics associated with the identified variability clusters. High yielding clusters were deeper to maximum depth of mollic colors, higher in total carbon (TC) and total nitrogen (TN) content at 0 to 25 cm, higher in Mehlich 3 phosphorus (M3P) and Mehlich 3 potassium (M3K) content at all depths and higher in clay content at 0 to 25 cm and 26 to 100 cm compared to low yielding clusters. At the 0 to 25 cm depth, 60% of yield variability was associated with TN, M3P and M3K and clay content. These four soil characteristics were positively correlated with yield (r= 0.73, 0.47, 0.41, 0.72, respectively). However, 55% of yield variability was associated with TC and TN content at the 26 to 100 cm depth. At the 101 to 120 cm depth, 33% of yield variability was associated with pH and M3P content. This study illustrates the ability to associate yield variability within fields on the Des Moines Lobe of Iowa with specific soil characteristics.