Soil properties and landscape factors affecting maize yield under wet spring conditions in eastern Canada.

Abstract

Spatial variation in soil properties is attributable in part to redistribution of moisture and topsoil and to soil compaction associated with water-logging. Digital Elevation Models (DEMs) can define topography at a fine scale where these processes may occur as related to field landscape positions such as footslope, depression, crest, and plateau. Our objective was to relate micro-relief (slope, elevation) to soil physical properties and to maize (Zea mays L.) yield under wet spring conditions and delineate soil management units in the field. The experimental 9-ha field was located in the St-Lawrence Lowlands, Quebec, Canada. A topographical survey was conducted using an all-terrain vehicle (ATV) equipped with high-precision Global Positioning System (GPS) receivers sampling 3-D coordinates to generate a grid as fine as 3 m. The GPS sampling method was compared to a total station (TS) for land survey on a 20-m grid. Soil samples and grain yield were collected and geo-referenced on a 30-m grid. Kriging, mapping, and correlation analyses were conducted. Topographic data sampled using a GPS receiver embedded on an ATV were accurate (<10 cm) compared to TS. Soil compaction indexes (bulk density, clay and clay/OM ratio) were the most closely related to yield (r2=-0.57 to -0.60). An independent dataset across the St-Lawrence Lowlands indicated that maize yield exceeding 10 t ha-1 in 1999 and 2000 was obtained with clay/OM ratios less than 12. The DEM and clay/OM ratio distribution maps were useful to locate five landscape positions of clay accumulation, water-logging, and soil compaction, forming two soil management units. In the footslope and depression management units, specific allocation of organic amendments may increase maize yield.