Site-specific seeding for maize production using management zone maps delineated with multi-sensors data fusion scheme

Abstract

Uniform rate seeding (URS) density may result in under- or over-optimal plant population that negatively affects crop growth and yield. Site-specific seeding (SSS) is one of the solutions to manage in-field soil variation by optimizing the input seed rate to match soil fertility. This study has evaluated the agronomic and economic response of maize to SSS compared to the URS using a data fusion scheme. Two fields of 5.5 ha and 10 ha in Belgium and France, respectively, were scanned using an on-line visible and near-infrared spectroscopy sensor to measure soil pH, organic carbon (OC), phosphorus (P), potassium (K), magnesium (Mg), calcium (Ca), sodium (Na), moisture content (MC), and cation exchange capacity (CEC). Crop normalized difference vegetation indices (NDVI) were retrieved from several Sentinel-2 images. Measured crop yield, retrieved NDVI and on-line measured soil attributes were then fussed using k-means clustering algorithm to delineate management zone (MZ) maps, whose classes were ranked based on their fertility level and crop yield. A parallel strips experiment was overlaid upon the MZ map, to allow comparing the performance of SSS against URS. Two SSS treatments were implemented in the strip experiment, e.g., the “Kings” approach that recommended the highest seeding density for the highest fertile MZ class and vice versa, and the “Robin Hood” approach followed the opposite principle. Results revealed that SSS treatments increased maize grain yield by 0.25–0.70 Mg ha−1 and thus improved gross margin by 26.7–92.67 € ha−1, compared to the URS. Besides, the SSS-Kings treatment outperformed the URS and SSS-Robin Hood in both fields, whereas the SSS-Robin Hood outperformed the URS treatment only in one field. Soil OC, MC, Mg, and pH revealed a positive correlation each with grain yield in the SSS-Kings treatment. The SSS-Kings treatment is therefore recommended to manage in-field soil variation, which can result in optimizing input seed rates for increasing maize productivity and profitability.