Crop sensor-based in-season nitrogen management of wheat with manure application

Abstract

It is difficult to predict the crop-available nitrogen (N) from farmyard manures applied to soil. The aim of this study was to assess the usefulness of the proximal sensors, Yara N-TesterTM and RapidScan CS-45, for diagnosing the N nutritional status of wheat after the application of manures at sowing. Three annual field trials were established (2014-2015, 2015-2016 and 2016-2017) with three types of fertilizer treatments: dairy slurry (40 t ha-1 before sowing), sheep manure (40 t ha-1 before sowing) and conventional treatment (40 kg N ha-1 at tillering). For each treatment, five different mineral N fertilization doses were applied at stem elongation: 0, 40, 80, 120, and 160 kg N ha-1. The proximal sensing tools were used at stem elongation before the application of mineral N. Normalized values of the proximal sensing look promising for adjusting mineral N application rates at stem elongation. For dairy slurry, when either proximal sensor readings were 60-65% of the reference plants with non-limiting N, the optimumNrate for maximizing yield was 118-128 kg N ha-1. When the readings were 85-90%, the optimum N rate dropped to 100-110 kg N ha-1 for both dairy slurry and conventional treatments. It was difficult to find a clear relationship between sensor readings and yield for sheep manure treatments. Measurements taken with RapidScan C-45 were less time consuming and better represent the spatial variation, as they are taken on the plant canopy. Routine measurements throughout the growing season are particularly needed in climates with variable rainfall. The application of 40 kg N ha-1 at the end of winter is necessary to ensure an optimal N status from the beginning of wheat crop development. These research findings could be used in applicator-mounted sensors to make variable-rate N applications.