Nitrogen decisions for cereal crops: a risky and personal business

Robert Farquharson1, Deli Chen1, Yong Li1, 2, De Li Liu3, Thiagarajah Ramilan1-4

1Faculty of Veterinary and Agricultural Sciences, University of Melbourne, Parkville, Victoria 3010, Australia, bob.farquharson@unimelb.edu.au 

2Chinese Academy of Science, 52 Sanlihe Rd., Beijing, China (100864) 

3NSW Department of Primary Industries, Wagga Wagga Agricultural Research Institute, Wagga Wagga, NSW 2650, Australia 

4Massey University, Palmerston North 4474, New Zealand

Abstract

Cereal crops principally require Nitrogen (N) and water for growth. Fertiliser economics are important because of the cost at sowing with expectation of a financial return after harvest. The production economics framework can be used to develop information for ‘best’ fertiliser decisions. But the variability of yield responses for rainfed production systems means that fertiliser decisions are a risky business. How do farmers make such decisions, and can economics give any guidance? Simulated wheat yield responses to N fertiliser applications show tremendous variation between years or seasons. There are strong agronomic arguments for a Mitscherlich equation to represent yield responses. Plots of the 10th, 50th and 90th percentiles of yield response distributions show likely outcomes in ‘Poor’, ‘Medium’ and ‘Good’ seasons at four Australian locations. By adding the prices for Urea and wheat we predict that the ‘best’ decisions vary with location, soil, and (sometimes) season. We compare these predictions with typical grower fertiliser decisions.  Australian wheat growers understand the yield responses in their own paddocks and the relative prices, so they are making relevant short-term fertiliser decisions. These are subjective or personal decisions. Myanmar smallholders grow rice and maize in the Central Dry Zone, with relatively low levels of fertiliser and low crop yields. They have pre-existing poverty, high borrowing costs and are averse to risky outcomes. A Marginal Rate of Return (MRR) analysis with a hurdle rate of 100% is illustrated for the Australian locations, and this approach will be tested in Myanmar.