Field Crops Research 139 (2012) 47–56 Contents lists available at SciVerse ScienceDirect Field Crops Research jou rn al h om epage: www.elsevier.com/locate/fcr A ‘small strip’ approach to empirically determining management class yield response functions and calculating the potential financial ‘net wastage’ associated with whole-field uniform-rate fertiliser application B.M. Whelan ∗ , J.A. Taylor, A.B. McBratney Precision Agriculture Laboratory, Faculty of Agriculture and Environment, University of Sydney, 1 Central Avenue, Australian Technology Park, Eveleigh, NSW 2015, Australia a r t i c l e i n f o Article history: Received 26 April 2012 Received in revised form 12 September 2012 Accepted 19 October 2012 Keywords: Management class Uniform-rate fertiliser Variable-rate fertiliser On-farm experimentation Gross margin analysis a b s t r a c t Site-specific crop management requires site-specific information. The amount and pattern of variability in natural resources and production output, the reasons for variability, the agronomic implications and the management opportunities ultimately need to be identified locally. In this context it is often desirable to obtain information on any variability in production response to different application rates of inputs across a site in order to help identify if worthwhile financial gains can be made by implementing spa- tially variable application rates. However there has been little development and promotion of practical, considered experimental design and analysis techniques that can be applied at a commercial farm scale to test variability in optimal input rates. Here a stratified, randomised, replicated ‘small strip’ experimen- tal design is proposed for use in commercial-scale cropping systems that employ a management class approach. The experimental design considers technical, economic, agronomic and non-spatial statistical constraints and seeks to find a compromise between these constraints such that any one constraint is not limiting to adoption or interpretation. A description and rationale for the experimental design and analysis is presented followed by the results from the application of this experimental design to 15 nitro- gen and 18 phosphorus response trials on various types of crops, in 3 different regions of Australia, over 6 seasons. Under these conditions, the results show that optimum response data can be obtained using the experimental technique. A gross margin analysis for each experiment was used to calculate the ‘net wastage’ in fertiliser and yield resulting from the customary, spatially uniform input management. A median ‘wastage’ of A$39/ha for nitrogen fertiliser and A$48/ha for phosphorus fertiliser was observed. Standardising the individual results to the total, seasonal cost of uniform fertiliser application in each experiment, determined that the median ‘net wastage’ value equated to 99% of the traditional fertiliser budget for phosphorus and 97% for nitrogen. These figures suggest that the potential financial benefit to be gained over a number of seasons by knowing more about the optimum rates of fertiliser for man- agement classes within a field could be equal to the amount of money traditionally outlaid on fertiliser. The technique does have some statistical limitations that require a practical, agronomic interpretation to be included in an assessment of the results. Although applied to nutrient response experiments in this study, it could be adapted for any crop input or management practice experiment. © 2012 Elsevier B.V. All rights reserved. 1. Introduction Site-specific crop management (SSCM) aims to identify, seek explanation for, and provide optimal solutions to local variation in production. Investigating the causes of variability and ameliorating Abbreviations: SSCM, site-specific crop management; VRT, variable-rate tech- nology; ANOVA, analysis of variance; REML, restricted maximum likelihood; OLS, ordinary least squares; MC, marginal cost; MR, marginal revenue; GM, gross margin; OR, optimal input rate. ∗ Corresponding author. Tel.: +61 2 8627 1132; fax: +61 2 8627 1099. E-mail address: brett.whelan@sydney.edu.au (B.M. Whelan). to restore/raise production potential where possible is a generic precursory step to exploring variable-rate nutrient management. If investigations indicate variable-rate nutrient management as a potential next step, then decisions on fertiliser requirements and application rate changes need to be made. Traditionally, crop fer- tiliser requirements are determined using district- or regionally derived yield response functions. This is despite the fact that crop production (yield and quality) and response to nutrient inputs has been shown to vary both within and between fields (e.g. Holford et al., 1992; Kitchen et al., 1995; Mamo et al., 2003; Whelan et al., 2009). In view of this situation, it would appear desirable for pro- ducers to determine the degree to which functions that describe yield response to a wide range of applied fertiliser rates may vary 0378-4290/$ – see front matter © 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.fcr.2012.10.012