Modelling the economics of herbicide treatment in wheat and barley using data on prevented grain yield losses

Losses in grain yield prevented by controlling weeds were measured in 59 fields of (southern hemisphere) spring-sown wheat (Triticum aestivum L.) (cv. Otane) and 45 fields of spring-sown barley (Hordeum vulgare L.) (cv. Corniche) in five consecutive growing seasons from 1988/89 until 1992/93 in the Canterbury region of New Zealand. The losses were measured as the differences in yield between weeded and non-weeded plots located in randomly positioned pairs in the fields. In the first 2 years. the weeding was by push hoe in 'organically' grown crops. For the last 3 years, the fields were under prophylactic herbicide regimens with nonweeded plots created by excluding commercial herbicide applications (made mostly in October for wheat and November for barley) with polyethylene sheets placed temporarily over the plots. For each season the distributions of yield losses were modelled using the normal distribution and probabilities of 'breaking even' on herbicide use derived by substituting cumulative probability density functions into a simple break-even model for herbicide use. The model assumed that herbicide application in the current crop has no flow-on economic effect for succeeding crops. The mean annual yield losses prevented by herbicide application were positively correlated with September and October rainfall for wheat and barley respectively. As a consequence, the probabilities of breaking even on herbicide use increased with increasing spring rainfall. Using historical rainfall records, probabilities of breaking even were estimated for each of the 48 years from 1947 to 1994. Averaging over these years, the analysis revealed that at current grain prices prophylactic use of the commonly applied herbicides is likely to be uneconomic in 24% (95% confidence limits 6% and 50%) of fields of average-yielding Otane wheat and in 26% (95% confidence limits 1% and 91%) of fields of average-yielding Corniche barley in Canterbury. It was concluded that there is potential for withholding herbicide treatments without jeopardizing profitability in these crops, particularly in seasons with low spring rainfall.