Visual assessment of herbage mass

An ability to accurately and quickly estimate farm grass supply helps greatly in the management of grazing pastures to enable high cow performance. The objective of this study was to evaluate the accuracy of visual assessment of herbage mass (HM). Measurements of HM were made on 13 dairy farms from November 1995 to October 1996. HM measurements were repeated over the following 12-month period. Farms were visited monthly from October to March and bimonthly from April to September. HM was estimated for a range of herbage masses on the farm. Herbage mass was visually estimated and then harvested by cutting a square (0.25 m²) quadrat to 40 mm above ground level. The variable of interest is the ratio of actual to estimated HM. In Year 1, when HM was < 750 kg dry matter (DM)/ha, the ratio of actual to estimated yield was biased upwards by between 0.06 and 0.40. The variance of the ratio declined with increasing estimated HM. The standard deviation of the ratio declined from about 0.20 to 0.12 as estimated HM increased from about 1000 to 3000 kg DM/ha. A model was fitted to the data with farm x visit as a random effect for all yields > 750 kg DM/ha. The increase in variability as HM increased was incorporated into the analysis. One farm's predicted ratio was significantly different from unity. The months of January, July and November had a predicted ratio significantly different (P < 0.05) from unity. The accuracy of the visual assessment method was confirmed by the results from Year 2. The results indicate that there is potential for precise estimation of HM by visual assessment.