Evaluation of techniques for estimating the yield of irrigated pastures intensively grazed by dairy cows. 1. Visual assessment

Abstract

A visual assessment technique was evaluated for estimating pre- and post-grazing pasture yields in experiments in which dairy cows rotationally graze irrigated pastures. Visual assessment, when associated with the harvesting of standard quadrats, accounted for 62.3 and 68.4% of the variability in pre- and post-grazing yield, respectively. The mean residual standard deviations associated with the pre- and post-grazing regressions were 490 and 395 kg dry matter/ha and, when the variable yields were taken into account, these represented coefficients of variation of 15.6 and 19.0%, respectively. By using the coefficient of variation as the criterion for precision, pre-grazing regressions were significantly more precise than post-grazing regressions. In practice, visual assessment by a double sampling procedure can be used to estimate pre- and post-grazing yields in dairy cow grazing experiments where herbage intakes and growth rates are required, provided sufficient readings are taken to allow an acceptable level of precision. It is suggested that about ten standards are required to formulate a calibration regression regardless of whether or not the pasture to be assessed has been grazed. Although the double sampling technique may be suitable for research purposes, it has been demonstrated that the completely subjective eye appraisal, often used by farmers and their advisers for feed budgeting, can result in many incorrect assessments. A more objective technique needs to be developed for this purpose.