A method for approximate on‐farm estimation of herbage mass by using two assessments per pasture

Abstract

Appropriate pasture management requires implementation of methods that reduce the time and labour required for assessing herbage mass. In this study, Monte Carlo simulations and a field study were used to evaluate a simple method for estimating herbage mass based on the corrected average of the single highest and lowest mass values in a pasture. The necessary correction factor for the method was objectively derived using a single nonlinear equation based on the scaled difference between the highest and lowest herbage mass regardless of the mean and variance of mass and pasture size. The accuracy of estimation using this method is acceptable for on-farm use: more than 90% of the estimates had error rates below 20% of the true value when the plot size assessed was larger than 64 m 2 in the simulation study; in the field experiment, approximately half of the estimates had error rates below 20% of the observed mean of random samples (R 2 = 0Á71). The advantages of the proposed method are that it requires only two assessments per pasture and that selection of assessment plots is straightforward. This method has potential for obtaining approximate on-farm estimates of herbage mass that could help improve pasture management.

A Simple Variance Estimation of Herbage Mass Based on Two Assessments Per Pasture

“…The values for b were defined using a previously reported average, 0.4 (McIntyre 1978). The defined values of the current study differed from those of the previous study (Nakagami 2016a) with respect to plot size because plot size ≥ 64 m 2 had been recommended to estimate mean herbage mass accurately from the comparison of plots that were 4, 16, 64, and 144 m 2 in size.…”

“…The relationship of the variance in herbage mass to the difference between the highest and the lowest mass values was analyzed using a Monte Carlo simulation that used random numbers under the assumption that the frequency distribution of herbage mass per unit area could be described by a gamma distribution (Shiyomi et al 1983(Shiyomi et al , 1984. The dataset used here was essentially the same as the one used in my previous study (Nakagami 2016a). In brief, a set of gamma random numbers was generated for all possible combinations of the following five conditions: (1) mean herbage mass, μ;…”

“…In my previous study (Nakagami 2016a), I proposed a simple method for making approximate on-farm estimates of mean herbage mass by modifying a study completed by Iwasaki (1976). The procedure involved: (i) a visual assessment to select two sampling locations (plots) that have the highest and lowest herbage mass, respectively, in a paddock; (ii) assessment of the mean herbage mass of each plot using a Rising Plate Meter (RPM); (iii) calculation of the standardized difference between the highest and lowest mass by dividing the difference by the sum of the two values; (iv) deriving a correction factor from the standardized difference using a single non-linear equation; and (v) estimating mean herbage mass by multiplying the average mass of the two plots by the correction factor.…”

“…By extending the previously proposed method for measuring mean herbage mass (Nakagami 2016a), it might be possible to estimate their variance. The standard deviation (SD) could be estimated from a range of random samples (Snedecor and Cochran 1967, Hozo et al 2005, Wan et al 2014.…”

“…The standard deviation (SD) could be estimated from a range of random samples (Snedecor and Cochran 1967, Hozo et al 2005, Wan et al 2014. Therefore, the difference between the highest and lowest herbage mass, which is used to estimate mean mass (Nakagami 2016a), could potentially be used to estimate the variance of the mass in place of a range of random samples. Iwasaki (1976) showed that the SD of herbage mass might be reasonably estimated from the difference between the highest and lowest mass using Tippett's d 2 :…”