The accuracy of a range of capacitance probe methods for estimating pasture yields

Abstract

SUMMARYNine different capacitance probe methods were used to estimate the herbage yields of 32 small pasture plots; ten meter readings were used in conjunction with the yield data from a variable number (n) of quadrats in each plot. Expected yields were determined by cutting ten quadrats per plot. Estimated yields were compared with expected yields for both the differing methods and the differing values of n. At each level of n the yield estimates from the different probe methods were compared with those obtained simply by cutting n quadrats.Using calibration data from only one quadrat per plot (n = 1), all probe methods except one gave greater accuracy than that obtained from the single quadrat yields only. For n = 2, the advantage of the probe methods was less pronounced and where n = 3 only one probe method (one which involved no regression computations) gave superior accuracy to that obtained merely by using the yield data from the three cut quadrats. At n > 3, only the one probe method gave increased precision; all other probe methods were markedly less accurate than the equivalent raw quadrat results.Only one probe technique (which combined data from markedly dissimilar swards) gave yield estimates which were noticeably biased.Results from an earlier study are presented to demonstrate the influence of season, species and irrigation on probe calibration.Recommendations are made on the various probe methods which may prove effective in some experimental situations.