אסיף מאגר המחקר החקלאי

Measurement error in the estimation of intake from herbage patches by non-fistulated heifers on apparently uniform swards

Year:

1997

Source of publication :

Authors :

Volume :

46

Co-Authors:

Ungar, E.D., Dept. of Agronomy and Nat. Resources, Agricultural Research Organization, Volcani Center, POB 6, Bet Dagan 50250, Israel
Ravid, N., Dept. of Agronomy and Nat. Resources, Agricultural Research Organization, Volcani Center, POB 6, Bet Dagan 50250, Israel
Genizi, A., Dept. of Stat. and Operations Res., Agricultural Research Organization, Volcani Center, POB 6, Bet Dagan 50250, Israel
Methorst, R., Dept. of Agronomy and Nat. Resources, Agricultural Research Organization, Volcani Center, POB 6, Bet Dagan 50250, Israel, Dept. of Animal Production Systems, Agricultural University, Wageningen, Netherlands

Facilitators :

Link :

Abstract:

We examined a field-based micro-sward methodology to study the plant-animal interface, using non-fistulated animals grazing small, carefully prepared patches of herbage. The study focused on the error introduced by the indirect estimation of herbage intake from a patch. In each of two experiments, 102 patches (height 0.3 m, area 0.5 m 2), each with an adjacent estimation area, were created in apparently uniform areas of a sown alfalfa sward. Of these, 72 patches were grazed (30 bites) individually by eight heifers, and 30 were used as a calibration set. Dry matter intake from a grazed patch was determined by the difference method whereby residual herbage mass was measured directly by clipping, and initial herbage mass was based on the mass of herbage clipped in the estimation area. Four methods were used to compute initial herbage mass of a patch; two incorporated information from the calibration patches in which both the patch and the estimation area were clipped. Measurement error was determined by applying the four computational methods to the calibration sets. The calibration set enabled us to detect and correct for a significant estimation bias. This resulted in very different bite weight estimates from those that would have been obtained otherwise. Measurement error was lowest for estimations of initial patch biomass based on a regression equation derived for the calibration sets. Measurement error estimated from the calibration sets was of similar order of magnitude to the total error (within-animal) variance component of the grazing trials.

Note:

Related Files :

Related Content

More details

You may also be interested in