Premium
Central Appalachian Hill Land Pasture Evaluation Using Cows and Calves. II. Comparison of Animal Days per Hectare Calculations with Statistical Evaluation 1
Author(s) -
Burns J. C.,
Harvey R. W.,
Giesbrecht F. G.,
Linnerud A. C.
Publication year - 1983
Publication title -
agronomy journal
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.752
H-Index - 131
eISSN - 1435-0645
pISSN - 0002-1962
DOI - 10.2134/agronj1983.00021962007500060006x
Subject(s) - grazing , zoology , hectare , pasture , mathematics , residual , stocking , beef cattle , statistics , standard error , biology , agronomy , ecology , algorithm , agriculture
In grazing experiments using variable stocking (put‐and‐take method), animal days ha −1 (from which gain ha −1 is computed) is calculated using the total digestible nutrient (TDN) method. This method has been satisfactorily applied to yearling cattle but not widely used when the tester unit is complicated by two animal types as with the cow‐calf unit. This study uses cow‐calf data from a 6‐year unbalanced grazing trial and compares two methods of computing animal days ha −1 and associated gain ha −1 . One method was the TDN procedure consisting of two types of calculations: 1) the standard procedure, using tester animal seasonal mean TDN (SMT) factors, and 2) the calculation of tester animal period mean TDN (PMT) factors. The second method was based on mean tester body live weights computed three ways: 1) converting all weights to a standard weight (453.5 kg) factor (STW); 2) calculating period mean weight (PMW) factors; and 3) calculating seasonal mean weight (SMW) factors. The five animal day calculations were compared for magnitude of year and field variance component and residual error. Relative and absolute values of treatment means were examined. The relative rank of the year and field variance components were generally similar among the five ways of calculating animal days ha −1 and associated gain ha −1 . A consistent difference was larger residual error(σ ∘ e 2 )for STW and PMT but smallest for the SMT calculation. The SMT animal days (standard used) and gain ha −1 were highly correlated with STW, PMW, and SMW (r ranged from 0.95**, significant at the 0.01 level of probability to 0.99**) but somewhat lower for PMT (r ranged from 0.82** to 0.88**). Absolute values (both days ha −1 and gain ha −1 ) from the SMT, PMW, and SMW calculations were very similar, while PMT values were variable and STW values greatly inflated. Results using the SMW and PMW calculations would be very similar to SMT values with little precision lost through increased variance. The use of live weight factors (PMW or SMW) in determining animal days ha −1 is much simpler and requires fewer assumptions than the TDN calcuations.