| Popis: |
The frequency and intensity of herbage defoliation have been the focus of grassland research because of their effect on herbage accumulation rate (HAR). The Gompertz equation describes standing herbage mass (HM) over time and its derivative describes the HAR. By combining those two relationships, a time-independent relationship can be developed between HAR and HM. The objectives of this research were: i) develop Gompertz growth curves and their time independent form, ii) investigate the seasonal pattern in those equation parameters, iii) use the time independent parameters to create a model to predict HAR based on HM, iv) validate the model, and v) investigate factors affecting the calibration of the rising plate meter (RPM) for measuring HM in pastures. In 2008, HM was measured weekly during 11 growing periods at Columbus and Coshocton, OH, and Arlington, WI. Those data were fit to Gompertz equations and time-independent HAR vs HM curves were calculated, and subsequently, the optimum HM at which HAR was maximized was determined. The optimum HM varied between 1600 and 4000 kg DM ha-1 and there was a seasonal pattern of the parameters of the Gompertz equation and its time-independent form. The parameters from the time-independent function of HAR vs HM were used to develop a model to predict HAR based on the standing HM, and a study was conducted from 20 May to 27 October near South Charleston, OH to calibrate the model parameters and validate the model predictions. Monoculture plots of tall fescue [Schedonorus phoenix (Scop.) Holub.] (TF), orchardgrass (Dactylis glomerata) (ORG), Kentucky bluegrass (Poa pratensis) (KBG) and a mixture (MIX) of those three species were subjected to four defoliation treatments: 3600 to 2800 kg DM ha-1 (Tall-Tall, TT), 3300 to 2400 kg DM ha-1 (Tall-Short, TS), 3600 to 1800 kg DM ha-1 (Hay) and 3000 to 1600 kg DM ha-1 (Variable, V). The proportion of variation explained by the regressions of observed on predicted model values for HM ranged from 0.48 to 0.88 and 11 out of 25 regressions were not significant different for m = 1 and c = 0. For treatments Hay and V, the model underestimated the standing HM, whilst overestimated HM in treatments TT and TS. The model showed promise in predicting the effect of defoliation on subsequent HAR; however, some refinement is required, especially when predicting herbage accumulation at residual herbage mass lower or about to 1600 kg DM ha-1. The effect of season and species on the RPM calibrations were studied in five Ohio environments: Columbus 2008 and 2009, two sites at Coshocton in 2008 and South Charleston in 2009. There was an effect of calendar week (P = 0.0021), environment (P < 0.0001), environment x week (P < 0.0001), and at South Charleston species (P = 0.0078) on HM. A pattern of rapidly decreasing of slope coefficient was observed in spring followed by increasing coefficients the remainder of the season, suggesting the existence of underlying processes responsible for the seasonal changes. |
