Mineral concentrations of fresh herbage from mixed grassland as influenced by botanical composition, harvest time and growth stage

The mineral concentrations in herbage from mixed grassland are variable which lead to challenging situations for animal's mineral supplementation. The aim of this study consisted in assessing macro and micro mineral concentrations in fresh herbage from mixed grassland, according to a classification method used for nutrients and nutritive values. This classification consists of six classes of botanical composition, two classes of harvest time and seven classes of growth stage. Two hundred and thirty six samples of fresh herbage from mixed grassland were collected from one experimental site during five seasons. The analysed mineral concentrations were modelled using botanical composition and harvest time as fixed effects and growth stage as linear and quadratic effects. Whereas the repeatability of the models for calcium (Ca), phosphorus (P), magnesium (Mg), potassium (K), sulphur (S) and copper (Cu) were similar to the ones for crude protein and fibre, the repeatability for sodium (Na), chlorine (Cl), iron (Fe) and manganese (Mn) were low. Cobalt (Co, 0.113 ± 0.173 mg/kg DM) and selenium (Se, 0.020 ± 0.022 mg/kg DM) concentrations were hardly detected. Except Fe, mineral concentrations were influenced (P < 0.05) by class of botanical composition. Herbage with increasing presence of grasses and thus a reduced presence of legumes and other herbaceous plants, had lower (P < 0.05) Ca, P, K, Cu and zinc (Zn) concentrations. Herbage with ryegrass predominance had lower (P < 0.01) Mg, S and Mn concentrations. Mineral concentrations, except Fe, were lower (P < 0.05) in the first seasonal harvest than in subsequent harvests. With increasing growth stage, P (P < 0.05, linear), Mg (P < 0.001, linear and quadratic), K (P < 0.001, quadratic), S (P < 0.001, linear) and Cu, Mn and Zn (P < 0.001, linear and P < 0.05 quadratic) concentrations progressively decreased. Sodium and Cl concentrations responded quadratically (P < 0.05) to the developing growth stages and Ca concentrations remained stable (P > 0.10). The developed models may be considered to assess reference values in order to take into account the native mineral concentrations of fresh herbage from mixed grassland when optimizing herbivore diets. This, is a contribution for improving the sustainability of mineral supplementation.