Levels of pasture substitution when concentrates are fed to grazing dairy cows in northern Victoria

This paper aims to provide information for farmers and their advisers to predict levels of substitution that might be occurring under various feeding conditions in northern Victoria. The approach taken involved compiling data from research conducted in northern Victoria and subjecting these to multiple regression analysis to define the key variables affecting substitution and marginal responses in milk production when concentrates are fed. A significant relationship was obtained between level of substitution (kg DM reduction in pasture intake/kg DM of concentrates eaten) and unsupplemented pasture intake (PI, kg DM/100 kg liveweight) when concentrates are fed. The regression relationship also included species composition of the sward being glazed (species: +1 grass, 0 clover), season of the year (season: +1 spring, 0 summer, -1 autumn) and concentrate intake (kg DM/cow.day). The equation is: Substitution = -0.34 + 0.16 (+/-0.035) PI + 0.16 (+/-0.053) species + 0.11 (+/-0.024) season + 0.03 (+/-0.014) concentrate intake [100R(2) = 50.9 (P < 0.01); r.s.d. = 0.14; CV = 37.7%]. Substitution increased by 0.16 kg DM/kg DIM for each increment of pasture intake. At any pasture intake, grass-dominant pastures, regardless of whether the grass was perennial ryegrass (Lolium perenne) or paspalum (Paspalum dilatatum), resulted in 0.16 kg DM/kg DM more substitution than white clover (Trifolium repens)-dominant pastures. In addition, substitution was 0.11 kg DM/kg DM higher in spring than in summer, and 0.11 kg DM/kg DM higher in summer than in autumn. Finally, substitution increased by 0.03 kg DM/kg DM for each additional kg DM of concentrates offered. Marginal returns in milk production (MR, kg extra milk/kg DM of concentrates eaten) were negatively related to substitution according to the following regression equation: MR = 2.62 - 0.80 (0.216) substitution - 0.28 (+/-0.084) season - 0.34 (+/-0.086) body condition [100R2 = 62.9 (P < 0.01); r.s.d. = 0.23; CV = 29.6%]. Marginal responses were 0.28 kg/kg DM lower in spring than in summer and autumn (season: +1 spring, 0 summer-autumn), and each unit improvement in body condition reduced expected marginal returns by 0.34 kg/kg DM. These relationships, together with those developed to aid estimates of unsupplemented pasture intake, can be used as background information in decision support systems to help farmers and their advisers make more informed decisions about feeding strategies when supplements are fed than has hitherto been possible.