UC Davis

The objective of this dissertation is to integrate information from a nonlinear ration formulation model, a thermal balance model, and emission and excretion models to evaluate the environment impact of dairy cattle and the effect of climate on dairy cattle at the farm level. The first chapter investigated the application of iterative linear programming (iteLP), sequential quadratic programming (SQP) and mixed-integer nonlinear programming based deterministic global optimization (MINLP_DGO) on designing feed formulation for dairy cattle based on NRC (2001). A simulation study showed that iteLP had limited capability to design least cost diets when nonlinearity existed in the constraints. Both SQP and MINLP_DGO were able to handle nonlinear constraints well, with SQP being faster but MINLP_DGO being more reliable. In the second chapter, a thermal balance model was developed to predict the body temperature and heat fluxes for Holstein dairy cattle under heat stress conditions. The model included five nodes representing the body core, skin and coat of a dairy cow. Heat production by the animal, heat conduction through the body core, skin and coat, and heat flows between the animal and the environment, including conduction, convection, evaporation and radiation, were calculated based on existing models and physical principles. Model evaluation suggested a likely overestimation of body temperature. Sensitivity analysis showed that heat production, surface area, air pressure and the parameters relative to respiration and sweating were the most sensitive. In the third chapter, environmental impact of dairy cattle was evaluated by considering relevant outputs simultaneously. Three multivariate Bayesian regression models were developed to predict enteric methane (CH4), carbon dioxide (CO2), water intake (Waterin), volatile solids (VS), biodegradable VS (dVS), fecal DM (FDM), fecal water (FW), fecal carbon (FC), fecal nitrogen (FN), total urine (Ut), urine carbon (UC) and urine nitrogen (UN) for lactating cows, nonlactating cows and heifers. Most equations predicted the response variables with reasonable accuracy, except Waterin and Ut equations for nonlactating cows and heifers. In the last chapter, a simulation study was conducted to evaluate the environment impact of dairy cattle and the effect of climate on dairy cattle at the farm level. The ration, body temperature, heat flows, greenhouse gas emission and manure excretion were predicted for two heifer herds, three lactating cow herds and one nonlactating cow herd.