Evaluation of an extant model for the excretion of phosphorus and nitrogen from swine fed diets with and without microbial phytase

摘要

An extant model was evaluated to assess its adequacy for nutrient management planning (NMP) from swine operations in Manitoba that includes predictions for major nutrients of interest, phosphorus (P) and nitrogen (N), and the land base for spreading of manure when standard and phytase supplemented diets are used. Furthermore, the effect of phytase on net greenhouse gas (GHG) emissions from manured dry and wet sandy loam soils was also determined. Data was generated from starter to finisher pigs where 20 pigs (10 per treatment), and sows where 18 (9 per treatment) were used with two dietary treatments. The control diet was formulated to meet the requirement of pigs for nutrients as per the recommendations of NRC (No-phytase) and the second diet was formulated with P level in the No-phytase diet reduced by an average of 0.1 percentage units and diet supplemented with microbial phytase at 500 FTU/kg (Phytase). There was no significant difference in average daily feed intake (ADFI) (P >0.25), average daily gain (ADO) (P >0.173) and feed conversion ratio (FCR) (P >0.084). Model tended to predict the P excretion in all the phases better in the No-phytase than the Phytase diet where in most cases the random variation (ED >50% except in phase IV) contributed the most towards the total error of prediction. However, the model was not satisfactory for predictions when phytase was supplemented to swine diets where error due to regression (ER) was more than 50% in all the phases except the second phase (14-25 kg body weight). Concordance correlation coefficient (CCC) analysis showed that the model was fairly accurate and precise in the starter to finisher phases of production. Evaluation of model adequacy in sow operations showed that most (60% and 50%) of the total error of prediction was associated with ER in the No-phytase diet and ED for the Phytase diets, respectively. The model over-predicted the excretion of P from sows by 32% in the No-phytase diet (mu =-1.06) and 23.5% in the Phytase diet (mu =-0.783). Furthermore, the model over-predicted the land base required for the spread of manure by 30% in the No-phytase and 21% in the Phytase diets. Generally, the assessment of model adequacy in the growing-finishing pigs showed that the model can be used to predict the excretion of P in the manure of growing-finishing pigs satisfactorily in No-phytase, but not in the Phytase diet. However, the model was not satisfactory for the predictions of both P and N excretion in sows. Application of manure originating from diets supplemented with microbial phytase showed a slight increase in the emissions of net GHG emissions mainly in the form of carbon dioxide (CO 2). Emissions were, however, mainly influence by the moisture content of soils rather than the type of manure where high moisture contents of soils resulted in a significantly higher emissions of both CO2 and nitrous oxide (N2O).