A fecal test for assessing phosphorus overfeeding on dairy farms: evaluation using extensive farm data

Predicting orthophosphate in feces and manure from dairy cattle

Dairy cattle excreta are a valuable source of orthophosphate (Ortho-P), an inorganic form of phosphorus (P) that is readily available for microorganisms, plant growth, and development. There is, however, a growing environmental concern about the potential negative environmental impact of excessive amounts of Ortho-P excretion, which can lead to the eutrophication of water bodies. As a result, the development of mathematical equations to quantify and manage Ortho-P excretion on dairy farms could prove valuable for environmental sustainability. This study aimed to use literature data to develop empirical predictions for Ortho-P (g/kg dry matter [DM]) excretion using total P (TP [g/kg DM]) content of dairy cattle feces (Ortho-Pf) and manure (Ortho-Pm). Data sets from studies that evaluated and characterized the different forms of P in feces and manure from dairy cattle were compiled. After outlier exclusion, the final retained database for feces included 37 treatment means from 4 published papers while the manure comprised 23 treatment means from 7 published papers. A linear-mixed model was used to develop the predictive equations, incorporating the random effect of the study. A leave-one-out cross-validation procedure was used to evaluate the predictive ability of the developed models, whereby studies were regarded as folds. The fecal equation was determined as Ortho-Pf (g/kg DM) = -2.447 (0.572) + 0.966 (0.083) × TP (g/kg DM) (R2 = 0.79) and resulted in a root mean square prediction error as a percentage of mean observed value (RMSPE, %) of 32.8% and error due to random sources of 97.6%. Additionally, the manure equation was determined as Ortho-Pm (g/kg) = -0.204 (0.446) + 0.590 (0.065) × TP (g/kg) (R2 = 0.77) and had an RMSPE of 43.3% with a random error of 93.9%. Both models revealed minimal mean and slope biases on feces and manure data. Findings suggest that these sets of equations can be used to estimate excreted Ortho-P from total excreted P, helping nutritionists and farmers to understand the impact of dietary P changes on the environment. Further, these equations can be incorporated into extant models such as the Cornell Net Carbohydrate and Protein System (CNCPS) to aid in understanding and mitigating P and Ortho-P excretion from dairy cattle and to clarify the portion of P that migrates more rapidly into watersheds.

Faecal excretion of total and acid extractable phosphorus in dairy cows fed rations with different levels of phosphorus

BACKGROUND

The phosphorus (P) originating from livestock operations causes eutrophication. Determination of acid extractable P (AEP) in cattle faeces has been proposed as a tool to identify excessive P feeding. The method has not yet been evaluated in controlled studies with cows subjected to individual recording of P intake. Thus the present study focused on investigating the relationship between different P fractions in faeces from cows fed rations with varying P content. The study also investigated whether AEP in faeces could be used to estimate dietary P intake in relation to the P requirement.

RESULTS

The results showed that acid extractable P predicted P overfeeding. P fed in excess of requirements was largely excreted as acid extractable P. The unavailable and/or inevitably lost P fractions in the diets were smaller than assumed.

CONCLUSION

This study demonstrates in experimental studies a positive relationship between measured AEP and indirectly calculated regulated P. Any P fed in excess of requirements was largely excreted as AEP. This fraction is thought to be the most mobile P fraction with regard to potential runoff losses. However, the unavailable and/or inevitably lost P fractions in the diets were smaller than assumed.