TY  - THES
T1  - Improved prediction of dietary protein use and nitrogen excretion in tropical dairy cattle
A1  - Salazar Cubillas,Khaterine
Y1  - 2023/11/06
N2  - The overall objective of the present doctoral thesis was to evaluate the adequacy (i.e., accuracy and precision) of existing laboratory methodologies and modeling tools, originally designed for temperate systems, in predicting the nitrogen (N) supply and excretion of cattle in tropical husbandry systems. It was hypothesized that the adoption of laboratory methodologies and modeling tools from temperate systems without validating and adapting them for tropical systems may result in inaccurate estimations of N supply, utilization, and excretion, which will hamper the assessment of N use efficiency. An in vitro study was conducted to evaluate the adequacy of the chemical method (Sniffen et al., 1992; Kirchhof et al., 2010; Valdés et al., 2011) to predict rumen-undegraded crude protein (RUP) of tropical forages grasses and legumes (n = 23). The adequacy of the predictions was assessed by comparing them with RUP proportions measured in situ at rumen passage rates of 2, 5, and 8% per hour. Results showed that the RUP of tropical forages estimated with the in situ method can be predicted using the chemical method. However, regression equations developed for temperate forages were not adequate enough to predict RUP proportions of tropical forages consistently for all rumen passage rates. Instead, developed equations in the present thesis can be used to predict RUP proportion of tropical forages with a similar chemical composition than the reference forage sample set. A second in vitro study was conducted to evaluate the adequacy of the chemical (Sniffen et al., 1992; Zhao and Cao, 2004) and in vitro methods (Steingaß et al., 2001) to predict post-ruminal crude protein (PRCP) supply of tropical forages (n = 23). The adequacy of the PRCP supply with the chemical and in vitro methods were tested against PRCP supply estimated from in situ measurements at rumen passage rates of 2, 5, and 8% per hour and digested organic matter. Results showed that the in vitro method can be used as an alternative method to estimate PRCP supply in tropical forages at moderate to fast rumen passage rate but not at slow rumen passage rate. Available regression equations developed for temperate forages were not adequate enough to predict the PRCP supply of tropical forages from concentrations of chemical crude protein fractions. Instead, developed equations in the present thesis can be used to estimate PRCP supply of tropical forages with a similar chemical composition than the reference forage sample set. A third study was conducted to assess the adequacy of modeling tools to predict N excretion of cattle in tropical husbandry systems. These models, namely model A (based on AFRC, 1993), model G (based on GfE, 2001), and model I (INRA, 2019), were selected to predict fecal N (FN), urine N (UN), and total N (TN) excretion as well as FN fractions of dairy cows, heifers, and steers kept under typical tropical husbandry conditions. The adequacy of the model predictions was assessed against reference values of UN (total collection and creatinine method) and FN excretion (total collection, internal and external markers) (n = 392 observations). Adjustments were made to the models with the greatest potential to predict N excretion. The adjustments were focused on the input variables driving the variability in N excretion predictions, identified through a sensitivity analysis. None of the tested models predicts adequately the excretion of UN, FN, and of different FN fractions of individual cattle kept under tropical conditions. Instead, model I in the present thesis, adjusted for increased efficiency of rumen microbial crude protein synthesis and reduced intercept of FN prediction, can be used to estimate FN and TN excretion of individual cattle kept under tropical conditions. The findings from the present thesis partially support our hypothesis. The adjustment of laboratory methodologies, such as the chemical method used to estimate the protein value of temperate forages, to tropical forages, results in more adequate estimates of the proportion of RUP and PRCP supply of tropical forages. Model I is, therefore, able to predict the N excretion of cattle more adequately in tropical husbandry systems, because it is sensitive to differences in the RUP proportion and PRCP supply. In addition to increasing the adequacy of these input variables, adjustments made to the microbial protein synthesis and intercept of the FN excretion of model I results in a more adequate prediction of N excretion by cattle in tropical husbandry systems. However, not all adjustments to laboratory methodologies and modeling tools from temperate systems yielded adequate predictions. Specifically, challenges remained in predicting RUP proportion and PRCP supply for tropical forage legume with slow rumen passage rates, as well as urinary N excretion in cattle with low N intakes. Consequently, further research is required to identify the factors contributing to their poor adequacy.
KW  - Stickstoff
KW  - Modellierung
KW  - Tropische Pflanzen
CY  - Hohenheim
PB  - Kommunikations-, Informations- und Medienzentrum der Universität Hohenheim
AD  - Garbenstr. 15, 70593 Stuttgart
UR  - http://opus.uni-hohenheim.de/volltexte/2023/2223
ER  -