Your browser doesn't support javascript.
loading
Prediction of drinking water intake by dairy cows.
Appuhamy, J A D R N; Judy, J V; Kebreab, E; Kononoff, P J.
Afiliação
  • Appuhamy JADRN; Department of Animal Science, University of California, Davis 95616. Electronic address: jaappuhamy@ucdavis.edu.
  • Judy JV; Department of Animal Science, University of Nebraska, Lincoln 68583.
  • Kebreab E; Department of Animal Science, University of California, Davis 95616.
  • Kononoff PJ; Department of Animal Science, University of Nebraska, Lincoln 68583.
J Dairy Sci ; 99(9): 7191-7205, 2016 Sep.
Article em En | MEDLINE | ID: mdl-27320675
Mathematical models that predict water intake by drinking, also known as free water intake (FWI), are useful in understanding water supply needed by animals on dairy farms. The majority of extant mathematical models for predicting FWI of dairy cows have been developed with data sets representing similar experimental conditions, not evaluated with modern cows, and often require dry matter intake (DMI) data, which may not be routinely available. The objectives of the study were to (1) develop a set of new empirical models for predicting FWI of lactating and dry cows with and without DMI using literature data, and (2) evaluate the new and the extant models using an independent set of FWI measurements made on modern cows. Random effect meta-regression analyses were conducted using 72 and 188 FWI treatment means with and without dietary electrolyte and daily mean ambient temperature (TMP) records, respectively, for lactating cows, and 19 FWI treatment means for dry cows. Milk yield, DMI, body weight, days in milk, dietary macro-nutrient contents, an aggregate milliequivalent concentration of dietary sodium and potassium (NaK), and TMP were used as potential covariates to the models. A model having positive relationships of DMI, dietary dry matter (DM%), and CP (CP%) contents, NaK, and TMP explained 76% of variability in FWI treatment means of lactating cows. When challenged on an independent data set (n=261), the model more accurately predicted FWI [root mean square prediction error as a percentage of average observed value (RMSPE%)=14.4%] compared with a model developed without NaK and TMP (RMSPE%=17.3%), and all extant models (RMSPE%≥15.7%). A model without DMI included positive relationships of milk yield, DM%, NaK, TMP, and days in milk, and explained 63% of variability in the FWI treatment means and performed well (RMSPE%=17.9%), when challenged on the independent data. New models for dry cows included positive relationships of DM% and TMP along with DMI or body weight. The new models with and without DMI explained 75 and 54% of the variability in FWI treatment means of dry cows and had RMSPE% of 12.8 and 15.2%, respectively, when evaluated with the literature data. The study offers a set of empirical models that can assist in determining drinking water needs of dairy farms.
Assuntos
Palavras-chave

Texto completo: 1 Coleções: 01-internacional Base de dados: MEDLINE Assunto principal: Água Potável / Lactação Tipo de estudo: Prognostic_studies / Risk_factors_studies Limite: Animals Idioma: En Revista: J Dairy Sci Ano de publicação: 2016 Tipo de documento: Article

Texto completo: 1 Coleções: 01-internacional Base de dados: MEDLINE Assunto principal: Água Potável / Lactação Tipo de estudo: Prognostic_studies / Risk_factors_studies Limite: Animals Idioma: En Revista: J Dairy Sci Ano de publicação: 2016 Tipo de documento: Article