In growing lambs, can body weight be predicted from the dorsal area?

Abstract Periodic assessment of body weight is one of the main parameters for evaluating the performance of a herd and the consequent efficiency of the production system. Based on this, the aim of the present study was to develop and evaluate different equation models for predicting body weight (BW) in growing lambs using the dorsal area (DA). The variables BW (27.56 ± 5.86 kg), hip width (HW), thoracic width (TW) and body length (BL) were measured in 275 growing male lambs (Pelibuey x Blackbelly x Katahdin) aged between 6 and 9 months and reared under humid tropical conditions. Dorsal area (cm2) was calculated using the mathematical formulae for calculating the area of a trapezium, taking into account HW, TW and BL in the calculation. The prediction models were constructed using linear and non-linear regression. Goodness of model fit was assessed using Akaike information criterion (AIC), Bayesian information criterion (BIC), coefficient of determination (R2), mean square error (MSE) and root mean square error (RMSE). Internal cross-validation (k-folds) was also used to evaluate the developed models. The root mean square error of prediction (RMSEP), R2 and mean absolute error (MAE) were used to assess the ability of the fitted models to predict the observed values. The model selected for the prediction of BW was: -8.47 + 0.06 × DA -2.29 × DA2 as it had the lowest MSE, RMSE and AIC values. Furthermore, for the relationship between BW and DA, this quadratic fit gave the highest R2 value (0.67). Therefore, with a reasonable level of accuracy, the quadratic model using the dorsal area may be suitable for the prediction of body in growing lambs. Based on results, dorsal area may be a moderate predictor of body weight in growing lambs, contributing to better property management.