Using body measurements to estimate body weight in gilts

Abstract

The absence of a scale on pig farms has led to indirect body weight (BW) estimation using regression models based on body measurements. The objectives of the present study were to (1) develop prediction equations for weight estimation in gilts using body measurements (FF: flank-to-flank distance; L: length; HG: heart girth; BF2: ultrasound backfat measurement; LD: loin depth; and BCS: body condition score) and (2) validate the use of an existing prediction equation for BW in gilts (HG2 × L × 69.3 = HGLW), only used for finishing pigs. Data set A (derivation, Large White × Landrace) included 42 gilts at first insemination, 45 gilts at the end of first gestation, and 58 gilts at weaning. Data set B (validation, Large White × Landrace) comprised of 14 gilts at first insemination, 15 gilts at the end of first gestation, and 19 gilts at weaning. Models were developed for each physiological state but a better BW prediction was obtained from an overall model, including an adjustment for physiological state (S1 and S2): −168.89 +1.06L +1.28HG +58.02S1 +33.03S2 +10.92BCS −1.10BF2 (adjusted R2 = 0.90). This model was validated under conditions found in the present study. Estimations using HGLW showed greater residual means than regression models.