Searching threshold temperature for farrowing performance traits of pigs in Japan using public meteorological data

For robust modelling of heat load, a plateau-linear regression analysis was performed using 52,668 farrowing records from 10,320 purebred Landrace female pigs reared in six farms in Japan obtained during 2007 and 2022 and public weather data. Daily maximum temperature data observed at Japan Meteorological Agency meteorological stations nearest from the farms were downloaded. Traits studied were total number born (TNB), number born alive (NBA), number stillborn (NSB), total piglet weight at birth (LWB), mean piglet weight at birth (MWB), and piglet survival rate at birth (SVB). Coefficient of determination of the plateau-linear regression model was obtained with varying the breakpoint value was changed from 10.0°C to 25.0°C at 0.1°C intervals, and the value showing the greatest model fitting was reported as the threshold temperature. Daily maximum temperature data at mating day was used for TNB, NBA, and LWB; those at farrowing day was used for NSB, MWB, and SVB. The threshold temperature detected was 17.7°C for TNB, 16.6°C for NBA, 22.9°C for NSB, 17.1°C for LWB, 10.0°C for MWB, and 22.9°C for SVB, and the estimated regression coefficient was negative for TNB, NBA, LWB, and SVB and positive for NSB and MWB. The analysis seemed not to perform well for MWB in this study. The results could give the fundamental information on developing the statistical model for national breeding value evaluation for pigs to cope with global warming.