Estimation of off-farm temperatures from meteorological observation records and their level of accuracy

Abstract

The purpose of this study was to devise an optimal method for estimating outside temperatures of pig farms using air temperature data obtained from multiple meteorological stations. Using daily average, daily maximum, and daily minimum temperatures recorded at nine selected Japan Meteorological Agency meteorological stations, we used data masking and inverse distance weighting to estimate temperature based on data from 1-8 stations in the vicinity of each point. The daily range was estimated from the daily maximum and minimum temperatures. Our results showed that daily average and daily maximum temperatures can be estimated with high accuracy by using data from the three nearest stations. The absolute value of the mean error and the root mean square error for all temperatures are low from June to October and the correlation coefficient is generally high in September. Based on these results, we conclude that the daily average and daily maximum temperatures can be estimated with high accuracy using this method, especially in hot weather.