Maximum permissible load for Kiso horses trotting over a short, straight course.

This study aimed to determine the load-bearing capacity of trotting Kiso horses using gait analysis. Ten Kiso horses with a height at withers of 128 cm were included. Their riders were fitted with a marker (70 mm in diameter placed on their chest) recorded by two digital DVD cameras while trotting along a short, straight course. In total, eight tests were performed for each horse: the first with a 70 kg load; six with randomly loaded weights ranging 80-130 kg; and then a final test again with a 70 kg load. Three-dimensional movement of the marker was analyzed using a motion capture system. The time series of vertical displacement of the marker underwent spectrum analysis by the maximum entropy method, and the autocorrelation coefficient was calculated. The first two peaks of the autocorrelation were defined as symmetry and regularity, and their sum was defined as stability. Regularity in the 120 kg test (0.54) was lower than that in the first 70 kg test (0.61), and stability in the 120 kg test (1.31) was lower than that in the first 70 kg test (1.42). We concluded that the maximum permissible load for a trotting Kiso horse is < 120 kg, which represents 31% of its bodyweight.

Method for estimating maximum permissible load weight for Japanese native horses using accelerometer-based gait analysis.

The aim of this study was to establish a method for estimating loading capacity for Japanese native horses by gait analysis using an accelerometer. Six mares of Japanese native horses were used. The acceleration of each horse was recorded during walking and trotting along a straight course at a sampling frequency of 200 Hz. Each horse performed 12 tests: one test with a loaded weight of 80 kg (First 80 kg) followed by 10 tests with random loaded weights between 85 kg and 130 kg and a final test with a loaded weight of 80 kg again. The time series of acceleration was subjected to fast Fourier transformation, and the autocorrelation coefficient was calculated. The first two peaks of the autocorrelation were defined as symmetry and regularity of the gait. At trot, symmetries in the 100, 110, and 125 kg tests were significantly lower than that in First 80 kg (P < 0.05, by analysis of covariance and Sidak's test). These results imply that the maximum permissible load weight is less than 100 kg, which is 29% of the body weight of Japanese native horses. Our method is a widely applicable and welfare-friendly method for estimating maximum permissible load weights of horses.