A study on the techniques used by untrained horses during loose jumping

Deterministic models1 developed for the jumping horseindicated the important factors involved when jumping an obstacle2. SVHS video recordings were obtained of 31 untrained horses (age: 3–5 years, height: 164.7±4.5 cm) jumping loose over a fence 1 m high by 0.5 m wide. The horses were designated to either a good group or a poor group based on a qualitative evaluation; good horses (n=18) cleared the fence with ease, and poor horses (n=13) consistently hit the fence. Video sequences were digitized to provide kinematic data on the horsesʹ center of gravity (CG) and carpal and tarsal angles. Twenty kinematic variables were examined from the approach to the landing. Analysis of Variance (ANOVA) revealed significant between-group differences for the horizontal velocity of the last approach stride (Good: 5.77±0.80 m.s−1; Poor: 6.42±0.95 m.s−1; p=0.046). Significant differences were found in the relative carpal angles at take off (Leading limb: Good: 1.02±0.19 rad, Poor: 1.25±0.28 rad; p=0.010; Trailing limb: Good: 0.92±0.21 rad, Poor: 1.06±0.15 rad; p=0.046). The height of the CG over the center of the fence was also a significant variable that differed between the groups (Good: 1.83±0.08 m; Poor: 1.71±0.13 m; p=0.002). Finally the horizontal velocity of the landing was significant (Good: 5.26±0.92 m.s−1; Poor: 6.27±0.84 m.s−1; p=0.004) along with the angle of the CG to the ground at landing (Good: −0.45±0.08 rad; Poor: −0.38±0.07 rad). The velocity and CG variables which distinguished good and poor horses are likely to be strongly influenced by a rider; therefore, it is unlikely that these data alone could be used to predict elite jumping horses. The carpal angle data, however, may indicate a certain natural tendency by the young horses in the good group to keep their legs clear of the fence.