Modeling the Influence of Crew Movement on Boat Velocity Fluctuations during the Rowing Stroke

Caplan and Gardner (2007: Journal of Sports Sciences, 25, 1025-1034) presented a rowing simulation which assumed that each rower moved as a single mass. Although validated against mean onwater shell velocity, instantaneous velocity was not modeled well. As any changes to rowing technique inherently influence the fluctuations in boat velocity throughout each stroke, the previous model was unable to fully explore the influences of technique on the progression of the boat during each stroke. The present study aimed to address this limitation by implementing a five-segment model of each crew member, in order to better simulate the movement of each body segment throughout each stroke, and the influence this movement has on boat velocity fluctuations. A Vicon motion analysis system was used to collect kinematic data of nine sub-elite club rowers rowing on a Concept 2 model C rowing ergometer, from which mean joint trajectories were extracted for the ankle, knee, hip, shoulder, elbow and hand. From this data, kinematic relationships were determined to model the motion of the rowers for input into the simulation. The single segment and five-segment models were validated against previously published on-water mean boat velocity. Both models showed a close agreement with the measured on-water data, although the agreement was improved for the five-segment model. When compared to instantaneous velocity during one stroke for onwater rowing, the five-segment model was seen to better simulate the main features of the on-water data, which were not seen with the single segment model. This development of the model could have practical applications to rowers, coaches and sports scientists as individual rower kinematics could be used to estimate the influence of their technique (or technique changes) on both mean boat velocity and the velocity profile during each stroke.