The Reconstitution and Modelling of the Work Capacity Above Critical Power Following Severe Intensity Cycling

Abstract

The two-parameter critical power model comprising critical power (CP) and W′ is well accepted as a mathematical model representing exercise in the severe intensity domain. CP represents the maximum work rate derived from aerobic metabolism and W′ the fixed capacity of work above CP. However, within competitive cycle sport few races are performed exclusively within this domain, instead stochastic efforts where W′ is repeatedly depleted and reconstituted typify race demands. Relatively little is known about the reconstitution mechanisms and kinetics of W′ hence the initial aims of this thesis were to develop a reliable method of assessing W′ reconstruction and evaluate likely underlying physiological contributors to the rate of W′ reconstitution. Thereafter, the aim was to develop a dynamic model of W′ reconstitution and depletion which could contribute to race planning and tactics leading to improved sporting performance. A repeated ramp test developed for Study 1 to measure the amount of W′ reconstituted following its full depletion was found to produce reliable results for 2-min recoveries at 50 W (ICC ≥ 0.859; TE ≤ 559 J; CV ≤ 9.2%). A slowing of W′ reconstitution following the repeated bout was evident in this and all subsequent studies. Study 2 found that W′ reconstitution after 2-min recovery was related to measures of aerobic fitness such as V̇O2max (r = 0.81) and CP (r = 0.52) in trained cyclists, whilst the reconstitution of W′ was more related to fat mass in untrained participants (r = -0.70). Studies 1 & 2 also demonstrated existing mono-exponential models of W′ balance did not fit the results obtained after 2-min recovery periods. Therefore, Study 3 compared existing mono-exponential models to a bi-exponential model, finding the latter a much superior fit (AICc bi-exponential: 72.2 versus bi-exponential: 30.2) of the temporal profile of W′ reconstitution of trained cyclists. The resultant model demonstrated that W′ comprised distinct fast and slow components that were unrelated to each other. Study 4 assessed the likelihood of a minimum recovery power output beyond which no further improvement in the rate of W′ reconstitution was apparent. Study 5 investigated the effects of different recovery intensities on W′ reconstitution. Fitting the results of the different recovery intensities into the bi-exponential framework of Study 3 allowed a full dynamic model of W′ reconstitution and depletion to be built allowing for both duration and intensity. The model can be customised to an athlete using the known parameters of CP and W′ together with a single additional test session including a 4-min recovery at 85% of CP. The model can be applied in real time for use by cyclists in competition to aid tactical decision making and optimising race performance.