What is a threshold?Before taking a look at the 'results' of our experiment, we should define what we're looking for to declare a 'threshold'. Traditionally a bar used to keep straw inside a house, the word now is used to mean a definite limit beyond which a process cannot continue. The most famous threshold is the 'anaerobic threshold' - intended to represent the point at which aerobic metabolism can no longer sustain exercise alone.
In theory, therefore, on one either side of a threshold we should expect to see marked changes in speed as we move from one physiological system to another. Experiments should be able to verify other changes associated with this change, and we should be able to construct some form of explanation as to why. With that in mind, we can approach the data.
|Running world record speed versus distance|
Phosphocreatine ThresholdOur first three world records (60m, 100m, 200m) were all run by the same guy, and all at about the same speed (the lower speed for 60m is almost certainly just due to reaction times and having to accelerate off the start). It's fair to assume that this represents the maximum speed humans are capable of - coming from the high-speed phosphocreatine system. Beyond 200m/20secs (first green triangle), we are reliant on replenishing our ATP another way - by one of the NAD-limited processes (of which only glycolysis can regenerate the NAD without oxygen). Traditionally we say 10-15 seconds of maximal effort for ATP plus phosphocreatine; it doesn't seem inconceivable that Mr Bolt can manage 20. So between 200m and 400m we have our first threshold.
VO2max ThresholdI wasn't sure if I could really call this VO2max - it appears to be somewhere around the 1000m to 1-mile mark (130 to 220 seconds), which surprised me as traditionally VO2max pace can be sustained for 6 to 10 minutes. In these short races, the athletes will be using oxygen at or close to their VO2max, but are running faster than their VO2max pace. The extra ATP must be coming from glycolysis - accompanied by a gradually worsening acidosis (and lactate, with no spare Krebs cycle capacity to use it). This acidosis is presumably what stops the athlete going any further, and the rate of glycolysis limits the speed.
Anaerobic ThresholdThe most famous threshold of them all - and in great use for training. particularly by cyclists as FTP. Traditionally pace at this threshold can be held for an hour - so we should be looking at the half marathon (21km), and expecting to see the graph flatten off - as pace just below this 'threshold' should be maintained for longer. In fact, we see nothing of the sort. Pace continues to gently drop until we hit that Blue Riband event - the Marathon (green triangle). That the pace there sits slightly higher than might be expected is, I believe, a product of focus on this event (over the 40k and 50k) rather than anything special going on physiologically.
Glycogen ThresholdI wasn't sure what to call the threshold that seems to exist at about 50k. Speed evidently takes a drop off after that, but it's difficult to tell how much we can blame on the fact that ultramarathon running simply isn't very popular compared with the shorter distances. Physiologically though, I wonder if one of the limiting factors around here is the fact we only store about 2000 calories of glycogen, so long races are likely to necessitate increasing amounts of fat burning. Whilst even the glycogen at this pace will be used aerobically (as we're well below VO2max), crucially the use of fat means that the lactate shuttle cannot come into play - so every muscle cell must be aerobic individually rather than allowing lactate to shuttle an 'oxygen debt' into a less hard-working cell. This may therefore be the true 'lactate threshold'!
Others?I hope this reasonably explains what thresholds I think can be identified from the data. I imagine beyond these there may be a 'sleep threshold' as races start to creep beyond 24hrs, and some kind of 'muscle breakdown threshold' for ones even longer than that!
Next time, I'll be looking at some mathematical thresholds.