…if I keep my HR in Z3 during the downhills. This will be a suicidal pace. Which I think is too much pounding for the body to handle.

Yes, I agree. I wouldn’t go at Z3 on the descents. BUT… if you’re racing and the race will include downhill sections, then you want to train them, but probably not by HR. I would start with 6x 30″ “fast” with easy recoveries. You can then progress the workout with more or longer reps.

I’ll also ask one of our coaches (Alison Naney) to comment. She actually likes running downhill (while I hate it…) 🙂

Or do I just let my HR drop back to Z1/Z2 and then accelerate as soon as possible? (And maybe add some time to the rest of the remaining time. If this is necessary?)

I much prefer doing all intensity above aerobic threshold on a treadmill by pace. I prefer the precision over the fresh air… 😉

But! When I’ve done these types of workouts outside, I would do them on one long continuous hill, going up for the work period, and then trotting down for the recovery interval. So if the hill is long enough, you can go up and down in shorter sections without reaching the top before you’re done.

Does that make sense?

I’ve done lots of treadmill training above aerobic threshold with different intensities varying in small increments between 100% and 130% of AeT pace. (130% is similar to a 4-minute time trial or “VO2” pace.)

In all cases, I’m sure that there is a duration where heart rate would reflect the actual pace. But due to heart rate lag, it’s impossible to estimate the pace by using heart rate.

For example, let’s say you wanted to do a workout with work periods at AnT and recovery intervals just under AeT with 6x 3′ work and 3′ recovery. On more than one athlete, the actual peak heart rates per set were:

* AeT -10
* AeT -5
* AeT -5
* AeT -5
* AeT
* AeT

So although the load for each work interval was right at anaerobic threshold pace, the lag in heart rate was at or less than aerobic threshold. So when athletes are training by heart rate and targeting an anaerobic heart rate with short durations, the actual load must be much higher.

Note: This would only apply to athletes with very small AnT/AeT spreads.

I don’t think that you’ll wreck your training by going by heart rate. It’s only a question of how precise you want to be. The imprecision of heart rate is only at high intensity, and if your thresholds are close, then you must be doing a good amount of easy volume (which is more important).

I first starting using treadmill intervals after reading a bunch about Roberto Canova’s training paces. I converted all of his paces to percentages of AeT and then did many of his workouts. That’s when I discovered that my heart rate had little in common with the intensity.