One 20k easy run versus two 10k easy runs on the same day

Nobody really knows. The general aerobic effect is the same and the two runs tend to be more tolerable (you can’t look at efforts in isolation. You need to look at them as part of a plan).

There is research that says 1 longer session gives more aerobic enzyme production. There are studies that suggest you will get more hgh and test. With 2 sessions. As I said, nobody can really tell you exactly which one is better.

There is also some research showing that doubles separated by at least 5-6 hours are beneficial for bone strength.

If you study professional runners (1500 - half). The very best do two easy runs. But they run so many kilometers each week, so it is hard to see what benefit they would obtain from a very long run. For hobby runners, it might be a different story

The fitness benefits you get from a run accrue more heavily in the early part of the run and diminish as the run goes on. So in that hypothetical 20k run most of the improved fitness came in the first 10k. That suggests that you'd get better overall fitness from that day's 20k by stopping after 10k and doing another 10k later in the day. There's a reason most elite runners run twice a day. Of course the same thing happens in those 10k runs. You get most of the benefits in the first half of them so why not do four runs of 5k rather than two of 10k or one of 20k? There has to be a minimum length of time spent on a run for it do anything.

That said, there are desirable things that happen on longer runs. There have been plenty of threads here about what those things are and no shortage of internet articles explaining some of those benefits so your best bet is to have a mix, days when you do the two 10k runs and days when you do the single 20k, or longer run. And it would be useful on that 20k day to do a second run of 10k or so. It seems like nearly everyone who asks this question wants to treat it as an either/or thing and it's not.

Aerobic improvement declines between 60 to 90 minutes then falls off a cliff after 2 hours.

In one of those “Dear Distance Gurus” articles on the front page, John Kellogg and Rojo talked about how running at least 60-65 minutes was incredibly beneficial. Best bang for your buck, so to speak. However, a more common trend these days, likely due to Jakob’s training, is to do two runs of 6 miles or 10k. Perhaps a better question is, which is better, 6 and 6 or 8 and 4? Or 45 and 45 or 60 and 30?

I read a quote from Nic Bedeau several years ago saying that if you ran for an hour every day you'd get quite fit. If you got one of those runs to 90 minutes or more you'd get even fitter and if you got a second of those runs to 90 minutes you'd be extremely fit. Ron Clarke thought that your main run should last at least an hour. I always had my main run last at least an hour when I was doubling with a second session of varying lengths but usually in the 30 to 60 minute range.

The lengths of runs that the OP asks about, 10 or 20 km, are not likely to take and hour but for a lot of people the 10 km will come close so I only mentioned those two distances in my response. But if I were going to start doubling again and was looking to get about 12 miles in on a normal day I'd probably go with 8-4 or 7-5.

A lot of the reason pro runners do this is because they already have such a large aerobic base. From what I understand, aerobic stimulus is enhanced from 1 run, and recovery is enhanced from splitting it into 2.

For many pros who split into 2 run, I don't think it's cause and effect, because they split into 2 runs they have become pro; it's because they already have the massive aerobic base to become a pro runner they can afford to split into 2 runs.

If you are new to running it is much better to do 1 run for all your mileage to better establish the aerobic base. (Also, from what I have seen if you are a low mileage guy racing 800/1500 splitting mileage is a lot more useful than it you are a slow twitch 5k/10k guy, assuming no doubles are needed for extra mileage)

I've done a fair amount of both. There are different advantages to both strategies. Which provides the "better" stimulus or is otherwise the most sensible strategy is probably more dependent on an individual's ability, training history, and lifestyle than the target event itself. Certainly, there is still more value to long singles as the target event distance increases, I just think that hypothetical is less relevant than just doing whatever allows us to get the most training done within the practical reality of our specific circumstances.

The two 10k runs are way easier physically and mentally to get done, and the frequency of training seems to help with improving durability. There's research that suggests that we get most of the musculoskeletal stimuli and hormone production at relatively short run durations, so with shorter and more frequent runs we get more of the "get more durable" stimuli relative to the amount of damage we take on.

The 20k singles are probably better for aerobic development and fat metabolism. For the mitochondria to increase their volume some research has suggested that they seem to really like long, fairly easy efforts. I've certainly experienced big jumps in aerobic ability after consolidating already high volume into less and longer runs.

An order of operations I've used recently and generally recommend to people who want to run a lot and get really fit:

1. Get used to training 2x /day at least 5 days /week with easy cross-training doubles
2. Replace those cross-training doubles with short runs.
3. Extend the shorter doubles until you have 2x solid but still very manageable runs at least 5 days /week, as needed get more volume by gradually extending your personal definition of 'solid but manageable'
4. Consolidate the same volume into longer singles or a different double scheme (i.e. go from 10/10 -> 16/6)

Take the safe/easy way to build up to running a lot, and then from there, you will have a level of fitness and durability that will allow some good experimentation to figure out what works best for you.

Doesn’t JK state there are special benefits for going over an hour compared to 45-50 minutes? Though I’m not sure how much is from the additional hour of running that’d be per week compared to the true aerobic benefits from the distance.

If you’re not otherwise training the aerobic system the way you do in longer runs, it will add a lot. Short runs alone don’t work.