Other methods are more applicable because we actually have data on what works and what doesn't with respect to soil loading, farmers really do not want to damage their land ('damage it in an hour, take a decade to recover') and have learned the hard way not to overload the soil already.
And secondly, 'sustainable' to a heart of wild beasts does not mean harvest every year - maybe the heard comes back in 20 years when soil has recovered.
Lastly, maybe they wheren't sustainable, after all they are extinct.
I mean, they existed for tens of millions of years (according to wikipedia), which is 1000's of times longer than the entirety of human farming; I think this easily classified as "sustainable".
Mass and ground loading is just one part of the picture. Roam area, root structure, etc make a difference.
If you can digest tree bark and have 5 square kilometers per person you can damage the soil as much as you want, something will still grow.
But if you want civilisation to survive, we need a regular harvest of 40 tons per hectare for potato, and if that number falls to 20 there is a famine.
You are the one advocating we live like Sauropods, so you should be telling us what was the roaming range of one - a male bobcat has roaming range of 20 to 70 sq. Kilometers
It would be fair to object that what was unsustainable by plants then might not be a problem for more modern plants, but I'm not sure what your comment about other methods aims at. The uncertainty comes from machines reaching previously unexplored weights, comparing with soil effects from lighter methods is unlikely to tell you if there's something bad coming.