He actively solicited thoughts on this topic and well, I just had to chime in. Long time readers will know that I don't fall in line with the ubiquitous assumption that larger dinosaurian herbivores, especially sauropods, were the large scale vegetative destructors they are often assumed to be. Here is a brief review of my arguments in favor of sauropods having a lighter touch on vegetation than generally assumed.
1) Weaker jaws and dentition than elephants- which routinely strip bark and girdle massive trees outright killing them.
2) Tree toppling is dangerous, even for large sauropods. An elephant has a much more compact build than a sauropod and is better equipped to topple trees. A sauropod, like camels, ratites, moas, and giraffes (all of which do not tree topple) would have simply reached up to grab the food it wanted with its neck.
3) Overbrowsing by sauropods on a massive extent was inhibited by their constant movements. In Camarasaurus migratory patterns have been found through isotopic evidence, But all sauropods would have been kept on the move by a combination of predators, parasites, water, food, and reproductive factors. Large tress were subjected to pressure at sporadic intervals but not constant pressure.
4) The largest sauropods were rare and even among more modest sized species the reproductive strategy of dinosaurs suggests a wide spectrum of sizes in a population at any one time- the vast majority of which would be less than full grown.
5) Very little in the adaptive arsenal of modern gymnosperms suggests that they sustained rampant overbrowsing in the past on a scale unsurpassed today. On the contrary gymnosperms tend to grow slower than angiosperms. Many, such as cycads and tree ferns, will outright die if their growing tip is severed.
6) African Elephants serve as ecosystem engineers whose actions of tree killing and brush clearing benefit a myriad of other, smaller critters. Many sauropods, on the contrary, existed within diverse browsing guilds of other sauropods. It does not make ecological sense that multiple species were all acting simultaneously as agents of tree/brush clearing in their environment. No, some species were much softer handed than generally assumed.
7) Keep in mind that sauropods generally lived on smaller island continents than African Elephants. And they did so in greater diversity and at greater size than elephants.Yet they did not eat themselves out of house and home.
8) Uric acid is less water soluble than urea. Mammals can simply "piss themselves to death" and are highly dependent on water compared to uric acid producing reptiles/archosaurs/birds. Elephants, beholden to water, will heavily browse vegetation surrounding water resources during dry times.
If you want to look at analogues for how sauropods interacted with the plants around them look at their closest herbivorous archosaur relatives- ratites and moas. These long necked browsers browse(d) across a wide swath of vegetation on a wide variety of plants. The nature of their feeding discourages heavy browsing. Camels and giraffes offer perhaps the next best analogy to sauropod browsing habits. While they do create browse lines, they are not known for tree toppling despite certainly having the strength and size to do so. While sauropods certainly had the capacity and strength to tree topple, i am dubious that it would have been an excessively common occurrence- especially since it would have been a lot easier to have used their necks to reach the food that they wanted.
Tree killing, slow reproducing, intelligent, water dependent, and resource intensive- Elephants offer perhaps the roughest approximation for sauropod behavior and ecological role in my view.