A series of articles addressing the argument that there are simply too many damned dinosaurs in some assemblages. Darren has put together 8 separate pieces dissecting each of the most common concepts discussed, specifically as they relate to the sauropod assemblages of the Jurassic Morrison Formation. I'm just going to add notes from all of them here. Full list of sources below:

• TMMD is often lumped together with the ontogenetic morphing hypothesis, the idea that some species are simply different stages in one phase of growth. There are some merits to these ideas, but overall applying it broadly seems to be a stretch.
• Part 2
• Fossil genera of mammalian megafauna have routinely been oversplit and later synonymised. That has happened to rhinos, brontotheres, and others. However, many of those supposed species were never truly considered. Take brontotheres: most species proposed in the early 20^th century were almost immediately questioned by specialists and most had been sunk into synonymy by the 1980s. Ditto American rhinos, which were being questioned from the '70s, it just took a while for a researcher to actually publish a review. Both groups also benefit from having easy markers within their morphology to test against, and both have highly unusual features (their horns) which led to gross oversplitting in the first place. Arguments on indricothere oversplitting fail to account for the fact that, even when regrouped, there are still half a dozen recognised species (though genera has been reduced significantly).
• Applying the logic of megamammals to Morrison sauropods fails. Unlike these groups, sauropod specialists have reviewed the Morrison fossils multiple times and not concluded that they are oversplit. That includes fairly detailed morphometric tests – like those undertaken on American rhinos – which failed to highlight synonymy. What's more, whilst several species in mammalian history were oversplit, even with these recombined those assemblages still contain dozens of contemporaneous megafauna.
• Part 3
• Arguments that giraffe diversity is a good analogue fail in several ways. First of all, that there is only one extant species of giraffe is now under debate, with potentially eight recognised from genetic studies. Further, those species are almost impossible to distinguish from skeletal remains, whereas sauropod species are clearly delineated based on morphometric analysis and skeletal features showing high degrees of variation. If anything, then, giraffes are an argument that we have too few damn dinosaurs. Oh, and of course modern giraffe ranges are a bad proxy for anything, as they have been greatly reduced by human activity. Plus, fossil evidence shows that before the Holocene multiple giraffe species did live alongside one another.

But however many giraffe species there are, the fact is that the variation reported for them thus far is substantially and dramatically less than that reported for Morrison Formation sauropod species and genera.

• Part 4
• The Morrison Formation spans a time period of around 7 million years. That's a lot of time! When you break down sauropod taxa by time brackets, there are rarely more than 2 species found in a given strata (though the most is 5 at Howe Quarry).
• The Morrison also spans a wide range of biomes, with evidence for wetlands, saline lake systems, deserts, floodplains, highlands, and more.
• Part 5
• What about sheer biomass limits? Was there enough physical space and food for multiple viable, large populations of megaherbivores the size of sauropods? It looks likely. The Morrison is roughly 1.2 million square kms; again, most species were not directly in competition but divided by habitat and time; and when modelling into metabolic requirements – assumed for both mammal-like metabolisms and varanid-like ones – there is enough vegetation and space for populations in the high thousands to low millions, which is more than enough for genetic diversity.
• On top of which, new palaeoclimate data indicates that we have historically underestimated the productivity of this region; it was not simply a perpetual savannah but contained dense forests for large periods of time as well.
• Sauropods also appear to have an unusual breeding strategy for megafauna, with egg clutches routinely being found with 20-40 eggs and often in close proximity to other nests, implying timed breeding events on a large scale. That implies that actual populations could have been predominantly juvenile and subadult, rather than the giants we imagine, and means that populations would have been fairly quick to recover from major die-off events like droughts. Evidence for this also comes from trackways and the prevalence of juvenile fossils in almost all sauropod-yielding locations.
• Part 6
• Sauropods also appear to have reduced ranges than would be expected when comparing them to mammalian megafauna. There are various reasons for this, but it actually appears to be a common trend across a lot of herbivorous dinosaur groups. As such, the justification that megafauna alive today tend to have continent-wide distributions may not apply to dinosaurs at all (also, I don't think this is even accurate; Africa has multiple species of elephant, for example, some of which until very recently practically overlapped, such as the Knysna elephants, and multiple rhino species, even with human habitat fragmentation).
  
• From the comments on part six: it's possible that nesting behaviour led to reduced species ranges. We know a lot of extant nesting species return to the same areas for centuries, if sauropods did the same thing then it would reduce their potential geographic area and drive allopatric evolution. Cool point.
• Also an interesting discussion around the nutritional value of wood. Today, most wood is "digested" by ants/termites and fungi, but there is evidence that these groups were much less effective (or absent entirely) in the Mesozoic. Whilst we know of fossil cockroaches and beetles that clearly digested wood, there may have been more available energy in bark than we credit palaeoecosystems with; hadrosaur coprolites also verify that at least some species of dinosaur did occasionally eat wood. Secondarily, we fail to account for the fact that humans burn wood routinely and have done for a long time. Apparently, if you include the energy we use in wood fires, we actually fall close to a megaherbivore in terms of biomass use.
• Part 7
• Were Mesozoic plants poor energy reserves? Cycads are, sure, but there were plenty of other plant groups around. Research has shown that fossil conifers, ginkgos, horsetails, and ferns were all highly nutritious; in fact, certain monkey-puzzles and conifers likely matched extant grass groups for energy value. We also often forget about the sheer volume of plant life even in modern-day fern meadows, such as bracken downs in the UK. Plus, sauropods (like most megafauna) likely had a "quantity over quality" feeding strategy combined with long, drawn-out digestion periods. Also, the sauropods clearly existed, even if they were all one taxa they would still eat the same amount of food... (okay, so lumping species would reduce the viable population size, but still there was clearly a lot of them kicking about).
• Part 8
• Basically, questioning the experts is rude. They are experts. Their theories have been tested by other experts. To date, no major flaws have been found.