I'm skeptical. Trees already are pretty efficient at making seed pods, what they lack is the ability to fend off predators from eating the seedlings.
Somewhere downhill and relatively nearby is a 'solved problem' for most trees -- but uphill, further afield, with appropriate soil depth for the seed size, in sufficient numbers, with less competition, etc. Aside -- ridges are one of the most useful places to focus reforestation efforts, in no small part because once established there, trees can generally propagate themselves across the rest of the slope.
Fending off predators is a concern, of course, but that's why you go for the large quantity approach when propagating by seed.
The claim these things can 'plant 100,000 trees a day' is clearly hyperbole. What they can do is plant 100,000 seeds a day, presumably at a much higher density than fully grown trees would be able to thrive, on the assumption that a useful number will actually establish.
While trees are mostly efficient at making new trees, that's not useful in areas that are heavily deforested -- and I suspect that's where this kind of technology would be most useful. Dungog in AU (as noted as a trial site in TFA) is a good example (gold rushes were especially bad for local flora, and in hot arid climates it can take centuries to recover). The breathtakingly vast plains west of Beijing that have been deforested, and now result in massive quantities of top soil being blown into the ocean each year would be a fantastic candidate for this kind of mass spray-and-pray planting.
Also, when re-foresting, pioneer trees are chosen - typically these are acacias (or similar) that have a relatively short life (10-20) years, are hardy, fast growing, and fix nitrogen. In degenerated older forests you may not find many of these around to produce seed naturally.
If the problem is deforestation, let's attack that problem first. Otherwise this is just pushing the dirt around (literally) and hoping this will fix it.
> The system’s engineers estimate that their method is about 10 times faster and only 20% of the cost of hand planting.
But is planting even the slow/expensive part, as opposed to other land-management issues? Is a "mass seedling boom" even a good idea for the emerging ecosystem in the first place?
> possible to plant in hard-to-reach areas that have no roads or steep, inaccessible terrain.
So then how'd they get deforested in the first place? Both agriculture and logging would leave infrastructure. Wildfires? Forest ecosystems have their own interesting lifecycles for that.
The maintenance plan helps to protect the new growth to be eaten e.g. by moose and elk.
This is in Eastern Australia, where we have native wombats, macropods, possums, as well as wild/feral pigs, goats, rabbits, and the occasional deer, along with parrots and fruit fly that tend to destroy fruit with potentially viable seed. But at least we don't have moose, I guess.
Fencing off an area to stop the ground-based pests has impressive observable results even after a couple of years - but it's very expensive, and fences decay within the lifetime of many of the pioneer trees.
So de jure reforestation, while a nice principle, isn't feasible in all environments.
Further complication in many areas desperately in need of reforestation is periodic drought -- Australia and the Americas have the famous El Niño / La Niña cycles to contend with, where we regularly have several consecutive years with minimal (or at best unreliable) rainfall. Again, this is an excellent candidate for a very cheap mass seed distribution system that can simply be re-run each spring until a favourable rainfall pattern produces a good result.
The traditional solution is seed balls. Functionally it's like burying the seed, but the balls can simply be scattered.
This is the scenario I see these drones as being useful for. The logistics of getting people on the ground in these remote areas can be difficult and costly.
Wouldn't have the same concerns for angry wildlife either!