[NotAtWork]

> a world with just farms and cities and the animals that can survive within is not going to bode well for us or any other species

In what way?

The main threat is disease if we shift to being large aggregations of essentially the same animals. We've run in to this problem before, with bananas as a particular case. However, we already know the solution to that. The answer is that you preserve several cultivars of the species, and intermingle them. We already see this kind of behavior with pets, gardens, zoos, etc. I fail to see why you think we're likely to end up with anything but more of the same kinds of pruning-but-not-completely-destroying-families behavior.

I'm also failing to see why you think we need thousands of kinds of lizards that are all basically the same, rather than a few hundred, and why such a bottle neck is either unnatural (hint: we've hit smaller choke points before) or why it would be particularly dangerous.

Ed: A key word.

[schiffern]

>In what way?

Because we are all completely dependent on ecosystem services provided by those species, such as erosion control (without which your country erodes away to desert), rainwater buffering (without which you experience catastrophic flooding), and transpiration (without which you lose almost all inland rainfall[1]).

Loss of individual species is just the symptom. The real problem—and what this article discusses—is wholesale destruction of wilderness.

Don't imagine that just a handful of species can provide these services either. Since ecosystems are complex adaptive systems, a particular species' role is almost always subtle and interconnected. See: the services provided by wolves in Yellowstone, which were never fully understood until they were removed and then re-introduced[2].

So why is wilderness important? Fundamentally, wilderness is arranged (and so, it functions) in completely different ways than human-tended landscapes. Now it's obvious that a suburb is different from a forest, but what's less obvious is that the way wilderness works is much more efficient. It's not dependent on a constant stream of material extracted from "somewhere else", but on average it produces far more economic value when ecosystem services are accurately accounted for.

Surprise, surprise: economics seems to suggest that those living on a spaceship shouldn't take a sledgehammer to the life support system…

(of course, the ultimate trick would be to design human landscapes that also function like wilderness ecosystems)

[1] http://www.nature.com/nature/journal/v496/n7445/abs/nature11...

[2] https://www.youtube.com/watch?v=ysa5OBhXz-Q

[NotAtWork]

I don't think your evaluation of human ability to cultivate environments realistic.

The vast majority of the external effects of forests and wildernesses is the effect of a few major species. In nature, those species are supported and interacted with by a complex network of other species. However, this established relationship isn't necessarily optimal for the few species of interest to us.

I find it likely that we're able to cultivate select main species of interest in forests if we set our minds to it, especially considering the large scale agriculture that we already do.

There's simply nothing to suggest that you can't construct an ecosystem with many fewer elements or the elements rearranged to, say, make room for houses, that has the same external effects as other ones. In fact, the success of many environments-in-a-bottle, indoor marijuana production, etc, suggests that we do have the ability to make relatively stable environments in which the necessary components of an ecosystem can thrive.

No one is suggesting that we do something silly like wipe out all the major predators while we let herd animals run free (which is the case with wiping out the wolves in Yellowstone), but rather that we can get away with a lot less moving parts and that we can tune the parts quite a bit to suite our fancy.

The only large scale human dependencies on plants relate to weather, oxygenation, water flow, and soil control. The last two we know we can do with intentionally seeded groves and other such constructs of plants we choose (using a reasonable selection), and don't need to replicate the full array of plants. In terms of oxygen, seeding the oceans with an algae would be far more efficient, but we really only need ferns, which are incredibly efficient at producing oxygen and are relatively hardy plants.

The final complexity is weather, which I must admit I know relatively little about, but am dubious there's any material reason it wouldn't work fine with planned forests.

Again, no one is saying "Fuck it, kill all the things". I just think we can get away with many fewer species - and sometimes who groups of species, where another can reasonably fill its role.

Or are you telling me we couldn't survive with 3,500 kinds of beetle... we really need all 350,000.

[dredmorbius]

we do have the ability to make relatively stable environments in which the necessary components of an ecosystem can thrive.

Only with the inputs of massive amounts of energy and additives -- fertilizers, pesticides, and cultivation supplied directly by humans (or our machines).

The energy intensity of modern ag is many, many times higher than of natural environmentments. Food production in the US requires ten calories of fossil energy for every calorie of food energy produced, in Europe it's closer to a 5:1 ratio.

A sustainable agricultural system would require that the output energy be greater than the input.

[kazagistar]

We have only been at it for so long, and barely gotten started in earnest. Plus... energy and additives are not "external" to the system: they are provided by a natural species in the system, us.

Yes, we do need to get to a point where we are not relying on expendable reserves to run the system. But it certainly is possible that we will engineer a better ecosystem (robot cultivators and solar panels included) that results in a more efficient net benefit for us.

[dredmorbius]

May I suggest Howard T. Odum's works, in particular Environment, Power, and Society and Energy Basis for Man and Nature.

In the former he argues strongly about the mechanisms by which humans have enhanced ag productivity in plants and animals.

Generally, there are the following methods:

Mechanical tillage, breaking up soil to make it easier for plants to grow and spread roots. This also, incidentally, increases topsoil loss to wind and water, such that many farms are effectively "mining" topsoil faster than it's being replaced.

Artificial irrigation. This varies from simply collecting and distributing water via gravity-flow reservoirs and irrigation ditches to transporting water and irrigation pumps and pipes to water mines which, again, deplete a resource faster than it is restored -- as is the case throughout the eastern Plains states in the US, much of China, and especially in the Sahara and Arabian penninsula where water tens of thousands of years old is used to irrigate crops, from underground reserves which aren't being replenished. Water availability itself is becoming a significant concern, with major droughts in the past 5 years disrupting crops in Russia, the United States, China, India, and of course, as is rather chronically the case, Africa.

Fertilizer. Nitrogen, fixed at great energy cost from the air using fossil fuels (mostly natural gas). Phosphorus, which is in extremely limited supply. Potash, rather more abundant, but still with only a century or three of reserves at present rates of use.

Selective breeding. Plants and animals have only so much metabolic budget. By diverting energy away from specific uses, especially immune response, physical activity, and foraging needs, more can be devoted to growth. This works to an extent, but is greatly facilitated by ...

Antibiotics and pesticides to reduce illness and parasites. Fun fact: the first virus identified wasn't a human illness but the tobacco mosaic virus. Antibiotics and pesticides mean that animals and plants need devote less of their own energy to competing in their environment. Unfortuately, both ultimately create resistance, a problem later to both the ag products themselves and quite possibly humans, especially in the case of antibiotics. Moreover, bred cultivars requiring such treatments don't compete where they're not available (similarly for fertilized crops, above).

Mechanical pest reduction. Removal of weeds, or native long-lasting plants which compete for ag lang productivity (e.g., natural plains, tropical rainforest).

Solar panels compete directly with plants for solar energy. At best you want to put them in regions plants cannot grow.

The history of ag enhancement is relatively brief, but it's all been accompanied either by vast investments of energy, or by the application of either materials or technologies themselves requiring or based on vast applications of energy. Even the father of the Green Revolution, Norman Borlaug, cautioned that he'd only provided at best a brief respite from hunger.

It's not so long ago that major famines still ruled the world, with major instances in the 19th century (Ireland 1845-52 killing 1.5 million, China 1850-73 with a population drop of 60 million), and 20th (1920s in Russia, 5 million, and China, 3 million, 1930s Ukrain Holdomor, 7-10 million and China, 5 million, and the Great Chinese Famine of 1959-61, 15-43 million). And that's just a set of highlights, see: http://en.wikipedia.org/wiki/List_of_famines

In many cases, 30% of regional populations died (or in some lucky cases, emigrated elsewhere, as in Ireland), in others historically 50-90% of populations were wiped out. I'd suggest you not think this cannot happen again.

I'll also advise you that this is a topic I studied, extensively, in school.

[NotAtWork]

Some of your problem predictions are 100+ years in the future. I generally regard these as nonsense.

Looking at the relative rates of technical growth and computing power, it's incredibly unlikely that we'll be able to accurately pick out what the future is going to bring.

Examples of technologies that are expected in the next 50-100 years: 3D printable organs which can be transplanted, based on your own stem cells; the first smarter-than-human general purpose AI; fusion power; the ability for bioengineering to be done with a home lab kit. (We're actually at the cusp of the first and last of these now.)

That level of bioengineering, computing prowess, and cheap power will have an incredibly hard to predict effect on issues like food production, ecosystem maintenance, etc.

So which of your warnings are only problematic at 100+ years?

[sk5t]

Mass agriculture works despite of our enormous ignorance of how the ecosystem works... you really shouldn't take our ability to manipulate corn or soybeans as indications that humans know how to create "relatively stable environments." We haven't and we don't.

[NotAtWork]

Neither of those were examples I cited, and I cited two other, specific examples.

I have the feeling you didn't read what I wrote.