I just saw this on phys.org and wanted to share. I find it satisfying that I am contributing in some very small way.
I scanned both General and Nature Talk and did not see this presented anywhere. If it is elsewhere, then I ask admin to delete. I may also not partake much in any discussions given some current health issues, but my silence does not mean I am not interested.
https://phys.org/news/2024-10-ai-empowers-inaturalist-california-unprecedented.html
You can run … but you can no longer hide.
“With remote sensing, almost every few days there are new pictures of Earth with 1 meter resolution,” Expósito-Alonso said. “These now allow us to potentially track in real time shifts in distributions of plants, shifts in distributions of ecosystems. If people are deforesting remote places in the Amazon, now they cannot get away with it—it gets flagged through this prediction network.”
So, will field botanists be no more necessary? Will floristic surveys be made by the AI? LOL …
Which reminds me of this other comment:
In some ways, we’re already living in a dystopian novel, we just don’t usually notice.
In a non-dystopian world, this is a tool for field botanists to use. To cover areas the budget does not extend to.
In this dystopian world Cape Town’s No Mow till November so our wildflowers can set seeds for next year is greeted with Mow Down ALL the WEEDS.
If only this were true. It would make biodiversity conservation work and studies vastly easier.
OTOH when they poached Cecil the white lion - they had to cut his head off - to remove the tracking collar.
How dim do ‘hunting’ customer and service provider have to be - to poach a collared lion. Office brain out on another assignment.
I often have doubts regarding certain floristic data made by humans, even more so I would be very dubious about a distribution made statistically by a software.
With species that define a specific kind of vegetation in a given area, climate and substrate it could be somehow reliable but, with generalist species that display a broad ecological spectrum, I think, at present, it would be like playing dice.
Anyway, compliments to the author of that paper for the sophistication and complexity of the methods employed.
You cut off the most important part of the quote; viz., “the world really does feel just suffocating.”
I didn’t include that because the reason the world feels suffocating to some is not because of the limited amount of wildlife monitoring that goes on.
It’s because of the monitoring of humans, and the increasing inability to just be left alone to do your own thing. That’s not a wildlife issue.
I’ve been toying with this idea for a while… Run a drone over an area and use AI to accurately map every single visible plant in the canopy. Should be doable.
Of course it will miss all the herbs and shrubs under the canopy, but would still be quite nice.
Should also work for grasslands/shrublands, etc. Just a case of being able to see the plants and having good enough definition for them to be distinguished.
You would receive only a hypothetical species list made on a statistical basis. Would you trust the AI as it is now?
It would obviously require some checking, but there’s no reason an AI couldn’t accurately determine canopy species based on even colour and texture alone.
I disagree. The word feels sufficating because there are entire charismatic species which exist only in managed reserves. This now-closed thread, Wild American Bison are captive? is just one example. In some African countries, most of the megafauna are in the same situation. If an elephant or other large animal escapes from Etosha National Park, it will be rounded up and returned. How about European megafauna? European Bison for instance? This is why the world feels suffocating.
IS a wildlife issue because it describes the situation of increasing numbers of wildlife species.
Well, canopy species is well different from a species list which is a much more complex task and, fortunately, far beyond from theAI possibilities. And I think it will always remain so.
Why fortunately? It would be invaluable to be able to get a fully representative species list via AI reviewing photos or video.
Because I think it is pretty obvious that it is better if humans will not be replaced by a machine. Of course, on the basis of my own values.
well, we all can feel suffocated by different things. Intensive wildlife monitoring isn’t something that bothers me either, nor does imperfect AI plant ID that is still on the same level as that of the many interns and seasonal techs that the belieguered field (sometimes unethically) relies on. the vast majority of wetlands in Vermont which is a rather well populated area have never had an ecological assessment. Many if not most aren’t even mapped despite some mapping improvements. Often the obstruction is politics and funding not ability to map them. I’ll be very happy if we ever get to them all, to me it’s the politics and ‘economic’ issues that feel suffocating. Among many other things.
Humans haven’t done a very good job watching over this planet given the tools we’ve found recently. Some humans and some groups do a better job but the predominant ones do not. But we are a tiny planet in a universe probably entirely full of life. If we kill outselves off and kill off a bunch of other organisms it’s a staggering loss for sure, but we still are just one tiny island in a vast wilderness either way.
But i get how when it feels like there is no mystery left, the world is suffocating. Any time i go back to California i feel suffocated by how far ecosystem collapse has progressed. I’m not that old, i am 45. And i’ve seen way way too much loss in that amount of time. So yeah. that does feel suffocating.
I wrote that comment after doing a cross-country road trip from Seattle to Moab with a friend of mine. That comment was directed toward charismatic megafauna, particularly after reading WSFW reports on game populations, as well as reports on our wolf packs.
The core of the problem is that the world is crowded with too many people and not enough wild space. I’m sorry that my comment was misinterpreted as a slight against conservation work, but that wasn’t my intent. I have to admit, I’m a little disheartened to see my words being chopped up and used in such a way.
Even with charismatic megafauna I would push back on the accuracy of your comment. There is far less direct monitoring of individuals than people think, and a lot more estimation and modeling to fill in the gaps, and there are plenty of large species that we simply do not have a good estimate of the numbers for.
There is no question that things are massively out of balance regarding the human population and the lack of both wild space and wild animals, but the idea that everything is monitored in such detail is just plain wrong.
I suspect that part of this comes from people not looking at the world as a whole. On platforms like iNat, Reddit, and the like the majority of the users are from the US, Europe, or Canada, and tend to assume that the sort of wildlife monitoring and management they’re familiar with in those regions is more-or-less the norm around the world. In the instance where they do hear about species in other countries it’s usually a handful of 5 or 6 extremely well known species, which is not at all representative of the full range of even charismatic megafauna, let alone wild animals as a whole.
The resources conservationists have at their disposal are massively over estimated by the general population, just as the resources necessary for doing conservation and wildlife monitoring are massively underestimated.