Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

USGS uses machine learning to show large lithium potential in Arkansas

usgs.gov

Post reply on HN
Print view View original

101–110 of 237 posts

Re: USGS uses machine learning to show large lithium potential in Arkansas

#101

Post by gottorf »

waveBidder wrote: post #86
declan_roberts wrote: post #70

Earlier quoted context omitted.

It seems backwards, but pretty much the only fuel that protects ecosystems on a large scale are fossil fuels and nuclear. Global reforestation is almost entirely the result of households switching from wood to coal in the 20th century.

> It seems backwards, but pretty much the only fuel that protects ecosystems on a large scale are fossil fuels and nuclear. This is ludicrously off-base for fossil fuels, even if we're only talking about local pollutants from the plants themselves, nevermind things like Exxon Valdez or the pipelines or the act of mining. Nuclear seems likely, though as the other commenter noted it's not a magic bullet either. > Globa…

> This is ludicrously off-base for fossil fuels, even if we're only talking about local pollutants from the plants themselves, nevermind things like Exxon Valdez or the pipelines or the act of mining.

The energy density of fossil fuels means that those side-effects would be worse with other sources of energy.

> is a result of urbanization mostly

Urbanization, made possible by the economical source of energy that is fossil fuels.

Re: USGS uses machine learning to show large lithium potential in Arkansas

#102

Post by Retric »

kylehotchkiss wrote: post #57
idontwantthis wrote: post #40

Say Lithium becomes essentially free because we find so much of it…would that drastically lower battery costs? Is our current supply of lithium limiting production?

Is sand essentially free because we have beaches and deserts full of it? It can be used to make concrete, a valuable material? (Don't forget the shipping, storage, and refining costs)

Logistics depends on where you are not the inherent price of the commodity. Plenty of things like air and are freely available but you still need ventilation systems in caves and whatnot. Moving free dirt around when building roads can be extremely expensive due even if it’s just being moved a few miles volume adds up.

So yea desert sand is essentially free, even if you pay for shipping.

Re: USGS uses machine learning to show large lithium potential in Arkansas

#103

Post by Animats »

There's also a big lithium deposit in Nevada, and preparations for mining are underway there.[1] General Motors put in $650 million for guaranteed access to the output of this Thacker Mine.

It's in a caldera in a mountain that I-80 bypassed to go through Winnemuca, Nevada. Nearest town is Mill City, NV, which is listed as a ghost town, despite being next to I-80 and a main line railroad track. The mine site is about 12km from Mill City on a dirt road not tracked by Google Street View.

Google Earth shows signs of development near Mill City. Looks like a trailer park and a truck stop. The road to the mine looks freshly graded. Nothing at the mine site yet.

It's a good place for a mine. There are no neighbors for at least 10km, but within 15km, there's good road and rail access.

[1] https://en.wikipedia.org/wiki/Thacker_Pass_lithium_mine

Re: USGS uses machine learning to show large lithium potential in Arkansas

#104

Post by levocardia »

lordgrenville wrote: post #99
folli wrote: post #21

From the paper's method section, a bit more about which type of ML algo was used: An RF machine-learning model was developed to predict lithium concentrations in Smackover Formation brines throughout southern Arkansas. The model was developed by (i) assigning explanatory variables to brine samples collected at wells, (ii) tuning the RF model to make predictions at wells and assess model performance, (iii) mapping spa…

So it turns out that there's no theoretical reason that gradient boosting will always outperform RF (which would violate the "no free lunch" theorem). But it does usually seem to be the case in practice, even with small and noisy data. I would hazard a guess that with better tuning, XGBoost would still have won. (The paper notes that the authors chose a suboptimal set of hyperparameters out of fear of overfitting - m…

That's been my experience. RF tends to do quite well out of the box, and is very fast to fit. It's less of a pain to cross-validate too, with fewer tuning parameters. XGBoost has a huge number of knobs to tune, and its performance varies from god-awful with bad hyperparameters to somewhat better than RF with good ones. Giant PITA with nested cross-validation, etc. though.

I haven't read in detail what their validation strategy is but this seems like the kind of problem where it's not so easy as you'd think -- you need to be very careful about how you stratify your train, dev, and test sets. A random 80/10/10 split would be way too optimistic: your model would just learn to interpolate between geographically proximate locations. You'd probably need to cross-validate across different geographic areas.

This also seems like an application that would benefit from "active learning". given that drilling and testing is expensive, you'd want to choose where to collect new data based on where it would best update your model's accuracty. A similar-ish ML story comes from Flint, MI [1] though the ending is not so happy

[1] https://www.theatlantic.com/technology/archive/2019/01/how-m...

Re: USGS uses machine learning to show large lithium potential in Arkansas

#105

Post by diggernet »

Animats wrote: post #103

There's also a big lithium deposit in Nevada, and preparations for mining are underway there.[1] General Motors put in $650 million for guaranteed access to the output of this Thacker Mine. It's in a caldera in a mountain that I-80 bypassed to go through Winnemuca, Nevada. Nearest town is Mill City, NV, which is listed as a ghost town, despite being next to I-80 and a main line railroad track. The mine site is about…

Your description of the location of this mine doesn't match your Wikipedia link.

Searching in Google Maps, Thacker Mine comes up as 40.58448942010599, -117.8912129833345. As you say, that is near I-80 and Mill City, and there is nothing there.

But Wikipedia says it's at 41.70850912415866, -118.05475061324945 in the McDermitt Caldera, nowhere near Mill City or I-80.

I'm thinking probably don't trust Google on this one. :)

Re: USGS uses machine learning to show large lithium potential in Arkansas

#106

Post by luckylion »

ianeigorndua wrote: post #98
declan_roberts wrote: post #96

Earlier quoted context omitted.

What do you think the people in cities burn for fuel to keep themselves warm?

Trash, mostly.

Not really. I live close to Hamburg, Germany. _Very_ eager to be eco-friendly and sustainable.

Currently: 64% coal, lots of nat gas, ~20 renewables.

The future plan is to use a lot more industrial waste heat. Burning garbage is done and planned, but nowhere near a major factor. Not to mention that the garbage would also need to come from something: plastics from oil, wood from trees etc.

Re: USGS uses machine learning to show large lithium potential in Arkansas

#107

Post by unusualmonkey »

gottorf wrote: post #101
waveBidder wrote: post #86

Earlier quoted context omitted.

> It seems backwards, but pretty much the only fuel that protects ecosystems on a large scale are fossil fuels and nuclear. This is ludicrously off-base for fossil fuels, even if we're only talking about local pollutants from the plants themselves, nevermind things like Exxon Valdez or the pipelines or the act of mining. Nuclear seems likely, though as the other commenter noted it's not a magic bullet either. > Globa…

> This is ludicrously off-base for fossil fuels, even if we're only talking about local pollutants from the plants themselves, nevermind things like Exxon Valdez or the pipelines or the act of mining. The energy density of fossil fuels means that those side-effects would be worse with other sources of energy. > is a result of urbanization mostly Urbanization, made possible by the economical source of energy that is f…

> The energy density of fossil fuels means that those side-effects would be worse with other sources of energy.

Can you expand on this? How does the density of fossil fuel make them a better source of energy than say wind?

Re: USGS uses machine learning to show large lithium potential in Arkansas

#108

Post by Dalewyn »

greenie_beans wrote: post #67

ugh i really don't want people to mine in the mobile basin. that's one of the most diverse ecosystems in north america. https://www.youtube.com/watch?v=8j9coyJeB4Q

A significant reason the real holders of power in the world today are Saudi Arabia and China is because we've refused to gather and use our resources while they have theirs.

It's high time we realize that Pax Americana is our era to lose, (re)start mining and (re)start development.

Re: USGS uses machine learning to show large lithium potential in Arkansas

#109

Post by moffkalast »

__MatrixMan__ wrote: post #53

Great, now ask the AI to engineer a fungal genome that'll help us purify it more easily: Frack in the substrate and spores, harvest fruit bodies on the surface, profit.

I mean, what could possibly go wrong?

Re: USGS uses machine learning to show large lithium potential in Arkansas

#110

Post by Dalewyn »

engineer_22 wrote: post #68
EA-3167 wrote: post #45

Earlier quoted context omitted.

The major limiting factor of lithium is not really availability so much as the cost of extraction. China is the leader in this field, not so much because of abundance or stellar technology, but out of a willingness to completely ignore environmental externalities (including those of the power generation involved in the whole process). As a result the price of Chinese lithium is low enough that it would be essentially…

> the price of Chinese lithium is low enough that it would be essentially impossible to compete with them... In an export model, yes. However, given their negative externalities (including geo-political factors), importing countries may place tariffs on Chinese lithium in order to make use of other sources. If the total embodied value of lithium in any particular product is small compared to the overall value of the…

Keyword there is may. Putting aside whether the sentiment is justified, it is currently extremely unpopular to impose Chinese tariffs.

It's also worth noting that Chinese prices are so low that certain tariffs can reach the stratosphere (eg: American 100% tariff on Chinese EVs), further making them unpopular with the commons.

Post reply on HN