The ‘solar canals’ making smart use of India’s space | Hacker News Reader

The ‘solar canals’ making smart use of India’s space

bbc.com

148 pointspalo370 comments

> The running water helps the panels to remain cool, which increases their efficiency by at least 2.5-5%

Pretty neat side effect.

This is an awesome idea that I hope SRP adopts here in Phoenix. We have plenty of sun, and canals crisscross the whole valley.

Aren't the metal structure that can bridge the gap more expensive than the ones planted on ground?

If land is expensive, then it might be worth it.

If it's a standardized width and they can be assembled and trucked in, the price difference might be minimal compared to ground mounted (which also need metal structure, just one with supports spaced closely together). Increase the span, use a bit heavier gauge square tube and have a 2d truss structure

Fun fact about those canals: many of them are close to 1,000 years old. The modern incarnation largely follows the old Hohokam canal system, though some of the newest, most expensive parts carefully deviate from it.

That's a fascinating observation.

I guess it makes sense, sort of like roads probably cross mountains where there used to be animal crossings, then human trails.

There was a proposal to do this to the California aqueduct, but it turns out they don’t lose much water from evaporation and it would be cheaper to put them on land.

https://green.blogs.nytimes.com/2011/04/20/could-the-califor...

CA aqueduct proposal was floating panels. Not suspended above the canal.

"don't lose much water" in CA... I doubt the CA case study lends itself too much to the Indian environment.

Yea, I think the parent was just discussing a similar project in California. They probably didn't mean to imply anything about the Indian project.

Sorry you're getting buried.

A quick google search found this PDF as my first result: http://watermanagement.ucdavis.edu/files/4114/3891/2385/A01_...

According to that paper, 700 miles of canals lose 9300 acre feet of water per day. They say the price per acre feet (in 2015) was $107. Almost a million dollars per day is lost according to that paper. And California hasn't been cooling down since 2015.

It looks like the paper looked at one section of the canal and extrapolated the evaporation loss to the rest of the canal system in California.

*this paper seems to be from a class at the university. Just using it as an example that it might be more evaporation than stated above

smaller width canals are more suited in terms of the cost of the structure to support the panels or so i feel ?

This is very cool. I love it. These canals are not particularly pretty so I think the panels are an upside in aesthetics. Look very futuristic.

What about the impact to birds insects and animals that access the canal for water?

Looks like they could fit through the gaps in the side. I doubt it would take them too long to find that, especially if they knew where the canal was

These are V-notch canals, made of cement and carry fast moving water. Not many creatures will try swimming or drinking in it. There are smaller mud canals that carry water from these to the field. Insects, birds, and animals can be found there.

If you look at the pictures, the canals are not enclosed, it's a roof with large openings between the panels as well as on the side.

Creatures like shade too.

Offtopic

> The emissions from travel it took to report this story were 0kg CO2. The digital emissions from this story are an estimated 1.2g to 3.6g CO2 per page view.

Isn't 1.2g per page view huge amount? I wonder if we can have api for this.

Well, you exceeded that in 5 minutes yourself by just existing.

Some envelope math. From https://www.eia.gov/tools/faqs/faq.php?id=74&t=11 , I get .99 pounds of CO2 per kWh. That converts to 449 grams. Splitting the difference between 1.2g and 3.6g, we get 2.4g. 2.4/449= 0.00534 or ~0.53% of 1 kWh consumed, or 5.3 watt-hours. Which does strike me as rather high.

The page took 20s to load for me. That's about 1kW to get the ~5Wh. That seems reasonable. Just my internet equipment consumes about 50W, and there are many servers involved in loading a BBC page.

Loaded in well under 2 secs for me (no JS, all ads blocked, not that there will be on the bbc)

they might have third party trackers though

Makes me wonder just how much power/carbon waste goes in to ads vs content on the modern web.

Tell you what, why don't you give it a try? It should take 30 secs to disable JS on firefox, then see how it feels, re-enable it afterwards. See what a snappy web can feel like (when it doesn't break, which is ~30% of the time).

I keep saying it, strange that nobody has ever said that they tried it. There seems to be some weird mental block around trying something new, even if easy to do and with possible benefits. Why is this?

I'm going to ask again if someone would PLEASE give this a try and tell me what you think. Just 15 minutes, no more, just so we can get an unbiased opinion (ie not mine).

If it turns out to be tolerable then you've got a trivially easy way of cutting carbon use, a little.

thanks

The BBC page didn't seem to load at all, clicking the link above.

BBC is a bad choice as they don't have traditional ads, just ones for their own content.

Another note: the preference to disable JS is not surfaced in Firefox unless you go to about: config, so it is a little fiddly.

Finally, some feedback! thanks.

The link I see is https://www.bbc.com/future/article/20200803-the-solar-canals... and it works fine for me. What do you see?

I'm very surprised about the JS, when I used firefox (I now use palemoon) it was under preferences. Noted, thanks.

Not the person you replied to, but to add to the subject, just uBlock Origin on Chrome and the page loads in a second or two, with 11 elements blocked. If I also disable Javascript(1), it loads even faster, but the difference between the two is on the order of half a second, so minimal in practice.

((1) I use the Quick Javascript Switcher extension https://chrome.google.com/webstore/detail/quick-javascript-s... )

Yep, they removed it

"Beginning from Firefox version 23, Firefox web browser from Mozilla has removed the UI to disable and enable the JavaScript (JS) on web pages from “Content” tab of web browser’s Options menu."

https://techjourney.net/disable-javascript-in-firefox/

Close enough to all/none if you ignore the cost of content production.

A pi-hole dns filter will typically block 10%-30% of all domain queries (for a fast and loose estimate).

According to Website Carbon Calculator, each page view produces 2.22g of CO₂. I think it would be lower if you use an ad blocker.

https://www.websitecarbon.com/website/bbc-com-future-article...

I really have no idea how they're calculating that other than a lot of hand-wavy math. It will vary greatly depending where you are. If I'm in Vancouver and I access CDN or hosting resources from a massive datacenter located in WA or OR, it's highly likely that 98% of the energy for my home, and all of the telecom/ISP sites in between, and the datacenter itself are all hydroelectric powered.

That power could be used elsewhere in your power grid. It probably makes sense to look at CO2/kWh averaged over the whole grid.

Despite our government repeating it over and over, it turns out that Hydro is actually worse than all fossil fuels in terms of GHGs (over double coal) depending on what you flood. If it's forest, it's the worst. For the W.A.C Bennett dam in BC, 350,000 acres of former forest land was flooded.

OK, but it's not being re-flooded constantly, it's already done and can't reasonably be undone.

So, long-term, it's a fantastic renewable resource with practically zero emissions.

See sibling comment. Practically zero emissions is false. It results in over double the GHG emissions per year than a coal fired plant producing the same amount of energy.

That is crazily misleading.

For the initial construction, yes. Ongoing? No.

At a point in time, the amortized emissions become zero.

It's not misleading. The vegetation that is flooded decomposes anaerobically and releases methane. So there's an ongoing release over the years that it decomposes. They've measured these emissions at different dams over time, and the worst (where forest is flooded) results in nearly 2.5 times the GHG emissions as the worst coal plants.

For instance the Balbina hydroelectric reservoir in Brazil was responsible for 39000 tons of methane each year (https://www.researchgate.net/publication/252089237_CO2_emiss...)

That was a 2011 paper on a dam that was finished in 1989.

That is still misleading, grossly.

More happens in the water than just releases of methane - there's plant and animal life producing oxygen and consuming decomposing materials.

Plus 1989 is a brand new dam. The Hoover Dam was completed in 1936[1], and is expected to have a service life of at least 10,000 Years.[2]

10,000 years at a minimum. It will have paid dividends in renewable energies so many times over by then, that it's not even worth discussing.

Do you think the solar panels installed on your roof will last 10,000 years? Will they even last 10 years without needing replacement or repairs?

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

[2] https://delzottoproducts.com/2017/03/15/long-will-take-concr....

How is it misleading? Explain please.

What I see is a 250MW dam built in 1989, that produces in 2011 the equivalent GHGs of a 1000MW coal plant.

And at a point will not produce equivalent GHG's, which your coal plant will continue to do so.

The coal plant won't even be able to operate in 10,000 years, this dam will continue to do so.

Do you even read your own links? There's no way the dam will be operating in 10,000 years time.

That would be a strong argument against new hydropower projects, but now that those dams are built, it seems like we have sunk costs (pun not intended) that would make it a bad choice to decommission, if we'd have to wait 100yr for the forests to begin to recover? That seems to jibe with critical studies I see [1].

1: http://blogs.edf.org/energyexchange/2019/11/15/long-consider...

That's not true. Decomposition of trees happens over a long period of time in a anaerobic environment. That's what causes the large methane emissions.

I see, that makes sense - though then it'd be a comparison of energy and emissions already put into the making of the dam, the power it outputs (which could otherwise be from fossil fuels) and the atmospheric impact of those methane emissions. My instinct would be the impact of pouring all that concrete etc. would be very large, so decommissioning would still be very costly, but it does strengthen the argument against new dam projects.

Honestly, compared to the ongoing methane releases from some of the worst offending dams, the concrete emissions are negligible.

> it turns out that Hydro is actually worse than all fossil fuels in terms of GHGs

That's a bad comparison. You can't simply compare them without any context.

Hydro has a one-time high carbon cost, while fossil fuels have a small ongoing cost. You can't simply state that one is better than the other, you need at minimum to restrict the comparison into a time-frame.

(By the way, nobody does a large hydro project without first cleaning out the previous vegetation nowadays. Historical data isn't good either.)

It's not a one time cost is what I'm saying. The trees in the forest that is flooded decompose slowly, and release methane the entire time. Specifically to BC, the entire ecosystem also got screwed, the fish in tributaries, up and down from the dams have mercury poisoning.

It is a constant amount emitted, for generating power during the entire lifetime of the plant. While fossil fuel plants emit an amount that is proportional to the power they generate.

How is that helpful though? They're basing it on the dams actual theoretical use. If they don't use it, the emissions will still occur, but we get nothing from it.

It's absurd to claim that every hydroelectric dam is causing emissions from rotting trees. The land flooded by the Grand Coulee dam was not forested. The area it flooded looks pretty much like the area you can see around it today in google earth, desert scrub land.

The area flooded by the Hoover dam was not forested.

The area flooded by the Glen Canyon dam was not forested.

I could go on...

I didn't make that claim. If your read what I wrote it specifically says "depending on what you flood."

You're also wrong about the Grand Coulee dam. Didn't bother checking the others.

"In late 1938, the Works Progress Administration began clearing what would be 54,000 acres (220 km2) of trees and other plants."

A much better example of multi-use space than solar roads.

>To maximise the absorption of energy, the solar panels should face south, but the canal’s direction cannot be dictated.

Why does the panel have to be tilted forwards along the row? Can't it be tilted in any direction relative to the others?

That sounds like a useful optimization.

Maybe it's a practical necessity for a uniform support structure that faces either upstream or downstream.

It would be interesting to see the math - it might only buy a few percent at certain times of the day and be a wash when the sun is overhead. It might also depend on the demand curve.

For every spot on earth there is an optimal angle of a solar panel / plane to capture the maximum amount of light (Even at any given moment in time). It is for this reason that panel for panel, the most efficient way of operating a given panel are with “duel-axis” trackers. [1]

In the absence of these, and factoring in the complications involved with an always operating mechanical component, you can have a “fixed axis” which is the theoretical best point for that place on earth. Most racking systems assume a rigid NxN rectangular grid, so having something that is commercially out there that follows the space exactly may be tough.

[1] https://en.m.wikipedia.org/wiki/Solar_tracker

Trackers are expensive and require maintenance. About 10 years ago The cost of the trackers exceeded the cost of land and panels almost everywhere, and so you Don’t see them anymore except in very unusual applications.

I would guess that it's easiest to manufacture a linear row of panels, and install N hundred copies of that row following the path of the canal. Otherwise each row of panels would need to be either made bespoke or adjusted on site.

This helps control evaporation losses as well, the water flow in canals is more predictable which helps. Also there is a great potential for floating solar especially behind existing hydro as the grid infrastructure almost exists.

I wonder how this would affect the climate, the rains especially.

Given that a square of 300X300km would produce enough electricity to power the WHOLE WORLD, and 300X300km square of ocean is such a minuscule thing, I'd say -- no measurable impact.

A good point from a dead comment : "I wonder how this would affect the climate, the rains especially."

Reducing water evaporation should have some kind of effect. Does anybody know of weather and/or rain effects due to these evaporation preventing floating black balls on reservoirs?

Less evaporation losses means more water will reach irrigation networks where it will evaporate anyways (either before or after temporarily becoming part of a plant)

The reduction of evaporation is negligible when wete talking about affecting local climate. These are man made canals after all.

You might want to check the numbers on the cost of enough solar panels to actually affect evaporation enough to make any change in local climate.

This is great in theory, but if they cover all the canals it will significantly hamper Knightboat's effectiveness.

https://www.youtube.com/watch?v=uoV1-fsFCmw

There's always a canal!

Reply on news.ycombinator.com