Drones that can plant 100k trees a day(weforum.org)
weforum.org
Drones that can plant 100k trees a day
https://www.weforum.org/agenda/2017/06/drones-plant-100000-trees-a-day?utm_content=buffer6368b&utm_medium=social&utm_source=twitter.com&utm_campaign=buffer
87 comments
> There are more trees now than there were 100 years ago.
You may be correct. However, coming from an avid guitarists who looks at this particular industry in isolation from most people; there isn't "old growth" which is a unique problem in my industry. And what trees there are, less of them are considered musical instrument grade.
Also, the types of trees matter a lot (for certain things). We simply can't use Brazilian Rosewood for new builds anymore. It doesn't exist, and what does is heavily, heavily protected (To the point where instruments made before the export ban -worth $$$$$$$- are getting confiscated. Others similar to rosewood have been confiscated simply for looking like it's made from the Brazilian kind).
I think, in general we were aware of our problem as consumers of trees. However, we underestimated (for musicians) how the age and growth patterns dictate what makes a good tree for X. What was untouched for centuries simply makes better wood.
Essentially, the niche markets are really, really feeling this squeeze. So much so that Fender has switched to an alternative Rosewood for most if not all of their guitars, Taylor often uses (and pioneers in this subject in many ways) alternative "tonewoods" and has for years, simply because the wood that exists for these instruments even 50 years ago is simply unattainable now.
Some boutique builders are forced to harvest "forgotten" logs which sunk to the bottom of lakes and rivers decades ago.
You may be correct. However, coming from an avid guitarists who looks at this particular industry in isolation from most people; there isn't "old growth" which is a unique problem in my industry. And what trees there are, less of them are considered musical instrument grade.
Also, the types of trees matter a lot (for certain things). We simply can't use Brazilian Rosewood for new builds anymore. It doesn't exist, and what does is heavily, heavily protected (To the point where instruments made before the export ban -worth $$$$$$$- are getting confiscated. Others similar to rosewood have been confiscated simply for looking like it's made from the Brazilian kind).
I think, in general we were aware of our problem as consumers of trees. However, we underestimated (for musicians) how the age and growth patterns dictate what makes a good tree for X. What was untouched for centuries simply makes better wood.
Essentially, the niche markets are really, really feeling this squeeze. So much so that Fender has switched to an alternative Rosewood for most if not all of their guitars, Taylor often uses (and pioneers in this subject in many ways) alternative "tonewoods" and has for years, simply because the wood that exists for these instruments even 50 years ago is simply unattainable now.
Some boutique builders are forced to harvest "forgotten" logs which sunk to the bottom of lakes and rivers decades ago.
Self replicating trees?? This technology needs to be put under regulatory control before we are all turned into trees.
I trust Monsanto is on it.
We'll use the blockchain to do a rights payment when a tree using our patented DNA grows.
There are also milestone payments: survives for 1 year - a small payment; survives for 2 - another payment...
There are also milestone payments: survives for 1 year - a small payment; survives for 2 - another payment...
back in the old days, we used to get a ring for each year.
It should be pretty obvious that trees can't reproduce anywhere near as quickly as we can pull them down. It's simple maths that the net loss due to human activity is 6 billion. Trees will reproduce at a higher rate than they die if they have the space to do so, but why do you assume that they do? The problem is those trees were pulled down to make room for other things and the trees can no longer replace themselves in that area.
> It should be pretty obvious that trees can't reproduce anywhere near as quickly as we can pull them down.
Of course not, but they do reproduce and can do quite fast for animal propagated tree species. We are talking about 3.04 trillion trees. Even if the natural germination rates are low, it would still be significant enough to make a difference in the assertion this article uses as introduction.
Like I said, a lot of things have to be taken into account, including local deforestation and its effect on desertification. Higher averages temperatures are forcing forests to move to higher latitudes, but they are also clearing areas of growth that are not yet inhabited by humans and were far too cold before for trees to propagate to.
I'm not suggesting that there's not a problem with the way we use our forest resources, just that the oversimplification of the starting premise of the article doesn't help at all.
Of course not, but they do reproduce and can do quite fast for animal propagated tree species. We are talking about 3.04 trillion trees. Even if the natural germination rates are low, it would still be significant enough to make a difference in the assertion this article uses as introduction.
Like I said, a lot of things have to be taken into account, including local deforestation and its effect on desertification. Higher averages temperatures are forcing forests to move to higher latitudes, but they are also clearing areas of growth that are not yet inhabited by humans and were far too cold before for trees to propagate to.
I'm not suggesting that there's not a problem with the way we use our forest resources, just that the oversimplification of the starting premise of the article doesn't help at all.
> Even if the natural germination rates are low, it would still be significant enough to make a difference in the assertion this article uses as introduction.
If self regeneration is low, it doesn’t matter what the absolute numbers are, they are still tiny compared to the level of deforestation.
Then there’s the attrition rate of forests, which is probably in the same order of magnitude as natural reforestation.
“Rate at which humans plant trees - rate at which humans destroy trees” is a good first order approximation of the general trends simply because the numbers are so large compared to all other factors.
In every environmental context, human action is the greatest impact by an order of magnitude simply because there are so many of us. For example, the greatest sources of methane in the environment are from human activities such as mining and cattle (64% according to USA EPA), way ahead of swamps and other natural sources.
Studies in Brazil suggest that natural reforestation is a process that takes between 60 years (some trees are back) to 4000 years (this patch has regenerated to be the same as the forest around it). It takes so much less time to clear a patch of forest than to regenerate, that the regeneration is a rounding error in calculations with two significant digits.
If self regeneration is low, it doesn’t matter what the absolute numbers are, they are still tiny compared to the level of deforestation.
Then there’s the attrition rate of forests, which is probably in the same order of magnitude as natural reforestation.
“Rate at which humans plant trees - rate at which humans destroy trees” is a good first order approximation of the general trends simply because the numbers are so large compared to all other factors.
In every environmental context, human action is the greatest impact by an order of magnitude simply because there are so many of us. For example, the greatest sources of methane in the environment are from human activities such as mining and cattle (64% according to USA EPA), way ahead of swamps and other natural sources.
Studies in Brazil suggest that natural reforestation is a process that takes between 60 years (some trees are back) to 4000 years (this patch has regenerated to be the same as the forest around it). It takes so much less time to clear a patch of forest than to regenerate, that the regeneration is a rounding error in calculations with two significant digits.
Can you give any sources?
Evolution of forest surface in western Europe in the last 100 years: https://www.washingtonpost.com/news/worldviews/wp/2014/12/04...
US Forests History and trends (FAO): www.fao.org/docrep/meeting/x4995e.htm
WorldBank has numbers for global trends but only with data starting in 1990 and for forest surface: https://data.worldbank.org
It shows a slight decline, but as FAO points out, forest density (trees pr sqmt) is also growing so forest surface area numbers don't tell the whole story.
US Forests History and trends (FAO): www.fao.org/docrep/meeting/x4995e.htm
WorldBank has numbers for global trends but only with data starting in 1990 and for forest surface: https://data.worldbank.org
It shows a slight decline, but as FAO points out, forest density (trees pr sqmt) is also growing so forest surface area numbers don't tell the whole story.
Seed pods ain't trees. There's a reason that when forestry companies replant trees, they are planting seedlings that have been cultivated for planting purposes. They aren't just shoving seeds into the ground and hoping for the best.
It's a very good idea to make planting trees cheaper. I've often pondered robotics for tree planting. But will it be as cheap as human labour in terms of stems that make it to maturity? Hand-planted seedlings have a fairly good survival rate and tend to be well spaced, even if the planter is relatively negligent. Typically, the planted trees will need to be thinned later on, because so many survive.
As it stands, it costs a licensee $2-$3 CAD per stem to reforest an area. Typical densities are 1400-1800 stems per hectare. That's actually pretty cheap. Mature trees at harvest are worth thousands.
Source: worked as a tree planter for three seasons, have friends in the forestry sector.
It's a very good idea to make planting trees cheaper. I've often pondered robotics for tree planting. But will it be as cheap as human labour in terms of stems that make it to maturity? Hand-planted seedlings have a fairly good survival rate and tend to be well spaced, even if the planter is relatively negligent. Typically, the planted trees will need to be thinned later on, because so many survive.
As it stands, it costs a licensee $2-$3 CAD per stem to reforest an area. Typical densities are 1400-1800 stems per hectare. That's actually pretty cheap. Mature trees at harvest are worth thousands.
Source: worked as a tree planter for three seasons, have friends in the forestry sector.
There was a discussion almost 3 years ago regarding drones, that are able to plant 36K trees per day. There was also a reference to the 1970s planting machines, that are much simple in mechanical design and are able to plant 10-20K trees per hour.
https://news.ycombinator.com/item?id=9444521
What about wheeled robot for this task? Something like Curiosity to access remote areas and use a drone to refill the seed container. Afaik it’s not enough to drop a seed on the ground and wheeled robot could drill small hole to put properly seed inside. Nice application, but who pays for this?
>A drone loaded with germinated seeds fires pods into the ground
My reading of this implies the seed ends up underground.
A drone seems like the best answer here to get maximum accessibility and speed. Reaching 100k a day with a wheeled robot would be tricky.
My reading of this implies the seed ends up underground.
A drone seems like the best answer here to get maximum accessibility and speed. Reaching 100k a day with a wheeled robot would be tricky.
We're really talking fleet of drones vs. fleet of wheeled robots here; wheeled robots, even if slower in deployment, will be able to carry more seed pods (reducing turnaround time) and will be orders of magnitude more energy-efficient - i.e. cheaper to operate and better for the planet.
As cool as flying is, I'm all for ground solutions on this one.
As cool as flying is, I'm all for ground solutions on this one.
Reforestation scenarios are always very rough terrain. Everything that isn't marketable, which is a lot, is left strewn about.
While I like the general idea here, to improve the environment with technology, it'd be better to use drones to pick up trash. Trees will plant themselves but trash just accumulates and gets ground up by erosive forces.
Also, people need jobs. Give them jobs planting trees and picking up trash.
/can't see the forest for the trees.
Also, people need jobs. Give them jobs planting trees and picking up trash.
/can't see the forest for the trees.
Nothing personal, but I disagree with pretty much everything you say :)
Firstly you're staging a false dichotomy. We can use both these kind of drones if we want to.
"Trees will plant themselves" < so they are idiots?
"people need jobs" is played every time we replace a brain dead job with some kind of effective innovation. There are no examples of us going back. Fortunately.
Firstly you're staging a false dichotomy. We can use both these kind of drones if we want to.
"Trees will plant themselves" < so they are idiots?
"people need jobs" is played every time we replace a brain dead job with some kind of effective innovation. There are no examples of us going back. Fortunately.
Just reducing the load of roaming cattle in natural areas and reintroduce predators if needed in natural parks, will do much more for much less money.
Birds plant trees much better than drones and wolves are critical to allow the saplings to survive their first five years. We could feed an army of planting tree drones and cover each inch of mountain with seedlings just to give an expensive yummy snack to sheeps and goats.
Birds plant trees much better than drones and wolves are critical to allow the saplings to survive their first five years. We could feed an army of planting tree drones and cover each inch of mountain with seedlings just to give an expensive yummy snack to sheeps and goats.
I don't understand why you are being downvoted. Well, maybe I do, but I want to believe that HN is not so techno-obsessed that they can't fathom what you are saying.
I agree. Let's not plant a lot of trees that goats and cows will just eat up.
I agree. Let's not plant a lot of trees that goats and cows will just eat up.
I'm just speculating about the economics here, and it kinda sounds like a high-school daydream solution, but I feel like you could artillery an area with some munitions filled with seeds and just rely on the natural rate of survival given a uniform-enough distribution of an airburst shell.
100 panderosa pine seeds is ~4g [1]
11kg means 275000 seeds / shell.
You'd plant pines at density of around 1000/acre [2]
The biggest question is seed survivor rates of being planted by artillery but let's consider a survivorship of 2%...
275000 * 0.02 = 5500 Surviving trees, or effective planting over 5.5 acres / shell.
A modern autoloading artillery can launch ~70 shells per hour. [3]
You'd have an effective planting rate of 385 acres / hour. Or 1.54 km^2 / hour. 13400 km^2 / year.
With a single artillery that has a radius of 25km you get an area of 1900km^2.
9,504,000 trees / day
13400 km^2 covered / year
But at 1%
4,752,000 trees / day
6700 km^2 covered / year
In any case 100k/day seems quaint when you can plant nearly 10 million.
Assuming 2% survivorship, covering 1900km^2 at 1.54km^2/hour you can plant the entire effective area in ~55 days, as long as you keep loading the thing.
It takes ~37 shells to plant 100k trees which means ~$19000 for munitions.
Let's say we want to finish this endeavour within 2 years, and assume our effective time spent planting versus moving and getting the project off the ground is 50%, so we get 1 effective year of dedicated planting time.
One artillery piece can plant 13400km^2 / year.
50000km^2 total area / 13400km^2 coverage per year = ~4 pieces of artillery.
Let's buy 3x what we need so we can just flip them out when they break down and have some spares in case repairs/maintenance take a while.
12 pieces of artillery is ~$51,000,000
You need ~46 shells to cover 1km^2
50,000*46 = 2,300,000 shells
2,300,00 shells at $500 / shell = $115,000,000 for munitions
Artillery + munition cost is $166 million.
Throw $10mil each at R&D + Testing, Logistics, and Administration/Staff and you're at $196 million with more than half of the fund left over.
It has some precedent: there have been some shotgun shells filled with seeds before that actually worked [7]
[0] https://en.wikipedia.org/wiki/M795 [1] https://www.fs.fed.us/pnw/olympia/kids/WASeeds.pdf [2] https://www.nrcs.usda.gov/Internet/FSE_DOCUMENTS/nrcs141p2_0... [3] https://en.wikipedia.org/wiki/Archer_Artillery_System [4] http://www.army-guide.com/eng/product2819.html [5] http://www.pmulcahy.com/ammunition/howitzer_rounds.html [6] https://www.weforum.org/press/2017/01/400-million-fund-launc... [7] http://www.businessinsider.com/flower-shell-shotgun-gardenin...
Some quick insomnia driven commentbox math
100 panderosa pine seeds is ~4g [1]
11kg means 275000 seeds / shell.
You'd plant pines at density of around 1000/acre [2]
The biggest question is seed survivor rates of being planted by artillery but let's consider a survivorship of 2%...
275000 * 0.02 = 5500 Surviving trees, or effective planting over 5.5 acres / shell.
A modern autoloading artillery can launch ~70 shells per hour. [3]
You'd have an effective planting rate of 385 acres / hour. Or 1.54 km^2 / hour. 13400 km^2 / year.
With a single artillery that has a radius of 25km you get an area of 1900km^2.
Survivorship is a major bound here
9,504,000 trees / day
13400 km^2 covered / year
But at 1%
4,752,000 trees / day
6700 km^2 covered / year
In any case 100k/day seems quaint when you can plant nearly 10 million.
Assuming 2% survivorship, covering 1900km^2 at 1.54km^2/hour you can plant the entire effective area in ~55 days, as long as you keep loading the thing.
Comparing costs to this drone
It takes ~37 shells to plant 100k trees which means ~$19000 for munitions.
Using that $400 million dollar fund entirely for this idea
Let's say we want to finish this endeavour within 2 years, and assume our effective time spent planting versus moving and getting the project off the ground is 50%, so we get 1 effective year of dedicated planting time.
One artillery piece can plant 13400km^2 / year.
50000km^2 total area / 13400km^2 coverage per year = ~4 pieces of artillery.
Let's buy 3x what we need so we can just flip them out when they break down and have some spares in case repairs/maintenance take a while.
12 pieces of artillery is ~$51,000,000
You need ~46 shells to cover 1km^2
50,000*46 = 2,300,000 shells
2,300,00 shells at $500 / shell = $115,000,000 for munitions
Artillery + munition cost is $166 million.
Throw $10mil each at R&D + Testing, Logistics, and Administration/Staff and you're at $196 million with more than half of the fund left over.
Hmmmmmmm......
It has some precedent: there have been some shotgun shells filled with seeds before that actually worked [7]
[0] https://en.wikipedia.org/wiki/M795 [1] https://www.fs.fed.us/pnw/olympia/kids/WASeeds.pdf [2] https://www.nrcs.usda.gov/Internet/FSE_DOCUMENTS/nrcs141p2_0... [3] https://en.wikipedia.org/wiki/Archer_Artillery_System [4] http://www.army-guide.com/eng/product2819.html [5] http://www.pmulcahy.com/ammunition/howitzer_rounds.html [6] https://www.weforum.org/press/2017/01/400-million-fund-launc... [7] http://www.businessinsider.com/flower-shell-shotgun-gardenin...
> It’s simple maths. We are chopping down about 15 billion trees a year and planting about 9 billion. So there’s a net loss of 6 billion trees a year.
This is a regional thing [0] though. We need to plant the trees in Latin America, Caribbean, and Sub-Saharan Africa. The rest of the world is gaining forests.
[0] http://www.telegraph.co.uk/news/2016/03/23/deforestation-whe...
[0] http://www.telegraph.co.uk/news/2016/03/23/deforestation-whe...
Now, the question is, which industry sector cuts the largest percentage of trees. Now, answering that question, is there a way to do the same thing without cutting trees?
Answering that question and then executing a business plan is probably worth billions.
Planting trees to combat 6 billion trees lost every year is a pure expense, no profit to be made at all. At least not if you won't cut them down a few decades later.
Answering that question and then executing a business plan is probably worth billions.
Planting trees to combat 6 billion trees lost every year is a pure expense, no profit to be made at all. At least not if you won't cut them down a few decades later.
"This Biodegradable Paper Donut Could Let Us Reforest The Planet"
https://news.ycombinator.com/item?id=16261101
I think this kind of thing is funny for us westerners. I recently found this tech has been used in arid countries for possibly hundreds or thousands of years.
https://duckduckgo.com/?q=ollas+gardening&t=lm&iax=images&ia...
I do realize the ollas are for more permanent gardens, but it's the same concept. I do hope the paper version gets used, there are plenty of places that could benefit from it.
https://duckduckgo.com/?q=ollas+gardening&t=lm&iax=images&ia...
I do realize the ollas are for more permanent gardens, but it's the same concept. I do hope the paper version gets used, there are plenty of places that could benefit from it.
> Then the most efficient planting pattern for that area is calculated using algorithms.
Code and algorithms you say?[0]
[0]: https://i.redd.it/ytdyqbyzjkky.png
Code and algorithms you say?[0]
[0]: https://i.redd.it/ytdyqbyzjkky.png
When my code crashes, it's usually because I forgot the algorithms
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.
Many (most?) trees are efficient at making seed, but what they really lack is the ability to get those seeds far enough away that they won't be out-competed for nutrients, water, and light by the parent tree.
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.
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.
Thanks for many of the good points. I don't disagree with your sentiment that deforestation is the problem, it's just that I feel like this is silly.
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.
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.
Exactly. Create the right conditions with little grazing pressure and reduced weed competition and regeneration will often happen naturally. There may be technical solutions to bump this along but for most situations this will not be it. Where replanting is being conducted in the third world and this drone is used instead of local planters it might increase community resentment and disengagement. Working with communities on more sustainable livelihoods and helping them understand the reasons why it is better for everyone if some areas are left un-grazed for periods and allowed to regenerate is not easy but without local buy-in nothing will last.
Yeah, this reads like one of those "technologies in search of a problem" pieces.
> 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 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 planting part could easily be solved by law. At least in Finland after cutting a part of forest, you'll need to re-plant it and create a maintenance plan for the forest to regrow.
The maintenance plan helps to protect the new growth to be eaten e.g. by moose and elk.
The maintenance plan helps to protect the new growth to be eaten e.g. by moose and elk.
I'm doing some regeneration work on a property in an occasionally semi-arid, very harsh region, with a combination of dry sclerophyll & cypress forest that are generally allelopathic.
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.
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.
>fend off predators from eating the seedlings.
The traditional solution is seed balls. Functionally it's like burying the seed, but the balls can simply be scattered.
https://en.wikipedia.org/wiki/Seed_ball
The traditional solution is seed balls. Functionally it's like burying the seed, but the balls can simply be scattered.
https://en.wikipedia.org/wiki/Seed_ball
You also need to water them and generally take care during early stages. You need to estimate how much of the 100K perish on a yearly basis due to these factors. Sounds like a kid put this together for a school project.
In Canada it's a semi-popular summer job to do tree planting for forestry companies. After the harvesters roll through and take process the mature trees you're left with a bit of a wasteland. Since the areas are highly managed there isn't a lot of natural young growth coming up to replace the harvested trees. Young people who aren't adverse to working long days and living/camping in tents for days on end walk these swaths of land with shovels on their back and bags of seedlings that they plant, and make good money doing so.
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!
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!
why couldn't we do this before with RC planes
VCs would not throw wads of cash at a business plan that uses RC planes. But have it use quad-copters, and boom! They make it rain.
It's the novelty of it, basically. We're not yet inoculated, as a society, to the charm of quad-copters.
It's the novelty of it, basically. We're not yet inoculated, as a society, to the charm of quad-copters.
At this point, it's not about quadcopters, it's just about the name. "Drones". The new sexy name for RC planes and helicopters/multirotors.
They sort of do, just with seeds instead of saplings
https://en.wikipedia.org/wiki/Aerial_seeding
https://en.wikipedia.org/wiki/Agricultural_drones
I think the real advancement here is the autonomous component of it.
https://en.wikipedia.org/wiki/Aerial_seeding
https://en.wikipedia.org/wiki/Agricultural_drones
I think the real advancement here is the autonomous component of it.
Because they can't hover
1 per second would only be 86,400 a day... and that assumes zero flying and positioning time, zero loading and reloading time, zero fueling time, and flawless, instant decisions.
I assume the "one per second" figure is calculated from the total tree output, not the other way around. As in, whatever rate the drones can plant per day divided by 86,400 and rounded up to one
Assuming the drone is only planting one tree at a time...
How many surviving trees do you propose to get out of one seedling pod?
(Hint: their own video says a max of 300 pods every 18 minutes, then reloading and repositioning. And it can only work in daylight.)
(Hint: their own video says a max of 300 pods every 18 minutes, then reloading and repositioning. And it can only work in daylight.)
One. That's why it makes sense to plant several in each pod - tree sapling survival rates aren't 100%. It makes sense to plant many more trees than you expect to grow to maturity.
Then I think they are cheating their numbers if they count multiple trees per pod. :P
It might not match your expectation that "number of trees planted === number of mature trees" but that's because you're assuming things that they're not saying. You can't reasonably make an incorrect assumption and then claim to be cheated when it's wrong.
I was being a bit snarky, but this is common practice. They could easily claim "100k" seeds, but full well knowing that only 10k will actually grow. But saying "we can grow 10k trees" just doesn't sound as impressive. It seems to be marketing numbers.
I hope they are not talking about those 15 to 20 min flight time drones.
I could see it work. They only need to plant, and go back to base... Could use multiple drones, some working, some getting recharged/batteries replaced.
I see a lot of skeptical comments on here and I suppose that's healthy, but generally I find news items like this encouraging. We need to find more ways like this to use technology to fight deforestation, pollution and the erosion of the natural environment.
I agree that it’s encouraging to see engineers and scientists focus their efforts on important societal issues. However, we need to find more thoughtful ways to use this technology. A good place to start is trying to deeply understand a context and need, before coming up with a solution. We just ran a workshop in Jordan with engineers who want to develop education technology for refugee learners. On the first day we took our teams to NGOs who are working with refugees and interviewed the learners. As a result most team scrapped their initial ideas. I wish learning to analyse a challenge through the lense of the user was a more integral part of engineering education.
What makes you think that these companies didn't go through that process?
https://news.ycombinator.com/item?id=16264127
Also, the usual rule of thumb: "drones". Flying means constantly expending energy to fight against gravity. Unless the particular problem solved requires flying (to e.g. reach otherwise unreachable spots), solutions involving flying machines are almost always bad. And I don't see anything in reforestation that requires flying.
This is doubly important if you're trying to make the world a better place, instead of just making a quick buck. If all you care about is money, you can make it so that someone else pays the energy costs. But if you truly care about doing something good, then you have to factor them in (they almost always involve non-renewable sources and pollution) and ask yourself, if your plan is even a net improvement for the world.
Also, the usual rule of thumb: "drones". Flying means constantly expending energy to fight against gravity. Unless the particular problem solved requires flying (to e.g. reach otherwise unreachable spots), solutions involving flying machines are almost always bad. And I don't see anything in reforestation that requires flying.
This is doubly important if you're trying to make the world a better place, instead of just making a quick buck. If all you care about is money, you can make it so that someone else pays the energy costs. But if you truly care about doing something good, then you have to factor them in (they almost always involve non-renewable sources and pollution) and ask yourself, if your plan is even a net improvement for the world.
I can see two alternatives to flying, sending a person, or a remote-control/autonomous ground based machine. The first one is not as productive, the second one is much more complex, and so more expensive.
A ground based drone is orders of magnitude simpler than a multirotor. There is nothing magical about multirotors except being the latest fad for tech geeks.
A ground based drone — eg a modified bulldozer — can carry far more plants as cargo, meaning planting a field of a thousand sapling might require one trip where a flying drone will require a thousand trips.
The only catch is access: for a remote region with no trails of any kind it might be difficult for a large vehicle to navigate through the forest. In this case clearing a path for the planting machine might be a lower cost to the environment than the energy required to fly a thousand sorties with flying planters. A ground based drone will also be able to plant more robust saplings, since payload weight is far less of a concern.
A ground based drone — eg a modified bulldozer — can carry far more plants as cargo, meaning planting a field of a thousand sapling might require one trip where a flying drone will require a thousand trips.
The only catch is access: for a remote region with no trails of any kind it might be difficult for a large vehicle to navigate through the forest. In this case clearing a path for the planting machine might be a lower cost to the environment than the energy required to fly a thousand sorties with flying planters. A ground based drone will also be able to plant more robust saplings, since payload weight is far less of a concern.
If cost and fuel are major concerns then sending a person is by far the cheapest option. Not the most productive but in the areas that need reforestation (i.e. where they burn or clear cut, Asia / Amazon) the labour is not that expensive. Though there are a lot of other factors involved.
These kinds of projects remind me of Juicero, where it combines a few things western (middle/upper class?) people care about with technology and expects a great result but end up a middling disappointment.
These kinds of projects remind me of Juicero, where it combines a few things western (middle/upper class?) people care about with technology and expects a great result but end up a middling disappointment.
Agreed! As a mathematician/software engineer I have been looking for environment/ecological projects to work on, but it is really hard to find opportunities.
Does anyone know of good places to find opportunities to work on projects like this one?
Does anyone know of good places to find opportunities to work on projects like this one?
All the large (and many of the smaller) civil engineering firms have an environmental/ecological division and GIS, remote sensing and mathematical modelling is kind of trendy in ecology right now, so that would be a good place to start looking. Ecologists that know GIS and can code are quite rare.
The downside is that environmental concerns are largely considered a cost for civil engineering projects so much of your work will involve putting together models the 'prove' that the project under consideration will have a low environmental impact. But occasionally you will get a chance to work on a more 'positive' project as well. Either way it's probably a good place to start and to get experience with the realities of the field.
The downside is that environmental concerns are largely considered a cost for civil engineering projects so much of your work will involve putting together models the 'prove' that the project under consideration will have a low environmental impact. But occasionally you will get a chance to work on a more 'positive' project as well. Either way it's probably a good place to start and to get experience with the realities of the field.
I remember reading a couple of years ago about swarms of drones that would be used to police an area, and I feared that some governments around the world would REALLY love this idea.. mass control, mass spying protests etc.
Reading about deploying a swarm of drones to make this planet a better place makes me restore a bit more faith over human kind.
Reading about deploying a swarm of drones to make this planet a better place makes me restore a bit more faith over human kind.
When I was in 8th grade in Oregon. We went to a clear cut site to plant trees. We had to dig a very small hole and put the seedling into it. They had little burlap bags around them with a bit of dirt around the root ball. There was twine tied to keep the bag closed.
The ground we had to walk through was like a world war I trench zone with massive holes, ripped up trees, rocks, pits, it was a mess and hard to climb through. It was fun as a kid. But there was no way you could just throw trees everywhere and have them grow.
That being said, I can see how drones could do this with a tech change in how they are planted. Each drone has a few seedlings, the seedlings are in a pod that is weighted (keep it upright during planting) and self-sustaining (ie, plenty of dirt, water retention polymers, etc..) so they could be set on _top_ of the ground. (maybe some hex-shaped carboard box thing? lol)
Then drones get brought in with a big giant ass truck, you have 1000 drones, they make many multiple trips, recharges. It's possible.
The ground we had to walk through was like a world war I trench zone with massive holes, ripped up trees, rocks, pits, it was a mess and hard to climb through. It was fun as a kid. But there was no way you could just throw trees everywhere and have them grow.
That being said, I can see how drones could do this with a tech change in how they are planted. Each drone has a few seedlings, the seedlings are in a pod that is weighted (keep it upright during planting) and self-sustaining (ie, plenty of dirt, water retention polymers, etc..) so they could be set on _top_ of the ground. (maybe some hex-shaped carboard box thing? lol)
Then drones get brought in with a big giant ass truck, you have 1000 drones, they make many multiple trips, recharges. It's possible.
The article says "a drone loaded with germinated seeds fires (!) pods into the ground at a rate of one per second". The video makes it look like the drone will just fly over the terrain.
Maybe the pods fall from say 10m and that is enough to get them into the ground (and not kill them)?
Maybe the pods fall from say 10m and that is enough to get them into the ground (and not kill them)?
Well, I missed that part somehow. :P But those are seeds not seedlings, maybe a downside to the change in planting technique from the past. I guess I can see how it would be cheaper/easier to fire seeds all over the place though.
I don't think 10m (or even 100m) would get them into the ground. In my experience, the ground is a lot of rocks, roots and some dirt all in a big mess. Especially where there's been logging. But there's also a lot of upturned dirt, so who knows.
The best way to find out is to give it a shot. Can't be hard to fund a trial run. I bet it cost more to make the 3d video than to actual try it out and record the drone doing the work. :P
I don't think 10m (or even 100m) would get them into the ground. In my experience, the ground is a lot of rocks, roots and some dirt all in a big mess. Especially where there's been logging. But there's also a lot of upturned dirt, so who knows.
The best way to find out is to give it a shot. Can't be hard to fund a trial run. I bet it cost more to make the 3d video than to actual try it out and record the drone doing the work. :P
If these are the same drones I've seen previously, the germinated seed is inside a hard(ish) molded nutrient pod with a pointy end, and the drone fires it into the ground with a compressed air cannon. Which makes me very dubious about the '1 per second' rate of fire - even if it's possible, the weight of the gear + seed pods + gas is going to severely limit how many can be planted before needing to reload and recharge.
So realistically you could just throw these seeds from the air using a plane and it would be even more efficient?
Modify some sort of sprayer to spray seeds over a large enough area. With enough seeds there should be sufficient trees to survive and grow.
I generally agree, it would seem easier to find a way to be able to "drop" a pod that could anchor itself, or not need to be burried... Something with barbs to hold it to the ground?
Well, I don't need to solve their problem for them, lol. I hope this works in some way, it's sounds neat in theory.
Well, I don't need to solve their problem for them, lol. I hope this works in some way, it's sounds neat in theory.
I mean why not use gunpowder?
Planting 100K trees a day is an awesome idea, has anyone thought where the water is going to come for these trees?
TL|DR: same place where the trees that were previously here were getting their water. I don't think the article is about planting trees in Sahara or Gobi. It would be more useful in the replenishment/repair after a fire.
Silly example, but in the U.K. (almost everywhere on this island) it rains 4629362619 days per year! (or it feel like this much:)
I do understand that U.K. forests are not diminishing as they are on Amazon but I think (not sarcastically) they are called Rainforests.
I guess it depends on the reason for the deforestation. If it is just for timber, then a cheap solution to repopulate the area would be taken up by the lumberjacks themselves (so they can go back in 10-20-30 years and collect again and repeat the cycle). If the deforestation is simple to create more land for crops/living areas, then this is irrelevant.
Silly example, but in the U.K. (almost everywhere on this island) it rains 4629362619 days per year! (or it feel like this much:)
I do understand that U.K. forests are not diminishing as they are on Amazon but I think (not sarcastically) they are called Rainforests.
I guess it depends on the reason for the deforestation. If it is just for timber, then a cheap solution to repopulate the area would be taken up by the lumberjacks themselves (so they can go back in 10-20-30 years and collect again and repeat the cycle). If the deforestation is simple to create more land for crops/living areas, then this is irrelevant.
Does this only have to be used in areas strapped for water? I mean they mentioned Oregon, which gets pretty good amounts of rain naturally.
Someone needs to explain to the target audience here (politicians, I guess, since this is coming from the WEF?) that "proprietary algorithms" is NOT a good thing. One might argue about if or if not it is necessary, but for the customer, it is always a limitation on what they can do with the product they supposedly bought.
Ah! So they're selling a service, not a tool.
Let's see 10x faster.
About 1/5th the cost as a person: http://www.johnswope.com/?p=83
And a 20% savings. looks like a 10x margin to me...
About 1/5th the cost as a person: http://www.johnswope.com/?p=83
And a 20% savings. looks like a 10x margin to me...
ehhhhhmmmm no. It's not simple math. Or if it is, those are the wrong numbers to pick. Trees can reproduce themselves without human help and there are around 3.04 trillion trees in the world so a lot of potential for these net loss figures to be waaay off.
As far as we know, the number of trees in the world is actually growing. Higher levels of CO2 and the arrival of plastics and aluminium have contributed to reforestation of large areas of the world (we used to make everything out of wood). There are more trees now than there were 100 years ago.
According to the Food and Agriculture Organization (FAO), in the US, forest growth nationally has exceeded harvest since the 1940s. Same goes for Europe. The story might be different in Asia and certain parts of South America but in aggregate numbers I'd bet that we have net growth.