Hacker News new | comments | show | ask | jobs | submit login
Fog-harvesting system could provide potable water for the world’s driest regions (mit.edu)
18 points by interconnector 1305 days ago | hide | past | web | 19 comments | favorite

This title is a bit of an over-exaggeration. As the article says, "Systems to make use of this airborne potable water already exist in at least 17 nations." The title implies that it's relatively new technology.

For some historical background, an early fog harvesting project started in Chile in 1987, and the success from that kicked off projects in other countries. (Based on a cursory search I just did.)

The linked-to article then clarifies the importance of this new work: "But the new research shows that their efficiency in a mild fog condition can be improved by at least fivefold, making them far more feasible and practical than existing versions."

Fog harvesting's a lot older than that. The Bimbache people used a communal fog-harvesting tree to provide water on an otherwise uninhabitable island in what's now the Canary Islands. They led the drip lines to gravity-fed channels, which directed water straight to their dwellings. Clever systems like this demonstrate what we already knew — that indigenous people were (and are) no dummies.



Interesting! Though in reviewing the second link I find very little that was definite. It recounts things second- and third-hand. Some claim a single tree could provide water for 8,000 people and their animals, another said for 80 families. Some say it was a single tree, others from multiple trees.

You say "otherwise uninhabitable land", but Brown's 1910 guide book points out that people lived there, grew grains, and pastured animals. As there are "practically no springs", people needed to build cisterns to catch rainwater. The translated Spanish document (from the last 1500s?) says there are three streams on the island, and that the tree (which was not seen by the Spaniard) is a third type of water source.

Since the 1910 inhabitants didn't use this tree as a water source, I presume that cisterns and those small springs were enough for them, and so would have been enough for the Bimbache.

For that matter, how did the first Bimbache live there if it was otherwise uninhabitable. Did they need to find the tree(s) before they could settle?

You write "which directed water straight to their dwellings" but Duret (1605) writes that the tanks of water were at the foot of the tree, and that the remnants from the 1612 storm were still there in 1753. But I don't see anything in those accounts about channels or pipes leading the water to people's homes. (The Wikipedia page says "The Bimbaches later filled water chaffs and water was transported to towns to all parts of the island", but as part of a legend, and without a citation.)

This is not me rejecting the idea that fog collection was done before the 1980s. Indeed, the Wikipedia page for 'Fog collection' lists several "natural or assisted" collection methods. Nor am I rejecting the idea that the Bimbache used trees as a fog collector.

What I'm commenting on is that the information I've read is suggestive, but not definitive, and that what I've read doesn't fully match with what you wrote here.

What I would like to see might include measurements of water run-off from one of those trees, growing in the same area and archeological evidence of an irrigation system, rather than carrying water from cisterns.

BTW, I really like one of the accounts in that second link about that sense of astonishment by the British scientists at Cape Town who, 110 years ago or so, found out that the fog around Table mountain provides a huge amount of water.

Yes, it seems I overstated their dependence somewhat. Obviously water is important, so you want redundant systems so you have water in both rainy and dry times. Multiple systems on different scales (family, village, and island) is a possible explanation for the disparate accounts. Even today most human systems come in different scales.

>Since the 1910 inhabitants didn't use this tree as a water source, I presume that cisterns and those small springs were enough for them, and so would have been enough for the Bimbache.

That depends on how many inhabitants and what they're doing. I can't find historical population numbers, unfortunately. Today's population of ~10,000 uses industrial desalination.

>But I don't see anything in those accounts about channels or pipes leading the water to people's homes.

Check out pp 156: "It falls into a pond made of brick, floored with stones very tight, by pipes of lead conveyed from the tree to it, and thence divided into several ponds through all the island. They which dwell up-hill fetch it in barrels."

All in all, a fascinating technology. It's amusing how the western accounts attribute everything to luck/God, but that's colonialism for ya. ;)

(If you can't tell, I'm a fan of Clarke's [revised] Third Law, "Any sufficiently advanced technology is indistinguishable from nature." It's even been proposed as a solution to the Fermi paradox: http://www.nextnature.net/2012/02/any-sufficiently-advanced-...)

I discounted the telling on p156 because the population size (enough water for 8,000 people and 100,000 animals), the description of the tree (withered in the day and dropping at night), and being able to fill 20,000 tuns (about 1,000 L/tun) in a single day.

I also wonder how they were able to get the lead for those pipe. Other reports imply that there was little contact between the islands, so they likely didn't trade for all that material. Figure a 4 cm across, and 5 km of pipe gives 6 cubic meters of lead or 70 tons of lead, which is a full load on a Spanish caravel of the 1400s. And I think I'm underestimating that number.

I tried to find more historical information. The Spanish Wikipedia page points out that a well with brackish water was dug between 1702 and 1704, as part of a search to find drinkable water near the coast. It also commented that Columbus stopped there on his voyages to the Americas.

So I looked up a translation of the log of his first voyage. The entry for 9 Aug 1492 says:

> The Admiral says that many honourable Spaniards, inhabitants of the island of Hierro, swore that they were on Gomera with Dofia Inez Peraza, mother of Guillen Peraza, who was afterward the first Count of Gomera, and that each year they saw land to the west of the Canaries

"Many" is not helpful, but at least a few hundred people on the island?

I looked backwards in time. Jean IV de Béthencourt conquered the Canaries for Spain in the early 1400. "There was no resistance offered by the scattered guanche population who were largely sold as slaves" says Wikipedia. Nice guy.

Your best bet I think is to look for records in Spanish. Oh! I found "Conquista y antiguedades de las islas de Gran Canaria y su descripcion: con muchas advertencias de sus privilegios, conquistadores, pobladores, y otras particularidades en la muy poderosa isla de Tenerife", dated 1676,

My Spanish is rather poor, but I think page 59 of http://books.google.com/books?id=cKvrI2UF-LgC&pg=PA59&img=1&... says there were 1,800 inhabitants of El Hierro.

Here's what I could copy from that text (updating the syntax to more modern Spanish):

La isla del Hierro esta diez y ocho leguas de Tenerife, es pequeña, tiene su Villa con buena Parroquia, y dos Beneficiados, que acuden a toda la isla y algunos Capellanes; ay vn Conaento de San Francisco; cogese much trigo, y ganado; rendra la isla mil y ochocientos vecinos, y muchos principales.

Here's my+Google's translation:

El Hierro is eighteen leagues from Tenerife, is small, has its Villa with good parish, and two Beneficiaries, who flock to the island and some chaplains; (also a convent of??) St. Francis; grows(?) much wheat, and cattle; the island (has?) 1,800 residents, and many (bosses? people who can represent your business interest?).

So, a 1670 population of about 1,800 people, under the Spanish crown, and before the 1702 well. I think this find greatly weakens the hypothesis that fog collection was necessary in order for a group of up to 2,000 people, and that they could have survived on the limited springs and cistern storage of rain water.

I am unable to read the text you linked to. :(

Sounds like the Bimbache people could support 4x the population that the Spanish could (8,000 vs. 2,000), which makes sense (water being the limiting resource in that landscape).

Why discount pp156 again? The numbers are perfectly doable with a carefully placed tree that uses the surrounding landscape to funnel fog to it.

Lead pipes wouldn't be needed for the entire length — clay-lined spillways would more than suffice. I read it as using lead just to the cistern.

Great research, thanks!

Before starting, let me say that I've found this conversation very interesting. Thank you for participating in it with me.

I discounted it for several reasons.

1) There's no corroborating evidence that the population was that high. I haven't found anything more than "sparse", but http://www.sharprazor.com/palma-history.htm gives some of the history of La Palma:

> The Guanches named their island Benahoare, and divided it into 12 kingdoms, each with its own ruler. Estimates of the Guanche population at the time of the conquest range from 1200 to over 4000.

La Palma is over twice as big as El Hierro (708.32 km2 vs. 278 km2) and it seems to support better agriculture and easier access to the sea. I find it difficult to believe that the population density of El Hierro was at least 4x more than that La Palma, when the best I can find is that the population of El Hierro was "sparse."

2) The numbers I see for modern fog harvesting are about 5 liters per square meter per day. The account says that a tree could fill 20,000 tuns/day, which is about 20 million liters. This needs about 4 million square meters. Assuming an 80 meter high tree and 32 meters across (the size of a larger redwood tree) gives a cross section of 2,500 sq. meters. There's no way these numbers are close to compatible.

3) Let's assume the tree can capture an excess of 1 liter per sq. meter cross section, even though there's no evidence that any tree can do that. Let's assume also that a single tree has a cross section of 2,500 sq. meters. That's 2,500 liters per day. Not bad, but people need about 2 liters per day as a minimum - and more if they do much activity. Even in this unlikely and extreme case, that's only enough to provide basic water needs for 1,000 people.

The WHO says "a minimum of 7.5 litres per capita per day will meet the requirements of most people under most conditions", so that's 330 people. And no animals.

(Another estimate is to look at the amount of water in fog. A cloud has about 0.5 g of water/cubic meter. However, this is too complex for me to figure out, because I don't know how much water you can extract with fog mining, I don't know what a tree or other plant can extract, I don't the wind speed, and I don't know the capture area.)

4) Why use lead pipes to lead to the cistern? Since clay is good enough for the cisterns, what's the advantage of using lead for those pipes?

Since each individual detail is not trustworthy, I find it hard to put any faith in the overall account. How are you certain that it isn't a tall tale, with little to no basis in reality?

As wiki says it's just a legend.

Sorry about that; I didn't check if wiki was actually accurate. That'll teach me to trust an uncited article. :\

Check out the other link.

If used on a large scale, I wonder how this would affect the environment around the collection sites. As is now, the fog goes somewhere and I'm sure some organisms rely on it getting to where it's going.

If the water is collected with enough efficiency, there could be significant unforeseen consequences.

Hmmm, that reminds of Arakis. Are they called windtraps ? ;)

Or moisture vaporators in Star Wars, which Luke's uncle and aunt owned.

If h2o is heavier than other components of air like o2 and nitrogen why can't we just run air through a giant centrifuge to get the water out?

Generally with these schemes the fog is from the ocean so you have unlimited saltwater to desalinate, the idea is to get water with little energy cost.

Because that's more expensive than a sheet?

That would not be applicable in desert regions.

How so? From reading around a bit, I found several sources that indicate fog is present in the desert.

There's no basis for the idea that there is never fog in the desert, as long as there's sufficient water vapor. But a very dry desert climate cannot fog up for lack of water vapor.

Remember that fog forms when temperature and dew point collide. But without water vapor, there's no meaningful dew point.


Quote: "The dew point is the temperature below which the water vapor in a volume of humid air at a given constant barometric pressure will condense into liquid water at the same rate at which it evaporates."


Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | DMCA | Apply to YC | Contact