Applying AI to the California Water Crisis?

Singularity University recently closed its 2015 Impact Challenge Contest, asking for “exponential technologies” to solve the California water crisis. (Winners will be announced in a few days from this writing, October 26th.) We decided to share our thoughts on the “crisis”, and the many solutions that exist.  This includes solutions that can apply “exponential technologies” like artificial intelligence and big data approaches to the problem. (We’ve previously discussed the concept of “exponential” technology growth in the context of the superintelligence. As most experts readily admit, exponential technological growth is not actually a new phenomenon.) In particular, there is one tried-and-proven 19th-century technology that today widely employs “exponential” technologies like artificial intelligence. For some reason, California is not (yet) applying this technology to the water “crisis” despite America’s widespread familiarity with it. We found over 16 water conservation technologies (some we just remembered from old textbooks) that have been successfully applied elsewhere in the world, but not in tech-savvy California. Normally, when you see a very effective 1960s technology like drip irrigation not applied here, it indicates some subsidies or some other government interference in the marketplace. Generally, if the government wants to use taxpayer dollars to support certain favored industries, there are better ways to do this than subsidies, which discourage conservation and innovation. It turns out we aren’t the only ones with these suspicions. It has been argued elsewhere (such as in this Reddit AMA with a prominent water economist) that the California water crisis is primarily political involving reluctance to end subsidies. If the problem is primarily political, is Futarchy a proven 19th-century way of applying so-called “exponential” technologies like artificial intelligence to the solve problem? 

Introduction

There is a very clear, proven “exponential” technology. This 19th century technology is widely used in the United States today. In modern times, it relies heavily on computers and artificial intelligence technologies. Although it is not currently much applied to California water shortage issues, this heavily computerized (and therefore “exponential”) technology can be adopted to the water crisis. This requires, however, an enlightened understanding of this technology on the part of government leaders and regulators, who are essential to its successful implementation. (There is some evidence that this enlightened understanding of existing exponential technologies may be lacking at the moment, which may be contributing to the crisis, as we shall see.)

Before describing this “exponential” technology, we must first convince readers that there are many old-school non-exponential technologies for dealing with the water crisis.. Most of these have been proven elsewhere in the world but have not been tried in California These proven but largely unimplemented and “non-exponential” technologies can be made exponential by incorporating them into an exponential planning system.

There are many of these “old school”, “non-exponential” technologies. Some of have been used very effectively in other parts of the world. They are not used in California because of historical concerns about their costs. (Compared with other parts of the world, water has historically been cheap and plentiful in California.) The key to solving the water crisis in an exponential way is to use exponential technologies to better coordinate, facilitate, and choose among these many effective but expensive “old school” water technologies.

Old-school water solutions largely unused in California

Therefore, let us begin this essay on “exponential” solutions to California’s water crisis by first considering the many “non-exponential” solutions that are not currently being used:

1. Agricultural best-practices, such as drip water irrigation systems. Agriculture accounts for 40% of water use in California but contributes less than 5% of Caliornia GDP according to Wikipedia. (Those are almost farm-industry figures. Critics of California industry claim agriculture uses 80% of the water but accounts for only 2.1% of the economy. Reference: LA Times) Primitive drip-irrigation systems have been in use in China since the 1st century BCE (just not much in techy California). Modern drip-irrigation systems, which use an inexpensive plastic pipe, were famously pioneered in the Israeli desert by the early 1960s. Although widely used in Israel and elsewhere, it is rarely used in California because water in California was historically inexpensive compared with Israel and other parts of the world (http://time.com/3063/california-drought-5-way-to-bust/). As noted above, when a proven cost-effective technology isn’t applied in a tech savvy place like California, it usually indicates subsidies or some other government interference in marketplace. (We aren’t alone in these suspicions.)

2. Dynamic shifting of crops. Within California agriculture, crop-selection remains somewhat traditional and not-water optimized. For example, despite the lack of water, California continues to grow significant amounts of water-intensive rice. Crops should be optimized dynamically given current and expected crop and water prices. Our “exponential” solution provides for dynamic optimization of crop selection by farmers. However, it should be noted that the current crop selection in California is sub-optimal and can be optimized even without our exponential technology.

3. Xeriscrapping, water recycling, and conservation (as already noted in the above-referenced Time article).

4. Extract water from air (“atmospheric water generation.”) This is a proven technology, and is in use in some rural areas, or under emergency circumstances. However, it is generally much more expensive than desalination. Large-scale use of this technology would likely modify the weather as well, and thus may be considered a form of weather modification. https://en.wikipedia.org/wiki/Atmospheric_water_generator

5. Improved desalination and/or increased use of desalination. California has a very long Pacific Ocean coastline, so this is a somewhat obvious solution. Although desalination is used in some areas of California, such as in San Diego, it has historically been considered too expensive. The principle cost in desalination is energy input, or electricity.

6. Thus energy and electricity technologies are closely related to desalination. Improved or less expensive sustainable power generation (energy) technologies, e.g., solar, wind, fusion, &c should be considered water technologies. California is on an ocean, so in principle any amount of water can be extracted via desalination, provided the energy is available. The primary cost issue (other than the immediate issue of constructing the facilities) is that desalination currently requires a lot of electricity/energy. Even if the amount of energy required were to be greatly reduced through improvements in desalination technology, there will always be a theoretical minimum energy required for desalination. This is because desalination moves matter from a more disordered (salt) state to a less disordered (fresh water) state. Consequently, the laws of thermodynamics require some energy expenditure to always be associated with desalination. Once this limit is approached, future cost improvements in desalination will mostly likely come with reduced energy costs. This is consistent with the Kardashev-Sagan scale for civilization, that links energy and information processing power to civilization and technological advancement.

7. Weather modification / geoengineering/astro-engineering. Amongst other things, make it rain, or modify the climate to make it rain more. We have already discussed atmospheric water generation as essentially a form of weather modification. Weather modification, of course, has all sorts of legal, political & ethical problems associated with it. If it rains in one place as a result of weather modification, is this “stealing” water from another region? What about ecological harm? Unintended consequences? On the geo-engineering side, if global warming is responsible for reduced rainfall, then engineering techniques to reduce CO2 in the atmosphere or otherwise alter the planet’s albedo might come into play. Geo-engineering techniques are extremely controversial and generally considered currently illegal under international law. This may change as these technologies improve or at least become more familiar and/or the world becomes increasingly desperate to address climate change. If climate change is responsible for California’s water shortage (as at least some scientists believe) geo-engineering technologies offer the possibility to kill two birds with one stone.

8. Improved groundwater sensing technologies (e.g., satellite, ground-penetrating radar). In addition to helping locate new sources of groundwater, this technology can also be used to determine appropriate sustainable levels of groundwater extraction. Thus, it might be used to calibrate extraction rates, reducing groundwater consumption today in favor of improved longer-term management of resources.

9. Deep-earth mining and/or chemical synthesis of water. (http://www.sciencemag.org/content/344/6189/1265 https://en.wikipedia.org/wiki/Beijing_Anomaly).( http://www.washingtonpost.com/news/morning-mix/wp/2014/06/16/study-deep-beneath-north-america-theres-more-water-than-in-all-the-oceans-combined/) There is three times more water below North America alone than in all the world’s oceans. However, this “water” is chemically “locked-up” in the form of hydrated minerals. (One wonders whether this mantle “water” is in some kind of geo-chemical equilibrium with the Earth’s oceans, and whether this under-reported and recently discovered phenomenon is in any related to the early emergence of life on Earth. The depth to which the biosphere extends into the mantle is unknown but thought to be significant due to extremophile archaea, thought to be related to the first earliest life forms. Deep-earth water, even if “locked-up” as mineral hydration, would likely further extend the biosphere downward, deeper into the Earth, permitting further geochemical planetary stabilization by microorganisms. Furthermore, if a geochemical equilibrium between oceans and mantel hydration did exist, it would affect early-Earth models of the heavy bombardment  and asteroid impacts, some of which have Earth’s oceans evaporating completely for extended periods.) Bottom-line: there is no shortage of water on this planet, although getting at some of it may prove prohibitive. Desalination should be much cheaper than deep-earth mining. Deep-earth hydration is nevertheless worthy of study, as there may be hidden relationships between ground-water and deep-Earth hydration that impact water management. Although the price-tag is prohibitive, we can expect additional deep-Earth wells to be dug as part of deep-Earth sensor efforts in fields such as earthquake and volcano prediction. As more resources become available due to exponential growth in other technologies, society will focus increasingly on threats from super-volcanos and earthquakes. As a result, we anticipate these sensors wells will be dug in the future despite their cost. Sensors for deep-Earth hydration will also be added, so these sensor wells can also be considered a groundwater-sensing technology.

10. Astro-mining of water. Astro-mining will indeed be needed for many other precious resources for which fewer practical terrestrial options exist, so the capacity to mine water in space will exist shortly. A number of companies, including some funded by Google (now Alphabet) are interested in developing astro-mining technologies. Although we now know of significant sources of water elsewhere in the Solar System, retrieving it back to Earth would be completely impractical and cost-prohibitive given all of these other options. Furthermore, water will be extremely useful to future space colonies in these regions, and so is best left there as a precious future resource. It is enumerated here merely in the interests of completeness.

11. Construction of additional aqueducts and other engineering measures to divert water from regions that have access water, such as Idaho or Canada. These are tried-and-proven water technologies but historically have required substantial planning and financing.

12. Converted oil ocean tankers to move water from Canada. This technology was successfully used in the 1980s to relieve Californians drought, and was apparently halted for primarily political reasons. Oil tankers (sea going or highway trucks) can again be used to transport large amounts of water from less expensive sources to California for agricultural use. Restrictions on export of freshwater may be any issue with this solution.(Reference: https://en.wikipedia.org/wiki/Droughts_in_the_United_States )

13. Electronic toilets. Unlike old-fashioned water-saving toilets, electronic toilets have the potential to become an ‘exponential technology.’ In addition to automating multiple-flush systems, a dual-water version of these devices might switch between sea and freshwater, with freshwater used for cleaning and automatic medical testing. Although commercially available for a number of years and popular in countries such as Japan, electronic toilets have not caught on in the United States outside of a few early adopters such as the Google headquarters. It is thought that the subject of toilets is taboo in the United States, and that attempts to market these products merely brings an emotional response such as a giggle rather than any serious consideration. This is unfortunate. From a consumer-standpoint toilets are self-cleaning, which means they can significantly reduce house-holding cleaning time and expense. From a government and insurance standpoint, these devices can provide early detection of a variety of serious illnesses, potentially adding billions to the economy. In addition to early warning of various metabolic illnesses, these ‘bathroom appliances’ should most notably provide early & complication-free warning, through continuous monitoring of stool, of colon cancer and pre-cancer, which is otherwise extremely difficult and expensive to timely detect. These devices also provide important hygiene benefits, further lowering health-care costs. And they do save water. Like dual-water systems, however, these devices require some planning in the form of revised construction codes. Specifically, the devices require installation of an GFCI electrical outlet. Since these outlets are expensive to retrofit, new construction should include them, which makes installing this kind of ‘bathroom appliance’ much more affordable.

This is an ‘exponential technology’ as both the on-board computer and laboratory-on-a-chip can be continuously upgraded to take advantage of Moore’s law. (Although the Moore’s law benefits would likely apply more to health monitoring that further water savings. However, an onboard computer, together with a dual-water system, could indeed continuously devise new water-saving strategies in response to current market conditions.)Unfortunately, outside of late-night comedy television, there will likely continue to attention given to planning for these devices.

14. Conversion of wastewater and stormdrain water to human use. Apparently, similar economics to desalination, but applicable in areas away from the ocean. With estimates suggesting residential use is only 6% of water use in California, not clear “human use” technologies will address the problem.

15. Satellite/aerial optimization of agricultural watering. This has been used successfully and cost-effectively in countries such as France.

16. Dual-water systems. Showers and toilets each account for roughly 20% of residential household water use in California. Water for both of these activities can instead be sourced from untreated or lightly-treated seawater. In addition, other household appliances, such as dishwashers and washing machines, could be modified to use seawater in some of their cycles. (For example, the first wash cycle in a dishwasher could use seawater.) Dual-use water systems are an old technology in the United States. The city of St. Augustine, FL, the oldest city in the United States, converted to dual-water several decades ago. (These systems were also widely installed in the former Communist countries in Eastern Europe, where brine from groundwater is used instead of ocean water.) Thus conversion to dual-water is more than economically feasible. With untreated seawater essentially free in California, dual-water systems could potentially reduce the residential freshwater bill by 50% or more.Dual-water systems do require some planning. We here talk about the current drought extending until 2030 or more. If that is the case, construction codes should be changed to mandate the installation of dual-water pipes in all new construction. (The salty-side of ‘dual-water’ need not be activated until later.) Manufacturers should be encouraged to develop dual-water dishwashers, washing machines, and electronic toilets. Existing household appliances such as shower heads and traditional toilets can safely run on salt water with little or no modification, but it may be desirable to have a valve on the decide to permit temporary use of fresh water, such as for cleaning or medical reasons.However, with such a large fraction of water in California going to agricultural and environmental use (although the exact percentages are debated), it is not clear if residential water-saving technologies are worth the expense. (Reference: https://en.wikipedia.org/wiki/Dual_piping  among others)

Futures markets

It’s already been noted that California’s crop selection is likely sub-optimal given current drought conditions. (For example, as of this writing California grows far too much rice, a water-intensive crop that can be efficiently sourced from many other rice-growing parts of the world.) On the other hand, if every farmer in California selected the same “water-efficient” crop, this would also lead to calamity.

Article continues next page.

1 2 3