Making The Desert "Blossom As The Rose."

A New Approach to Arid-Land Agriculture

An essay in hypertext by Scott Bidstrup

"The wilderness and the wasteland shall be glad for them; and the desert shall rejoice, and blossom as the rose"
--Isaiah 35:1

Why This Essay

As I approach the end of my life, I have begun to realize that my economic circumstances are such that I'll never be able to achieve some of my dreams.

One of these dreams has been to find a way to do agriculture in the desert wastes of the American west.

This has been a lifelong dream for me. Even as a child, I would ride through the sagebrush steppe of the Snake River Valley of southern Idaho where I was growing up, and think to myself what a waste it is that this land is suitable only for grazing of cattle, and then, with only very poor productivity! When I would ask my parents why no one was farming this land, the answer I always got was simple - one that would be obvious to anyone traveling this land in July or August: This land would be highly productive farmland except for the lack of water. The lack of even groundwater in vast areas of this high desert mean that otherwise fertile land sells for as little as $100 per acre, or even given away by the Bureau of Land Management to those who can show that they can make the land agriculturally productive through "irrigation." Yes, you can still homestead desert lands, and this may well be a way to do it!

In considering this problem, I have begun to realize that the apparent lack of water is actually a solvable problem. In the high deserts of the west, where nothing much but sagebrush, greasewood and saltbush grows today, there is an agricultural bonanza for those willing to use creativity to solve the problem of limited rainfall and a lack of surface and ground water. Yet this land actually does receive water - about 7 to 11 inches of moisture, mostly rain, each year, with relatively good reliability. The key to using this water is to collect it, use it, and carefully conserve it, so that it is as productive as possible. In the process, the winter cold of these deserts is moderated to the extent that winter agriculture is not only possible, but highly productive.

In this paper, I will outline a method of conserving this water with sufficient efficiency that it will quadruple or quintuple the use of it - making a farm with 11 inches of rain per year effectively have as much water available to growing plants as a ordinary farm that receives 40 or 50 inches per year, with the advantage that the application of the water will occur precisely when the farmer needs it - reducing one of the great uncertainties involved in farming. The reduction in uncertainty will change farming from an industry resembling gambling into something more resembling manufacturing. In addition, the isolation imposed by this system between neighboring farms growing similar crops, means that organic growing methods will become much more feasible than in conventional agriculture.

If, by now, you have figured out that I'm talking about greenhouses, you would be right. But I'm not talking about ordinary greenhouses, but greenhouses with a difference - a big difference. In fact, two big differences.

My first extended visit to a greenhouse came when I was a freshman in college. I was working part time, in the dead of winter, as an electrician's apprentice, and had been sent by my employer to a greenhouse being built near Rexburg, Idaho, where I was attending school. The wiring I was doing was for the gas heaters that were intended to keep the greenhouse warm at night during those arctic Idaho winters. When I asked the farmer if it wasn't rather expensive to use gas to heat a greenhouse, he replied that it was his highest single operating cost, and only the high price of hydroponic local tomatoes in the winter would be sufficient to make it profitable. Nevertheless, I was struck by the warm, cozy atmosphere of the greenhouse in the daytime, even on the cloudy winter day I was there working on the wiring. In fact, the ventilation fans came on periodically while I was there - dumping that valuable heat into the Idaho winter outside, knowing that just a few hours later, the farmer would be burning natural gas to replace it.

This made quite an impression on me. I realized that the cost of heating is clearly the barrier to winter agriculture in greenhouses, and in summer, heating is largely not needed - in fact, cooling is the priority. This is why greenhouses are so uncommon in Idaho and other areas with severe winters, and where they do exist, they serve only high-margin narrow niche markets.

Well, recent developments have led to solutions to both of these problems - the lack of rainfall and the need for supplimental heating at night. Not only is it now possible to heat greenhouses without significant energy input from fossil fuel sources, but it is possible to collect, conserve and utilize rainwater so efficiently that supplimental water sources should never be necessary.

An additional benefit to this system is that it provides almost free cooling in the summer months - the farmer can maintain the optimum growing temperature and humidity all summer long, regardless of the ambient outdoor temperature or humidity

Surely by now you have some curiousity about how this integrated greenhouse system works. It works through two modifications to the way greenhouses are built.

Dealing With The Water Problem

The diagram at right shows the basic method of conservation of rainwater. The greenhouse is built over an artificial aquifer, which consists of an impoundment lined with a membrane to make it waterproof. The key is that it is filled with river sand, so that water can permeate it and the pores between the sand grains will serve as the water reservoir.

Why build the greenhouse over the reservoir? For several reasons. From the standpoint of water conservation, the presence of the greenhouse will capture moisture lost through capillary action, and allow it to be recovered through the HVAC system and re-used, as will be explained later. Additionally, any excess water applied to the crop that filters down through the soil, will be captured by the reservoir. This is crucial. Another reason is it will become an essential part of the HVAC system that will be discussed below.

How deep does it need to be? The short answer is that it needs to be deep enough to store all the water that is intended to be stored, divided by the porosity of the sand. If a given volume of sand will store one fourth its volume in water, then the reservoir will need to be four times the depth of the water to be stored.

How much water will need to be stored? At minimum, it should be equal to the roof area of the greenhouse, times at least twice the depth equal to the annual precipitation rate. Since the area of the roof equals the area of the reservoir, the total volume of water the reservoir will need to store will simply be twice the six inches or 12 inches of water (this is to take advantage of wet years to accomodate the need for additional water during drought years). Since the sand in our example will store water at one fourth its volume, the depth of the sand in the reservoir will need to be at least four times that depth, or at least four feet.

Will this be enough water to supply the needs of the plants being grown? Yes. Does the water consumption requirements of any given species being grown need to be considered? Not really. Water-loving plants typically transpire a great deal - they give up their excess moisture to the atmosphere, and in copious quantities. But since we're going to recover the moisture from our atmosphere, the fact that they do so is of no consequence to us. We'll see later, in the HVAC section, how this will happen automatically for us.

Water entering the system will meet one of four fates: First, and ideally, it will be absorbed and used by the crop, and will be lost to the greenhouse only when the crop itself is removed from the greenhouse. This, however, represents an insignificant loss we can neglect in our calculations. Second, water will be transpired from the crop, through evaporation from its leaves, in which case it finds itself in the air in the greenhouse, where it will be recycled through the HVAC system. Third, it will fall on the soil in the greenhouse and be evaporated into the air in the greenhouse. Fourth, excess water falling on the soil will percolate through the soil and back into the reservoir. Of these four, the second and third represent a problem. Since the air in the greenhouse is gradually exchanged with the outside air, any moisture it contains is lost to the system. How much is lost? It depends.

The amount of moisture lost from the greenhouse to the outside air is a function of three parameters: the relative humidity of the inside air being lost, the relative humidity of the outside air replacing the inside air, and the rate of exchange. I am going to assume that great care has been taken in the design and construction of the greenhouse, and that it is sufficiently tightly constructed that only four complete air changes occur each day. Since I'm not venting air to the outside, nor drawing it in, it should be feasible to construct a greenhouse this tightly.

How much is lost per exchange? Well, it turns out that at 80 degrees f., with 100 percent humidity, there is a gallon of water in roughly each 4,343 cubic feet of air. Let's assume our greenhouse has a width of 28 feet, a depth of 96 feet and an average roof height of 10 feet - this means the interior volume is 26,880 cubic feet of air. Divided by 4,343 cubic feet, this means that there is 6.1 gallons of water in the form of water vapor in the air at 80 degrees and 100 percent humidity. It means we will lose 6.1 gallons of water with each exchange of air to the outside (assuming the air that replaces it has no moisture in it at all). Since the outside air will have to be heated to the ambient temperature of the greenhouse, and its relative humidity will fall as it is heated, we can probably figure that outside air will have only 5 percent humidity when heated to 80 degrees. This means it will replace only 5 percent of the moisture we have lost. Therefore, the net loss per air change is 6.1 gallons, minus 5 percent, or a net loss of 5.87 gallons per air change. Of course, on wet and cloudy days, the outside air raised to 80 degrees will have considerably more residual relative humidity, and that's a bonus to us. But let's take the worst case and assume four air changes per day of nominally dry air, all year round, we are losing 23.5 gallons of water per day through air exchange. But how much can we afford to lose?

If we assume six inches of rainfall per year, our roof, which has a collecting area of 2688 square feet will collect 2688 times one half, or 1344 cubic feet of rainfall. That works out to 11,209 gallons per year. If that's our sole source for water, we can afford to use 11,209 gallons in a year, or 30.7 gallons per day. We're losing 23.5 gallons per day through air exchange, so we've got a surplus of roughly 7 gallons per day. We should have plenty of rainfall, at six inches per year, to meet our needs and then some!

Since we're recycling our water, we are also recycling our fertilizer. This has three advantages to us; first, none of our expensive fertilizer is lost to the environmental pollution, and second, since none is lost, we don't need to worry about regulatory problems. Third, our cost for fertilizer is greatly reduced. This will more than compensate for the cost of pumping the water out of the reservoir and sprinkling it onto the crop. This is essentially a hydroponic growing system, but with real soil rather than simply a rooting medium. Of course, care must be taken to ensure that the quality of the water is maintained, and that toxic solutes don't leach out of the soil and build up. This is where our surplus water comes in - we use it to dilute toxic solutes that may accumulate in the reservoir water, which can be discharged to an evaporation pond outside, where the solids can be readily disposed of.

Dealing With The Heating And Cooling Problem

The problem of heating and cooling the greenhouse is one that has been vexing greenhouse owners since the invention of the greenhouse itself. Cooling has traditionally been accomplished by simply venting hot air to the outside and allowing it to be replaced by (hopefully) adequately cool air. To say this has always hit and miss, is putting it charitably - it has been one of the main uncertainties of the greenhouse business. In the summer, particularly in warm summer areas, adequately cool air is not always readily available from the ambient air outside the greenhouse.

Heating, particularly in cold night and cold winter areas, is an even greater concern. As noted above, it's always been the limiting factor on the commercial viability of greenhouse operations.

Well, again, new developments have solved that problem. What has that new development been?

Surprisingly enough, it is the introduction of cheap plastic pipe.

Figure 2 shows how I propose to utilize cheap plastic pipe to provide our heating and cooling requirements.

The introduction of cheap plastic pipe has the potential to bring agriculture to many areas of the world where agriculture is not currently possible. Why is it so revolutionary?

It is revolutionary for the simple reason that it makes possible the easy and cheap use of geomass thermal reservoirs for solar heating and cooling. Here's how it works. In this case, air is pushed (2)into some large - 15" PVC - sewer pipes (3) buried in the water reservoir. The water is an excellent thermal reservoir - so the pipes are located right at the bottom of the reservoir. That ensures that they'll always be wet on the outside - facilitating heat transfer between the sand and water and the air being pushed through the pipes.

Since the water reservoir by itself doesn't provide a large enough heat reservoir, we are going to use four separate ones, arranged on the outside of the greenhouse area (figure 3). The water reservoir exchanger (10) will serve only as an emergency heat dump or source. The outside four exchangers are considerably larger, and will serve as the principal heat sources and sinks. There will be two large blowers (8) which will force air through two of the four heat exchanger grids, and which ones will be determined by the position of the Modulflow dampers (9).

When the cooling occurs below the dew point of the air (which it will be nearly all the time that cooling is occuring), moisture condenses out and is made available for recycling.

During the cooling season, the four outside reservoir systems will be cycled, two at a time, until they are full of heat. In the event of a long stretch of unusually hot weather that leaves all four of the thermal reserviors full of heat, and they can't store any more heat from the greenhouse, the system will switch to drawing and blowing air through the fifth reservoir (10), located in the water sump, until it's full of heat. If necessary, it will then cycle at night to draw air from the outside and dump the heated air to the outside at night to cool down the reservoir as much as possible, and then resume cooling in the morning. It will continue this behavior until one or more of the outside four heat reservoirs falls in temperature enough to maintain a safe temperature in the greenhouse. That's why the Moduflow dampers (1, 8). They're an emergency means of dumping heat without losing moisture from the greenhouse. Acting in this manner by itself, the reservoir should have sufficient heat transfer capability to maintain a safe temperature in the greenhouse to get through the day and make it safely through the hot spell.

The heated thermal reservoirs represent the stored heat that will be used during the heating season. Since it is conducted through the ground very slowly, the heat will be adequate to maintain a warm greenhouse all winter long, when the heat the greenhouse will absorb from the sun during the winter months is taken into consideration.

The thermal reservoir pipes will collect condensate whenever the surface temperature of the pipe is below the dew point of the air being pushed through them. This is a good thing - it enables us to recycle the water that is transpired from the plants. By putting the pipes on a carefully designed gradient, the condensate will run to low spots where it can be collected and pumped back to the reservoir by sump pumps. This will serve to maintain a reasonable humidity level in the greenhouse, keeping plant diseases to a minimum.

One might think that running water through copper pipe would be superior to using a much larger plastic pipe carrying air for the heat exchange, but it's not. On the contrary, it turns out that the much larger plastic pipe carrying air is much superior. The surface area of a one inch copper water pipe one foot long is 45 square inches; for a four inch plastic pipe, it is 158 square inches. The difference is enormous - in spite of the lower thermal conductivity of the plastic pipe, each foot of 4" plastic pipe has more than three times the thermal transfer efficiency than the 1" copper water pipe! Use of air moving through a large diameter plastic pipe also eliminates the cost and complexity of using a heat exchanger system. One simply blows the greenhouse air through the pipes directly to accomplish the heating or cooling required.

Knowing that we want to do heat exchange to the geomass, we need to calculate the total amount of geomass we need to couple to, and how big our heat exchange network needs to be to accomplish the heat exchange we need to accomplish.

Through a complex series of calculations, I've determined that each pipe will readily exchange heat with a three-foot radius of geomass surrounding it (over the span of a few days), and much more slowly (over a period of a few weeks) with a radius about three feet beyond that. knowing that, I put together a quick and dirty feasibility spread sheet that enabled me to do the calculations based on the mean temperatures and insolation (solar energy) values for my home town of Idaho Falls, Idaho. Remembering it as a place with a perfectly awful winter climate, I figured it would be a good place to test the viability of this system.

What I discovered was that if I capture solar heat during the summer, store it in geomass during the fall and use it as I need it during the winter, I not only have plenty of heat stored up to heat the greenhouse during the winter, but have lots left over to lose to the geomass surrounding our thermal reservoir through conduction. In fact, in our example greenhouse design, the greenhouse will gain about 510 million BTU more in the summer than it will lose in the winter - heat we'll ultimately have to get rid of. Should it be necessary, the emergency cooling system discussed above should be more than capable of dissipating the excess heat.

The main parameter that drives cost in this system is the cost of installing the geomass thermal pipes, and therefore the closeness of temperature regulation that I am trying to maintain. The less the differential, the greater the mass that I have to couple heat transfer into. This is because the highest temperature I can achieve in the geomass reservoir is the same as the greatest temperature that I can tolerate in the greenhouse. Since the effective thermal storage is the mass storing the heat times the temperature differential between the heat source (geomass) and the heat sink (greenhouse). Therefore, allowing a greater temperature differential between the geomass (at a maximum of, say, 95 degrees in the geomass), and 55 degrees in the greenhouse, each parcel of dirt in the geomass will yield three times the heat capacity than it will if the greenhouse is only allowed to rise to 85 degrees and fall to 75. Assuming the average greenhouse temperature remains the same in both cases, one can quickly see that the ability to store heat in each parcel of soil is directly proportional to the maximum allowable temperature differential. By plugging in different numbers for the allowed differential, tab three on the spreadsheet will allow you to see the difference in the costs. If you are contemplating the use of the system for a delicate crop (such as ornamental flowers), you can quickly determine whether the additional cost is reasonable.

The thermal reservoir is buried under a six-inch deep layer of polyurethane foam (6). This foam layer serves two purposes: first, it is thermal insulation that prevents the escape of reservoir heat through conduction and convection to the atmosphere where it is lost. The second purpose it serves is to collect moisture on those rare days when the soil is saturated, and direct it to the thermal exchange pipes where it is needed to facilitate heat transfer. It also prevents that soil mosture from being lost, so that the pipes are kept constantly moist. The cheapest foam available that is stable in soils, polyurethane, has an R-value of 7 per inch; this yields an insulation value of R-42 between the thermal mass and the soil directly above it. This should be sufficient to prevent a significant amount of heat from escaping from the reservoir. Since heat, even in soil, tends to rise, it is necessary to only insulate the reservoir on the top. The soil below simply acts as additional free reservoir over longer periods of time. As the system is used, the soil beneath the reservoir mass will tend to slowly rise to the mean annual temperature of the reservoir; this will actually improve the performance of the reservoir system during the heating season.

There are small breaks in the foam directly above each of the heat transfer pipes. This is to allow soil moisture to seep through. It is important for this to happen; if the soil adjacent to the 4" heat transfer pipes dries out, heat transfer is greatly impeded. For this reason a vertical paper barrier should be placed along the foam so that ground water can soak into the paper and down through the foam insulation, and seep around the pipes. The paper will quickly rot away, leaving a very thin crack through the thickness of the insulation, through which the water can seep. The foam layer should be slightly crowned in the middle between the 4" pipes to force the moisture to run to the cracks.

Financial Considerations

While I didn't calculate a full-blown business case for this system, as that involves so many variables that I can't even estimate reasonably, I have looked into how much more it would cost to build this system than it costs to build a comparable greenhouse without it. Those costs can be seen on the third tab of the spreadsheet. You can add that to the cost of a standard 28x96 greenhouse and determine for yourself if it makes sense.

What you should consider is that you can provide fresh local soil-grown produce to a local market that is some distance from sources of winter vegetables now. The most obvious first target for this system would be near the bigger cities of the interior west: Albuquerque, Phoenix, Reno, Las Vegas, Salt Lake City, Ogden, Provo, Denver, Boise, Pocatello, Idaho Falls, Cheyenne, Laramie and Helena. With the possible exception of Denver, all these cities have vast tracts of land nearby that sell for very little money. All of these cities should provide ready markets for produce during the winter months if it is fresh, table ready and vine ripe. While the cost of the construction of this system is substantially higher than conventional greenhouse systems, bear in mind that this system has several advantages:

As you can see, this system offers some considerable economic advantages. Obviously, the longer one amortizes the cost of the heating system, the cheaper it becomes as part of your yearly operating cost. In the example given in the spreadsheet, a cost of $50,000 can be amortized over 20 years, leading to an amortized cost of $2500 per year for the operation of the greenhouse. After that, its free. Since this is less than the typical cost of heating a greenhouse, even if a natural gas supply were available, it should be a winner economically. Since nearly all the cost of operating the greenhouse is the amortization of the investment, an additional advantage is realized - long term costs are known to a high degree of certainty, meaning that long-term financing should be much easier to arrange.

Practical Design And Operation Considerations

When looking for a site, there are a number of considerations that one should look into. Distance to markets and suppliers is, of course, an important consideration as it will drive recurring costs in your business plan significantly.

The National Weather Service can provide guidance on the annual rainfall and variability. One should look at the thirty-year record and observe the lowest amount of rain to occur during that period. If it was part of several years of running drought, one should take the running average of that drought to be safe in terms of calculations of rainwater availability.

Beyond these, there are some gotchas that have occurred to me when writing this article, that may not be obvious.

First, a site should be selected that has sufficient subsoil depth to allow for the excavation work that will be required. That's obvious, but many desert soils in the western U.S. are not as deep as one may think. The western states tend to be active geologically, and as a result, soil depths tend to be shallow. Investigate this before purchasing a site!

Second, when calculating the depth of the required water reservoir, the sand porosity should be tested to determine the total amount of sand volume that will be required. Since sand varies in its porosity, it is important to test it, and make this determination.

Third, when designing the rainwater collection system, it is important to include some filtering to keep dust out of the sand in the reservoir. A simple rapid-sand filter, constructed of a trough of sand from which the top layer can be manually scraped off and replaced periodically should be quite sufficient. Failure to do this will result in the sand in the reservoir eventually getting plugged up with airborne dust washed in from the roof and into the sand; this would necessitate a very difficult and expensive replacement or cleaning of the sand. The diagram above shows the water collection pipes inside the greenhouse for a reason - freezing. It will be necessary to install heat tapes on the rain gutters outside to ensure that the gutters remain sufficiently free of ice that they do not allow precious moisture to escape. Running the collection pipes on the inside, along with the sand filter, will ensure that the water loss due to ice plugs is minimal.

Fourth, in designing the geomass heat exchanger piping, it is necessary to ensure that condensate will run to and be collected at intentional low spots. A sump pump should be placed at the low spot(s) to remove accumulated condensate. This must be returned to the water reservoir or retained for use where highly pure water is needed (such as nursery seedlings, etc.). Failure to do this will result in stagnant water accumulating in areas where it cannot be removed; such water will eventually cause the growth of undesirable fungi and other organisms. The standard one inch of gradient per ten feet should be sufficient. Air intakes should have furnace filters placed over them to prevent dust accumulation within the pipes. Once dust is in there, it's there for good, as there's no way to clean it out. Since deserts are dusty places, that's inevitable, but it can be forestalled by filtering the air before it goes in.

Fifth, when assembling the heat exchange piping, it is vitally important to ensure that all glued joints are absolutely watertight. The reason for this is to prevent the loss of recycled condensate, and intrusion of ground water into the heat exchange pipes. In desert soils, ground water often contains large amounts of alkali, and should alkali intrude into the condensate collection system, it will contaminate the reservoir water. This cannot be allowed to happen - it will pollute the water and require its entire replacement - an expensive proposition.

Sixth, in using fertilizer, you need to be aware that since a congressional act passed in the first Bush administration, it is perfectly legal for fertilizer companies to lace their product with toxic wastes to "enhance" the "micronutrient content," regardless of the presence of other toxics in the additives. This means that fertilizers these days often contain harmful, persistent toxics, such as boron, selenium, cadmium, lead, arsenic, as well as dioxin, furans and other inorganic and organic nasties. The act specifically protects the manufacturer of the fertilizer from lawsuits, as long as he labels the material to state that it is the responsibility of the user to determine the suitability of the product for his intended use - and with nearly all fertilizer bags now carrying that warning, you're right to assume that most fertilizers these days contain substances that will do long-term damage to your soil. Because of this, and because the water in this system is constantly recirculating, and not washing away toxics, any periodically-added toxics will build up, and you cannot remove them economically. Toxics cannot be assumed to be washed from the soil by rain, as in conventional agriculture. The result is that your investment can be very easily ruined, and your hard work and investment turned into a toxic waste site, for which the cleanup responsibility is yours! Take great care in selecting the source and type of fertilizer you use! In most cases, the use of ammonium sulfate or ammonium nitrate from a reliable source should be sufficient - potash and phosphorus from the initial soil fertility should be sufficient and will constantly be recycled. Sold for fertilizing lawns, ammonium sulfate or ammonium nitrate of good purity is relatively easy to obtain, and may well end up being the only fertilizer you'll ever need. A quick but rough test for the purity of ammonium nitrate or sulfate is to look for snow-white granules with no discolored or off-color granules mixed in; it should quickly dissolve completely in water, with no residue whatever if it is pure.

Seventh, water pH must be carefully monitored. Rainwater is highly acid, to a degree from natural sources of acidity, but mostly from man-made sources, such as the airborne sulfur dioxide and hydrogen chloride generated by coal-fired factories and power plants. This means that rain can sometimes be as acid as lemon juice - and that would have very destructive consequences for our plants. Since our plants need potash as a fertilizer anyway, I would suggest that a potassium hydroxide solution be added to the inflowing rainwater, sufficient to bring the pH to between 6.8 and 7.4 - as is easily determined by a simple aquarium or swimming pool test kit. The potash should be added to the inflowing water stream, never to the reservoir directly, since the water in the reservoir is held by the sand and does not mix quickly. As this system becomes popular, I suspect that eventually some enterprising company will develop potash blocks that can simply be tossed into the rainwater filtering trough (as mentioned above), and they will dissolve as needed. But until that day comes, the system operator will have to maintain tight control on the reservoir pH manually, by adjusting the potash inflow and monitoring the results.


By now, it should be obvious that it is possible to turn some of the most barren wasteland in North America into productive farmland. Not just do so, but at a reasonable cost, both environmentally and economically.

In short, I wish to see the fulfillment of a lifelong dream - Isaiah's miracle - to see the desert of my youthful experience truly blossom as the rose.

Source URL:

Scott's Home Page
Write to Scott

Copyright 2003, by Scott Bidstrup. All rights reserved.