In this guide
Somebody buys five acres in the San Luis Valley and the first thing they picture is a garden. Not a farm. A garden: tomatoes, beans, a row of carrots, maybe fruit trees in a few years. It is one of the most common reasons people call us about Colorado and New Mexico ground.
It is a reasonable thing to want, and it is doable. But three things decide whether it works, and they are not the three most people prepare for. In rough order of how often they stop somebody: the well permit, the length of the season, and the chemistry of the dirt. Technique is fourth, and a long way back.
Start with the well permit, not the seed catalog
This is the part that surprises people, and in Colorado it is written into statute.
Colorado allows certain small wells to be permitted outside the usual water court process. Those exempt wells can legally serve a long list of uses: ordinary household purposes, fire protection, watering poultry, domestic animals and livestock, and irrigation of not more than one acre of home gardens and lawns. That is the list most people find when they search, and it sounds like a green light.
The catch is in the next subsection. The statute grants a presumption that a well will not injure other water rights in three situations, and the first two are the ones that matter here. One is a well that is the only well on a residential site and is used solely for ordinary household purposes inside a single-family dwelling and is not used for irrigation. The other is a well that is the only well on a tract of thirty-five acres or more.
Most of what we sell in Colorado is in Costilla County at around five acres. That lands on the wrong side of the line for a well-fed garden. It does not mean you cannot grow food there. It means the water has to come from somewhere other than a household well, and you should know that before you buy rather than after you drill.
Two routes are worth looking at instead. Hauled water into a cistern is the common one and it is what a lot of people out there actually do. Rainwater catchment is the other, and Colorado has allowed residential rooftop collection since 2016, within limits worth reading before you build around it.
The season is shorter than the pictures suggest
The San Luis Valley floor sits between about 7,500 and 7,800 feet. The USDA's ecological site description for the valley floor gives a frost-free period of 70 to 80 days, averaging 80.
That number moves depending on how you count. The National Weather Service office in Pueblo puts the average last spring freeze at Alamosa on June 9 and the average first fall freeze on September 14, which is about 97 days between them. The difference is partly definitional, since a freeze at 32 degrees and a freeze at 28 degrees are different events and both get published. Either way you are planning around something closer to seventy or eighty usable days than the hundred and twenty a seed packet assumes.
Precipitation has the same spread. The NRCS site description says 10 inches a year with a range of 6 to 10. The Weather Service's 1991 to 2020 normal for Alamosa is 7.39 inches. The Colorado Climate Center puts the middle of the valley near 7. Call it 7 to 10 inches, driest in the center, and understand that none of those numbers supports a garden without irrigation.
Your soil is alkaline, and you are not going to fix that
Colorado soils commonly run pH 7.0 to 8.3. Arizona is mostly 7 to 8.5. New Mexico's caliche soils run 7.5 to 8.5. Utah averages around 8.0. This is normal for the region and it is not a problem you created.
What makes it permanent is free lime. Arid soils hold calcium carbonate that never got rained out. Utah State puts a typical figure at 15 to 40 percent lime by weight, and each one percent is roughly ten tons per acre in the top six inches. So a typical soil is sitting on a 150 to 400 ton reservoir of the exact material that resists acidification. Their published conclusion is blunt: attempts to lower calcareous alkaline soil pH are often futile or impractical.
Elemental sulfur is the right material if you are going to try, and it is worth knowing what it actually does. Utah State's field work found no measurable pH change until 12 to 18 months after application. On a low-lime site, 1,000 pounds per acre moved pH from 7.0 to about 6.0. On a site with 38 percent lime, pH barely moved at all. Montana State's garden trial applied 225 pounds per thousand square feet, which is an enormous amount, and moved pH from 7.9 to 7.5.
So: do not add lime, which is the amendment for the opposite problem and which your soil already has in quantity. Do not count on sulfur to rescue a genuinely calcareous site. Pick plants that tolerate high pH, or build the growing medium rather than trying to convert the one you have.
The soil test to order, and the one that will lie to you
If you order a standard soil test from a lab that serves the whole country, there is a good chance the phosphorus number you get back is meaningless on your ground. We have written separately about what a soil test actually gates, and the short version is that the septic permit and the garden want different readings off the same sample.
The Bray P1 test was designed for neutral to acidic soils. On calcareous soil it routinely under-reads, and an Idaho study found cases where Bray detected no phosphorus at all in soils whose Olsen values were relatively high. The sodium bicarbonate method, called the Olsen test, is the one that correlates with what plants can actually get at high pH, because the bicarbonate pulls calcium out of the system and frees the calcium-bound phosphorus the Bray extractant leaves behind.
Where the switch happens depends who you ask. Montana State says above pH 7. USDA's NRCS and New Mexico State say 7.2. A figure of 7.4 circulates widely online and we could not find it in any authoritative source, so treat it as folklore. The practical rule: if your soil has free lime, ask the lab for Olsen phosphorus by name.
Interpretation ranges differ too, and it is worth knowing the spread rather than trusting one chart. Montana State's home garden guidance calls Olsen P below 4 parts per million very low and says no phosphorus fertilizer is needed at or above 16. New Mexico State's general scale runs higher, calling 20 to 40 sufficient. Numbers from a lab using a different extractant are not comparable to either, which is a trap when you are reading two reports side by side.
| What to order | Why it matters out here |
|---|---|
| Olsen phosphorus, by name | Bray can read near zero on soil that has plenty. Ask for Olsen and say so on the form. |
| pH and free lime (calcium carbonate) | Tells you whether amendment is realistic or futile. Above about 3 percent lime, pH is locked in. |
| Salinity as ECe, saturated paste | The number that quietly limits what you can grow. See the next section. |
| SAR or exchangeable sodium percentage | Sodium wrecks soil structure and infiltration. Different problem from salt, different fix. |
| Texture and organic matter | Sets your water-holding capacity and how much compost the ground actually needs. |
One thing not to bother with by default: New Mexico State notes the boron test is often offered and seldom needed, and should wait until plants show symptoms or tissue testing points at it. Boron toxicity is real in parts of the arid west, but it is a local problem rather than a universal one.
Salt is the one that actually kills gardens
Salinity is measured as electrical conductivity of a saturated paste extract, written ECe and reported in decisiemens per meter. A soil is formally classified as saline above 4. That classification line is where most interpretation charts start, and it is misleading for a vegetable garden.
The reason is that the most sensitive garden crops start losing yield at ECe 1.0, which is a quarter of the way to being called saline. Beans and carrots both have a threshold of 1.0. Onions are 1.2. A soil report that comes back at 2.5 and says slightly saline is, for beans, already costing you about a third of the crop.
| Crop | Threshold ECe | Yield lost per additional point |
|---|---|---|
| Bean | 1.0 | 19 percent |
| Carrot | 1.0 | 14 percent |
| Onion | 1.2 | 16 percent |
| Lettuce | 1.3 | 13 percent |
| Spinach | 2.0 | 7.6 percent |
| Beet | 4.0 | 9 percent |
| Asparagus | 4.1 | 2 percent |
| Barley | 8.0 | 5 percent |
Those are the Maas and Hoffman threshold values as published by New Mexico State and the USDA's Agricultural Research Service. Other extension services publish different sets, and they are not interchangeable: Utah State lists beet at 5.3 and asparagus at 5.0, and the Pacific Northwest bulletin publishes the level at which yield has already fallen 10 percent rather than where it starts. Do not mix numbers from two charts.
Sodium is a separate problem that often arrives with salt. A soil is sodic when the sodium adsorption ratio is above 13 or exchangeable sodium is above 15 percent, and the symptom is structural rather than chemical. Sodium is a far weaker flocculator than calcium, so the crumbs that make soil drain come apart, the pores clog, and you get crusting and water that will not soak in. The San Luis Valley floor is mapped with extensive slick spots, which is exactly this.
Gypsum is the right amendment for that, and for the right reason: it supplies soluble calcium that displaces sodium off the clay, and unlike lime it does not change pH. It also does nothing without drainage and water to carry the displaced sodium out of the root zone. Gypsum on a soil with good structure is money burned.
Test the water before you pipe it to anything
Water that is fine to drink can still be wrong for a garden, and the difference is not obvious. We cover the drinking side in testing a well before you buy a filter. For irrigation the panel worth asking for is salinity as EC or total dissolved solids, SAR, chloride, boron, nitrate and bicarbonate.
New Mexico State's thresholds for irrigation water: EC below 0.7 is no problem, 0.7 to 3.0 brings increasing problems, above 3.0 is severely limiting. Total dissolved solids below 450 milligrams per liter is fine and above 2,000 is severe. Bicarbonate above 2.5 milliequivalents per liter is severe, because bicarbonate reacts with calcium in the soil solution and strips out the calcium that was holding your structure together.
SAR is where the published guidance genuinely splits. Colorado State's irrigation criteria treat SAR of 1 to 9 as a low hazard. New Mexico State treats anything above 3 as the start of problems and above 9 as severe. For a garden on sodium-prone ground, the conservative reading is the one to plan around.
Boron is the one that catches people, because the toxic range sits close to the normal range. Sensitive crops are affected between 0.5 and 0.75 parts per million in irrigation water. Tolerant ones go to 4. There is not much daylight between a water that is fine and a water that is quietly poisoning your stone fruit.
Caliche, if you are in New Mexico or Arizona
Caliche is soil cemented together by calcium carbonate. It shows up as light-colored lumps or as a solid layer anywhere from a few inches to several feet thick, and the hardness ranges from crumbly to effectively rock. It is common through southern Arizona and across New Mexico, and the depth varies by soil series, from four inches down to several feet.
It causes three separate problems at once. Roots cannot get through it, so plants stay shallow and drought-stressed. Water cannot get through it either, so the root zone waterlogs and salts concentrate above the layer. And the carbonate drives pH up, which locks up iron and produces the yellowing between leaf veins that people misread as a fertilizer problem.
The University of Arizona's guidance is practical. Dig planting holes that go completely through the layer so water has somewhere to drain. Where that is not realistic, punch a narrower drainage hole through the caliche into normal soil below. Where neither works, build upward: get to two feet of total depth over the root zone, or at least eight inches of good topsoil for a lawn. Before you plant anything, fill the hole with water and watch it. New Mexico State wants to see at least an inch an hour.
What actually works out here
None of the above says do not grow food on arid land. It says the constraints are specific, and most of them are settled before you turn any soil.
Work through it in this order.
Ask the state about the well first.
In Colorado, call the Division of Water Resources with the parcel number and ask what a permit on that specific tract would allow. Everything else depends on the answer.
Get a soil test with Olsen phosphorus, ECe and SAR on it.
Thirty dollars or so, and it tells you whether you are gardening or building a growing medium from scratch.
Test the water you intend to irrigate with.
Same sample can usually cover drinking safety and irrigation suitability. Ask for boron and bicarbonate explicitly.
Dig a hole and fill it with water.
Free, takes an afternoon, and it finds caliche and hardpan that no map will show you.
Size the season honestly.
Seventy to eighty frost-free days rules out a lot of the catalog. Short-season varieties, transplants rather than direct seed, and some form of cover are not optional at 7,500 feet.
Build up rather than fighting the pH.
Raised beds with imported soil, compost at a sane rate, and drip rather than overhead. You are not going to acidify a calcareous soil, so stop trying.
The thing that makes this work is that none of it is exotic. It is the same sequence any careful buyer runs before they commit: find out what the law allows, find out what the ground is, find out what the water is, and then plan inside the answers rather than around them.
Where we fit
We sell owner-financed rural land across the west, and a good share of our Colorado inventory is in Costilla County at around five acres. That is ground where the garden question has a real answer, and the answer involves hauled water or catchment rather than a well. We would rather tell you that before you buy than have you find out after.
If growing food is the point of the purchase, say so when you call. Parcel size, basin and soil all move the answer, and a Land Specialist can pull what is known about a specific parcel before you are committed to it. Our homestead and cabin ground is the place to start, and what we have in Colorado is where this particular well rule applies.
The desert does grow food. It grows it for people who found out what the permit allowed, what the salt reading was and where the hardpan sat before they bought the seed.
Start from what you want the ground to do. If it is a homestead with food on it, the parcel you want may be a different parcel than the one you were looking at, and that is a much cheaper thing to learn now.
