A correctly restored desert usually looks worse in photographs than the damaged ground beside it.
That is the practical problem with desert restoration as a funded activity. In nearly every other biome, the restored state is visibly greener than the degraded one, so the before-and-after photo does the reporting for you. In a desert it does not. Restore a Mojave bajada properly and you get widely spaced perennials, a hard undisturbed surface, and a great deal of open gravel. Bare ground between shrubs is not missing cover. It is the cover pattern. This is written from the North American deserts; the distinction travels to any dryland that was never broken.
what desert restoration actually means
In the practitioner's sense — the one used by crews working for the Bureau of Land Management, the National Park Service, Tribal natural-resource departments, state agencies, and desert land trusts — desert restoration is the recovery of a native dryland plant community and its soil surface on ground that has been physically disturbed. It is not the addition of water or canopy to ground that was always dry.
What that looks like in a work plan:
- Closing and decompacting unauthorized routes, then vertical mulching them — dead shrub skeletons set upright so the scar catches seed and stops reading as a road.
- Salvaging topsoil and its native seed bank ahead of a planned disturbance, and putting it back.
- Outplanting nursery-grown natives — creosote, white bursage, saltbush, blackbrush — with supplemental water for establishment only, then cutting it off.
- Removing invasive annuals like red brome and Sahara mustard before they carry fire into a community that did not evolve with it.
- Keeping vehicles off the site and waiting.
That last line is the largest single intervention and the hardest to invoice. Most of a desert restoration budget buys two things: a soil surface that behaves again, and time with nothing driving across it.
Notice what is absent. Permanent irrigation is not on the list. Canopy is not a target. The supplemental water has an end date, and the end date is the tell — if the planting still needs water in year ten, the project has stopped being restoration and started being agriculture.
three jobs, one phrase
"Greening the desert" and "restoring the desert" get used as synonyms by people who mean entirely different work. They produce different objects.
| what was done | what you are holding | what it costs to keep | how it scores on a greenness index |
|---|---|---|---|
| Tree planting on dryland | A young plantation, or a failed one | Water, replacement stock, fire management | High, if the stems survive |
| Irrigated greening | An irrigated landscape — orchard, turf, green belt, oasis farm | A permanent water subscription | Highest |
| Recovered dryland | The native community returning on disturbed ground | Access control, invasive removal, decades | Low |
| A living desert that was never broken | The desert | Not converting it | Lowest |
Read the last column again. The four rows are ranked almost exactly backwards from how much desert you are holding.
Greening is a description of color. Restoration is a claim about which living system came back. Only one of those shows up in a free vegetation index, and it is the wrong one.
the measurement asymmetry that funds the wrong row
Greenness is cheap to measure. Satellites have produced continuous vegetation indices for decades, the data is free, and a rising line is legible to anyone on a board. Desert condition is expensive to measure: shrub composition and spacing, soil surface stability, wash and channel function, biocrust extent, the presence or absence of a handful of invasive annuals. That takes people on the ground on a repeat schedule.
So a healthy desert scores badly on nearly every remote index that restoration money is easy to raise against, and an irrigated green belt scores beautifully.
You might be thinking that no serious funder confuses an orchard with a desert. Agreed — nobody writes that sentence down. The confusion happens one layer below the sentence, in the indicator. A program states an ecological goal, then selects greenness or canopy as the progress metric because it is the only one that reports on time and on budget. From then on the money optimizes the metric. Nobody has to believe anything false for the wrong row to get funded.
This is not a conspiracy and it is not a reason to stop measuring. It is a reason to be specific about what the indicator is standing in for, and to notice when a desert is being graded on a curve built for a forest.
some drylands really are broken
This post is not an argument that dry land should be left alone on principle. Land in arid, semi-arid, and dry sub-humid regions can genuinely lose the living cover and soil structure that let it catch water, and it can be brought back — that process has a name, desertification, and reversing it is real work with real returns. the desert isn't destiny owns that argument here, and rewater the land owns the hydrology. Go there for it; this post will not recap it.
The line between the two is the whole point:
Desertification is a dryland that stopped working. A desert is a dryland that works because it is dry.
A degraded field that sheds every storm needs its sponge back. A Sonoran bajada carrying saguaro, palo verde, and ironwood needs nothing of the sort. Treating the second one as a candidate for the first one's treatment is how a functioning desert gets classified as an opportunity.
the clock nobody prices
The other reason greening substitutes for restoration is that greening is fast and desert communities are not.
At Saguaro National Park, a saguaro normally does not grow its first arm until it is 60 to 75 years old, and in drier areas it can take up to 100 years. Average lifespan runs 150 to 175 years, and some plants are believed to pass 200. Creosote is slower still: the U.S. Forest Service puts average longevity at roughly 625 to 1,250 years depending on site, and the largest known clone in Johnson Valley, California is estimated at about 9,400 years. Joshua trees and blackbrush stands operate on comparable terms. The soil crust that holds the surface between all of them recovers on its own timescale — the crust that holds the desert together owns that story.
Restoration finance runs on grant cycles, crediting periods, and reporting years. Creosote runs on centuries.
This is why "restore it later" is not the same option in a desert that it is in a wet biome. Clear a temperate forest, walk away, and a recognizable young stand appears inside one human career. Clear a creosote flat and the community that returns is not the one you removed, not in your lifetime and not in your grandchild's. The invasive annuals that move into the gap in the meantime change the fire regime while you wait — the Sonoran version of that story is buffelgrass, and its cousin is covered in cheatgrass is a fire cycle.
The asymmetry that follows is uncomfortable and worth stating plainly: in a desert, holding an intact acre is worth far more than the promise of restoring a broken one, because the restoration promise is written against a clock no funder can hold.
what a funder is actually buying
For an allocator, an agency, or a land trust board, the question is not "is planting good." It is: what object is on the report at the end of this?
A planting receipt is a count of stems. It does not settle survival, and it does not settle whether the biome changed. An irrigation scheme is a functioning agricultural or amenity landscape, and it settles nothing about the desert except how much groundwater the new landscape now requires. A perpetual conservation easement is real protection of use; a term easement is the same with a renewal date. Neither is a statement of condition. Each is a legitimate purchase. None of them is the same as the condition of a living dryland, held now, on ground you can name.
That last object is what ensurance is built to fund. A certificate issued under specific ensurance routes money to a named place, and what it pays for is condition on the ground — the repeat survey work the crews already do — native composition, surface stability, invasive load, the wash still doing its job — rather than a greenness figure that an orchard could also produce. A valuation sits underneath that, because capital needs a number to move. The number is a bridge to the desert. It is not what the desert is worth.
Our own desert doors are live and small: deserts.ensurance and the named ecoregion agents for the Sonoran and Mojave. Early volumes, stated as such.
keep reading
The planting-versus-condition argument has cousins in other biomes, and each owns its own ground:
- a tree on a prairie is not a grassland — the same conversion, on native grass
- a planted tree is not a forest — why a full bag is not a stand
- do trees actually cool the planet — where dryland planting backfires on physics, not just ecology
- the desert isn't destiny — the cousin that owns desertification and getting the rain back
- rewater the land — catch, hold, recycle on land that genuinely lost it
- what grasslands actually are — the prairie version of "this is not empty land"