Most cities end where it stops being worth building. This one ends where the land stops being available. That difference propagates into every part of the housing market, from lot sizes to the economics of renovation, and it is the single most useful fact to hold when reading local housing argument.
Desert hydrology, briefly
Water in a basin like this arrives as mountain precipitation, moves through gravels beneath the valley floor and historically emerged at a few low points as springs. The quantity involved is small, it recharges slowly, and pumping it faster than it recharges lowers the water table and, where the sediments compact, lowers the ground itself.
A settlement of any size therefore has to import water from outside the basin. Once it does, the constraint on growth becomes the allocation of that imported supply rather than local hydrology, and the question shifts from how much water exists to how much of it this place is entitled to take and how efficiently it uses what it takes.
The bounded land supply
Because most of the land surrounding the developed area is federally held, the outward edge does not move continuously. Land becomes available for development through a deliberate disposal process, in defined parcels, at times decided administratively.
The market consequence is discreteness. Instead of a smooth outward creep responding to price, supply arrives in lumps, and between lumps the only ways to accommodate additional households are to build at higher density inside the existing edge, to accept longer commutes at the far edge, or to bid up the existing stock. A bounded supply does not make a market expensive by itself; it makes it more responsive, so demand shows up in price faster than it would elsewhere.
What this does to the housing stock
Constrained land raises the value of what sits on it relative to the building, which changes the arithmetic of renovation and redevelopment. When the site is the larger share of value, replacing a small old house with a larger one becomes rational sooner, and the older stock in accessible locations comes under pressure that has nothing to do with its condition.
It also puts a premium on the fine-grained older grid, which is the part of the valley where such change is legally and physically easiest, and it makes the recorded rules of newer communities, which frequently prohibit exactly this kind of change, a more significant feature of the property than they appear.
Landscape as a policy variable
In a place where outdoor irrigation is a large share of residential water use, the treatment of the ground around a house is not only an aesthetic choice. It is the part of household consumption most responsive to policy, which is why desert landscaping moved over a few decades from unusual to expected, and why the front garden of a valley house is a reasonable indicator of the decade its landscaping was last renewed.
How to read growth argument
Local argument about growth tends to conflate three separate constraints: how much water is available, how much land is available, and how much infrastructure the built area can carry. They have different mechanisms, different timescales and different decision-makers.
Separating them makes most of the argument tractable. A water constraint acts on consumption per household and is addressable by efficiency. A land constraint acts on the location of new dwellings and is addressable by density or by disposal. An infrastructure constraint acts on timing. Any claim that treats all three as one thing is usually making a rhetorical point rather than a factual one.