Nobody wants to do that, and there are all sorts of complex legal issues around water rights. But push come to shove, California could survive just fine without its agricultural sector.
Instead these will need to be shipped from overseas by marine transport which is an intense source of pollution.
You can grow carrots and strawberries almost anywhere in the US. Peaches need things a bit warmer, but again most of the US is fine.
It makes sense to shift water intensive crop production to states with water.
To see why you have to go back to the Dust Bowl, and then the prospect of severe famines in Europe at the end of WW II. It turns out that if you have 10% more food than people need, food is cheap. But make that 10% less and food becomes very expensive, people starve, and you get political unrest.
Pinch off ag in enough places, and you'll find it's got some underappreciated benefits.
The alternatives to crops being raised in Calfornia aren't limited just to the lost ag revenue, but to the expense of transporting in food, the loss of fresh fruits and vegetables (California specialises in "table" crops (vegetables), fruit, and grapes especially, knock-on effects of worsened nutrition and reduced food quality, price increases resulting from withdrawal of California ag land, etc.
Simply figuring out where you'd source that food from becomes problematic. Seattle is the only West Coast port with significant capacity and growing regions near it (Portland might do in a squeeze), plus growing area. There's no assurance that California crops, particularly greens and vegetables, would be suited to more northern climates and growing conditions, though adaptations might be made.
And there's considerable room for adjustment in California's crop mix. No need to grow alfalfa in the desert, as a perennial crop, that's easy to fallow in dry years. Tree crops are multi-decadal investments, and will not simply fail to bear fruit without water, but will die -- one of the real crimes of the almond-planting spree around the state. You lose cropping flexibility.
(There's a study I turned up some months back, either IRS or Federal Reserve, on tree crop asset lifetimes -- some bearing orange trees in Southern California are approaching or over 300 years old.)
A straight market-based analysis does poorly.
That is: Not ag.
(Ag water generally comes from canals, irrigation districts, and in some cases pumps or aquifers.)
Or at least fewer of them.
If food were grown by magic, without employing any people, and it could all be sold at cost, it would contribute 0% to the state's economy. Does that mean that it would make economic sense to immediately stop growing all that food? Of course not! The fact that growing food is so cheap is a good thing!
My quick calculation is the biggest of these desal plants would only feed 25,000 acres of almonds. There are 1M acres +/- of almonds in California, if we wanted to double nut protein production in California (the leading global exporter of almonds) you would need about 50 of the plants described not just three.
Who wants to build the giga-desal plant to feed the Clifton Court Forebay?
This is a superficial analysis. It assumes that California's agriculture could be easily replaced. Not to mention that whatever replaces it may face the same issue with desalination. In actuality the entire country is dependent on California agriculture to a large extent and "The loss of California’s output would create a dire situation for at least a decade."[1]
[1] http://www.slate.com/articles/health_and_science/explainer/2...
What if your current comparative advantage gets lost due to development elsewhere? Your exports will no longer earn you the foreign currency you require to import food (look at Venuzvela for a recent example or the disaster of Haiti where IMF prescribed food imports or India's famines of the 1960s)
This is the reason nations heavily subsidize agriculture even when they can rely purely on imports.