What’s the Colorado River failure mode? By which I mean, with as much specificity as we can muster, when the water begins to run short, who will get less? I’m pushing this question because of a frustration with the doom scenarios generated by simply extrapolating the diverging supply and demand curves forward in time and invoking Anasazi/Hohokam portents of doom.

I’ve pointed a couple of times recently (here and here) to the risk faced by Arizona and Nevada. Juliet McKenna and her colleagues have usefully added to the discussion with a more detailed look at the pecking order within users of Arizona’s Central Arizona Project supply of Colorado River water:

M&I (municipal and industrial) subcontractors are in a relatively secure position for the next 25 years, while agricultural and other users of “excess” CAP water should continue to prepare for frequent, and perhaps sustained, reductions.

This gets to the critical question of how, in a very practical manner, shortage will spread through the basin. If there’s not enough water in the river to meet all needs, the “law of the river” will determine allocation among the various states. Each state will then be responsible for allocating shortage within its boundaries. McKenna’s analysis shows that ag holds the short straw in Arizona.