Water Models

In the comments to this post, MEBates mentions CALVIN (for CALifornia Value Integrated Network), a economic-engineering water model* created and overseen by Jay Lund (Civil and Environmental Engineering), Richard E. Howitt** (Agricultural and Resource Economics) and Marion W. Jenkins (Civil and Environmental Engineering) at UC Davis.

So what does CALVIN do? It "reproduces" the engineering structure of California's water system (sources and sinks and the conveyance in-between), combines that system with the economic demands for water, and then -- using 70+ years of hydro data -- allows one to simulate the effects of changing economic or engineering parameters.

In one famous example [PDF], CALVIN was used to simulate the removal of O'Shaughnessy Dam at Hetch-Hetchy reservoir, which would be the first step towards "restoring" the "lost Yosemite" -- an idea I dislike.***

So, the good side of CALVIN is that it allows us to contemplate how changes in one part of the system might result in changes elsewhere. Since California's water system is so complicated, it's nearly impossible to understand implications, side-effects, etc. without something like CALVIN (there are other models out there).

The bad side of CALVIN is that it may give users a false sense of security (e.g., "we know that this would happen"), especially if the model is missing something important or fails to achieve meaningful resolution. Critics of CALVIN (and its ilk) point out that it misses the institutional friction that can drive or impede water use. I personally know that CALVIN's resolution in urban Southern California is too crude to capture sources and sinks. CALVIN doesn't do a good job at simulating water markets because it misses small resolution, critical features.

This critique is general: Bob Murphy points out the dispute among climate modelers can be traced to the "faith" that some have in macro models (that make big assumptions and derive conclusions) and micro models that take complex interactions (soil composition?) into account.

Bottom Line: Howitt likes to say "all models are wrong, but some are more useful than others," and CALVIN needs to be used with that fact firmly in place, i.e., for BIG changes like adding dams. (Lund has said many times that California needs more conveyance, not more dams.)

---

* CALVIN won the 2008 Hugo B. Fischer Award for (1) the development, refinement or innovative application of a computer model or (2) furtherance of the effective use of models in planning or regulatory functions.

** Howitt covered water economics (among other things) for my dissertation.

*** That idea has gone nowhere, because it would cost about $5 billion and enviros can't convince politicians to spend other people's money on it.