Jay Lund published a clever satire of California water rhetoric today, and he’s right that policy-based evidence-making occurs in our water debates. I worked with Jay and other Delta Independent Science Board members in 2021 to save the Delta Independent Science Board from defunding, and I value his truthtelling.
But Jay’s framing misses something critical happening at the Department of Water Resources. There’s a difference between rhetoric and documenting actual methodological failures in climate risk assessment.
Context: The 2023 Delivery Capability Report’s Hidden Methodology
DWR’s modelers have claimed the 2023 Delivery Capability Report incorporates climate change using sophisticated CMIP6 global climate model projections downscaled with LOCA2.
What they actually did: extracted linear temperature and precipitation trends from the climate models, then fed these into a stochastic weather generator (WGEN) that randomly reshuffles weather patterns from 1948-2018. The generator implicitly assumes past atmospheric patterns will repeat forever. This is a stationarity assumption that’s scientifically indefensible, given documented regime shifts in California’s climate.
Most remarkably, the Cornell scientists who developed this weather generator documented a 30% increase in Pacific ridging patterns (which block storms) and created a “dynamic scenario” to account for this change. It exists in the WGEN code as a module. DWR chose not to use it in the 2023 Delivery Capability Report, opting instead for the “thermodynamic” (stationary) scenario that only incorporates projected temperature increases.
This isn’t rhetoric. It’s a methodological choice with profound implications for water planning. With that context, I wanted to propose the following:
California’s Institutional Water Rites (the ones Jay missed):
- Claim you’re incorporating climate change while assuming historical patterns are continuing:
- Correlative rite 1.1: Have a “dynamic scenario” option available but choose not to use it.
- Insist your models are sophisticated because they use “CMIP6 downscaled with LOCA2” when you’re actually just extracting two numbers (∆T and ∆P) to drive a weather generator.
- Blame model limitations for not analyzing failure scenarios when the real reason is “we know things break, it’s not helpful to analyze.”
- Assert your modeling represents “95th percentile risk” without mentioning you ignored a 30% increase in blocking ridges documented at 1% statistical significance.
- Co-opt the watchdogs by prioritizing funding for “collaborative” projects and putting advocacy groups’ names on papers so they defend rather than scrutinize your methodology.
- Use complexity – make the methods so technically opaque that only specialists can identify the stationarity assumptions buried in the code.
- Defer the hard questions – promise to explore uncomfortable scenarios “in a future report” after the Bay-Delta WQCP update and Sites Reservoir and Delta tunnel hearings.
- Validate against mixed statistics from historical periods containing climate regime shifts, then claim accuracy when your model reproduces the average statistics for the entire period.
Yes, there’s plenty of California water rhetoric to satirize. But lumping valid critiques with rhetoric creates false equivalence. Some of us have repeatedly documented actual methodological failures, like choosing stationary models when dynamic options exist, with real consequences for California’s water future.
That’s not a “rite.” That’s science. Perhaps we need a new taxonomy that distinguishes between political rhetoric and technical whistleblowing.
Further Reading:
The Original Post: Lund, J. (2025). “Nine California Water Rites.” California Waterblog, Nov 30, 2025. https://californiawaterblog.com/2025/11/30/eight-california-water-rites/
On the Stationarity Problem: Milly, P.C.D., Betancourt, J., Falkenmark, M., Hirsch, R.M., Kundzewicz, Z.W., Lettenmaier, D.P., and Stouffer, R.J. (2008). “Stationarity Is Dead: Whither Water Management?” Science, 319(5863), 573-574. https://www.science.org/doi/10.1126/science.1151915
On California’s Changing Ridging Patterns: Swain, D. (2018). “New insights into the Ridiculously Resilient Ridge & North American Winter Dipole.” Weather West. https://weatherwest.com/archives/5982
Swain, D. (2021). “The Ridiculously Resilient Ridge Returns Again.” Weather West. https://weatherwest.com/archives/8692
The Weather Generator Used by DWR: Najibi, N. and Steinschneider, S. (2023). “A Process-Based Approach to Bottom-Up Climate Risk Assessments: Developing a Statewide, Weather-Regime based Stochastic Weather Generator for California, Final Report.” Cornell University for California Department of Water Resources. https://water.ca.gov/-/media/DWR-Website/Web-Pages/Programs/All-Programs/Climate-Change-Program/Resources-for-Water-Managers/Files/WGENCalifornia_Final_Report_final_20230808.pdf
DWR’s New Climate Change Methodology: California Department of Water Resources (2023). “Risk-Informed Future Climate Scenario Development for State Water Project Delivery Capability Report.” Technical memorandum. https://data.cnra.ca.gov/dataset/a3bb1ddd-624b-4c3d-95e7-2aa6b3bf2b5b/resource/e41f531d-dace-4d37-b52e-35a6ddd2224e/download/risk-informed-future-climate-scenario-development-for-swp-dcr_final122023.pdf
That was fabulous and applies to any subject.
1. Claim you are incorporating data while assuming …
2. Insist your models are sophisticated because … jargon …
3. Blame model limitations when issue isn’t incorporated in model
4. Assert your model (whatever) when you explicitly (counter whatever)
5. Co-opt the critics by adding them as authors
6. Use complexity to obfuscate
7. Defer the hard questions
8. Validate against limited data
The framework for so many technical publications!
I enjoyed this.
It is an important perspective extending my more whimsical critique with more substance. There is a common tendency to add complexity to modeling, particularly for climate change. Sometimes this is an earnest but unavoidably imperfect attempt to address a difficult issue, sometimes it is a rhetorical device to indicate you are working on a perceived major problem, and sometimes it is an effort to avoid or even distract from more important (and often difficult) issues. Sometimes it seems to be a mix of motivations and effects.