Risk Rating 2.0: A first look at FEMA’s new flood insurance system

By Ryan Miller, Peter Hansen, and Nicholas Pinter

Risk Rating 2.0 has been called the Federal Emergency Management Agency’s (FEMA)’s most significant reform in 50 years.  Roughly 77% of customers of the National Flood Insurance Program (NFIP) nationwide will see increases in their premiums, while the other ~23% will see reductions or no change.  FEMA will formally introduce Risk Rating 2.0 on October 1, 2021, and most premium increases will kick in on April 1, 2022.

In brief, Risk Rating 2.0 moves the NFIP away from its heavy reliance on in-or-out flood zones, in particular in-or-out of the so-called “100-year floodplain,” and towards an individual assessment of risk for each property (as promoted in this Blog [6/16/2019]).  FEMA’s 100-year flood area will not go away – in particular, this will still be the basis for whether a property owner with a federal mortgage must buy flood insurance – but the premiums for each property will be determined based on individual factors that include flood risk from an ensemble of three privately developed flood models. 

The Data

Recently, FEMA published summaries of NFIP premium changes by state, with details down to individual zip codes.  This first glimpse of Risk Rating 2.0 is a limited one; for example it shows only first-year changes (increases are capped at 18%/year) and lumps the most extreme increases or decreases into bins of “greater than $100” per month.  Nonetheless, our group took the zip code-level data and “reverse engineered” that information to provide a first look at what policyholders and others interested in flood-risk management can expect from Risk Rating 2.0. 

The focus of this review is California.  Over the history of the NFIP, California has been a donor state, contributing hundreds of $millions more in flood insurance premiums than it has received in claims (California Water Blog 12/14/2016).  Under Risk Rating 2.0, 73.2% of California NFIP policyholders will see premiums increase, and 26.8% will see decreases. 

We encourage readers to explore Risk Rating 2.0 premium changes across California using our interactive map of average premium changes by zip code (Link to Interactive Map).

Premium Changes and Patterns across California

Statewide, the majority of Californians with NFIP policies can expect to see premium increases of less than $10 per month (Figure 1; Table 1).  That being said, the total amount of premium dollars paid by California policyholders as a whole is actually set to decrease.   The overall distribution of premium changes across California is bimodal – a large majority of policyholders will receive small rate increases, but some policyholders will receive larger discounts. Of the >20,000 policies receiving discounts of ≥$20 per month, nearly half are >$70/month. We cannot precisely calculate the total net impact of Risk Rating 2.0 on California because FEMA’s data masks the exact dollar amount of changes at the extreme tails (8 policies with increases >$100 per month and 7421 properties with decreases >$100 per month). For the purposes of this analysis, we conservatively assumed a $100 change for the extreme tails (+/- >$100/month). Using this method, the total change in premiums in the first year of Risk Rating 2.0 is an aggregate decrease of at least $1.4 million per month (= at least $16.76 million per year statewide), or ≥9.9% of total California NFIP premiums (FEMA, 2021b). About half of this estimated total discount discount will go to single-family homes (~165,000 policies), and the other half going to the remaining non-single-family and non-residential NFIP properties (~49,000) in California.

The overall downward shift in premiums is consistent with our earlier findings that NFIP seemed to be overestimating flood risk in California.  The state has been hit by more than its share of disasters, but several decades of flood data were not adding up to the flood hazard suggested by NFIP premiums until now.  In shifting to a more physical, individualized formula, Risk Rating 2.0 seems to be recognizing this earlier overestimate, although not correcting the formula as much as may be warranted. 

This bimodal distribution of premium changes is also interesting when mapped across California (Figure 2). Increases and decreases are scattered across the statewide zip codes, but a few regional patterns emerge. First, the California Central Valley – including the Sacramento Valley north of Sacramento and the San Joaquin Valley to the south – is characterized by average premium decreases. under Risk Rating 2.0 (greens in Fig. 2). 

Figure 1. Count of monthly premium changes to all California NFIP policies.

Stakeholders in the Central Valley have long complained that NFIP was overcharging them; for example, pressing the Government Accountability Office in 2014 to study perceived biases against agriculture baked into the NFIP (https://www.gao.gov/products/gao-14-583).  In addition, the California Central Valley has broad areas protected by levees, some of which meet the 100-year protection standard, some not. 

Figure 2. Approximate change in NFIP premium cost by California zip code, for all zip codes with at least 5 NFIP policies listed in that area.  Values shown are not true averages because policies listed by FEMA as changing by >$100/month were calculated as +/- $100.

Figure 3 maps out premium changes in the Sacramento area. Four of California’s top five zip codes with the most NFIP policies are located around Sacramento. Under Risk Rating 2.0, three of those (95833, 95834, and 95835 [Natomas]), will see modest declines in premiums. Conversely, 95831 (the Pocket area) will see premium increases. All four of these are levee-protected areas along the Sacramento River.  The US Army Corps of Engineers recently completed significant levee improvements around Natomas[1], and is commencing improvements to the Pocket area.[2]  Until now the NFIP handled levees as all-in or all-out, a practice widely criticized from all sides.  Under Risk Rating 2.0, the Corps has supplied FEMA with levee data, but estimates of levee reliability is notoriously difficult to quantify, and the new algorithm for levee protection is currently a black box that needs scrutiny.

Our own zip codes here in Davis illustrate some of the odd shifts under Risk Rating 2.0.  The 95616 zip code, including western Davis and the university, will see average decreases of almost $4.00 per month, while 95618 to the east will see increases >$4.00. Both areas lie at similar elevations and are largely outside of FEMA’s mapped 100-year floodplain.  The whole area is protected from flooding of the Sacramento River and other tributaries by levees and bypasses.  We initially suspected that the private-sector flood models embedded in Risk Rating 2.0 are favoring levee-protected areas, making this change in Davis puzzling, but a lack of transparency in underlying data obscures the mechanisms driving such changes.  

Figure 3. Average change in NFIP premium cost by ZIP Code, Sacramento area

Figure 4 shows Risk Rating 2.0 premium changes in the Los Angeles area. Within this region, changes are most pronounced in the foothills and coastal areas in Malibu, the Santa Monica foothills, and the Hollywood Hills, with policyholders in zip code 90265 (Malibu) set for an average discount of >$40/month. This represents the steepest discount for any zip code in California with >1,000 current policies. Elsewhere in L.A., the densely populated San Fernando Valley and main L.A. Basin area, on average, will see modest increases in monthly premiums.

Figure 4. Average change in NFIP premium cost by ZIP Code, Los Angeles area

The large premium decreases in Malibu, the Santa Monica foothills, and the Hollywood Hills are surprising.  FEMA is heavily marketing Risk Rating 2.0 as “Equity in Action.”  Malibu’s median household income is >$150,000 per year (2019), 2.5 to 3 times higher than the $50,000-$60,000 in most San Fernando Valley zip codes. 

Risk Rating 2.0 does include one important fix that promotes economic fairness.  Under the current NFIP pricing system, premiums are based on the insured value of a structure and not the actual value of that building.  And currently, coverage up to $60,000 is charged at a higher rate than coverage above that threshold.  FEMA originally did this to encourage policyholders to insure the full value of their structure and contents (up to allowable caps).  But the policy is regressive as well as counterintuitive, as high-value structures are also more likely to incur expensive damages.  Under Risk Rating 2.0, premiums will be priced based on a structure’s replacement cost. 

Beyond the clear fix in premium pricing above, it is unclear how much Risk Rating 2.0 lives up to its “Equity” label, either in intent or in its impact.  By all accounts, Risk Rating 2.0 was designed primarily and from the start as a new pricing structure to fix NFIP’s perennial funding shortfall.  As the Congressional Research Service put it, “Risk Rating 2.0 will continue the overall policy of phasing out NFIP subsidies.”  Fixing the perennial hemorrhaging of NFIP may be a worthy goal, as is making insurance rates that better reflect risk, but rolling out these changes under the banner of “Equity” is disingenuous. 

To preliminarily assess the economic equity of Risk Rating 2.0, we compared premium changes across California to median household incomes in the same zip codes (Table 2). There is no clear correlation between neighborhood income level and average premium change. Premium decreases were largest in the lowest-income category (averaging -$9.25/month for neighborhoods with incomes <$50,000).  Premium decreases were smallest for middle incomes (averaging -$3.43/month for neighborhoods with incomes $70,000-$80,000). Premium changes in all other income categories, including areas with mean incomes >$100,000, were intermediate, averaging -$6.50/month.

Other questions about Risk Rating 2.0 focus on the modeling and formulas used to calculate risk and premiums.  Some details have been provided (e.g., the “Milliman Report,” April 2021), but a lot of Risk Rating 2.0 is a black box.  Our research group is trying to obtain better, clearer data to assess questions like how coastal flooding and future climate change are reflected in the new RR2.0 premiums.  In the meanwhile, Figure 5 shows zip code-level premium changes for the San Francisco Bay Area. Here, it appears that many low-lying costal areas are set to experience premium increases, while foothill areas are set for significant discounts. Coastal zip codes like 94080 (South San Francisco), 94044 (Pacifica), and 94030 (Millbrae) are will see premium increases in the $5-%7 range, while mountainous, suburban areas like 94605 (Oakland Hills) and 94583 (San Ramon) are set for premium decreases >$20 per month. However patterns in the San Francisco Bay Area appear to run counter to the Los Angeles area: average premiums in low-lying coastal areas of San Francisco going down, versus scattered decreases in Los Angeles. Deconvolving the overlapping hydraulic, climatic, engineering, and actuarial changes embedded in Risk Rating 2.0 will require much more openness and data sharing on the part of FEMA and its partner modeling contractors.

Figure 5. Average change in NFIP premium cost by ZIP Code, San Francisco area


Risk Rating 2.0 is FEMA’s effort to bring about significant reform to the NFIP, with an eye toward incorporating more accurate flood risk information and, purportedly, more equitable premiums proportionate with that risk. Limited pricing data released by FEMA suggest that most California policyholder will see modest increases, whereas a smaller number of policyholders will see larger discounts which bring down the total NFIP bill to California by at least 10%.  Mapping these data highlights how these increases and decreases vary widely across the state, and FEMA continues to mask the “extreme tails” of the distribution (changes >$100/month, i.e., >$1200 per year) as well as changes beyond the first year of Risk Rating 2.0.  Also importantly, the premium decreases across California are smaller than the documented imbalance between what the state has paid into the NFIP as premiums and what its policyholders have received as payouts over the history of the program (California Water Blog, 12/14/2016). 

Answering questions about Risk Rating 2.0, like the one above, requires independent scrutiny of the modeling and assessing assumptions on which the new insurance premiums are based.  High on that list of important questions is how data provided to FEMA by the Army Corps of Engineers characterize flood risk behind different levees.  “Levee fragility” and “residual risk” behind levees are notoriously difficult to assess and quantify, and slightly different assumptions can mean huge differences in the insurance bills that policyholders are soon to receive. 

Starting on October 1 of this year, flood insurance customers eligible for renewal can lock in the new premiums under Risk Rating 2.0.  Policyholders should check with their insurance agent whether their bills are slated to go up or down, and if down, by all means they should reset as soon as possible.  On April 1, 2022, the new changes will apply to all new and renewing policies.   

While the apparent net decrease in California’s flood-insurance premiums is a welcome change, important questions remain about how private-sector flood models and assumptions about local risk (e.g., levee protection) are weighed in Risk Rating 2.0.  We call for more transparency and dialog with FEMA, including the sharing of additional information and data at a fine-grained spatial resolution.

Ryan Miller is a Ph.D. student in the Geography Graduate Group at UC Davis, researching flood and fire risks, climate change, and urban planning. Peter Hansen, MA is a geospatial specialist and IT consultant in the College of Behavioral and Social Sciences at CSU, Chico. Nicholas Pinter is the Shlemon Chair in Applied Geosciences in the Department of Earth and Planetary Sciences at UC Davis and is Associate Director, Center for Watershed Sciences.

Further Reading

California Department of Water Resources (DWR), 2021.  Central Valley Flood Protection Plan.  https://water.ca.gov/Programs/Flood-Management/Flood-Planning-and-Studies/Central-Valley-Flood-Protection-Plan

FEMA, 2021a.  Risk Rating 2.0: Equity in Action.  https://www.fema.gov/flood-insurance/risk-rating

FEMA, 2021b (Data as of: 07/31/2021).  Policy Information by State and Community.  Available from https://nfipservices.floodsmart.gov/reports-flood-insurance-data

First Street Foundation, 2021.  Find Your Home’s Flood Factor.  https://floodfactor.com/

Pinter, N., R. Hui, K. Schaefer, and D. Conrad, Dec. 14, 2016.  California, Flood Risk, and the National Flood Insurance Program.  California Water Blog, https://californiawaterblog.com/2016/12/14/california-flood-risk-and-the-national-flood-insurance-program/

Pinter, N., 2019.  The Problem with Levees.  Scientific American, Observations (Scientific American Online OpEd), Aug. 1, 2019.  https://blogs.scientificamerican.com/observations/the-problem-with-levees/

Pinter, N., Aug. 28, 2013.  The New Flood Insurance Disaster.  Op-Ed in The New York Times; http://www.nytimes.com/2013/08/29/opinion/the-new-flood-insurance-disaster.html

Schaefer, K., and N. Pinter.  Flood Mapping in California: The Good, the Bad, and the Ugly.  California Water Blog, Jun. 16, 2019.  https://californiawaterblog.com/2019/06/16/flood-mapping-in-california-the-good-the-bad-and-the-ugly/

U.S. Army Corps of Engineers, Sacramento Division, 2021a.  Natomas Basin.  https://www.spk.usace.army.mil/natomas/

U.S. Army Corps of Engineers, Sacramento Division, 2021b.  Reducing Flood Risk in Sacramento.  https://www.spk.usace.army.mil/Missions/Civil-Works/Sacramento-Levee-Upgrades/

U.S. Government Accountability Office (GAO), 2014.  National Flood Insurance Program: Additional Guidance on Building Requirements to Mitigate Agricultural Structures’ Damage in High-Risk Areas Is Needed.  https://www.gao.gov/products/gao-14-583

[1] https://www.spk.usace.army.mil/natomas/

[2] https://www.spk.usace.army.mil/Missions/Civil-Works/Sacramento-Levee-Upgrades/

Posted in Uncategorized | Leave a comment

Could California weather a mega-drought?

Source: National Resource Conservation Administration
Source: National Resource Conservation Administration

By Jay Lund

“Mega-drought” has become a frightful “thing” in public and media discussions.  In the past 1,200 years, California had two droughts lasting 120-200 years, “megadroughts” by any standard. Could the state’s water resources continue to supply enough water to drink, grow crops and provide habitat for fish with such an extreme, prolonged drought today?

Clearly, some ecosystems and rural communities would be devastated by such a drought, and it would certainly affect all California residents.  But with careful management, California’s economy in many ways could substantially withstand such a severe drought.

The UC Davis Center for Watershed Sciences explored this question a few years ago using computer models. We constructed a drought similar in scale to the two extreme ones found in California’s geological and biological records of the past 1,200 years (Harou, et al. 2010). We created a virtual 72-year-long drought with streamflow at 50 percent of current average rates, with all years being dry, as seen in the paleo-drought record.

We then explored the simulated drought using a computer model of California water management that suggests ways to minimize the economic costs of water scarcity for populations and land use in the year 2020.

Not surprisingly, the model results showed that such an extreme drought would severely burden the agriculture industry and fish and wildlife, and be catastrophic to some ecosystems and rural towns. The greatest impacts would be felt in the Central Valley.

However, if well managed, such a mega-drought would cause surprisingly little damage to California’s economy overall, with a statewide cost of only a few billion dollars a year out of a $2+ trillion-a-year economy.

The key to surviving such a drought lies in adaptive strategies such as water trading and other forms of water reallocation. These strategies would be essential to improving the flexibility of California’s water supply and demand system during such a prolonged drought.

Interestingly, most reservoirs we have today would never (yes, NEVER) fill during a decades-long drought.  So expanding surface storage capacity for managing megadroughts would be futile.

California has a very flexible water supply system that can support a large population and economy under extreme adverse circumstances — provided it is well managed.

In adapting to the climate warming and changes that are upon us, the most important thing for California is to be well-organized and led for effective water management.  Panic or complacency generally lead to poor decision.  Good management of such a complex system will require serious and reasoned analysis and discussions, plus a political will to make reasoned decisions, even when ideal solutions do not exist.

Jay Lund is a professor of civil and environmental engineering and co-director of the Center for Watershed Sciences at UC Davis.

This article originally ran April 12, 2011. Some text has been updated,

Further reading

Harou, J. J., J. Medellín‐Azuara, T. Zhu, S. K. Tanaka, J. R. Lund, S. Stine, M. A. Olivares, and M. W. Jenkins (2010), Economic consequences of optimized water management for a prolonged, severe drought in California, Water Resources Research46, W05522, doi:10.1029/2008WR007681.

MacDonald, G.M. (2007), Severe and sustained drought in southern California and the West: Present conditions and insights from the past on causes and impactsQuaternary International, 173-174: 87-100.

Stine, S. (1994), Extreme and persistent drought in California and Patagonia during medieval timeNature, 369, 546–549, doi:10.1038/369546a0.

Posted in Uncategorized | 10 Comments

Lessons from Three Decades of Evolution of Cropland use in the Central Valley

by José M. Rodríguez-Flores, Spencer A. Cole, Alexander Guzman, Josué Medellín-Azuara, Jay R. Lund, Daniel A. Sumner

California’s Central Valley is the source of more than $30 billion of farm value. It produces more milk than any state outside California, and dominates national production of dozens of fruits, vegetables, tree nuts and rice. The valley has two main parts: the Sacramento Valley (north) and the San Joaquin Valley (south); each has particular distinguishing agricultural features (such as soil, hydrology, climate, and economy) that have driven how agriculture and water infrastructure have developed. This post reviews the evolution of the major crops and crop categories produced in the Central Valley of California from 1990 to 2019.

Figure 1 shows cropland use in the Sacramento Valley across three decades. Rice acreage has covered about half of the land used for crops in the valley for many decades. The price and yield of rice have been relatively stable in this period, and acreage has temporarily declined mainly during severe drought periods.  Figure 1 shows sharp drought declines in rice harvested area, such as in the early 1990s and the 2013-2015 period, which recovers with water availability in wetter years. Three decades ago, other grains were second to rice in land use, but now nut crops mainly walnuts, have more acreage than the grain crops. There has been an upward trend in more profitable tree nut crops such as almonds, pistachios and walnuts since the early 2010s, even during the 2012-2016 drought. These perennial crops have shown technology-driven increases in yields and higher prices in the last ten years, but are somewhat more water intensive (requiring 3 to 4 acre-feet per acre) and allow much less flexibility in water use from year to year.

Figure 1: Evolution of the most important crops in the Sacramento Valley from 1990 to 2019, prepared using data from County Ag Commissioners’ Data Listing.

In the San Joaquin Valley (SJV) almonds, pistachios, and walnuts are currently among the dominant commodities by acreage (Figure 2) and production value. These crops now cover over 50% of total irrigated area in the SJV, where Fresno and Kern counties are the main almond producers. Cotton was the most planted single commodity for during the middle of the 20th century in the southern SJV until its production began a sharp decline in the 1990’s as subsidies were reduced and nut acreage began its rapid expansion. Miscellaneous vegetables, melons, tomatoes, fruit and hay and other field crops make up most remaining crops, with their acreages also declining in the last decade.

Figure 2: Evolution of most important crops in the San Joaquin Valley from 1990 to 2019 prepared using data from County Ag Commissioners’ Data Listing.

The crop mix in the SJV has changed with expected future crop demands, relative prices, production costs, and water conditions. For example, alfalfa acreage declined steadily since 2009 because of lack of growth in the California dairy industry, and low returns relative to tree nuts and some annual crops such as tomatoes. The San Joaquin Valley is now dominated by tree nut crops, that climate-limited ranges which align with California’s Central Valley climate.

Central Valley agriculture is supported by California’s array of intertied water supply infrastructure, including surface and groundwater storage, and conveyance, which provides a delivery system for perennial crops with inflexible water demands over the growing season and from year to year. These factors, along with significant long-term improvements in crop yields for several crops and higher expected prices, have encouraged expansive growth in almonds, pistachios, and walnuts.

In dry periods, cropping decisions are limited by water availability and sometimes increased water costs from transfers or groundwater pumping. Apart from old tree removals, acreages of perennial crops (tree nuts, grapes, and citrus trees) remain roughly constant during droughts due to high upfront costs involved in cultivation and higher revenue per harvested acre or acre-foot of water. Most land use reductions occur in crops with more flexibility and for which net return per unit of water is relatively low such as grain, hay and other field crops.

Almonds, pistachios and walnuts grew substantially even in the 2012-2016 drought due in part to high prices and continued expectations of high future prices which outweighed costs of water limitations and concerns about future water access. With the Groundwater Sustainability Plans (GSPs) required by the 2014 Sustainable Groundwater Management Act, some orchard retirement to cope with water scarcity may become a more frequent drought adaptation tool as GSP mandated cutbacks represent a longterm reality in many sub-regions.. For example, corn grown for silage might be expected fall severely in dry periods, but actually maintained substantial acreage in recent droughts because of its key role in local dairy rations and because it is too expensive per unit of value of haul more than 50 miles or so. Corn acreage provides a useful area to spread cow manure, which is unsuitable to distribute through drip fertigation systems. In contrast, alfalfa, which is also water intensive and typically used as a dry roughage feed crop, saw a downward trend in acreage during droughts (Figure 2) as has been regularly shipped into the San Joaquin Valley from Northern California and other states. Alfalfa hay can also use reduced water in drought years at the expense of fewer cuttings and therefore lower yields per acre.

Historically, cropping decisions differ between the Sacramento Valley, where rice production in the north declines somewhat and in the San Joaquin Calley, where hay, corn silage, grain and cotton decline when water is scarce. While many Groundwater Sustainability Agencies have not yet fully implemented groundwater use limitations, such reductions will be a major part of a portfolio of water management actions to achieve local and regional balances in abstraction and recharge by 2040.

Readers interested in visualizing and downloading DWR data on historical water and cropland use and County Agricultural Commissioners’ Report data on crop prices, land use, crop yield and production value, visit the UC Merced – Water Systems Management Lab website in the agricultural data section: https://wsm.ucmerced.edu/agricultural-data/.

Further reading

Hanak, E., Jezdimirovic, J., Green, S. & Escriva-Bou, A. Replenishing Groundwater in the San Joaquin Valley. (2018).

Hanak, E. et al.Water and the Future of the San Joaquin Valley (2019).

Lund, J., Medellin-Azuara, J., Durand, J. & Stone, K.Lessons from California’s 2012–2016 Drought”. J. Water Resour. Plan. Manag. 144, 04018067 (2018).

Mall, N. K. & Herman, J. D. “Water shortage risks from perennial crop expansion in California’s Central Valley”. Environ. Res. Lett. 14, 104014 (2019).

Lund J.Why California’s agriculture needs groundwater managementCalifornia WaterBlog. (May 26, 2014).           

Josué Medellín-Azuara and Jay Lund,“Jobs and Irrigation during Drought in California”.  California WaterBlog. (June 6, 2021)

José M. Rodríguez-Flores and Spencer A. Cole  are respectively Ph.D. candidate and M.S. student at the Environmental Systems graduate program at UC Merced, Alexander Guzman is a former Junior Research Specialist and Lab Manager of the Water Systems Management Lab at UC Merced, Josué Medellín-Azuara is an associate professor at the Department of Civil and Environmental Engineering at UC Merced, and an associate director at the UC Davis Center for Watershed Sciences, Jay R. Lund is a Professor of Civil and Environmental Engineering at UC Davis, where he is also Co-Director for UC Davis Center for Watershed Sciences, Daniel A. Sumner is the Frank H. Buck Jr. Distinguished Professor in the Department of Agricultural and Resource Economics at UC Davis.

Posted in Uncategorized | 4 Comments

Dammed hot: California’s regulated streams fail cold-water ecosystems

Putah Creek, seen from the outlet of Monticello Dam (photo: Ann Willis)

by Ann Willis, Ryan Peek, and Andrew L. Rypel

Given the current drought, it’s no surprise that California’s dams are struggling to provide cool water habitats to support native freshwater ecosystems. But what if they were never able to support them under any conditions?

New research shows how current stream management fails to provide the patterns of cool water that California’s native ecosystems need. The challenges stem from two issues: an oversimplification of stream temperature targets and the assumption that dam regulation can replicate desirable cold water patterns.

We analyzed available long-term (> 8 years) daily stream temperature data records for 77 sites throughout California. We then used an algorithm to see whether the sites could be sorted into groups with similar thermal regimes. The sites were sorted based on annual average temperature, annual maximum temperature, and the day when the annual maximum occurred. Aside from these three temperature metrics, no additional information was used to influence how the algorithm grouped the sites.

Results revealed that California’s streams naturally form five distinct classes (Figure 1). Two of the five classes were unique to groundwater springs, whose temperatures are notably constant throughout the year (stable warm and cold). Of the remaining classes, one included only sites not influenced by dam regulation (variable cool), one included sites heavily influenced by dam regulation (stable cool), and the last showed sites whose temperatures were mostly affected by weather (variable warm).

(A) California’s streams classify into five thermal regimes when grouped using their (B) annual mean, (C) day of annual maximum, and (D) annual amplitude (maximum).

In ecological terms, “cool” and cold” are typically used to describe species that rely on relatively lower temperatures to support optimal growth and survival. These terms are often simplified to mean a specific threshold temperature associated with a species. In the Sacramento River, 56°F is a common target threshold used to sustain salmon. However, our research clearly shows how single temperature metrics like this do not capture the full thermal dynamics of California’s cold-water ecosystems. We must learn to manage towards multiple natural thermal regime endpoints; parallel approaches have greatly improved stream flow management. For example, flow management has transitioned away from single, minimum flow targets to environment flow management that recognizes the importance of diversity in magnitude, frequency, timing, rate of change, and duration of flows. Stream temperature management must make the same change.

When mapped throughout California’s hydrologic regions, the futility of sustaining cold-water ecosystems through regulations is starker (Figure 2). Results show that even when we can heavily manipulate flows or water temperature released from dams, we rarely approach the complexity endemic to natural thermal regimes. Rather than resetting temperature patterns by releasing cooler water than would naturally occur (or warmer, during the winter), dams create an artificial thermal regime that disrupt natural seasonal patterns and create novel thermal habitats – sometimes for tens of miles.

California’s thermal regimes mapped onto their hydrologic regions.

Except for Shasta Dam, no other regulated river showed the kinds of natural temperature patterns that cool- and cold-water ecosystems need. It’s the exception that proves the rule of how poorly dams perform when operated for environmental temperatures. While a few systems may provide opportunities for management of new or novel thermal regimes (which may provide some potential for thermal climate refugia or resilience) these sites are few and far between.

Conservation planning for cold-water species is a risky investment in California. Located at the southern edge of the geographic range of cold-water and anadromous species, California’s freshwater fauna is extraordinarily vulnerable to human-dominated ecosystems. Extinction is likely for most (74%) of California’s native salmonids; though altered or degraded thermal regimes are a major stressor, they are not the only limitations. Bold conservation actions are required to reverse the trend towards extinction. 

Ann Willis is a civil engineer and senior researcher at the UC Davis Center for Watershed Science. This research was part of her dissertation on water temperature management and stream conservation. Ryan Peek is an ecologist and senior researcher at the UC Davis Center for Watershed Science. Andrew Rypel is Co-Director of the UC Davis Center for Watershed Sciences and Peter B. Moyle and California Trout Chair in Coldwater Fish Ecology in the Department of Wildlife, Fish & Conservation Biology.

Further Reading

Willis et al. (2021). Classifying California’s stream thermal regimes for cold-water conservation.

Steel et al. 2017. Envisioning, quantifying, and managing thermal regimes on river networks.

Isaak et al. (2020). Thermal regimes of perennial rivers and streams in the western United States.

Maheu et al. (2016). A classification of stream water temperature regimes in the conterminous USA

Grantham et al. (2014). Systematic screening of dams for environmental flow assessment and implementation

Lusardi et al. (2021). Not all rivers are created equal: the importance of spring-fed rivers under a changing climate.

Moyle et al. (2017). State of the Salmonids II: Fish in Hot Water

Poff et al. (1997). The Natural Flow Regime

Yarnell et al. (2015). Functional flows in modified riverscapes: hydrographs, habitats, and opportunities

Posted in Uncategorized | Tagged , , , , , , | Leave a comment

2021 Drought in California – in one page

by Jay Lund

Droughts and this drought in California

  • California has more hydrologic variability than any state in the US, meaning that we have more drought and flood years per average year than any other state.  This is a problem, but has also meant that we have designed for droughts, which are always testing us.
  • 2021 is the 3rd driest year in more than 100 years of precipitation record.  2020 was the 9th driest year in the precipitation record. 
  • Much warmer temperatures are further reducing streamflows and aquifer recharge, and has lengthened and deepened the wildfire season.
  • Large reductions are occurring in surface water available for agriculture, especially in the San Joaquin Valley, but also in the Sacramento Valley and smaller river valleys statewide.
  • Much increased groundwater pumping greatly reduces agricultural impacts, but affects rural wells.
  • Major forest and aquatic ecosystem impacts are occurring, especially for wildfires and salmon runs, particularly for winter-run salmon downstream of Shasta Dam.
  • A growing number of small communities and towns are being affected, in addition to more common problems for rural household and community wells. Santa Clara Valley (San Jose area) is the most-affected major urban area, seeking 30% water use reductions.


  • If next year is also dry, agricultural and environmental impacts will increase and urban impacts will expand.
  • Warmer temperatures from climate change are worsening droughts, reducing the amount of precipitation that arrives at reservoirs and aquifers, lengthening wildfire seasons, and worsening conditions for cold-water fish species, such as salmon.  We need to further adapt water, land, and environmental management for these changes.
  • Another dry year is likely. Very dry watersheds, very low reservoir levels, falling aquifers, and higher temperatures mean more precipitation is needed to make next year not dry.
  • Under SGMA, farmers will need to repay additional groundwater pumped during the drought, meaning some reductions in lower-valued crops in wetter years so aquifers can recover to sustain permanent crops in future droughts. Few basins can sustain aquifers with managed aquifer recharge alone; many will need deep reductions in aquifer pumping in wetter years.
  • Sizable long-term reductions in irrigated area seem unavoidable in parts of the San Joaquin Valley. Urban water conservation statewide is helpful, and still more conservation will help a bit.
  • A more formal state water accounting system is needed to support tighter surface water right administration, SGMA planning and implementation, and environmental uses. Water right curtailments are likely to become routine in more basins.

Further Reading

Arax, Mark (2021), “The Well Fixer’s Warning,The Atlantic, August 17.

Lund, J.R., J. Medellin-Azuara, J. Durand, and K. Stone, “Lessons from California’s 2012-2016 Drought,” J. of Water Resources Planning and Management, Vol 144, No. 10, October 2018.

Pinter, N., J. Lund, and P. Moyle. “The California Water Model: Resilience through Failure,” Hydrological Processes, Vol. 22, Iss. 12, pp. 1775-1779, 2019.


Jay Lund is Co-Director of the Center for Watershed Sciences and Professor of Civil and Environmental Engineering at the University of California – Davis.

Posted in Uncategorized | 7 Comments

Living with non-native fishes in California requires using the right words

Redeye bass in the Cosumnes River: a non-native species that deserves the title of Alien Invader because it has eliminated native fishes from most of the river. It was introduced assuming it would have a mutualistic relationship with people (Moyle et al. 2003).

by Peter Moyle

Everywhere you go in California, people live in landscapes where non-native species are conspicuous:  European grasses turning the hills golden, earthworms tilling our garden soil, exotic trees providing shade, bullfrogs jumping into backyard ponds, starlings making tight maneuvers overhead. In this blog, I want to describe the language of our relationships with non-natives and the nature of those relationships as biological phenomena, using fishes and other aquatic organisms as examples.  Reconciling our relationships with non-native species requires a vocabulary that reflects our attitudes towards them and their management.


In a recent WaterBlog, Stompe et al. (https://californiawaterblog.com/2021/07/04/home-is-where-the-habitat-is/) argued that non-native fishes thrive in California because we have created habitats much like the ones to which they were native. These habitats, alas, are mostly quite different from the ones to which our native fishes are adapted.  To some extent, native and non-native fishes can form coalitions (novel ecosystems) where the resources available are divided up among the species, much as it supposedly is in undisturbed habitats (e.g. Aguilar-Medrano et al. 2019).  But, in general, non-native fishes are replacing natives as habitats change, mostly as we have changed them. How do we live with this change but still save native fishes?  Let’s start with language.

  • I use non-native as the general, seemingly neutral, term for species from elsewhere that have extended their range into California, thanks to being brought here by people. Most are introduced species, another neutral term.
  • The word exotic is a somewhat positive term, bringing to mind exotic people and places, as well as nature specials on television.  Today, it is most often applied to fish and plants in the aquarium trade.
  • Non-indigenous is the term often used by government agencies because it is neutral, hard to pronounce, and can be bureaucratically abbreviated to NID species.
  • Invasive is the term for non-native species that are harmful and/or spreading.  Unfortunately, it is widely applied to all non-native species, even though many are not demonstrably harmful.
  • Naturalized can be used to describe non-native species that were introduced so long ago they have adapted to California’s distinctive environment and are integrated into the biotic communities. Fish examples include striped bass, American shad, and common carp.
  •  Alien is a negative term for non-native species, enhanced by science fiction movies.  The worst thing you can call a non-native species is alien invader.  This term fits realistically the northern pike, highly predatory species, on which California was willing to spend millions of dollars to eradicate before it spread widely, devouring native fishes.
Examples of non-native fish species in California. Upper left: Shimofuri goby, USBR (60 mm, commensal).   Upper right, American shad USFWS (adult, naturalized, mutualist); Lower left: Mississippi silverside NANFA (75 mm, mutualist turned alien invader); northern pike UC ANR (adult, alien invader, eradicated).


The diversity of words we use to describe non-native species reflects the ambiguity in our attitudes towards them.  This partly reflects the diverse symbiotic relationships that we humans have with other species, especially non-native species.  Symbiosis in its simplest sense means ‘living together’ and usually includes mutualism, commensalism, parasitism, competition, and predation.

The symbiotic relationship looked upon most favorably is mutualism, where both species benefit from a relationship.  Most sport fishes brought into California by agencies could be considered as mutualists, when people benefit from the fishery and the fish benefit from having their range expanded, important in the game of long-term evolution and survival. Of course, as human attitudes change and protection of native species from non-native species becomes a priority, the relationship between humans and non-native fish becomes more complicated.

For some species, the relationship with people is commensalism, where one species benefits but the other species neither benefits nor is harmed.  An example is the shimofuri goby, a small fish with a life cycle that permits it to be carried across oceans in the ballast water of ships. It became established in the Delta in the 1980s and is now a common species, enjoying life amongst the rip-rap and trash, feeding primarily on non-native invertebrates that are not eaten by other fishes.  This benign relationship with people may change if it manages to invade (which it probably will) estuaries in southern California where it can compete with/prey on the endangered tidewater goby.

A poorly understood (for California fish) symbiotic relationship is parasitism, including disease, where the parasite benefits from the relationship but the host is harmed.  For example, whirling disease, scourge of both hatchery and wild populations of rainbow trout, is native to Europe and was brought to North America in trout from Europe.  It has forced some California trout hatcheries to shut down with considerable impact on recreational fisheries.

Competition and predation are often not listed as forms of symbiosis but they are interspecies interactions of great consequence. Humans, for example, compete with freshwater fish for water (a resource in short supply), with the fish usually losing, regardless of native vs non-native status, unless people choose to back away from the relationship.  Likewise, people are the biggest predators on fishes, via fisheries, management of which can favor non-native species (e.g. largemouth bass) over native fishes (e.g. Sacramento pikeminnow).

Regardless of the initial reason people and non-native species wind up living together, the relationship often becomes complicated. The Mississippi silverside was introduced into Clear Lake for an assumed mutualistic relationship, to control pestiferous gnat populations, which it did. It also became abundant in the process, an expected result of the relationship.  But what was not expected was that silversides were carried from the lake, down Cache Creek and into the Delta, where they likely prey on the eggs and larvae of native fishes, such as delta smelt. They also serve as prey themselves for native birds like herons and egrets. They are now abundant in reservoirs in southern California, with unknown effects.

All these types of interactions between people (a non-native species) and fish show that understanding the biology of each species and our relationships with them is important. Like it or not, we humans need to have this understanding for living symbiotically with non-native species so we don’t have to consider ourselves at war with them.

Living with non-native species.

The widespread colonization of highly altered freshwater environments by non-native species is resulting in increased biotic homogenization. This is especially true in California and the American west because of intense development of water for people has resulted in massive changes to our waterways. A few natives can survive in new habitats such as reservoirs, but most cannot.  Our inability to stop the juggernaut of habitat change and non-native species demonstrates “…we are losing the battle against extinction…and against the seizure of aquatic ecosystems by alien invaders (Moyle 2021, p 70)”.  To take this militaristic analogy further, one of the best ways to save native fishes seems to be through creation of a system of fortresses, aquatic preserves that are bulwarks against species invasions.  For non-native fishes, a triage system for management can be recognized: eradication, control, acceptance. 

Eradication should be the main option for new arrivals that have a limited enough distribution that eradication is possible (e.g., Northern pike in Lake Davis). It is also a good option for populations of non-natives confined to treatable areas, such as brook trout in Sierra Nevada lakes where eradication via gill nets is possible, one lake at a time.  This strategy is important to restore native amphibians and invertebrates to the lakes.

Control can be an alternative where complete eradication is not possible. This approach has been used to create stream refuges for golden trout in California: a barrier is first built and then non-native trout eliminated above the barrier using a degradable fish poison. Golden trout are then re-introduced.  Such control of non-natives is rarely permanent, however.  In regulated streams, some control over non-native species can result from a tightly managed flow regime that favors native fishes.

Acceptance of non-native fishes is hard for those of us engaged in native fish conservation, but reality dictates that there is often little choice.  This does not mean they should not be managed at all, but instead managed in ways that minimize negative impacts on native species.  It is not worth spending time and energy on ‘saving’ native fishes where environments are so severely altered that native species will not persist whether or not non-native species are present (e.g.  most reservoirs).  On the other hand, where major positive environmental change is likely, as through dam removal, its effects on both native and non-native fishes will need to be considered.  Sometimes environmental restoration to favor native species will instead favor non-natives (e.g., Williamshen et al. 2021).

An approach that I think is helpful in resolving the native vs non-native fish dilemma is reconciliation ecology.  Reconciliation ecology is defined as “the science of inventing, establishing, and maintaining new habitats to conserve species diversity in places where people live work and play” (Rosenzweig 2003, p 7). This concept acknowledges that people dominate most ecosystems today, which means we determine what species they will contain as time goes on.  Many, if not most, ecosystems can support both native and non-native species. If we understand the role of non-native species, we are more likely to also keep native species as significant parts of California’s unique ecosystems.

This blog is based, in part, on Moyle (2020).

Further reading

Aguilar-Medrano, R., J. R. Durand, V.H. Cruz-Escalona and P.B. Moyle.  2019. Fish functional groups in the San Francisco Estuary: understanding new fish assemblages in a highly altered estuarine ecosystem. Estuarine, Coastal and Shelf Science 227106331 https://doi.org/10.1016/j.ecss.2019.106331

Moyle, P.B., 2020.  Living with aliens: nonnative fishes in the American Southwest. Pages 69-78 In D.L. Propst, J.E. Williams, K.R. Bestgen, and C.W. Hoagstrom, eds., Standing Between Life and Extinction: Ethics and Ecology of Conserving Aquatic Species in North American Deserts. Chicago: University of Chicago Press.

Moyle, P. B., P. K. Crain, K. Whitener, and J. F. Mount. 2003. Alien fishes in natural streams: fish distribution, assemblage structure, and conservation in the Cosumnes River, California, USA.  Environmental Biology of Fishes 68: 143-162.

Rosenzweig, M.L., 2003. Win-win Ecology: How the Earth’s Species Can Survive in the Midst of Human Enterprise. Oxford University Press on Demand.

Williamshen, B.O., T.A. O’Rear, M. K. Riley, P.B. Moyle, J. R. Durand. 2021. Tidal restoration of a managed wetland in California favors non-native fishes.  Restoration

Ecology. Society for Ecological Restoration. doi: 10.1111/rec.13392 12 pages

Peter Moyle is Distinguished Professor Emeritus at the Center for Watershed Sciences, UC Davis

Posted in Uncategorized | 2 Comments

The California Water Model: Resilience through Failure

by Nicholas Pinter, Jay Lund, Peter Moyle

This is a slightly-edited re-posting from May 5, 2019.

Sacramento 1862

Figure 1.  1861-62 flood in Sacramento.

A review of 170 years of water-related successes in California suggests that most successes can be traced directly to past mistakes.  California’s highly variable climate has made it a crucible for innovations in water technology and policy.  Similar water imperatives have led to advances in water management in other parts of the world.  A close look at California’s water model suggests that “far-sighted incrementalism” is a path to progress.  Given the complexity of water management systems, better scientific information and new policy tools must be developed coherently and collaboratively over time.  A history of learning from previous failures can guide progress towards stable, secure, and resilient water systems worldwide.  This includes learning from other regions and other “water models” – the one option clearly superior to innovating in response to your own mistakes is learning from the errors of others.

This post summarizes an article published in Hydrological Processes,  https://onlinelibrary.wiley.com/doi/epdf/10.1002/hyp.13447

Average runoff in California is about 100 km3/yr, but our ecosystems and parts of our economy have been water-limited for decades.  Part of the state’s challenge comes from the variability of its climate.  Average years are unusual, and instead long droughts are punctuated by years of heavy rain or snowmelt and flooding.  Nonetheless, the state has managed to thrive, with 40 million people, agricultural production exceeding $45 billion/year, and the world’s sixth largest economy.  California’s droughts and floods and tension between economic growth and environmental protection have pushed it to develop a diverse toolkit for managing water.

The toolkit consists of an integrated system of infrastructure, laws, institutions, and economic tools. This system, the ”California water model,” has evolved from the first Spanish settlement, through the gold mining era, the ascendancy of agriculture and major cities, to the recent broad mix of objectives that includes strong environmental protection. California has steadily adapted its water management by making mistakes and then learning from those mistakes. In 2017, for example, California avoided major flooding despite the winter being one of the wettest on record and major spillway failures. This was partly due to luck. Reservoirs began low from a drought and winter storms were widely spaced. And most flood infrastructure, particularly the flood bypasses, functioned well. California’s water model offers broad lessons for water managers, particularly in arid regions.

Water in California is framed by the state’s Mediterranean climate.  Summers are long and dry and most precipitation comes during the winter.  Historically, much of the water supply comes from mountain snowmelt (the state’s largest surface reservoir), reservoirs, and groundwater.  In addition to this seasonality, wet years often follow droughts, and vice versa, high variability accentuated by climate change. This near-perpetual alternating water crisis forces Californians to find innovative solutions.  Whereas other US states and other countries may have decades to settle into a false sense of security, California’s hydrologic extremes accelerate innovation.

In 2017, California emerged from a severe five-year drought.  The drought’s effects on agriculture were limited because past droughts had led to more flexible water markets, and farmers greatly expanding groundwater pumping.  Although the state lost about a third of its water supply, agricultural revenue losses were only 3%, and only about 6% of the land was fallowed.  This was in part because producers of lower-value crops sold or transferred their water to producers of higher value crops such as fruit, nuts, and vegetables and to urban water users.  The expanded groundwater pumping raised the visibility and impacts of long-term groundwater problems, which in turn led to passage of California’s Sustainable Groundwater Management Act, which will regulate groundwater in the future.

At the other extreme, California also has long history of damaging floods, and flood risk remains widespread today.  Winter storms of 1861-62, turned much of the Central Valley into an inland sea, and frequent levee breaches through the 19th and 20th centuries resulted in high costs to landowners and to the state. In less variable regions, the decades between major floods lead to a “hydro-illogical cycle” in meaningful steps avoid flood damage are forgotten in intervals between disasters.  Early on in California, repeated flooding led to construction of Yolo and Sutter Bypasses, which remain a world model for basin-scale flood management.  A costly 1986 levee failure (and national headlines from New Orleans in 2005) sparked new legislation and investment that has upgraded many California levees from some of the worst in the nation to some of the best.  Repeated flood disasters have kicked the state in the right direction, although much work always remains.  The near-disaster at Oroville Dam in February of 2017, where two spillway failures led to major evacuations, sparked scrutiny and investment at Oroville Dam and for aging water infrastructure across California.  Other regions with large dams, or contemplating new dams, should include Oroville’s lessons in their textbook.

sjv land subsidence

Figure 2.  Unchecked groundwater overdraft has brought ground-surface subsidence.  California’s San Joaquin Valley’s severe subsidence over the past century, continues locally today.  Photo courtesy of Michelle Sneed, US Geological Survey.

Despite successes, California’s water management faces continued challenges.  High on this list, protecting endemic aquatic species remains a vexing challenge.  Despite legal protections under federal and state regulations, California’s native fishes are in rapid decline, with 80% of species on paths towards extinction.  California will need to expand its toolkit – such as by accepting “reconciliation ecology” as a new model for maintaining natural diversity in the face of human pressures and a changing climate.

A prerequisite for providing and maintaining healthy aquatic ecosystems and adequate supplies of clean water is “far-sighted incrementalism” among water managers and political leaders.  Incrementalism involves addressing seemingly intractable problems by small forward-looking steps.  “Far-sighted,” at least in California, has involved forward-thinking planning among scientists, managers, and leaders during and after each water-related crisis.  The common response after a damaging flood is reactive – repair the levee breach and rebuild floodplain neighborhoods.  Far-sighted leaders see opportunities in such a crisis to move the system forward, usually incrementally, in a longer-term strategic direction (usually too controversial or difficult to achieve in one step).  California must continue to support organized and independent learning from and adapting to disasters and extremes.

 Lessons for managing water in a thirsty world

By 2050, an additional 2.3 billion people worldwide will face severe water stress, especially in Africa and southern and central Asia.  Already, 2.1 billion people worldwide lack access to safe drinking water. Three out of four jobs worldwide depend upon access to water and water-related services.  Water-limited regions and populations must prepare for changes in water management, addressing existing and emerging weaknesses and learning from mistakes, if possible from other areas, without repeating those errors.

Water management successes often rest on past failures – failures from which scientists, managers, and leaders learn and adapt.  This is especially true for California, where hydrologic variability frequently tests water systems and water policy.  As the world, especially the arid to semiarid world, looks for water solutions, the failures and lessons from California’s turbulent history can provide guidance for future global water resilience.

Nicholas Pinter, Jay Lund, and Peter Moyle are faculty in the Departments of Earth and Planetary Sciences, Civil and Environmental Engineering, and Wildlife, Fish, and Conservation Biology (respectively) and work together at the Center for Watershed Sciences at the University of California, Davis.  Email: npinter@ucdavis.edu; jrlund@ucdavis.edu; pbmoyle@ucdavis.edu

Further Readings

Auerswald, K, P. Moyle, S.P.Seibert, and J. Geist. 2019. HESS Opinions: Socio-economic and ecological trade-offs of flood management – benefits of a transdisciplinary approach. Hydrology and Earth System Sciences 23: 1035-1044.  https://www.hydrol-earth-syst-sci.net/23/1035/2019/  Open access.

Dettinger MD, Ralph FM, Das T, Neiman PJ, & Cayan DR. 2011. Atmospheric rivers, floods and the water resources of California.  Water, 3: 445-478.

Faunt, C., and M. Sneed, 2015.  Water availability and subsidence in California’s Central Valley.  San Francisco Estuary & Watershed Science, vol. 3, available fromhttps://ca.water.usgs.gov/pubs/2015/FauntSneed2015.pdf

Grantham, T.E., R. Figueroa, and N. Prat, 2013.  Water management in mediterranean river basins: a comparison of management frameworks, physical impacts, and ecological responses.  Hydrobiologia, 719: 451–482.

Independent Forensic Team, 2018.  Independent Forensic Team Report, Oroville Dam Spillway Incident, Jan. 5, 2018, https://damsafety.org/article/oroville-investigation-team-update

James, L.A., and M.B. Singer, 2008. Development of the Lower Sacramento Valley Flood-Control System: Historical Perspective, Natural Hazards Review, 9(3): 125-135.

Kelley, R., 1989.  Battling the Inland Sea, University of California Press, Berkeley, CA.

Konar M, Evans TP, Levy M, Scott CA, Troy TJ, Vörösmarty CJ, Sivapalan M. 2016. Water resources sustainability in a globalizing world: who uses the water? Hydrological Processes, 30: 330-336.

Lund, J.R., J. Medellin-Azuara, J. Durand, and K. Stone, “Lessons from California’s 2012-2016 Drought,” J. of Water Resources Planning and Management, Vol 144, No. 10, October 2018. (free download)

Lund, J., 2016.  You can’t always get what you want – A Mick Jagger theory of drought management.  California Water Blog, https://californiawaterblog.com/2016/02/21/you-cant-always-get-what-you-want-a-mick-jagger-theory-of-drought-management/.

Moyle, P., R. Lusardi, P. Samuel, and J. Katz. 2017. State of the Salmonids: Status of California’s Emblematic Fishes 2017.  Center for Watershed Sciences, University of California, Davis and California Trout, San Francisco, CA. 579 pp. https://watershed.ucdavis.edu/files/content/news/SOS%20II_Final.pdf

Multi-Benefit Flood Protection Project, 2017.  Projects, http://http://www.multibenefitproject.org/projects/.

OECD Organisation for Economic Co-operation and Development, 2012.  OECD Environmental Outlook to 2050: The Consequences of Inaction.  OECD Publishing, Paris.  http://dx.doi.org/10.1787/9789264122246-en

Opperman, J.J, P.B. Moyle, E.W. Larsen, J.L. Florsheim, and A.D. Manfree. 2017 Floodplains: Processes, Ecosystems, and Services in Temperate Regions. Berkeley: University of California Press.

Pinter, N., J. Lund, and P. Moyle. “The California Water Model: Resilience through Failure,” Hydrological Processes, Vol. 22, Iss. 12, pp. 1775-1779, 2019.

Pinter, N., A. Damptz, F. Huthoff, J.W.F. Remo, and J. Dierauer, 2016.  Modeling residual risk behind levees, Upper Mississippi River, USA.  Environmental Science & Policy, 58, 131-140.

Pisani, D., 1984. From the Family Farm to Agribusiness: The Irrigation Crusade in California, 1850–1931. Berkeley: University of California Press.

Soulsby, C, Dick J, Scheliga B, & Tetzlaff D. 2017. Taming the flood—How far can we go with trees? Hydrological Processes, 31: 3122–3126.

Vahedifard, F., A. AghaKouchak, E. Ragno, S. Shahrokhabadi, and I. Mallakpour, 2017.  Lessons from the Oroville dam.  Science, 355: 1139-1140.

Van Lanen HAJ, et al. 2016. Hydrology needed to manage droughts: the 2015 European case.  Hydrological Processes, 30 https://doi.org/10.1002/hyp.10838

WHO & UNICEF World Health Organization and the United Nations Children’s Fund, 2017.  Progress on drinking water, sanitation and hygiene: 2017 update and SDG baselines. Geneva: World Health Organization (WHO) and the United Nations Children’s Fund (UNICEF).

Posted in Uncategorized | 1 Comment

Experimental Habitats for Hatchery Delta Smelt

by Peter Moyle

Borrow pit on Bouldin Island that is now a pond maintained by ground water inflow. If expanded and deepened, and carefully managed, such ponds could become temporary rearing habitat for delta smelt. Photo by Peter Moyle.

The Delta smelt is either extinct in the wild or close to it; in the past year only a handful have been caught, with great effort. In contrast, the UC Davis Fish Conservation and Culture Laboratory (FCCL) has considerable success spawning and rearing the smelt in captivity. This coming winter, the FCCL will have as many as 40,000 smelt ready for release, when temperatures are low and the smelt are likely to spawn naturally. Such releases will be ‘experimental’ so not subject to the take provisions of the federal Endangered Species Act. In this blog, I support the concept that success of re-establishing smelt in the wild requires using multiple approaches.  In previous blogs and papers (Börk et al. 2020, Stompe et al. 2021), the idea of planting hatchery smelt in selected reservoirs was discussed. Here I first explore the problems of releasing fish back into the Delta and then describe an experimental reintroduction project taking advantage of the characteristics of Delta islands.

What do you do with 40,000 smelt?

It is a big question as to what to do, exactly, with 40,000 hatchery smelt. This is the job of the interagency Culture and Supplementation of Smelt Steering Committee (CASS) that has also has to deal with obtaining the many permits needed for any kind of release (CEQA, ESA etc.). Permitting will be a barrier to any action unless CASS decision makers are able to make risky decisions in short periods and get rapid action on the permits. My impression is that CASS prefers to use most of the fish for experimental releases, either as a one-time introduction of all fish or as series of smaller introductions made at selected sites where conditions appear favorable.  There are a number of problems with this approach:

  • The current Delta environment seems unsuitable for wild smelt. Cultured smelt will face the same problems as wild smelt, resulting in low survival. Börk et al. (2020) show that the factors put forth in the USFWS Biological Opinion as supporting recovery of smelt populations don’t hold up under close scrutiny.
  • Keeping track of even 40,000 small smelt in the big Delta is extremely difficult, if not impossible, given the small size of the smelt.
  • Hatchery smelt are genetically and behaviorally adapted for life in an aquaculture facility, so it is doubtful they will be able to quickly adjust to being released into a more natural environment. 
  • A strategy focusing on experimental releases into the Delta does not diversify risks but instead seems to be “putting all the eggs in one basket”.

Alternative: island ponds

Figure 1. Conceptual model of island experimental Delta smelt pond (diagram by Kelly Schmutte).

Mount and Twiss (2005) published a provocative analysis that showed most Delta islands (polders) were below sea level and were continuing to subside. Given weak levees, these polders are subject to sudden flooding, with huge costs to reclaim the land. But viewing flooded islands as habitat for fishes, including delta smelt, struck a number of biologists, including me, as a vision worth exploring. One variant of this idea was to deliberately flood islands and have gated openings so the water could be managed. This variant was not original, in that a private company called Delta Wetlands had actually purchased islands to flood them for water storage. While these big schemes never came to fruition, the idea of using flooded islands for fish conservation continued to be of interest. The version presented here proposes smaller, more manageable habitats, essentially large ponds, on subsided islands. This version developed through several Zoom meetings and involved scientists from Metropolitan Water District, UC Davis, and US Geological Survey.

The basis for this project is the need to spread risks for smelt introductions and to build a controlled food-rich habitat for conditioning hatchery smelt for improved survival in the wild. If this project is successful, it may be possible to propagate smelt in more or less natural ponds until the Delta ecosystem, or parts of it, has improved enough to support smelt on their own. The project is based on research, including findings of Jon Burau and others at USGS, on the hydrodynamic interactions between marsh and pelagic environments. The project includes ponds which are hydrologically-connected to artificial tidal marshes of varying sizes within a 43-acre plot on the northwest corner of MWD’s Bouldin Island (Figure 1, above).

Under current plans, one marsh-pond complex will be operated with water levels that will be raised and lowered for maximum night-time cooling; preliminary modelling by MWD indicates that under present conditions, water temperatures during the hot days of summer can be kept at or below 20°C. A second marsh complex will be operated to maintain maximum residence time for food (zooplankton) productivity. The ponds will be supplemented with cold, aerated groundwater (usually <17° C) during hot months in order to maintain water temperatures sufficient for a Delta smelt growth and survival. Temperature modelers at MWD are confident that sufficient design features can be developed to maintain appropriate conditions for smelt. Furthermore, bioenergetic studies of fish show that a robust food supply, which the project will include, can help compensate for the elevated metabolism of fish living in warmer water than might seem optimal. Finally, this site allows for the research and monitoring necessary for successful smelt supplementation efforts and Delta-wide tidal marsh design parameters. For example, the project could be useful for evaluating acclimation and release methods (particularly of early life stages, hatching frames, and mesocosms) and practices to reduce domestication selection through rearing in a more natural environment.

Figure 2. Drone photo of Bouldin Island, showing site of potential ponds for rearing Delta smelt. The existing facilities in the lower right foreground are eight experimental pools for examining (among other things) how floating mats of tules could provide a supply of invertebrates as smelt food.  New pools could be constructed immediately adjacent to these floating marshes as a near-term, cold weather conditioning facility while a permanent marsh-pond complex is being constructed in the center of the photo. The field in the center could be used in various ways to create ponds for Delta smelt rearing. The lighter colored area upslope of the pond site and edged by the dark green growths of blackberries, could be converted into marshes that drain into the ponds, as in Figure 1. The island is subsided so the ponds are about 20 feet below sea level. Sea level is the surface elevation of the water in the surrounding channels. Photo from Russ Ryan, MWD.

While MWD developed and funded the initial ‘proof of concept’ project, smelt stakeholders and experts will need to work together for the planning, funding, permitting, and monitoring of the project if the project is to succeed. It is urgent that a collaborative effort be made so that a project using 2000 or so hatchery smelt can be possible by the end of this year.   Assuming smelt are available, the project would be on-going in 2022. Time is of the essence, given uncertainties about the status of the smelt.  However, it is worth noting that the facilities and information produced as part of this project can also be applied to saving other native fish species in decline: splittail, Sacramento perch, longfin smelt, hitch, and tule perch.


This pond project here has been presented in a positive light. But it is a high risk project for smelt because it involves creating new habitats by moving dirt and water around, on a subsided island. Such islands have a history of being flooded through levee failure, with additional risks caused by sea level rise, major flood events, and earthquakes. Climate change and drought may make water temperatures warmer than expected.  One way or another, it is likely that the created habitats will be invaded by non-native fishes such as Mississippi silverside. But any project in the Delta, including a straight-forward smelt re-introduction project, faces these same or similar problems; they will have to be dealt with in creative ways. The alternate path of doing nothing or doing too little leads to extinction.


There is well-established precedence for the use of ponds in supplementation strategies of listed fish species as well as for Delta Smelt experimentation. The endangered Rio Grande Silvery Minnow is propagated in earthen ponds prior to release in the wild. The endangered Razorback Sucker is also grown out in seminatural and natural ponds before stocking, a practice which has improved growth rates and subsequent survival in the wild. Aquaculture ponds at the U.C. Davis Center for Aquatic Biology and Aquaculture have been employed extensively in research on Delta Smelt physiology and field trials in species-specific enclosures. 


Re-establishing Delta smelt in its native Delta using hatchery smelt is an extremely difficult task, given how completely the habitat has been altered (Stompe et al 2001). Releasing fish directly into the wild is very risky and success will be hard to determine. Alternative projects need to be developed to spread risk. The Polder Pond Project proposed here is one such project. It proposes to rear Delta smelt in large ponds on a Delta island, on natural foods, which should prepare the fish better, at larger sizes, for release into the wild. The project also entails some risk for the Delta smelt needed for the project (ca. 2,000/year) but even if the smelt fail to adapt well to the ponds, useful information will be obtained on restoring native fishes to the Delta.


Bixby, R., and A. Burdett. 2013. Annual report 2011-2012; Resource utilization by the Rio Grande silvery minnow at the Los Lunas Silvery Minnow Refugium. Available at: ose.state.nm.us

Börk, K., Moyle, P., Durand, J., Hung, T., Rypel, A. L. 2020. Small populations in jeopardy: delta smelt case study. Environmental Law Reporter, 50(9), 10714-10722

Caldwell, C.A., Falco, H., Knight, W., Ulibarri, M., and W.R. Gould. 2019. Reproductive potential of captive Rio Grande Silvery Minnow. North American Journal of Aquaculture 81:47-54.

Day, J.L., Jacobs, J.L., and J. Rasmussen. 2017. Considerations for the propagation and conservation of endangered Lake Suckers of the western United State. Journal of Fish and Wildlife Management 8:301-312.

Mount J. and R. Twiss, 2005. Subsidence, sea level rise, seismicity in the Sacramento-San Joaquin Delta. San Francisco Estuary and Watershed Science. Vol. 3, Issue 1 (March 2005), Article 5. http://repositories.cdlib.org/jmie/sfews/vol3/iss1/art5.

Stompe. D., T. O’Rear, J. Durand, and P. Moyle. 2021 Home is where the habitat is. California WaterBlog  https://californiawaterblog.com/2021/07/04/home-is-where-the-habitat-is/

Watson, J.M., Sykes, C., and T.H. Bonner. 2009. Foods of Age-0 Rio Grande Silvery Minnows (Hybognathus amarus) reared in hatchery ponds. The Southwestern Naturalist 54:475-479.

Peter Moyle is Distinguished Professor Emeritus at the Center for Watershed Sciences, University of California, Davis.

Posted in Uncategorized | Leave a comment

California’s Missing Forecast Flows in Spring 2021 – Challenges for seasonal flow forecasting

by John Abatzoglou, Anna Rallings, Leigh Bernacchi, Joshua Viers, Josué Medellín-Azuara

California’s 2021 water outlook became grimmer this spring as the state did not get fabulous February or miracle March precipitation. Unsurprisingly, spring streamflow forecasts from snowfed basins in the Sierra were far below average. For example, early April forecasts from California DWR called for April-July runoff to be between 59-70% of normal. Bad, but not terrible. Then came April, bringing little additional precipitation. To compound matters, April brought very warm temperatures for much of the state that led to rapid ablation (evaporation) of the Sierra snowpack. April began with snowpack around 60% of normal, but by month’s end was 20% of normal.

In early May a significant revision of forecasted spring flow estimated substantial reductions from forecasts of just one month earlier – in some cases a 28% reduction. Aggregated over the Sacramento River Basin, total forecasted flow for April-July dropped by 0.8-1.0 million-acre feet. This reduction in forecasted water supply turned a bad water year into a dreadful one with an amplified conundrum of long-standing water conflicts.

What led to such a drop-off in forecasts? Where did all that snow go if not into flow? While there is an official review of the procedures that led to the differences in forecasts underway, in this blog we look at five possible culprits.

Suspect #1: Lack of April Showers

April was dry. Much of the state had 10-25% of ‘normal’ April totals. April is never a drought buster, but the shortfall of April precipitation only worsened the developing drought. Aggregated over the Sierra Nevada, we estimate a shortfall of around 3” of precipitation in April 2021 that could otherwise contribute to streamflow, acknowledging that some of this became evapotranspiration. Certainly, some of the streamflow forecast revision could be from lack of rain, but unlikely enough to warrant such a large downgrade by itself.

Figure A: April 2021 precipitation percentiles relative to 1895-present period. Data source: West Wide Drought Tracker (Abatzoglou et al., 2017).

Suspect #2: Sublimation of water from snow to the atmosphere

There has been speculation that high rates of sublimation in April facilitated the rapid ablation of snow – limiting how much of the snow became streamflow. Sublimation – direct loss of water from the snowpack to the atmosphere – can be an important fate for snowpack, particularly in locations with strong winds, very low dewpoint temperatures, snow in tree canopies, and where snowpack energetics limit melt. In practice, sublimation is challenging to track. To get a sense for how important sublimation generally is for snowpack ablation, we look at long-term data from the Variable Infiltration Capacity macroscale hydrology model. While these data suggest that sublimation can be quite high in the interior West, up to 40% in the semi-arid areas of the West including in the eastern Sierra, it is much lower in more maritime locations such as the Sierra western slope where sublimation is generally below 10%.

Data source: Variable Infiltration Capacity model output.

In April, dewpoint temperatures were 1℃ below 1991-2020 normal which would have hastened sublimation a bit, and wind speeds were not unusual. Given the rapid pace of snowpack ablation, it is unlikely that sublimation was the primary culprit behind the missing flow. More generally, hydrologic model simulations show that a warming climate generally reduces sublimation and increases infiltration.

Suspect 3: Evapotranspiration

Evaporation from melted snow at the surface and overall evapotranspiration is another likely factor reducing flow. There is strong evidence suggesting April 2021 reference evapotranspiration (Eto) was near record high values for much of the region. The combination of warm temperatures, unusually clear skies, and low dewpoint temperatures facilitated this increase in atmospheric thirst for the month. But this would only be about 1” more than average ETo for April.

Figure C: April reference evapotranspiration (ET) aggregated broadly over the Sierra Nevada during 1979-2021. Data source: gridMET meteorological data and Climate Engine (Huntington et al., 2017).

Suspect 4: Dry antecedent conditions

Torrid dry conditions last year, depleted soil moisture through the summer of 2020, and an autumn with subpar precipitation had left soil moisture in the Sierra Nevada at near record low levels by April. Observations of soil saturation from the SNOTEL network show very low soil moisture throughout last winter and into early spring, with a distinct increase in late March and early April coincident with the drop in mountain snowpack.

<!– /* Font Definitions */ @font-face {font-family:”Cambria Math”; panose-1:2 4 5 3 5 4 6 3 2 4; mso-font-charset:1; mso-generic-font-family:roman; mso-font-pitch:variable; mso-font-signature:0 0 0 0 0 0;} /* Style Definitions */ p.MsoNormal, li.MsoNormal, div.MsoNormal {mso-style-unhide:no; mso-style-qformat:yes; mso-style-parent:””; margin:0in; margin-bottom:.0001pt; line-height:115%; mso-pagination:widow-orphan; font-size:11.0pt; font-family:”Arial”,sans-serif; mso-fareast-font-family:Arial; mso-ansi-language:EN;} .MsoChpDefault {mso-style-type:export-only; mso-default-props:yes; font-family:”Arial”,sans-serif; mso-ascii-font-family:Arial; mso-fareast-font-family:Arial; mso-hansi-font-family:Arial; mso-bidi-font-family:Arial; mso-ansi-language:EN;} .MsoPapDefault {mso-style-type:export-only; line-height:115%;} @page WordSection1 {size:8.5in 11.0in; margin:1.0in 1.0in 1.0in 1.0in; mso-header-margin:.5in; mso-footer-margin:.5in; mso-paper-source:0;} div.WordSection1 {page:WordSection1;} –>
Figure D: Soil saturation data at 8” depth for the Echo Peak SNOTEL station. The black line shows that soils were parched through mid-March when snowmelt commenced. Data source: NRCS.

Likewise, hydrologic models depict a similar fate of soil moisture being very dry throughout much of the winter. This suggests some portion of snowmelt was taken up in soil moisture given the far-from-saturated state of the soils when the 2021 water year began.

Suspect 5: Snow Water Equivalent was overestimated

The 60% snow water equivalent (SWE) numbers on April 1st came from DWR-based snow stations. Measuring  snow over large areas is challenging and even a well-designed operational network might not adequately capture the distribution of snow across complex terrain. So perhaps the snow water volume in the Sierra was less than 60% of long-term average.

The single notable atmospheric river to hit California in 2021 in late January, bringing exceptional precipitation and snowfall to parts of central California and the Central Sierra. Given that it occurred at climatologically the coldest time of the year, it dropped heavy – in some cases record – snowfall in lower elevation mountains. For example, Calaveras Big Tree State Park (4800’ elevation) saw 76.5” of snow, over ¾ of their total for 2021 during this event. For this site, the 3-day snowfall total was the highest in the observational record. This provided a huge bump for snow at lower elevations but left higher elevations in a sizable deficit that expanded thereafter.

Colleagues, including Dan McEvoy, from the Desert Research Institute have been tracking snow droughts (less than usual accumulations of snow). Reductions in mountain snowpack have been sizable in recent years and are expected to increase in the coming decades. The ratio of SWE and cumulative precipitation since Oct 1st shown in the plot below, effectively tracks snow water storage efficiency. Curiously, for a snapshot of these data in late February all sites with SWE values above 85% of median SWE were at lower elevations (<7500’), despite seeing well below normal precipitation. Increased aerial surveying of snow, such as from Airborne Snow Observatories, may improve estimates of SWE beyond those with traditional snow survey and automated snow sensors.

Figure E: Distribution of percent of median snow water equivalent by elevation for SNOTEL stations in the Sierra at 23rd Feb 2021. The colors of the circles depict the percent of precipitation for the water year at each site. Data source: Dan McEvoy.

So, whodunit?

In a classic whodunit, five suspects are in the line-up for the case of the missing flow.

Lackluster April precipitation certainly contributed to the downward revisions in forecasted flow. In addition, the prime suspects of parched soils from 2020 and dry atmospheric conditions, acted together to steal snow from their typical streams. Michael Dettinger finds similar shortfalls in streamflow relative to precipitation since fall of 2019 across California and Nevada. Tessa Maurer and colleagues exploring catchment level feedbacks between ET and subsurface storage found that nonlinearities in the water balance can decouple precipitation and runoff during droughts. They also identify several confounding factors that may limit our understanding of hydrological feedbacks, including the loss of high-elevation runoff, often considered a drought mitigator, from temperature fueled increases in ET at high elevations or lateral redistribution of precipitation excess from higher elevations to unsaturated soil at lower elevations.

Suspects we can manage through forecasting

While we have focused on identifying the prime suspects for the missing streamflow, such scientific detective work may ultimately help improve forecasting accuracy and early seasonal warning systems for managing water. We highlight two fronts to improve forecasts.

First, modeling studies are showing that old rules of thumb are becoming less reliable for anticipating water resources in a changing climate and demand significant updates. Improved understanding of mountain processes that involve snow, soil, and vegetation may help improve forecasts. Previous studies show substantial changes in how water years play out with climate change in California – including more frequent dry and critically dry water years. Back-to-back snow droughts – like we experienced in 2020 and 2021 – are projected to become increasingly likely in the Sierra Nevada with continued warming. Likewise, we expect new types of water years that we have not seen in modern times that will challenge operational water forecasting and allocation decisions. More critical calculations will support better understanding to improve forecasts, allocations, and flexible management.

Second, incremental advances in sub-seasonal to seasonal climate forecasts have potential to inform water forecast and allocation decisions. For example, April forecasts issued in mid-March called for very dry and warm conditions for the state. Incorporation of such climate forecasts into water allocation forecasts may aid decision-making.

Further Reading

Abatzoglou, J. T., McEvoy, D. J. & Redmond, K. T. (2017) The West Wide Drought Tracker: Drought Monitoring at Fine Spatial Scales. Bull. Am. Meteorol. Soc. 98, 1815–1820

He, M., Anderson, J., Lynn, E., Arnold, W. (2021) Projected Changes in Water Year Types and Hydrological Drought in California’s Central Valley in the 21st CenturyClimate9, 26.

Huntington, J. L., Hegewisch, K. C., Daudert, B., Morton, C. G., Abatzoglou, J. T., McEvoy, D. J., & Erickson, T. (2017). Climate Engine: Cloud Computing and Visualization of Climate and Remote Sensing Data for Advanced Natural Resource Monitoring and Process UnderstandingBulletin of the American Meteorological Society98(11), 2397-2410.

Livneh, B., Badger, A.M. (2020). Drought less predictable under declining future snowpack. Nat. Clim. Chang. 10, 452–458

Marshall, A. M., Abatzoglou, J. T., Link, T. E. & Tennant, C. J. (2019). Projected Changes in Interannual Variability of Peak Snowpack Amount and Timing in the Western United States. Geophys. Res. Lett. 46, 8882–8892

Maurer, T., Avanzi, F., Glaser, S. D., and Bales, R. C.: Drivers of drought-induced shifts in the water balance through a Budyko approach, Hydrol. Earth Syst. Sci. Discuss. in review, 2021.

Null, S. E., & Viers, J. H. (2013). In bad waters: Water year classification in nonstationary climates. Water Resources Research, 49(2), 1137–1148.

Williams, A. P. Williams, A.P., Cook, E.R., Smerdon, J.E., Cook, B.I., Abatzoglou, J.T., Bolles, K., Baek, S.H., Badger, A.M. and Livneh, B. (2020). Large contribution from anthropogenic warming to an emerging North American megadrought. Science 368, 314 LP – 318

John Abatzoglou is an associate professor in the Management of Complex Systems Department at UC Merced.  Anna Rallings in a staff scientist and lab manager of the VICElab at UC Merced. Leigh Bernacchi is the program director of the Center for Information Technology Research in the Interest of Society and the Banatao Institute at UC Merced. Josué Medellín-Azuara is the associate director at the UC Davis Center for Watershed Sciences and an associate professor at the Department of Civil and Environmental Engineering at UC Merced. Joshua Viers is Campus Director of the Center for Information Technology Research in the Interest of Society and the Banatao Institute, Associate Dean for Research, School of Engineering at UC Merced and professor in the Department of Civil and Environmental Engineering at UC Merced.

Posted in Uncategorized | Leave a comment

California isn’t running out of water; it’s running out of cheap water

by Wyatt Arnold

A California water myth which becomes especially pernicious in droughts is that California is “running out of water” (Hanak et al. 2009). Viewing California’s supply and demand pressures in terms of fixed water requirements perpetuates this myth and invariably places undue attention on building additional supply infrastructure. Instead, managing water as a scarce resource suggests a balanced portfolio of water trading, investments in conveyance, smart groundwater replenishment, and demand management. With such a balanced portfolio, 1) California’s water supply situation is not broadly dire, and 2) California’s vast and interconnected water infrastructure and groundwater resources can minimize most problems from the state’s highly variable climate.

An economics-driven model of California’s water system, the California Value Integrated Network (CALVIN), has provided such insight from several perspectives, including climate change, groundwater, water markets, and reservoir operations. But in many of these studies, authors lamented an unrealized potential to capture the impact of hydrologic variability more realistically. With perfect foresight, CALVIN was run with complete foreknowledge of 82-years of hydrology – giving exactly optimal solutions to managing reservoir over-year (“carryover”) storage through multi-year droughts, for example. Now, with access to high performance computing resources, a limited foresight carryover storage value function (COSVF) method (Draper 2001) has been applied to California’s entire system – more than 26 surface reservoirs and over 30 groundwater basins (Arnold 2021, Khadem 2018).  These model runs are the most comprehensive and realistic analyses of the potential for broad integrated portfolios of actions across water agencies to address California’s water supply problems.

So, what do these new limited foresight CALVIN results tell us about California’s water supply? Here are three things to get started:

  • From the standpoint of long-term average marginal economic value of water, perfect and limited foresight closely agree; however, limited foresight is more relevant for the risk averse, who prefer to minimize larger but rarer shortages at the cost of average performance.
  • Limited foresight results also suggest that, in general, increasing the storage capacity of reservoirs in California has a very low marginal economic benefit relative to other infrastructure investments like conveyance and groundwater pumping capacities.
  • A large range of carryover storage and conjunctive use operations yield similar statewide economic performance (summing water operation and scarcity costs statewide over 82 years of wet and dry conditions). Consideration of a broad portfolio of conjunctive use, trading, water conservation, and local infrastructure options may not significantly change major surface reservoir operations.

Economics of Carryover Storage

Many reservoirs in California have been drawn down to near record low levels in the current drought. People are alarmed that reservoir storages are so low after only two dry years and are speculating whether prudent decisions were made about storage management. Climate change is making these decisions riskier, yet modeling the historical record remains important as a frame of reference.

Here, I focus on Shasta, Oroville, and off-stream San Luis carryover operations suggested by the new limited foresight optimization results and compare with carryover storage simulated by two versions of the State’s reservoir system model, CalSim-II, to shed light on how and why the system’s carryover storage is so volatile.

Figure 1 shows the time series of carryover operations modeled through two of California’s notorious droughts. The first thing to notice is that total carryover storage varies a lot from year to year for all models. Second, all models tend to quickly draw down carryover storage in dry years – only perfect foresight CALVIN, knowing the exact length and depth of drought in advance, maintains and draws down carryover storage in the final year of drought. Third, CalSim-II simulated carryover storage – both the Water Storage Investment Program run and the recent Delivery Capability Report 2019 run – lie just above the sampled range of limited foresight runs, suggesting near-optimal operations (blue shaded region) during dry years and droughts.

Without a crystal ball, it is not economical to maintain too much carryover or drought surface water storage. The probability of refilling the following year is high, both due to the volume of carryover storage capacity relative to annual runoff and the low year-to-year correlation of annual runoff. Lower carryover storage raises groundwater pumping in the latter year(s) of drought, but also reduces average groundwater use and pumping costs and helps reduce groundwater overdraft (see Figure 2). Also, higher reservoir releases tend to reduce shortages where access to groundwater is limited, which lowers average shortage costs. Sustaining higher carryover volumes (upper end of the limited foresight range and CalSim-II alike) provides more surface supply in the latter year(s) of a drought, which reduces maximum shortage costs; however, the more risk-averse operation raises long-term average costs and the marginal value of groundwater that would eliminate overdraft. Limited foresight modeling (both optimization and CalSim-II) tends to use more groundwater in drier years (Figure 2), which points to groundwater’s importance as a buffer against hydrologic uncertainty.

Figure 1. Total carryover storage (ending September) of Shasta, Oroville, and San Luis as modeled by perfect and limited foresight CALVIN and CalSim-II Delivery Capability Report 2019 (CS-II DCR 2019) and Water Storage Investment Program (CS-II WSIP) historical runs. Drought periods of 1976-77 and 1987-92 are shaded in light tan. The limited foresight range is based on 26 near-optimal statewide solutions.
Figure 2. Total groundwater pumping volume in the Central Valley as modeled by CALVIN for perfect foresight, limited foresight, and with reservoir carryover storage fixed to CalSim-II Water Storage Investment Program historical run outputs.

Other Considerations for Carryover Storage

Water supply is not the sole objective of carryover storage operations. Federal and State operators of Shasta and Oroville reservoirs seek to maintain storage reserves for environmental requirements. For example, Shasta’s carryover storage objectives include maintenance of a cold-water pool to support Salmon habitat in the Sacramento River. Other economic objectives include recreation and hydropower. CalSim-II’s higher carryover storage relative to limited foresight CALVIN are partially attributable to these objectives in addition to Federal and State contractual water supply obligations to Sacramento and Feather River water rights holders. While CALVIN incorporates minimum environmental flow constraints, more complex environmental requirements such as cold-water pool management and some Delta operational constraints are less well represented. Nevertheless, the limited foresight CALVIN results provide a more realistic representation of the economic value of carryover storage in California’s multi-reservoir conjunctive use system.

Concluding Thoughts

Aggressive use of carryover water storage in California’s major reservoirs is economically prudent and reduces overall groundwater reliance. Water supply risks of lower carryover storage are further mitigated through greater system integration such as increased water trading, groundwater banking, and drought water use reductions. The higher risks of having low carryover storage, although not quantified here, appear to fall on California’s stressed ecosystems. A warming climate, expected to continue through at least mid-century even with aggressive global greenhouse gas mitigation, is changing runoff timing, magnitude, and frequency in ways that will make managing carryover storage more challenging. Future work should focus on this aspect and incorporate alternative hydrologic traces reflecting expected climate changes.

Further Reading

Arnold, Wyatt. 2021. The Economic Value of Carryover Storage in California’s Water Supply System with Limited Hydrologic Foresight. [MS, UC Davis]. Available at: https://watershed.ucdavis.edu/shed/lund/students/WyattArnoldThesis2021.pdf

Draper, A. J. 2001. Implicit Stochastic Optimization with Limited Foresight for Reservoir Systems. [PhD, UC Davis]. https://watershed.ucdavis.edu/shed/lund/students/DraperDissertation.pdf

Hanak, Ellen, Jay Lund, Ariel Dinar, Brian Gray, Richard Howitt, Jeffrey Mount, Peter Moyle, et al. 2009. “California Water Myths.” Public Policy Institute of California.

Khadem, M., C. Rougé, J. J. Harou, K. M. Hansen, J. Medellin‐Azuara, and J. R. Lund. 2018. Estimating the Economic Value of Interannual Reservoir Storage in Water Resource Systems.” Water Resources Research 54 (11): 8890–8908.

Wyatt Arnold recently completed a master’s degree in Civil and Environmental Engineering at the University of California – Davis.  He currently works for the California Department of Water Resources in the Climate Adaptation Program.

Data from CalSim-II runs are available on the California Natural Resources Agency OpenData site: 1) Water Storage Investment Program model (1995 Historical Detrended run) available at https://data.cnra.ca.gov/dataset/climate-change-projections-wsip-2030-2070, 2) Delivery Capability Report 2019 run available at: https://data.cnra.ca.gov/dataset/state-water-project-delivery-capability-report-dcr-2019 

Posted in Uncategorized | 6 Comments