When do water bonds pass? Lessons from past elections

By Cassidy Craford and Hannah Safford

Californians cite drought and water-supply challenges as some of the most important environmental issues facing the state today. A whopping 85% of California voters agree that water supply is a “big problem” or “somewhat of a problem” in their region. Population growth, dated infrastructure, and climate change are combining to strain water systems statewide.

The California legislature has long sought to mitigate water stress by passing water bonds that compensate for gaps in water-related funding. Such bonds were relatively modest throughout most of the 20th century rarely exceeding $2 billion (in 2018 dollars) in any given year.[1]

The size of these bonds skyrocketed as California’s water problems became more pressing—and perhaps because raising money through a simple majority vote for a statewide water bond was politically easier than raising money through additional taxes or utility-rate increases. Despite the higher sticker price, the public remained on board. California voters have approved nine state bonds worth a total of $27.1 billion over the past 25 years.

That was the story until late last year. During the November 2018 election, California voters were asked to consider Proposition 3, an $8.9 billion water bond that would fund water-related infrastructure and environmental initiatives throughout the state.  Given California’s record of voter support for water bonds and expressed public concern for water safety and security, Prop 3 seemed likely to pass. But it failed by 0.65%—a slight yet impactful margin.

The obvious question is “Why did Prop 3 fail?” Multiple commentators have suggested answers. But exploring “Where did Prop 3 fail?” provides additional insights. The results are sometimes counter-intuitive…and deepen our understanding of how voters think about water in California.

Conventional wisdom predicts where water bonds should enjoy support

Multiple studies show partisanship affects how individuals vote on (ostensibly nonpartisan) ballot measures. Democrats tend to favor more government spending and greater government intervention than Republicans on nearly all issues—and environmental issues in particular. So we should reasonably expect water bonds to be more popular among Democrats, and more likely to succeed in Democratic-leaning areas.

This has certainly been the case in the past. In 2014, Proposition 1—a $7.1 billion bond to improve water quality, supply, and infrastructure—passed in all but one of California’s 20 most Democratic-leaning counties, but failed in half of California’s 20 most Republican-leaning counties. In 2018, Proposition 68—a $4 billion bond for parks, environment, and water projects —passed in all of California’s 20 most Democratic-leaning counties and failed in all of California’s most Republican-leaning counties.

Voter sentiment towards Prop 3 fell along similarly partisan lines leading up to the November 2018 election. Polling found that in July, 72% of likely Democratic voters were planning to vote “yes” on Prop 3, while only 43% of likely Republican voters said the same. But on Election Day, several counties bucked historical trends. Marin, Mendocino, Solano, and other strongly Democratic counties opposed the measure, while Republican-leaning Tulare and Kings counties supported it.

So if parties didn’t decide the fate of Prop 3, what did? A deep statistical analysis is beyond the scope of this piece, but we can look at four likely factors.

WaterBondResultsFactor 1: Funding priorities

The most obvious explanation is that not all water bonds are the same. Props 1, 68, and 3 had different funding priorities. For example, consider Central Valley voter interests in three measures. Props 1 and 3 earmarked millions of dollars for Central Valley water issues. Prop 1 allocated $34 million to the Tulare/Kern hydrologic region and $2.7 billion in funding for water storage projects (a traditional rural and recent Republican interest). Proponents claimed this funding would augment agricultural water supplies as well as help rural communities update failing wells and meet water quality standards.


Map showing the Friant-Kern Canal and other water infrastructure in Fresno, Kings, and Tulare counties.

Prop 3 proposed doing more. The Friant-Kern Canal sends water from Millerton Lake to San Joaquin Valley, serving approximately 18,000 farms and 160,000 families and businesses along the way. But, overreliance on groundwater has caused serious land subsidence which has damaged the canal and threatens future water availability. Prop 3 would have allocated $750 million to the Tulare/Kings/Fresno area, in part to repair and strengthen the Friant-Kern Canal. The sweeteners Props 1 and 3 offered the Central Valley were enough to get voters in Republican-leaning Tulare and Kings counties to go against party fiscal doctrine and support the bonds. But Prop 68—which did not contain such special measures—was predictably voted down in the region.

Factor 2: Timing

Countless researchers have demonstrated the importance of timing on policy success. This is illustrated by the fate of Prop 3. Not only was Prop 3 the most expensive measure on the November 2018 ballot, it also came quickly after Prop 68 in June 2018. Voters who historically supported water spending may have been turned off by another, larger financial ask arriving so soon (They want more money for water from me already?!). Moreover, the California drought was substantially over in 2018, compared to 2014. Absent an imminent water crisis, voters may have felt no need to support two water bonds in the same year.

Timing also may have affected voter confidence in the institutions that manage state water resources. One notable example comes from the region surrounding Lake Oroville, in light of the February 2017 spillway crisis. A survey of Oroville-area residents in October 2017 found 70% of respondents claimed that the Department of Water Resources had “lost all public trust with the communities downstream from the Oroville Dam.” Area voters who would have supported a $17.8 billion water bond in 2014 may have felt by 2018 that state agencies had proven incapable of handling such a large amount of funding. The Oroville data point could represent public sentiment more broadly.

Factor 3: Path to the ballot

Another problem with Prop 3 was its procedural history. There are two ways a state proposition can be put on the ballot in California. The first is by the state legislature—as was the case with Props 1 and 68. The second is by individual citizens through the “initiative process.” The initiative process allows any registered voter in California to propose a constitutional amendment or state statute, and for that proposition to go before voters on the state ballot as long as the proponent has a minimum number of supporting California voter signatures (typically 5% of the number of votes cast for governor).

One problem with the initiative process is that because California is so big and populous, the resources of at least one major institution or interest group are typically needed to fund the gathering of requisite signatures. The upshot is that “no initiative resulting purely from a volunteer drive has reached the ballot in three years.”

Opponents framed Prop 3 as an egregious example of “pay-to-play” politics that can shape the initiative process. Sierra Club California wrote that Prop 3 “[flew] in the face of good governance by being written behind the scenes.” The Sacramento Bee reported that California Assembly Speaker Anthony Rendon thought that Prop 3 was characterized by “a lack of oversight…and a surplus of special interest projects.” Such criticism may have swayed voters, especially those already turned off by the “insider baseball” side of politics.

Factor 4: The media

Media coverage is important in shaping election results. This is particularly true for the down-ballot policy proposals for which voters tend to have fewer preconceived notions. It is even more true in California, where voters are typically asked to consider more proposals than voters in other states.

Multiple studies have shown that even in the social media era, traditional newspaper endorsements can still significantly affect voter behavior. There is evidence that endorsements carry more weight when they (1) come from small, independent newspapers, (2) oppose public-benefit measures, and (3) concern economically relevant—rather than purely ideological—issues.

The theory suggests that media coverage helped determine who voted for and against Prop 3. The Bakersfield Californian and The Fresno Bee—two smaller newspapers serving Central Valley—both supported Prop 3. The emphasis that these papers placed on the specific regional benefits the bond stood to offer likely did much to win over even fiscally conservative voters in Tulare, Kings, and Fresno counties.

On the other hand, several papers serving Democratic areas strongly opposed Prop 3. The Sacramento Bee opined that Prop 3’s “list of beneficiaries” was “not enough to deserve voters’ support,” while the Marin Independent Journal underscored the problematic way that Prop 3 made it to the ballot. And indeed, neither Marin, Sacramento, nor neighboring Solano county ultimately supported Prop 3. Opposition from other papers—including the Los Angeles Times, the Monterey Herald, the Santa Barbara Independent, the San Francisco Chronicle, and the Mercury News—likely also did much to cut reduce support, even in counties that voted “yes” overall on Prop 3. Among the 20 most Democratic-leaning counties in California, county-level support for Prop 3 was, on average, 11.5% less than support for Prop 68 just a few months earlier.

The takeaway: Californians are willing spend on water…to a point.

California voters almost always say yes to bonds. Statewide bond measures pass around 90% of the time and California’s public has approved billions of dollars of water bonds specifically many times in the past.

But Prop 3’s failure proved that voters are willing to go only so far.  There is no simple way to predict when a voter will resist. Partisan lean is a predictor of the county-level vote on water bonds, but is by no means definitive. Funding priorities, timing, procedural history, and media attention are some of the factors that may also be at play.

Policymakers and bond proponents would do well to consider such factors. Voters have shown that they’re unwilling to just “throw money” at state water problems. To secure California’s water future, we must have policies to secure California’s vote.

Cassidy Craford is pursuing a B.S. in sustainable environmental design at UC Davis and is a student assistant at the UC Davis Policy Institute for Energy, Environment, and the Economy.

Hannah Safford is a Ph.D. student in environmental engineering at UC Davis and a researcher at the UC Davis Policy Institute for Energy, Environment, and the Economy.

[1] The major exception is a massive bond authorized in 1960 to finance construction of the State Water Project. This bond, though, was repaid with water-sale revenues, whereas recent water bonds are being repaid with general-fund revenues.

Posted in Uncategorized | 2 Comments

Portfolio Solutions for Safe Drinking Water – Multiple Barriers

by Jay Lund


Only some parts of the world have safe drinking water almost ubiquitously, and only in the last century.  (We lucky few!)  In these countries, drinking water safety relies on a complex portfolio of actions and accountability by individuals, industries, and diverse layered units of government.  The provision of safe drinking water is another example of portfolio approaches to water management.

Safe drinking water usually occurs through a so-called “multiple barriers” approach, an effective way to structure actions when failures have especially dire consequences.  A multiple barriers approach attacks water contamination at several points between water sources and water users.  Because no single barrier to contamination is perfect, erecting multiple barriers can increase the stability and reliability of drinking water safety at a lower cost.

Table 1: Multiple barriers to waterborne diseases

Multiple-barriers Infrastructure Multiple Accountability
1. Banned/regulated chemicals and activities Local water utility, elected boards
2. Source protection: Rivers, lakes, reservoirs, groundwater Public health agencies
3. Drinking water treatment State regulators
4. Distribution system Federal regulators
5. Public health system Professional societies
Universities, NGOs, media

The left column of Table 1 lists common layers of barriers to prevent drinking water contamination. These wide-ranging activities include:

  1. Banning and regulating risky chemicals and activities. These barriers suppress or regulate potential contaminants in the entire economy and watershed, making them less available to enter drinking water systems.  Some chemicals are banned (e.g., DDT ) from the economy and some are limited or regulated (some pesticides, lead, arsenic, and SO2 air emissions) so they are less likely to exist or escape into watersheds and water sources.
  2. Water source protection. These activities focus on reducing the availability and frequency of contaminants in streams, reservoirs, and groundwater which supply a drinking water system.  These can include upstream wastewater treatment, septic system regulations, erosion management, and managing lake and stream water quality to help keep contaminants from arriving at drinking water system intakes.
  3. Treatment of water entering drinking water infrastructure. Drinking water treatment often includes several layers of treatment from settling (and enhanced settling), filtration (sometimes through several filter types and layers), and one or more forms of water disinfection.
  4. Distribution system. These barriers prevent new contaminants from entering the water distribution system by maintaining positive pipeline pressures, regulating cross-connections, and other measures.  In addition, a residual disinfectant (usually chlorine-based, sometimes with disinfection boosters in large distribution networks) is usually maintained to attack any surviving or entering pathogens in the water distribution system.  Local flushing also is sometimes used to reduce residence times of treated water in parts of the distribution systems.
  5. Public health system. If previous barriers fail, it is important to have a public health system which detects any waterborne disease and investigates to find and correct its origins.  The public health system also can organize and apply medical treatment of any illness not prevented by the drinking water system.

Contaminants often have different properties and likelihoods of being introduced into different water system parts, having managing multiple barriers is more likely to intercept contaminants.  Also, not all contaminants are known and it is usually impossible to monitor for all possible contaminants in a timely way. Overall, having multiple barriers increases general water purity and safety.

Barriers for drinking water safety have advanced tremendously.  From ancient times until the 19th century, only selecting better-quality source water, preventing contamination of stored water, and boiling water just before use were widely available to protect drinking water (Frontinus 97 AD).  Only in recent times have new technologies and institutions greatly improved and diversified contamination barriers for drinking water safety (Tarr 1984; McGuire 1999).

A second side of multiple barrier approaches is the multiple layers and paths of accountability for drinking water safety, summarized in the second column of Table 1.  People and institutions are imperfectly reliable, so having multiple layers of institutions and people, often with interacting legal responsibilities and authorities improves overall system safety.  The result is a diverse and mutually-reinforcing ecosystem of institutions and people with means and responsibilities for drinking water safety.  These groups of people span several levels of government as well as professional groups without official drinking water responsibilities.

In most US states, drinking water safety rests first on local water utilities (mostly with locally-elected governing boards).  Local water utility management is directly accountable to water users for reliability, quality, and cost.  But some aspects of safety are poorly suited for local accountability (such as water toxicity or pathogens not related to water taste, odor, or color) – so state regulators (California’s SWRCB) become responsible for setting state standards, secondary inspections, and follow-up regulation and enforcement of state and federal standards.  State regulators, in turn, are overseen by federal regulators (USEPA) who set and can ultimately enforce national drinking water standards for quality, testing, and reliability.  This system of federal regulation delegated partially and supplemented by state regulators is a feature of the US Safe Drinking Water Act.  Both state and federal regulators ultimately also report to independently-elected officials, accountable to the public.

If these direct hierarchical legal water authorities prove insufficient, a separate hierarchy of local, state, and national public health authorities and professionals can come to bear.  These include local and state health departments, and federal public health and disease control agencies.  These public health authorities report to many of the same elected officials, but can act independently, providing an additional path to water system problem identification, accountability, and action.

If this second set of governmental paths of accountability is insufficient, several unelected professional groups and organizations also sometimes become involved in drinking water safety.  These include water quality-related and medical practitioners and their professional societies, including environmental engineers, science, and health professionals in private and public practice.  In addition, nongovernmental environmental and health organizations, university researchers, and the media also sometimes become involved.  In addition to investigating potential waterborne disease outbreaks, these organizations also help develop drinking water safety regulations and standards.  The lead poisoning failures in Flint, Michigan required several of these deeper forms of back-up organizations and individuals to highlight problems neglected by those with governmental responsibilities for drinking water safety.

This diverse portfolio of institutional and personal responsibilities, technologies, and actions has been quite effective.  Most major drinking water safety failures are averted through normal drinking water system utility, standards, inspection, and regulation processes.  But with tens of thousands of water systems nationally, hundreds (even thousands) of problems remain, mostly in smaller rural systems.

Independent testing at each stage, and across stages, with multiple and substantially independent pathways to identifying and correcting problems have greatly improved drinking water safety.  But drinking water system safety is imperfectable and requires vigilance.  Drinking water safety failures can be hard to detect and correct because some contaminants can incur damages and illness for days to years before people become aware of problems.

The complexity of safe drinking water management provides many people with opportunities, interests, and responsibilities for paying attention and acting effectively, especially in a democracy where many ways exist to bring problems to elected officials and the public.  This process continues, even today in California, for some of our most difficult drinking water safety issues.

Further Reading

AWWA Organisms in Water Committee. Committee Report: Microbiological Considerations for Drinking Water Regulation. J. Am. Water Works Assoc. 1987, 79, 81– 88 DOI: 10.1002/j.1551-8833.1987.tb02848.x

Federal-Provincial-Territorial Committee on Drinking Water (2002), From Source to Tap: The multi-barrier approach to safe drinking water, Canadian Council of Ministers of the Environment.

Lund, J. (2019), “Portfolio Solutions for Water – Flood Management,” 3 March, CaliforniaWaterBlog.com

Lund, J. (2019), “Portfolio Solutions for Water Supply,” 10 March, CaliforniaWaterBlog.com

Lund, J. (2019), “Shared interest in universal safe drinking water,” January 13, CaliforniaWaterBlog.com

McGuire, M.J. (2013), The Chlorine Revolution: The History of Water Disinfection and the Fight to Save Lives, American Water Works Association.

Tarr, JA (1984), “Water and wastes: a retrospective assessment of wastewater technology in the United States, 1800-1932,” Technology and Culture 25 (2), 226-263

Summerscales, I.M and E. A. McBean (2011), ”Incorporation of the Multiple Barrier Approach in drinking water risk assessment tools,” Journal of Water and Health  9.2  2011

Marron, E.L., W.A. Mitch, U. von Gunten, and D.L. Sedlak (2019), “A Tale of Two Treatments: The Multiple Barrier Approach to Removing Chemical Contaminants During Potable Water Reuse,” Chem. Res., 2019, 52 (3), pp 615–622, DOI: 10.1021/acs.accounts.8b00612

White, Gilbert (1966), Alternatives in Water Management, Publication 1408, National Academy of Sciences – National Research Council, Washington, DC, 52pp.

Jay Lund is a Professor of Civil and Environmental Engineering at the University of California, Davis.

This is the third in a series of blog posts on portfolio approached to water management. Flood and water supply management portfolios were summarized previously.


Posted in Uncategorized | Leave a comment

Climate Warming Brings New Water to California’s Delta

Ice GG

First ice berg test for fitting under the Golden Gate Bridge:Photo by Mustafa Dogan

April 1, 2019

By Nestle J. Frobish

The California Department of Water Resources is working to employ the ongoing break-up of the Antarctic ice cap to provide a vast supply of water for California.  Current plans are to employ ocean tugs to bring ice bergs into San Francisco Bay for docking in the State Water Project’s Clifton Court Forebay.  Several propulsion designs are being explored.

The resulting meltwater will provide a salt-free source of water in the south Delta for local and Delta export water users, cold water for Delta Smelt, and a summer winter-sports recreation activity in the southern Delta.  Satisfying the entire roughly 7 maf/year of total Delta export water demand will require roughly 8.6 billion tons of ice berg annually (9.2 cubic kilometers of ice).

“We decided that we needed a new source where we could be the senior appropriator, and use the water to help contribute cold environmental flows in a warmer climate.  In this case, as soon as the ice drops off the Antarctic coast, it is ours to divert,” said Russell Berg of DWR.

“The operating costs are high, but this drought-proof source will mean we don’t need expensive new Delta conveyance,” said Bob Apple of the Southern California’s Metropolitan Water Authority.  Other savings also are likely.  “With this immense new water source, we can stop studying cloud seeding, desalination, and fog capture,” said a spokesman for LBL.

Environmental experts are particularly excited about adding cold water to the Delta.  “Before this solution, we haven’t had a way of keeping parts of the Delta cool enough for Delta Smelt with a warmer climate,” said Allison Ick of UC Davis.  As an adaptation to climate change, rising sea levels will allow deeper ice bergs to be brought through the Golden Gate and into the Delta.

ReSnore the Delta, a formerly-sleepy local environmental group, has raised concerns about the impacts of deepening Delta channels to allow the bergs to be brought into the Delta.   However, the Byron Chamber of Commerce welcomes the prospect of summer-time winter sports for the local economy.  “The southern Delta will be a place for winter sports in the summertime. We already sell ice, bait, and beer for bass fishermen.  Renting ice-axes, crampons, and skates for winter sports will bring additional summer business,” said spokeswoman Bethany Brentwood.

Restore Hetch Hetchy expressed enthusiasm for the idea as a substitute form of cold water storage for the people of San Francisco.  “SFPUC can park pure glacier water at Pier 39, and no longer need Hetch Hetchy reservoir,” said a spokesperson.

Titanic Cruise Lines has raised safety concerns that transporting ice bergs to San Francisco Bay might bring hazards to navigation.

ICE-MAR corporation, the world’s largest cultivator of ice fields, also working to bring smaller ice bergs to just outside of Bakersfield for aquifer recharge.  “Infiltrating another couple of cubic kilometers of ice could end groundwater overdraft in the San Joaquin Valley, and be quite an attraction during our hot summer,” said a company spokesperson.

In the nearer term, calving bergs from glaciers in nearer-by Alaska and British Columbia could be used, although ice bergs being shed at growing rate from Antarctica are seen as a more sustainable supply.

Water from the ice bergs would return to the ocean as Delta outflow, urban wastewater, and precipitation from agricultural evapotranspiration, and so would be unavailable for combatting sea level rise.  “More research is needed,” said Jay Lund of UC Davis.

Nestle J. Frobish, former chairman of the Worldwide Fair Play for Frogs Committee, is Innovation Chair at the UC Davis Center for Climate Adaptation.

Further reading

“The facts about iceberg towing,” https://www.theatlantic.com/technology/archive/2011/08/the-many-failures-and-few-successes-of-zany-iceberg-towing-schemes/243364/

“A Tug on a Frozen Straw,” https://www.fastcompany.com/1755444/watch-tugboat-drag-arctic-iceberg-parched-people-half-world-away-video

“Running ice water,” http://web.mit.edu/12.000/www/m2012/finalwebsite/solution/glaciers.shtml

“Ice for African Development,” https://www.cbsnews.com/news/south-africa-cape-town-water-crisis-plan-drag-iceberg-from-antarctica/

“Ice Bergs for Middle East Peace!,” https://www.huffingtonpost.com.au/2017/05/18/uae-company-wants-to-tow-an-iceberg-from-antarctica-for-drinking_a_22086611/



Posted in Uncategorized | 20 Comments

Groundwater and agriculture: a comparison of managing scarcity and droughts in France and California

By Josselin Rouillard

Overview of French and Californian agricultural groundwater management

France and California face a common challenge of managing overdraft in intensively exploited aquifers. As of 2018, large areas of France and California have overexploited groundwater (see maps below). And both regions have passed landmark groundwater legislation, the Loi sur l’Eau et les Milieux Aquatiques (LEMA) of 2006 in France and the Groundwater Sustainable Management Act (SGMA) of 2014 in California. The LEMA and SGMA both aim to eliminate long-term imbalances between extraction and recharge in priority aquifers. They also both rely on multi-level governance where local stakeholders are given primary responsibilities to define and reach sustainable yield, but state action is possible if local managers do not implement adequate plans to reach sustainability.


Priority basins in France (left; in yellow are priority groundwater basins, in blue are priority surface water basins, in orange are both priority surface and groundwater basins) and California (right). Source: DREAL-DRIEE –Sandre, 2018; PPIC, 2018

France and California have very different physical realities and histories of managing water resources (see table below). With a predominantly Mediterranean climate, California has built a colossal water infrastructure, with 53 billion m3 of surface water storage capacity and large water transfer operations from wetter northern regions to drier southern regions. Few large water transfer schemes exist in France.

Early on, California established a permitting regime for surface water use in 1914, but it never extensively regulated groundwater extraction. The SGMA introduces for the first time local “Groundwater Sustainable Agencies” (GSA) which have extensive powers to implement more management actions in their Groundwater Sustainability Plans (GSP), including for example local licensing schemes to regulate extraction.

Water extraction in France also was not strictly regulated historically. However, the 1992 Water Law included groundwater extraction in its national water licensing regime. Since then, the state has defined priority aquifers where it can enforce bans on drilling new wells. In addition, maximum extractable volume and allocations for each use category (public water supply, agriculture, industry, etc.) are to be defined. LEMA then requires the creation of “Organisme Unique de Gestion Collective” (OUGC) to allocate water among irrigators.

Summary comparison of French and Californian agricultural groundwater management

  France California
Key physical characteristics (annual)
Average annual rainfall 900 mm 530 mm
Total agricultural area 29 million ha 17 million ha
Total irrigated land 2 million ha 4 million ha
Average water extraction for irrigation 3 km3 35 km3
Average groundwater extraction for irrigation 1 km3 10 km3
Number of farm businesses 515 000 80 000
Key legislative and policy characteristics
Predominant water rights Water as national heritage commonly owned Individual water rights defined by common law and permits
Legislative target Good chemical and quantitative status Avoiding six undesirable results
Exemptions On the basis of socio-economic assessments Exemption on pre-2015 impacts
Key organisation OUGC (various legal status) GSA (public)
Key measures set in legislation Extraction licensing, extraction cap and user-based allocation GSAs have extensive powers including setting up caps and allocation system
Other key measures (existing and potential) Water efficiency, off-line reservoirs, crop change and higher value chains Water efficiency, land fallowing, conjunctive use, groundwater artificial recharge, water markets

The rest of this blog post reviews in more depth some similarities and differences in implementing LEMA and SGMA. Corentin Girard provides more background on French water and groundwater management.

Between preventing deterioration and restoring: comparing legislative aims and levels of ambition

In California, SGMA aims to avoid six “undesirable results” of bad groundwater management. Several of these objectives are similar to the European WFD objectives. However, SGMA includes an up-front exemption – authorities are not required to address impacts existing prior to 2015, when the legislation took effect. For example, this could mean that, where past extractions disconnected groundwater from surface water bodies before 2015, groundwater levels may not need to be restored to levels that secure environmental flows in rivers.

The LEMA is France’s implementing act under the European Union Water Framework Directive. The Water Framework Directive requires achieving good chemical and quantitative status for all groundwater bodies (see figure below). For aquifers connected to surface water bodies, European legislation additionally requires that imbalances in aquifers should not impair the ecological and chemical status of surface water bodies nor the integrity of groundwater dependent ecosystems. The Water Framework Directive sets an ambitious goal of restoration towards good status and a presumption towards reconnecting groundwater and surface water systems.

In reality, the Water Framework Directive allows for justified exemptions to good status. Reasons can include overriding public interest, technical feasibility, or disproportionate economic costs. Thus, it is legally possible to maintain low aquifer levels affecting surface water bodies and groundwater dependent ecosystems, but this should be justified on technical and socio-economic grounds.


The objectives of the European Union Water Framework Directive. Source: EEA, 2018

Accounting for ecosystem, society and economic needs in a variable and uncertain environment

Under SGMA, “minimum thresholds” are conditions for a specific parameter, such as aquifer levels, below which impacts become “unreasonable”. These minimum thresholds should not be breached even in dry years. However, “unreasonable” is not defined in the law and, unlike France, no uniform priority system among uses is set in the law. This means that what counts as acceptable impacts from low aquifer levels on e.g., fisheries, wetlands or drinking water supplies may differ between GSAs.

In France, LEMA focuses on structural, chronic water deficits. It requires setting annual maximum extractable volumes, which become the basis for groundwater allocations. Maximum extractable volumes are defined as the annual volume of water available for water uses without affecting minimum environmental flows at least four years out of five on average. Setting annual volumetric targets however does not help deal with imbalances from very dry years or multi-year droughts. So LEMA is complemented with parallel drought management rules. The risk of not reaching minimum aquifer levels or river flows in any one year leads to increasing restrictions on industrial and agricultural water extraction. Minimum aquifer levels and river flows are defined as levels which ensure sufficient water for environmental flows and “priority” uses (e.g., drinking water, national defence infrastructure, fire services).

Experience in France shows the reiterative nature of defining maximum extractable volumes and minimum aquifer levels and river flows. Basins initially lack full quantitative analysis of their groundwater and appropriate statistical and modelling tools. Existing studies are thus currently limited in scope – for example they do not account for impacts of climate change and do not always consider field-tested minimum biological flows.

Expecting these limitations, LEMA requires regular assessments and revisions of water extraction caps. A study led by the French Geological Surface compared the methods used in France. At EU level, technical groups involving Member States are in place, although most work to date has focused on building common methodological frameworks for quality parameters rather quantitative ones.

Between a rock and a hard place: the difficult implementation of reallocation policies

In California, stakeholders have historically resisted using strict groundwater (re)allocation policies to reach groundwater sustainability. Nevertheless, California has famous adjudication cases where water users under the supervision of a judge settled on hard extraction caps and individual allocations. Many stakeholders I met see SGMA as a structured approach to reach allocation agreements, the goal being to avoid costly adjudications while achieving similar outcomes. This is a challenging task as past adjudications have often taken decades to complete. Decisions relied on expensive technical assessments and legal analysis, which are currently absent in most groundwater basins.

The inclination in California to protect individual water rights challenges the possibility of any reallocation policy occurring outside courts. Legal experts have stressed the importance of accounting for the specific combination of existing rights that may exist in each groundwater basin when implementing SGMA.

In France, LEMA requires agricultural irrigation extraction caps. The quantities of water available for irrigated farming in any groundwater basin are shared among irrigators using allocation rules developed by OUGCs. The agricultural sector initially actively resisted the imposition of groundwater extraction caps, resulting in long delays in implementing the law. However, with the creation of OUGC controlled by farming organisations, allocations are now being made. Ongoing research on allocation rules show that farmers not only consider existing allocations and historical groundwater use when realising reallocations under LEMA but also issues of equity, economic efficiency and technical feasibility.  Past research has shown that allocation rules usually reflect different logic and philosophies of social justice.

To date, no allocation decisions in France have been challenged in court. One reason may be a legal basis that emphasizes the common ownership of water (res comunis ominium) and a longer history of groundwater co-management between stakeholders and the state. Nevertheless, it is clear that bargaining, negotiated compromise and pragmatism were necessary in implementing the first groundwater extraction limits. Future conflicts are however likely since further restrictions will be needed to fully align total allocations with extraction caps.

The broader policy mix: designing an innovative combination of supply and demand actions

Historically, California built a large surface water infrastructure network to deal with regional water shortages and drought. However, it is widely recognised that options for increasing supply are limited. Interviews with groundwater managers suggest that “conjunctive use” will be an important tool for SGMA implementation. Conjunctive use aims to optimise the patterns of extractions and storage between surface water and groundwater systems to account for periods of surplus and shortage in each water system. It also implies the wider use of artificial groundwater recharge techniques to restore groundwater stocks in times of surface water surplus and prepare for drought.

The second main tool mentioned during interviews are groundwater markets. As reaching groundwater sustainability will require significant agricultural transformations including land fallowing, some practitioners valued the potential of water markets to reduce costs by favouring higher value farming – a view supported by recent research focusing on the San Joaquim Valley. However, groundwater markets will require an initial allocation of water rights. As mentioned above, this will remain highly contentious.

Water markets do not exist in France and recent research has shown a general reluctance by agricultural users and stakeholders to implement water markets in France. Direct transfer of quotas between farmers is not permitted. Requests from irrigators for additional volume or for a new allocation are examined by the OUGC in charge of reallocating water. The intention is to encourage collective decisions and compromise on how to distribute available water equitably with the objective of securing competitive farming systems and benefiting rural communities at large.

Conjunctive use is not commonly practiced as it requires a level of planning and flexibility between surface water and groundwater pumping that does not yet exist in France. Much attention is on constructing additional surface water storage in the form of offline, medium size reservoirs built outside the riverbed to capture winter run-off and winter aquifer surplus. These projects face significant opposition due to their visual and potential environmental impact.

France has a long history of water saving programs in agriculture incentivised through European and regional subsidies. Measures focus on more efficient irrigation, changes in crop practice, alternative crop varieties, crop diversification and agri-environmental measures. Some schemes aim to plan coordinated changes across a large number of farms to achieve larger water savings across whole catchments and aquifers.

An important concept in French water and rural development policies is the notion of “territoire”, which can best be described as the “common living space” defined by a specific set of environmental, cultural and economic conditions. Water being a common good, its allocation is seen as a tool for the development of the “territoire”. As a result, new planning strategies, called “projet de territoire”, are currently designed in several regions across France. They aim to enhance the coherence between water availability, allocation decisions, crop production systems and agricultural-food value chains.

Some concluding remarks

Groundwater managers in France and California face challenges in transforming complex systems towards sustainability with incomplete or contradictory knowledge, involving many actors and with potentially large economic consequences. The delays in implementing LEMA and ongoing resistance in France show the difficulty of engaging the agricultural sector in these reforms. In such “wicked” situations, an adaptive approach is warranted, which encourages nested experimentation and multiple cycles of implementation and revisions.

To achieve an openness to change, a culture of learning needs to be nurtured amongst GSAs and their constituents. The initial stages of GSP preparation will be fundamental to create a sense of common ownership of the problem, and establish a planning process that encourages negotiated compromises. The ambitious deadline for developing GSPs (e.g. by 2020 in critically overdraft basins) might eventually be the biggest barrier to that learning process if GSAs are tempted to rush through the initial trust-building stages.

The strong focus on local solutions in California provides a favourable ground for experimentation. However, this also runs the risk of very disparate interpretations of SGMA provisions. When implementing the Water Framework Directive, the European Union developed a range of early guidance documents to avoid unfair competition between European countries. In California, early guidance or a common framework may be needed, for example on what counts as “undesirable” and how to set “minimum thresholds”, to guarantee a minimum level playing field between GSAs.

Sustainability cannot be reached with a single instrument, but will require multiple strategies and approaches. France offers an original case of collective groundwater management focused on negotiated, user-driven reallocation decisions. California provides cases of costly but successful adjudicated basins as well as a variety of solutions optimising supply actions and economic instruments. Rather than antagonistic, these different strategies are complementary in reaching sustainable groundwater management. Future work could make more structured comparisons and encourage additional knowledge exchange to inform implementation in both regions.


Dr. Josselin Rouillard works for the French Geological Survey (brgm) on groundwater allocation and agricultural irrigation systems. He is an associate of Ecologic Institute on European water and agriculture policy. More information on his current research project can be found here.


This post arises from the author’s observations and discussions in California funded by the University of Montpellier through the MUSE program and the European Union H2020 Marie Skłodowska-Curie Action under Grant Agreement no. 750553. I thank the many Californian groundwater practitioners, policy-makers and academics who kindly shared their thoughts and experiences on implementing SGMA. 

Further Readings

Arnaud L. (2016) -Estimation des volumes prélevables dans les aquifères à nappe libre: retour d’expériences sur les méthodes utilisées, identification des problèmes rencontrés, recommandations. Rapport final, BRGM/RP-64615-FR. 107 pages, 42ill., 1 annexe.

Babbitt, C., Dooley, DM., Hall, M., Moss, RM, Orth, DL., Sawyers, GW. (2018). Groundwater pumping allocations under California’s sustainable groundwater management Act. Considerations for groundwater sustainability agencies. Environment Defense Fund-NewCurrent Water and Land, LLC. https://www.edf.org/

Babbitt, C., Gibson, K., Sellers, S., Brozović, N., Saracino, A., Hayden, A., Hall, M., Zellmer, S. (2018). The future of groundwater management in California: lessons in sustainable management from across the West. Environmental Defense Fund and Daugherty Water for Food Global Institute at the University of Nebraska.

Blomquist, William. Dividing the waters: governing groundwater in Southern California. ICS Press Institute for Contemporary Studies, 1992.

European Commission: http://ec.europa.eu/environment/water/water-framework/

EEA (European Environment Agency) 2018. European Waters. Assessment of status and pressures 2018. EEA Report 7/2018. ISSN 1977-8449. https://www.eea.europa.eu/publications/state-of-water

Figureau, A.-G., M. Montginoul, and J.-D. Rinaudo (2015), “Policy instruments for decentralized management of agricultural groundwater abstraction: A participatory evaluation,” Ecological Economics, Volume 119, November 2015, Pages 147-157

Hanak, E., Escriva-Bou, A., Gray, B., Green, S., Harter, T., Jezdimirovic, J., Lund, J., Medellín-Azuara, J. Moyle, P., Seavy, N. (2019). Water and the Future of the San Joaquin Valley. Public Policy Institute of California, California.

Langridge, R., Brown, A., Rudestam, K., Conrad, E. (2016). An evaluation of California’s adjudicated groundwater basins.

Nelson, R. (2011). Uncommon innovation: developments in groundwater management planning in California. Woods Institute for the Environment, Stanford University.

Rinaudo JD., Moreau C., Garin P. (2016) Social Justice and Groundwater Allocation in Agriculture: A French Case Study. In: Jakeman A.J., Barreteau O., Hunt R.J., Rinaudo JD., Ross A. (eds) Integrated Groundwater Management. Springer

Rouillard, J. (2019). The role of sectoral policies to restore groundwater balance: a study of European agricultural policies and their impact on irrigation water demand in France. In Sustainable Groundwater Management : a comparative analysis of French and Australian policies and implications to other countries, Rinaudo, JD., Holley, C., Montginoul, M., Barnett, S. Springer, in preparation.

Posted in Uncategorized | Leave a comment

The Collapse of Water Exports – Los Angeles, 1914

by Jay Lund


Collapse of Los Angeles aqueduct pipeline through Antelope Valley from a major flood in February, 1914 (3-months after the aqueduct’s official opening). LADWP historic photo archives.

“In February, 1914, the rainfall in the Mojave Desert region exceeded by nearly fifty per cent in three days the average annual precipitation.

Where the steel siphon crosses Antelope valley at the point of greatest depression, an arroyo or run-off wash indicated that fifteen feet was the extreme width of the flood stream, and the pipe was carried over the wash on concrete piers set just outside the high water lines. The February rain, however, was of the sort known as a cloud-burst, and the flood widened the wash to fifty feet, carried away the concrete piers, and the pipe sagged and broke at a circular seam. The water in the pipe escaped rapidly through the break under a head of 200 feet, and the steel pipe collapsed like an emptied fire hose for nearly two miles of its length. In some places the top of the pipe was forced in by atmospheric pressure to within a few inches of the bottom. The pipe is ten feet in diameter, and the plates are 1/4 and 5/16 of an inch thick. Many engineers pronounced the collapsed pipe a total loss, and advised that it be taken apart, the plates re-rolled and the siphon rebuilt.

The damage was repaired, however, by the simple expedient of turning the water on after the break was mended, relying on the pressure to restore the pipe to circular form. The hydraulic pressure, under gradually increasing head, restored the pipe to its original shape without breaking any of the joints or shearing the rivets, and a month after the collapse the siphon was as good as new. The total cost of repairing the siphon was only $3,000. It would have cost about $250,000 to take it apart and rebuild it” (LABPSC 1916).


LADWP historic photo archives.

Water management and policy has always faced challenges, even unexpected ones following great technical triumphs. California’s water problems have never been easy.

But sometimes challenges require only creative solutions based on fundamental insights and a willingness, occasionally driven by desperation, to venture forth and adapt.


Jay Lund is the Director of the Center for Watershed Sciences and Professor of Civil and Environmental Engineering at the University of California – Davis.  This is a re-posting from May 2016.

Further reading

Complete report on construction of the Los Angeles aqueduct, Los Angeles Board of Public Service Commissioners, Los Angeles, CA 1916. (pp. 20-21)

Water and Power Associates. Construction of the Los Angeles Aqueduct

LADWP historic photo archives

YouTube – Construction of the Owens Valley Project

Posted in California Water | Tagged | 5 Comments

Portfolio Solutions for Water Supply

by Jay Lund

“Water problems in the western United States, when viewed from afar, can seem tantalizingly easy to solve: all we need to do is turn off the fountains at the Bellagio, stop selling hay to China, ban golf, cut down the almond trees, and kill all the lawyers.” – David Owen (2017), Where the Water Goes: Life and Death Along the Colorado River.

Given California’s long dry seasons and tremendous variability in annual rainfall, its water supplies for cities and agriculture are surprisingly reliable and inexpensive.  This reliability has not been easy to achieve and requires constant attention (Lund et al, 2018).  In recent decades this reliability has been due to portfolio approaches employed by California’s most reliable water supply systems.

Water supply portfolios are usually considered somewhat differently than portfolios for flood management.  Water supply portfolio activities, summarized in the table below, are usually divided into water supply activities (which deliver water to users) and activities which manage or lessen demands for water use (including water conservation and water allocation actions).  However, since the time of Frontinus (97 AD), it is clear that successful water supply systems also requires cooperation from many individuals and groups who manage supplies and demands, so today’s taxonomy adds a category of incentives that encourage people involved in a water system to work well together.  (Others will propose different, perhaps better, taxonomies.)

water supply portfolio

Water supplies almost always begin with precipitation in some form.  Rarely, management actions grab additional precipitation by cloud seeding or almost never fog capture.  There is some discussion of modifying watershed to enhance runoff and make more water available to supply. These are prohibitively expensive in almost all practical applications.  Precipitation is the predominant source of water for streams and aquifers, and precipitation varies greatly seasonally and across years.  Even fossil water in aquifers originated from past precipitation.  Wastewater is increasingly thought of as an additional source of water (for reuse).

Water from these sources is rarely at the time and place when people want to use water, so it must be conveyed or stored for use, or to improve the reliability of supplies for water use.  Water is heavy and bulky, so conveyance and storage involve costs and inconvenience.  Many storage and conveyance approaches are available, and they often operate as an integrated system.

Water quality is also vitally important for many uses, so the protection of source water quality is always a concern.  Water quality is often improved with treatment, making unsuitable water suitable for additional uses.  Many forms and contexts of water treatment are available, and has become increasingly prominent.  Some American cities  now treat wastewater for potable reuse.

Water supply systems have many components which must operate well together.  Substantial improvements in costs and reliability often can be achieved by more effective operations.  In California, operation increasingly includes conjunctive use of surface and ground waters.

Water demands also can be managed.  Ideally, water use is reduced and shifted from times and places when the costs of providing additional water are not worth the value of the additional water use.  Usually we seek to reduce or shift water use from less convenient or expensive times and locations.  This is often done with water efficiency actions which modify technology (such as low-flush toilets) to provide equivalent service with less use of water.  At other times, we seek to modify behavior to change water use, such as by shortening showers or watering landscaping less.  Demand management activities can be varied for permanent, hourly, seasonal, or ad hoc reductions in water use to make water deliveries more reliable and economical.  Being able to conserve additional water during drought is a useful asset.  In principle, demand management can and should apply to all demands for water supply.

Everyone is part of and relies on a water supply system and most water systems function only if many people and interests work together.  Customers must pay water bills, maintain their plumbing, not steal or over-use water, maintain water quality, and reduce use more during droughts or other shortages.  Local water utilities and their contractors must safely, effectively and efficiently operate distribution infrastructure.  A host of regional water wholesalers (e.g., MWDSC, SWP, and CVP in California), water sellers, and a variety of service, material, equipment, and operating contractors are essential to most water systems in California.  This need for many people and organizations to work well together requires suitable mutual expectations, inspections, standards, and enforcement of approximate compliance with mutual expectations.  Any water system will collapse without effective incentives to work well together, enforced mutually and by governmental powers.

Portfolio approaches that artfully combine these many elements cannot eliminate conflicts among water users and conflicts across water management purposes (such as among water supply, flood, and ecosystem purposes).  Indeed, portfolio solutions will sometimes cause some new conflicts and trade-offs.  However, water supply portfolio solutions should reduce overall water supply problems and provide greater reliability at less cost and conflict than would likely occur otherwise.  Indeed, adopting portfolio solutions for all major water management purposes would likely reduce conflicts across purposes, as portfolio solutions usually are far more flexible and adaptable.

Portfolio solutions are more complex than simple and less adaptable water supply solutions in the past.  These more complex solutions require more complex institutional arrangements and analysis, using computer modeling, to provide assurances that components will work well together over a range of conditions.  The additional noisiness and controversy from these analysis and negotiations belies the typically greater reliability of portfolio management – it is often the sound of relative transparency and people paying attention.

Effective water supply portfolios also vary with time and conditions.  California’s San Joaquin Valley is going through painful portfolio changes arising from the state-mandated end of groundwater overdraft, increases in environmental flows, and the expansion of profitable tree crops (Hanak et al. 2019).

One last point is the role of portfolios within each sector for making agreements to improve performance across water management purposes.  An example is the agreement for operating Folsom Reservoir outside Sacramento, California for both water supply and flood control.  The dam operator, the US Bureau of Reclamation – mostly concerned with water supply, contracts with a local flood control authority (the Sacramento Area Flood Control Agency) to lower the reservoir more in winter to reduce flood risk, and is compensated for water deliveries lost in years when the lower winter storage results in less water supply being available.

There is a common saying in California water these days that, “There is no silver bullet, only silver buckshot.”  But effective water management is unlikely to result from a shotgun blast of disintegrated actions.

Further Reading

Frontinus, Sextus Julius (97 AD), The Water Supply of the City of Rome.  An 1899 translation by Clemens Herschel was published by the New England Water Works Association (1973)

Ellen Hanak, Alvar Escriva-Bou, Brian Gray, Sarge Green, Thomas Harter, Jelena Jezdimirovic, Jay Lund, Josué Medellín-Azuara, Peter Moyle, and Nathaniel Seavy (2019,” Water and the Future of the San Joaquin Valley, PPIC, San Francisco, CA, February.

Lund, J. (2019), “Portfolio Solutions for Water – Flood Management,” 3 March, CaliforniaWaterBlog.com

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.

Maven’s Notebook, More reliable water supplies for California: Building a diverse regional water supply portfolio.

White, Gilbert (1966), Alternatives in Water Management, Publication 1408, National Academy of Sciences – National Research Council, Washington, DC, 52pp.

Jay Lund is a Professor of Civil and Environmental Engineering at the University of California, Davis.


Posted in Uncategorized | 4 Comments

Portfolio Solutions for Water – Flood Management

by Jay Lund

The tweet below, shows slight (but still frightening) levee overtopping this week on Cache Creek, just north of Woodland, California.  It also illustrates the combined operations of flood preparation and response, with a simultaneous floodplain evacuation order.  Integrating a range of preparations and responses have made the Sacramento Valley much safer from floods.

One often hears, “If only we did X, we would solve this problem.”  Alas, effective solutions are rarely so simple or reliable.  Most robust solutions for problems involve a diverse and complementary portfolio of actions, developed over time.  When a set of diverse actions are carefully crafted to work together, they often provide more effective, adaptable, and reliable performance, at less expense that a single solution.

The Sacramento Valley’s flood management system is a good example where a portfolio of actions has greatly reduced flood damages and deaths, with relatively little management expense and attention in a highly flood-prone region.  This case also illustrates how the many individual flood management options presented in the table can be assembled into a diversified cost-effective strategy involving the many local, state, and federal parties concerned with floods.

flood portfolio

Portfolio strategies usually include actions which work in different ways over different times.  Flood management portfolios usually include actions that prevent flooding (such as levees) complemented by actions that reduce the need for more expensive flood prevention (such as flood evacuations).  Actions which protect areas from flood waters (often structural actions) are distinguished from actions that reduce vulnerability to damage and death if flooding occurs.  Levees, bypasses, and reservoirs are all designed and operated to support each other in the Sacramento Valley to reduce the extent of flooding.  Floodplain land management and flood warnings and evacuations have greatly reduced the property and people exposed to flooding.

Because most floods occur and pass quickly, most flood management is in preparing for floods and flood recovery, rather than in response during actual floods.  As with fire-fighting, elections, and war, flood management time is more than 99% preparation and recovery and less than 1% actions during flood events.  Pre-flood preparations to contain floods (with levees, reservoirs, and bypasses), reduce flood damage potential (with evacuations, building codes, insurance, and floodplain zoning), and prepare for rapid flood operations and evacuations (with education, warnings, and training) are crucial.  Making and coordinating investments, training, and education are all prominent before floods, making urgent flood operations more effective (and less panicked).  Post-flood response also can reduce flood damages and help prepare for the next flood.  There is always a next flood.

Portfolio solutions require overcoming some challenges, however.  Because portfolio solutions involve a range of actions usually controlled by different authorities and groups, they require more social and political organization than single-action silver-bullet solutions.  This can take time, motivation, and leadership to bring together.  But such diversification of responsibility for implementing portfolio can help spread expenses and provide useful perspectives of attention to details and effectiveness.  In flood management, local residents and land owners, local governments, regional governments, and state and federal agencies all specialize in different elements of regional flood management.  This specialization helps lower costs, increase attention to detail, and diversify political support.  This can lead to a mutually-reinforcing ecosystem of institutions that are collectively more effective at attending the problem and innovating than would a single larger bureaucracy.

It took decades for California’s Sacramento Valley to build its current flood management portfolio.  Even so, some flood problems remain.  Small towns and some rural industries remain vulnerable – and there is always vulnerability to bigger floods and infrastructure failures.  There will always be residual flood risks, as flood solutions are never perfect or complete.

Flood problems also change as the economy, population, social expectations, and now climate change.  We now see floodplains and flood bypasses are closely linked to California’s environmental solutions, bringing new objectives in our flood discussions. Portfolio solutions can often better incrementally adapt to change because of their supporting diversified institutional network and operational flexibility.

Successes with water problems (and other areas) often comes from developing or evolving portfolio solutions.  An integrated range of institutions supports this management, mixing the advantages of centralized and decentralized governance and finance to make more effective and adaptable solutions at less expense.  No single person or institution can usually solve such problems.

A series of blog post essays will explore the use and development of portfolio solutions for major water problems.  Successes and challenges will be discussed, as well as the problem of coordinating portfolios of actions across problems – such as managing a common water infrastructure for floods, water supply, and ecosystems – which traditionally have separate solution portfolios.

Further Reading

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

Gilbert F. White (1937), Notes on Flood Protection and Land-Use Planning, Journal of the American Institute of Planners, 3:3, 57-61, DOI: 10.1080/01944363708978728

Independent Forensic Team (2018). “Independent Forensic Team Report: Oroville Dam Spillway Incident”. January 5, (2018).

Jeffres, C.A.; Opperman, J.J.; Moyle, P.B. (2008), Ephemeral floodplain habitats provide best growth conditions for juvenile Chinook salmon in a California river. Environ. Biol. Fishes, 83, 449–458.

Kelley, R. (1989), Battling the Inland Sea; University of California Press: Berkeley, CA, USA.

Lund, J.R. (2012), “Flood Management in California,” Water, Vol. 4, pp. 157-169; doi:10.3390/w4010157.

Jay Lund is a Professor of Civil and Environmental Engineering at the University of California, Davis.


Posted in Uncategorized | 8 Comments

Tough Fish in a Harsh Place: Red Hills Roach

by Peter B. Moyle


Red Hills roach. The upper fish is about 60 mm (2.5 inches) total length. Photo by P. Moyle

Red Hills Roach are small (adults are 60-70 mm in total length) bronzy minnows that live in a challenging environment. They survive in a few small streams that start as seeps in a hot dry landscape, the serpentine outcrops of the Red Hills, at about 1200 ft in elevation (Tuolumne County).  The streams flow through a hot landscape in summer, only lightly shaded, and a pool with more than a foot of water is regarded as deep.  The water of the streams is likely laced with magnesium, iron, and other minerals leached from the serpentine deposits.  Because the land through which the streams flow is of low value, in the past it had been mined, heavily grazed, and run-over by off-road vehicles. The region is now managed by the Bureau of Land Management (BLM) as the Red Hills Recreation Management Area.  Of the 11 varieties of fish that are labeled as roach (Hesperoleucus), the Red Hills roach has the most restricted distribution, so is the most vulnerable to extinction (see California water blog for February 10, 2019; Baumsteiger and Moyle 2019).

My guess is that the Red Hills roach has survived mainly because BLM designated part of their habitat as an Area of Critical Environmental Concern, which was established to protect rare species of plants that mainly grow on serpentine soils. The plants include the purple-flowered California verbena (Verbena californica) that, like the roach, is found only the Red Hills, in association with seeps that feed the stream. By protecting the native plants, the managers also protected the only species of vertebrate that is endemic to a serpentine-dominated landscape: the Red Hills Roach, Hesperoleucus symmetricus serpentinus.  This is a classic endangered species, a small, ordinary-looking fish with a peculiar common name, for a fish, which is followed by a very long, if somewhat poetic, scientific name (try saying it out loud a few times). It is perhaps appropriate that this unusual California fish is associated with the official state rock, serpentine.


Red Hills Roach Habitat, July 15, 2010. Photo by P. Moyle.

For a good description of the Red Hills geology, flora, and fauna see the Wikipedia account. To get a good idea of the habitat of the Red Hills roach watch the video that was filmed and narrated in 2015 by Kit Tyler (5 minutes): https://youtu.be/11kDPGr7hCw

Also read Amber Manfree’s blog account of the search for Red Hills roach (and other roaches) during the last drought.


The Red Hills roach is currently listed as a Fish Species of Special Concern by the California Department of Fish and Game based on its limited distribution (1-2 km of small stream) and small population (<500 adults).  The CDFW account also provides a description of threats to the fish and well as recommended management actions.  https://www.wildlife.ca.gov/Conservation/SSC/Fishes#

Now that the Red Hills roach has a formal scientific name, it should be listed as both a state and federal threatened species, with the extra protection such designations provide.  Moyle et al (2011) rated its status as of High Concern (status score of 2.1 on a scale of 1.0-5.0), while it was rated as “critically vulnerable” to extinction because of climate change by Moyle et al. (2013).

Map of RHR

Map of the Six-bit Gulch watershed, which drains the Red Hills.  The Red Hills roach is found primarily in the upper half of the gulch.  The lower reaches and headwaters are typically dry in summer.  Map by Amber Manfree.

Further reading

Baumsteiger, J. and P. B. Moyle. 2019. A reappraisal of the California Roach/Hitch (Cypriniformes, Cyprinidae, Hesperoleucus/Lavinia) species complex. Zootaxa 4543 (2): 2221-240. https://www.mapress.com/j/zt/article/view/zootaxa.4543.2.3  (available as open-access download).

Moyle, P.B., J. V. E. Katz and R. M. Quiñones.  2011. Rapid decline of California’s native inland fishes: a status assessment.  Biological Conservation 144: 2414-2423.

Moyle, P.B., J. D. Kiernan, P. K. Crain, and R. M. Quiñones. 2013. Climate change vulnerability of native and alien freshwater fishes of California: a systematic assessment approach. PLoS One. http://dx.plos.org/10.1371/journal.pone.0063883

Moyle, P.B., R. M. Quiñones, J.V.E. Katz, and J. Weaver. 2015.  Fish Species of Special Concern in California.  3rd edition.  Sacramento: California Department of Fish and Wildlife. https://www.wildlife.ca.gov/Conservation/SSC/Fishes#

Peter B. Moyle is a UC Davis Professor Emeritus of fish biology and an associate director of the Center for Watershed Sciences.


Left: typical habitat, 2014, photo by M. Ogaz.; Right: Jacob Katz, Red Hills Roach habitat, 2010,photo by P. Moyle.

Posted in Uncategorized | Leave a comment

The sociology of science in environmental management: Reflections on “Fields and Streams”

WaterwaysRestoration-Tools.jpgby Jay Lund

Most readers of this blog are water management wonks who toil in the bureaucracies and professions of water management, the water-industrial complex, so to speak.  We mostly work on technical issues and internal and inter-organizational rules and controversies.  Despite the daily “firefighting” foci of our activities, almost everyone understands something of the organizational and multi-organizational nature of water and environmental management.  But few actively study the larger organizational and methodological sociology of our work.

Some excellent sociology, history, geography, and political science exists on water institutions (such as Maass 1950; Tarr 1984; Walker and Williams 1984; Kelley 1989; Blomquist 1992; Lund 2015).  But there is much less insightful writing on the sociology of the professions and methodological trends (and fads) which often dominate environmental activities and institutions.

The 2013 book Fields and Streams, by Rebecca Lave, is a worthwhile and thought-provoking work on the field of stream restoration, its development as a profession and implementation by agencies and NGOs.  The book particularly traces the development of the “Natural Channel Design” (NCD) approach to stream restoration.  This NCD approach had largely non-academic origins and came to dominate stream restoration in practice, with a set of internally consistent and coherent definitions, procedures, and methods that are sometimes effective in the field.  The approach, its origins, and its mixed results have fed controversy between its largely consulting and agency staff proponents and opposing academic and government researchers.

The book describes the history of the NCD approach and how its growth was driven by a growing regulation-driven market for pragmatic approaches for stream restoration, the ability of non-academic professionals to develop and market coherent pragmatic advice (even when flawed), and the difficulties of more scholarly researchers producing and delivering practical advice.

The book’s story and findings are interesting in themselves, but lead me to some broader reflections on the development and application of technical ideas and knowledge by government agencies with mandated missions:

  1. Agencies which lack strong organized engagement with science can be attracted to whatever seems like conveniently organized science, seems legitimate, makes sense, and is not too expensive. The author’s wonderful history of fin-de-millennium stream restoration seems to mirror recent environmental flow regulation controversies, where a real problem is met with less than coherent academic and professional advice, so agencies and advocates take more convenient advice from earnest environmental professionals and business-seeking firms.
  2. Academics are not the sole source of wisdom and methods used in practice. Practicing engineers and professionals know this well.  Nevertheless, we live in times when academic work can be influential, and sometimes is.
  3. Academic work is more often useful for large practical problems if organized externally. For stream restoration, academics were largely unsuccessful at organizing themselves to shape practice.  Academic ideas tend to be used more in practice when agencies or firms focused on a problem organize and sponsor academic involvement (such as the Manhattan Project of the 1940s, Cold War military applications, some past Federal and state water infrastructure development, water and wastewater treatment technologies under US EPA leadership and regulations, university cooperative extension, underwriters’ laboratories, and the internet).  Indeed, modern engineering academia arose in the 1700s and 1800s from the French State’s need for an academic enterprise to educate its national and regional road and waterway staffs.
  4. So, government agencies that want effective science-based management often must lead in organizing technical and scientific input and insights. This can take many forms, but requires fundamental sustained agency commitment to organized problem solving.
  5. In a world of changing problems and expectations, such as increased regulatory and market demand for stream restoration, it is a persistent struggle for agencies to move from doing what sounds good to doing what works. If institutions are not organized to develop, test, and improve approaches and methods scientifically, they will expend their resources on what sounds like it might work.

Political economic structure always shapes environmental management and environmental professions, even without capitalism and neo-liberalism.  Ancient China’s deforestation (mentioned by Mencius), Mesopotamian soil salinization, invasive species spread by Polynesian and European seafarers, the Soviet Union’s Aral Sea, and India’s air pollution, are a few examples of the human-induced degradation globally throughout history, unrelated to liberalism.  Excellent scholarship on political-economic effects on modern environmental management and professions includes histories of flood control (Kelley 1989) and sewers (Tarr 1984).

The effects of political economy on environmental management is broad and eternal.  Even without neoliberalism, there has always been profit in being a capable courtier to kings.

To me, the great value of Fields and Streams is in highlighting the importance of unfortunately rare sociological studies of the development and use of technical knowledge in complex multi-institution problems.  We mostly blunder through sociological thinking on environmental management.  The book highlights the costs of this blundering in terms of environmental efficacy, distraction and waste of human time and resources, and expansions of controversy for already-hard environmental problems.

Many of our water and environmental management and regulation systems need some fundamental rethinking to more effectively achieve sometimes conflicting objectives in changing times.  This book’s partial anatomy of stream restoration’s water-industrial complex is a thought-provoking start on this difficult and important aspect of contemporary environmental and ecosystem management.

Further reading

Lave, R. (2013), Fields and Streams: Stream Restoration, Neoliberalism, and the Future of Environmental Science, University of Georgia Press, Athens, GA, 170 pp.

Two useful reviews of Lave, R. (2013), Fields and Streams: https://www.tandfonline.com/doi/full/10.1080/2325548X.2017.1366833, https://agupubs.onlinelibrary.wiley.com/doi/pdf/10.1002/2013EO090013

Blomquist, W. (1992), Dividing the Waters: Governing Groundwater in Southern California. San Francisco: ICS Press.

Kelley, Robert (1989), Battling the Inland Sea: Floods, Public Policy, and the Sacramento Valley, University of California Press, Berkley, CA.

Lund, J.R., “Integrating social and physical sciences in water management,” Water Resources Research, Volume 51, Issue 8, Pages 5905–5918, August 2015.

Maass (1950), Muddy Waters, The Army Engineers and the Nation’s Rivers, Harvard University Press, Cambridge, MA.

Tarr, Joel A. (1984), “Water and Wastes: A Retrospective Assessment of Wastewater Technology in the U.S., 1800-1932,” Technology and Culture, Vol. 25, No. 2 (April), pp. 226-263.

Walker, R.A. and M.J. Williams (1982), “Water from Power: Water Supply and Regional Growth in the Santa Clara Valley,” Economic Geography, Vol. 58, No. 2 (April), pp. 95-119.

Jay Lund is a professor of Civil and Environmental Engineering at the University of California Davis, who has never recovered from his exciting graduate school foray as a social geographer.

Posted in Uncategorized | 1 Comment

Roaches of California: Hidden Biodiversity in a Native Minnow

by Peter B. Moyle



If you inspect small streams in northern California, including those that seem too small or warm for any fish, you will often see minnows swimming in the clear water. Chances are you are seeing a very distinctive native Californian, usually called California roach.  This fish is a complex of species that occurs as far north as Oregon tributaries to Goose Lake and is widespread in tributaries to the Sacramento and San Joaquin rivers, as well as in rivers along the coast from the Eel River to Monterey.

“California Roach” is the name originally given to some minnows collected in 1854 from the San Joaquin River.  When the great Stanford ichthyologists David Starr Jordan and Barton Warren Evermann put these fish into their grand monograph Fishes of North and Middle America, they decided it looked like the roach (Rutilus rutilus), a common minnow in England and Europe.  They then gave it the scientific name Rutilus symmetricus.  While the relationship to European roach was dismissed by John O. Snyder in 1913, the unfortunate common name of “roach” stuck.  Snyder placed California Roach in its own genus, Hesperoleucus, and divided it into six species, based on body shape and counts of fin rays and scales (see Table).  His species were also based on the isolation of their home waters from other watersheds, which would prevent interbreeding.

Because roaches are small inconspicuous fishes, little formal attention was paid to their taxonomy (or status).  By the 1950s, there seemed to be a general consensus that Snyder’s species were at best subspecies and the California roach was back to one species.  This was reflected in the classification presented in my 2002 book, Inland Fishes of California, although the species was divided into eight subspecies.   Then, Andres Aquilar and Joe Jones (2009) looked at populations that were part of this ‘species complex’ using mitochondrial and nuclear DNA. Their analysis indicated that two of Snyder’s species, northern roach and Gualala roach, were strongly supported as ‘good’ species.  The other six subspecies I listed in 2002 were at least supported as distinct genetic units by their analysis.

To clarify the relationships among the species more firmly, new techniques in genomics were brought to play.  This effort was led by Jason Baumsteiger, a postdoctoral scholar at the Center for Watershed Sciences and in the genomics laboratory of Mike Miller.  He performed restriction-site associated DNA (RAD) sequencing on roach samples collected throughout California to discover and genotype thousands of single nucleotide polymorphism (SNPs) (see Baumsteiger et al.  2017). This detailed examination of the genomes of roaches from throughout their range allowed determination of how much each population had diverged from other populations.  Among other things, it allowed for ‘rules’ to determine which populations were species, subspecies, or distinct population segments.

Distinct population segment (DPS) designations are based on the use of DPS designations under the national Endangered Species Act; they are isolated populations that are distinctive, but not quite different enough so to be called species or subspecies. DPS designations are widely used for determining whether or not salmon and steelhead populations are eligible for protection under the ESA.

The application of genomics to the taxonomic relationships of roach populations (Baumsteiger and Moyle 2019) resulted in our recognition of five species, four subspecies, and 5 distinct population segments (Table 1). The five species each have distinctive, interesting features.

The California roach is the most widespread species, historically found in streams throughout the Central Valley, with many opportunities for adaptation to local conditions, such as those found in the Kaweah River (hence the Kaweah roach DPS). It appears to be losing these locally-adapted populations rapidly, however, as they become increasingly isolated by dams and damage to streams, and by invasions of their small stream refuges by green sunfish and other non-native predators.

The Clear Lake roach is a bit of mystery because it a perfect hybrid between coastal roach and California roach.  This fits the geologic history of the region, which has been alternately connected to the Russian River and to the Sacramento River. Presumably representatives from both watersheds made it into the Clear Lake basin at times and hybridized.  The hybrid was apparently superior to either parent species in its ability to persist in streams tributary to Clear Lake.  Today, the Clear Lake roach is more isolated than ever, because the lake is full of non-native predatory fishes.

Hybridization also has led to the development of new species in the northern roach.  This roach inhabits small streams and springs of the upper Pit River basin and looks like other roach species.  So we were surprised when the genomics study showed that about 80% of the genome was like that of the hitch, a related species in a different genus (Lavinia exilicauda).  This seems to have been from an ancient hybridization, perhaps when Sacramento Valley fishes invaded the Pit River region thousands of years ago. Curiously, we also found that the roach-like fish abundant in Hetch-Hetchy Reservoir, on the upper Tuolumne River, also are hitch-roach hybrids even though they were introduced into the reservoir by persons unknown.

The southern coastal roach is also known to hybridize with hitch, where the two species occur together naturally, but these hybrids seem unimportant to the populations of both species. The presence of subspecies and DPSs in the coastal roach distribution reflects the isolation of coastal watersheds from one another with enough connections in the past to keep populations from differentiating enough to be labeled species.  This also makes the Gualala roach a bit of an anomaly, given that watersheds on both sides of the Gualala River contain coastal roach.   The northern coastal roach also shows how rapidly a species can spread when introduced into new watershed, in this case the Eel River. These roach, probably introduced in the 1960s, now occupy most of the accessible habitat in the Eel, one of California’s largest watersheds; the genomic study indicates that they came from fish in the Russian River roach DPS, just to the south, so were pre-adapted for conditions in the Eel River.

This study of small fishes demonstrates again the high endemism in fishes that are adapted to the special, often harsh, conditions in California streams.  This surprising diversity is another example of what makes California special and needing of a well-supported, state-wide conservation strategy. The roach species complex is also good example of hidden biodiversity revealed by new genetic techniques.  Modern genomics can support conventional taxonomic methods to designate species, subspecies, and DPSs and should improve our ability to conserve California’s richness of fishes.


Northern roach. Photo by Stewart Reid


Common name Scientific name Snyder 1913 Moyle 2002 Notes
California Roach H. symmetricus H. symmetricus H. symmetricus Name applied to all roach by Moyle 2002 and others
Red Hills Roach H. s. serpentinus H. s. subsp. Serpentine endemic; Tuolumne County
Central California Roach H. s. symmetricus H. symmetricus H. s. symmetricus Tributaries to Central Valley
Kaweah  Roach H. s. symmetricus H. s. symmetricus DPS, Kaweah River
Clear Lake Roach H. symmetricus x venustus H. s. subsp. Hybrid that behaves like a full species; tribs. to Clear Lake
Coastal Roach H. venustus Originally multiple species/subspecies
Northern Coastal  Roach H. venustus navarroensis Introduced into Eel River.
Russian River Roach H. venustus navarroensis Lumped with Clear Lake Roach DPS, introduced into Eel River
Navarro Roach H. venustus navarroensis H. navarroensis H. s. navarroensis DPS, Navarro R.
Southern Coastal Roach H. venustus subditus
Tomales Roach H. venustus subditus H. s. subsp. DPS, Tomales Bay streams
Monterey Roach H. venustus subditus H. subditus H. s. subditus DPS, Salinas-Pajaro watersheds
Northern Roach H. mitrulus H. mitrulus H. s. mitrulus Pit River; originated as hybrid with Hitch.
Gualala Roach H. parvipinnis H. parvipinnis H. s.  parvipinnis Gualala River

Further readings

Baumsteiger, J. and P. B. Moyle. 2019. A reappraisal of the California Roach/Hitch (Cypriniformes, Cyprinidae, Hesperoleucus/Lavinia) species complex. Zootaxa 4543 (2): 2221-240. https://www.mapress.com/j/zt/article/view/zootaxa.4543.2.3  (available as open-access download)

Baumsteiger, J., P. B. Moyle, A. Aguilar, S. M. O’Rourke, and M. R. Miller. 2017. Genomics clarifies taxonomic boundaries in a difficult species complex. PLoS ONE 12(12): e0189417. https://doi.org/10.1371/journal.pone.0189417 (available as open access download)

Moyle, P.B. 2002. Inland Fishes of California.  University of California Press, Berkeley.

Peter B. Moyle is a UC Davis Professor Emeritus of fish biology and an associate director of the Center for Watershed Sciences.


Classic California roach habitat.  Dye Creek, Tehama County, July 2014

Posted in Uncategorized | 3 Comments