Murky Waters: the mixed legacy of water policies, infrastructure, and law in California

As the United States acquired its Western lands between 1803 (Louisiana Purchase) and 1898 (the annexation of Hawaii), it increased the nation’s climate and topographic heterogeneity. While there are many different climate types in the American West, approximately 40% of the Western region is classified as arid or semi-arid, which is very different from the eastern side of the 100th meridian divide. Aridity is essentially a measure of dryness expressed as a ratio of precipitation to evapotranspiration (i.e., water that leaves the surface of the soil or plants). The driest areas are mostly found in the Southwest deserts and the interior West plains while the wettest are found at high elevations and the northwest. In addition to greater aridity, Western as opposed to eastern climate is characterized by more seasonality (little or no rain in the summer) and higher variance between rainy and dry years. All of this means that settlers migrating from the east to the West had to deal with scarcer and more variable water resources across both time and space.

For centuries before the arrival of the Europeans, Native American tribes dealt with this variability by moving their settlements when conditions worsened — what we now call strategic retreat. However, the US government in this expansion period was looking to populate and develop its Western lands permanently to strengthen and secure their claims on them, which meant building water delivery and storage systems that would solve the problems of distribution over different climate zones and during longer time intervals between precipitation events in the American West. Initially, Western infrastructure efforts were predominantly privately financed, but the urgency and expense of creating the infrastructure for extensive growth development during the late nineteenth and early twentieth centuries required the Federal government’s financial resources and expertise.

The Corps of Engineers was formed in 1802 followed by the Department of Interior in 1849, the Department of Agriculture in 1862 and the Bureau of Reclamation in 1902. These agencies played a major role in developing the American West both before and then after World War 2, when there was a second wave of government induced by the Korean, Vietnam, and Cold War conflicts. Federal assistance for building Western infrastructure generally (e.g. highways, dams, reservoirs, railroads, etc.) was most forthcoming during periods of territorial expansion or military mobilization and then waned by the 1970s when the frontier had long been closed and Pacific Ocean security threats abated.

Water is essential for human consumption of course, but also for agriculture, mining, and energy. While artists like John Muir and explorers like John Wesley Powell who visited the West appreciated the beauty of the region’s lakes, waterfalls, forests, and wildlife, the Federal government was primarily intent on populating and developing the land. Massive amounts of water were used for hydraulic gold mining until the practice was finally banned in 1884. Large corporate farms growing crops like cotton and alfalfa flourished with government subsidized water and state enforced water rights. The premise underlying this activity was that land had value only after it was mixed with labor. This Age of Reason assumption was still deeply enmeshed in America’s political culture in the nineteenth century and undergirded the country’s expansionist logic.

The assignment of water rights, which will be discussed at greater length later, was a key element in a broader strategy to incentivize natural resource and economic development in the newly acquired Western region. To induce people to move into water scarce and more dangerous lands, the government gave ownership rights to the land and the resources on it through a series of bills such as the Homestead Act (1862), the Timber Culture Act (1873), and the Desert Land Act (1877). Water cannot be owned in the same way as minerals, trees, or plants: it can only be used but not possessed for the lifetime of a water molecule, and new water cannot be grown since the amount of water on earth is fixed minus some atmospheric escape. Water has multiple uses: it supports fish and other wildlife and is essential to energy production and other industrial activities. And unlike electrons, water molecules can be aesthetically pleasing. Water’s multiple attributes lead to multiple interests and perspectives, all of which eventually percolates into politics and conflict. In the words of Mark Twain (purportedly), “whiskey is for drinking, water is for fighting.”

This highly utilitarian view of land and resources placed less emphasis on water’s intrinsic environmental value. Early US environmentalists like John Muir staked an opposing view to this position. Small diversions from rivers, lakes, and streams for commercial purposes or human consumption do not typically cause much environmental damage. As the Western regional population grew and industrial processes became more advanced in the late nineteenth and first half of the twentieth centuries, water was increasingly stored in large dams and reservoirs. Filling in canyons with massive bodies of stored water destroyed the existing ecosystem. The legacy of wanting to use water to support settlement and commerce created serious tensions with environmental advocates and non-profit organizations that persist to this day.

Consider the case of the Hetch-Hetchy Dam. The construction of the Hetch-Hetchy dam was a harbinger of two contemporary controversies: one over the merits of different water uses and the other over so-called gray infrastructure like dams and reservoirs. Even before the highly destructive 1906 earthquake, San Francisco leaders had qualms about the adequacy of ground and surface water supplies in and near San Francisco and were searching for augmented supplies in the Sierras. The shady land acquisition dealings and extortionist pricing of the private local water companies increased public support for public ownership of the water system. The fires that followed the 1906 earthquake heightened the importance of ensuring an adequate supply of water for protecting and serving the rebuilding city.

After considering and discarding various alternatives, the San Francisco Public Utilities Commission [SFPUC] chose the Hetch-Hetchy valley, which was located inside the boundaries of the recently designated Yosemite National Park (1890). The Hetch-Hetchy valley had many advocates for both preservation and recreational uses. Camping and hiking had become very popular. While some wanted the Hetch-Hetchy Valley preserved for its pure wilderness value, others wanted to promote outdoor recreation, including the Sierra Club which derived revenue from the camping fees. This three-way tension between preservation, conservation (i.e., balancing preservation with utility), and maximum utility (i.e., filling a valley with water that would be transported to the Bay area for consumption and commercial use) has persisted to this day. The movement to restore Hetch-Hetchy is alive and well, but to date it has neither persuaded a majority of the SF voters nor the courts.

With the supporting advocacy of several other local Bay Area communities, SF leaders lobbied the Wilson administration and Congress, finally securing passage of the Raker Act in 1913. Even without the formidable obstacles of modern day permitting, it would take until 1923 to complete the construction of the dam itself and another 11 years before the system of tunnels, pipes and reservoirs could deliver water to SF. The engineer they hired, MM O’Shaughnessy, had extensive experience with building tunnels, bridges, reservoirs and rail lines and overseeing other municipal projects. The dam initially had what most people thought at the time was more than enough capacity for San Francisco but was subsequently raised from 227 to 312 ft, enough to accommodate many other Bay Area communities as well. The project’s ambition was apparently partly inspired by the grandeur of the Roman Empire water system. In retrospect, the decision to build beyond what San Francisco County needed contributed to the rapid population growth around the Bay. Many of the 26 water agencies in the system today get their water primarily from the Hetch-Hetchy.

Whether the State Water project that was built in segments between 1961 and 1997 to or the huge San Pedro Reservoir in 1971 would or could have eventually provided the necessary water for the Bay Area is a subject of ongoing dispute. The alternatives would likely have been more energy dependent and expensive (i.e., less sustainable) in the long run. Hetch-Hetchy water does not require as much cleaning as most stored reservoir water requires because its reservoir is enveloped in a granite canyon within the boundaries of protected Federal parkland. Moreover, unlike the State Water system that uses massive pumps and huge amounts of energy to convey the water over the Tehachapi mountains, the Sierra water flows to the Bay area mostly by gravity. The Hetch-Hetchy system also produces hydroelectric power, a form of clean energy that is even more valued now than it was then because of the need to reduce carbon emissions. The Hetch-Hetchy legacy is a mixed one: a sustainable exemplar on the one hand and despoiler of natural beauty and habitat on the other.

Leaving aside for the moment the trade-off between water for utilization versus water as an environmental good, the legacy of gray infrastructure left modern Westerners several contentious issues. The Western water infrastructure system is varied and complex. The critical components are very large. According to the National Inventory of Dams, there are 19,700 “large” dams in the US West (defined as over 8 m in height with storage capacity at or above 16.2-acre feet). The 10 largest reservoirs in the United States are in states that are located entirely or at least partially to the West of the 100th meridian. The scale of water infrastructure has had socio-economic repercussions. The Hetch-Hetchy dam fostered large scale population and commercial growth throughout the Bay Area counties in the twentieth century. The population of the seven Colorado River basin states, enabled by the system of dams and reservoirs along the river — including the largest reservoir in the United States, Lake Mead — grew to 10 times its original size between 1920 and 2020.

Despite the formidable size and solidity of these structures, they are vulnerable in various ways. They can fail structurally, although infrequently. One estimate suggests that there have been 2,543 dam failures in the United States for a very low annual rate of 0.00045. The rate is even lower for concrete as opposed to wooden or earthen dams (Dupuis, 2025). Nonetheless, dams in general become less safe as they age, and the big dams require large sums of money to be maintained and repaired. This point was demonstrated recently by the failure of the Oroville Dam emergency spillway. Working around the clock, it took a year and half and 1.1 billion dollars to repair the damage (California Department of Water Resources, 2018). In addition, the storage capacity of reservoirs behind the dams can diminish with the accumulation of silt in them. Removing the silt can be a complex and expensive undertaking with dredging costs typically ranging from 8 to 20 dollars per cubic yard.

A dam’s design hinges on two key assumptions. One is about expected future demand due to population growth. As noted already, the original dam was heightened in 1938, enough to accommodate not just San Francisco but also other communities in the Bay Area. The other assumption is about the climate conditions that support the water supply. In the case of the Hetch-Hetchy system, the implicit climate premises were that the snowpack would accumulate in the mountains during the winter and the snowmelt would flow to the reservoirs in the Spring (i.e., a timing assumption).

Global warming can affect both assumptions. A rise in the average temperature can result in insufficient snowpack in the mountains, especially during extended periods of drought. That alters both the supply of water for consumption and for hydroelectric power (i.e. a too little water problem). In the rainy years, larger winter flows followed by lower spring flows can strain the capacity to capture and regulate the flow optimally, (i.e. a too much water problem). Other related problems include more extreme flooding and wildfires, both of which can exacerbate sediment and water quality issues. In short, legacy water infrastructure struggles to meet modern climate expectations.

Fixing these problems and replacing existing dam and reservoir infrastructure is not just expensive but complicated by the modern political ambivalence towards gray infrastructure. Environmental groups would like to see more dams removed to restore natural habitat and traditional fish migrations. Some dams are too filled with silt to offer much storage capacity. Consequently, there have been at least 2,025 dams removed since 1912, 11% of them large dams. In the Bay Area, within the span of 12 months (2024–2025), expansion projects at three major reservoirs — Vaqueros in Contra Costa County (Cronin, 2024), Pacheco south of San Jose (Valley Water, 2025), and Nicasio in Marin County (Wu, 2025) — have been called off as approval processes dragged on and costs ballooned.

While there are ways to store water that are less environmentally harmful such as by replenishing aquifers or recycling used or stormwater water to reduce surface water demand, it must be done at comparably large scale to replace what the legacy gray infrastructure now provides. There have been remarkable achievements in water efficiency in recent years, but that has been offset to some degree by increased and hardened demand due to continued economic and residential development in the American West. This is most vividly demonstrated by the urban growth in the upper basin of the Colorado River. Strategic retreat from water scarcity is not a politically viable option. Limiting further encroachment on water deficient lands is a more realistic option, but by no means easy to accomplish politically.

Aquifers in the West have been used traditionally as the backup system by many farmers during drought years when surface water supplies were low. Some of those aquifers can be recharged, but farmers must be incentivized to undertake the expense of doing so. The soil in some aquifers compacts after the water is withdrawn. The soil conditions for potential new aquifer storage are very specific with respect to porosity and the absence of natural contaminants. Even when the land is appropriate for storing water by these criteria, new underground storage in many instances still requires conveying groundwater from it where it is stored to where it is needed, which means new piping. With enough money and political will, these problems are solvable. But these conditions have shifted the calculus of water policy now as compared to the nineteenth and early twentieth centuries. When the West was expanding, water infrastructure led population and commercial growth. Now to a greater degree, existing growth defines the demand for water infrastructure.

Another critical aspect of the earlier water infrastructure legacy in the West is the shift in the politics that initially sustained it to the current more complex situation. The most expansive period of large gray water infrastructure construction was led by the Federal government in the mid-twentieth century for reasons of security and economic development. Absent a salient national security threat and the closing of the frontier, it is unclear whether the Federal government will step up in the same way in the modern era.

The most ambitious era of large water infrastructure was between 1900 and 1970. This list of achievements in this period is impressive: in California alone, the Owens River Aqueduct (1913), O’Shaughnessy Dam (1923), Hoover dam (1931), the Central Valley Project (1935), Colorado Aqueduct (1940), Shasta Dam 1944, Delta Mendota Canal (1951), State Water Project (1960), Trinity Dam (1962), California Aqueduct (1966), and Oroville Dam (1968). The immediate postwar World War 2 period in California — referred to as the Pat Brown era — brought aggressive pro-growth policies built around water infrastructure and higher education expansion that were meant to attract new residents.

However, by 1970, a countermovement aimed to slow urban growth that was contributing to increased urban traffic congestion, smog, and environmental degradation. This curbed the state’s rationale for infrastructure-led growth although not growth per se. This has led to the rate of new water infrastructure lagging population growth and commercial development. Also responding to the rising environmental concerns, Federal legislation such as the National Environmental Policy Act (1969) strengthened and lengthened the environmental review of Federal projects while similar state measures such as California’s CEQA did the same for state and local projects. The Federal government’s Clean Water Act (1972) and Endangered Species Act (1973) plus additional new state environmental laws such as California Wild and Scenic Rivers Act (1972) put water infrastructure projects under stricter procedural requirements and increased the complexity and delay of permitting times. Modern environmental concerns about the harmful impact of creating large dams and reservoirs reinforced a growing reluctance in Northern California to allow its water to be diverted to the south to enable yet more growth there. This led to the rejection of the peripheral canal in 1982 and subsequent squabbling over a Bay Delta settlement. The gray versus green water infrastructure debate has grown to more prominence ever the last 50 years.

The large gray infrastructure system relied heavily on state or federal funding for initial building. It also created subsidies for agricultural users that hardened the demand for both the water and the gray delivery system infrastructure. It also created disparities between what residential customers and farmers pay for water, which could create considerable tension if climate change worsens. The San Diego County Water Authority, for instance, charges $1,929 per acre foot for treated water to its retail customers while farmers in the Imperial Valley pay as little as $20 per acre foot for Colorado water. The cost of desalinated water is as much as $2,367 per acre foot. The Imperial Irrigation District gets 80% of the Colorado River water delivered to California. Historically, the costs to the region’s farmers are low because the storage and delivery of water was heavily subsidized by the Federal government. Even more subsidies are now needed to induce farmers with treaty water rights to conserve their water use. Most recently, to increase water conservation in the lower Colorado River basin, farmers have been offered general financial incentives to conserve water and leave it in the ground (Zulauf, 2023).

Subsidies, low and easily forgiven loans, and federally covered crop insurance have encouraged persistent inefficient practices along the Colorado River such as growing low value crops like alfalfa or crops that could be grown in the American south where water is more abundant and climate less arid. The path dependency of growing cotton in dry hot places like Arizona where evapotranspiration and aridity rates are high is illustrative. During the civil war, the blockade of southern cotton incentivized farmers in Arizona to grow cotton for the European market. New strains of cotton developed in North Africa, subsidies and ample groundwater encouraged farmers to get into the cotton growing business. The demand for cotton was later boosted during World War 1 when it was used for truck tires and airplane wings. By the 1950s, southwestern cotton farmers were able to produce twice as much cotton as the rest of the south because of water irrigation advantage. In doing so southwest farmers use two to four times more water per acre than their competitors, an inefficiency that will likely worsen as droughts become more prolonged due to global warming (Lustgarten et al., 2015).

Residential and commercial customers are hurt by this entrenched arrangement for agriculture. Squeezed by the expense of paying farmers for a portion of their water and the cost of finding alternative sources, the more heavily populated coastal counties in California have increasingly relied on recycled and desalinated water. Not only is the treatment and delivery of that water expensive, as mentioned earlier, but the fixed charges associated with it are essentially a regressive tax on consumers, which impacts lower income communities especially. The legacy water infrastructure system cannot be cheaply replaced or easily fixed. Something will have to give in the future as the votes and the money are ultimately not on the farmers’ side.

The ability of California landowners to farm the state’s highly fertile but dry land depends not just on its highly visible, gray water storage and conveyance infrastructure but also on its hidden underground water resources. In addition to a legacy of reservoirs, dams, and canals heavily subsidized by federal and state governments, there is a legacy of above and below ground water rights created to encourage the expansion, settlement and economic development of the American West. Water rights in California are based on both constitutional and statutory authority. Article X Section “A Legacy of Large Gray Infrastructure” was added as an amendment to the California Constitution in 1928. It established the usufructuary foundation of the state’s water system: i.e., that water is not owned by any individuals or entities, but there are rights to use it (Attwater and Markle, 1988, p. 979). The Constitution provides the overriding doctrine of reasonable and beneficial use, and statutory law under the Water Commission Act of 1914 sets out the permitting system for surface water appropriations (i.e., water uses for those who do not have riparian rights by owning property that adjoins a body of water). Appropriative rights are divided into pre- and post-1914 rights, that determine the level of administrative oversight and priority given to water allocations in a drought (Attwater and Markle, 1988, p. 972). Many farmers to this day retain senior water rights and access to extensive reservoirs and canals granted to them long ago.

When the surface water supply becomes insufficient to support the normal level of agricultural demand due to extended droughts, farmers in the American West depend on the groundwater underneath their lands. The endemic uncertainty associated with Western surface water deliveries has increased over time due to both global warming (hence more frequent and severe droughts) and to environmental restrictions that mandate water delivery cutbacks to preserve water ecosystems. Across the state’s agricultural regions, farmers have made up for emergency surface water cutbacks through more groundwater pumping. Since the landmark case of Katz v. Wilkinshaw in 1903, the right of an overlying landowner to pump groundwater has been sacrosanct (Leahy, 2016, p. 6). This meant that for many decades, groundwater pumping was largely unrestrained — and free — in most agricultural regions of the state. Finally, in 2014, California established statewide standards for managing this critical resource despite staunch resistance and alarmingly large groundwater overdrafts resulting in dry wells and substantial land subsidence.

As already noted, climate change has made groundwater storage an even more critical resource. Thanks to “climate whiplash” — i.e., rainfall and temperature patterns that are extreme even in a region that has long been accustomed to droughts and floods — California, like many Western states, has a chronic water storage deficit (Swain et al., 2018, p. 431, Public Policy Institute of California, 2018). As more precipitation arrives in very wet years, it must be stored to weather the growing number of dry years. As rising temperatures increase evaporation in reservoirs and decrease snowpack, the state’s above-ground storage capacity is becoming smaller and less reliable. Water managers are finding that the key to climate resilient water supplies lies underground. In other words, the state must shift away from (although not completely abandon) its legacy dependence on above-ground gray infrastructure and rely much more heavily on aquifer storage. The additional storage space available in California’s groundwater basins is 13–20 times greater than that of the state’s surface water storage capacity (California Groundwater Update 2020, 2023, H-7). A statewide strategy to build water supply resilience in light of climate change calls for, among other things, increasing groundwater recharge by 500,000 acre-feet each year (California Natural Resources Agency, 2022, p. 6). Extensive aerial surveys have helped map where soil and rock conditions below ground allow for the best recharge (Dlubac et al., 2024, p. 21).

However, institutional and legal norms — particularly the legacy of water rights established over a century ago — have not evolved as fast as climate change has proceeded. California’s water management institutions and policy are highly decentralized. Local control is highly prized, and state regulation is strongly resisted. Statewide management of groundwater has been especially challenging, representing the third rail of California water policy that no one has been willing to touch. Over a century ago in 1914, when the state legislature first acted to establish a water permit system for the state, regulation of groundwater was removed from the legislation, resulting in a Water Commission Act that regulated only surface water (Leahy, 2016). In decades since, landowners in certain local areas of the state — mostly coastal regions with significant urban water use and facing the potential for seawater intrusion — came together to either establish a local groundwater management agency, such as the Orange County Water District in 1952, or to settle their disputes in court (Blomquist, 1992; Leahy, 2016). However, groundwater pumping in the Central Valley — the center of agricultural activity in the state — continued unimpeded. A series of droughts — 1976–1977, 1987–1992, 2007–2009 — brought bad impacts such as dry wells, subsidence and ever lowering groundwater levels. This led to state efforts to introduce pumping restraint. Progress until recently was incremental at best, limited to voluntary planning and better data collection. Nothing emerged that required limits on pumping (Leahy, 2016, p. 16–26).

In 2014, the dam finally broke. The passage of the Sustainable Groundwater Management Act (SGMA) represented a dramatic leap forward in terms of institutional and policy change. Several factors combined to make this possible. The state was (again) experiencing one of the worst droughts on record. But this time, then Governor Jerry Brown used the budget process to force parties to the table by including funding for new staff at the State Water Resources Control Board to “protect groundwater basins at risk of permanent damage until local or regional agencies are able to do so” (Full Budget Summary, 2014, p. 116). Brought to the table at last by this credible threat to regulate, during the spring and summer of 2014, major players such as the Association of California Water Agencies and the California Farm Bureau participated in negotiations over SGMA’s design.

The resulting legislation finally established statewide standards for managing groundwater, but without discarding long-standing norms for local control. Under SGMA, local agencies were required to establish and implement plans to reach sustainability by 2040 or 2042, depending on the basin. It is the local agency’s responsibility to ensure goals are reached, or else face the possibility of state intervention (California Legislative Analyst’s Office, 2024). SGMA’s designers were betting that when given the choice, landowners will far prefer a local regulator to a state one. So far, this bet is paying off. Of the nearly 100 groundwater basins that fall under SGMA’s purview, as of September 2025, only six are at some stage of the state intervention process (State Water Resources Control Board, 2025a).

SGMA’s ultimate success, however, also rests on another bet: that the legacy of an individual right to use water under the farmer’s land will not be used to overturn the collective policy achievement of preventing the overuse and depletion of groundwater for all those who depend on the underlying aquifer. The law made no changes to water rights. Even though SGMA endows Groundwater Sustainability Agencies (GSAs) — 250 of which have been formed since SGMA’s passage — with authority to regulate groundwater, including instituting both fees and usage restrictions, all groundwater users retain their rights. Planning processes and local ordinances, motivated by SGMA’s requirements, merely constrain their use. So, if one or more groundwater users believe their rights are not being adequately upheld, they can seek a remedy in the courts. The political bet is that policies developed at the local level with input from stakeholders will prevail over lengthy legal battles. The jury is still out on whether this bet will pay off.

SGMA has substantially altered the institutional landscape for groundwater management. Landowners are now facing new fees and restrictions on groundwater use, which had long been their back-up supply during increasingly frequent periods of surface water shortage. This makes farming more costly and uncertain, even for those with surface water rights. Thus, farmers now have incentives to reduce water use, use their land for groundwater recharge, and even to take their land out of farming altogether (Hanak et al., 2023, p. 3), assuming the law is not over-turned at some point in the future.

There is also a growing effort to expand underground water storage. Groundwater Sustainability Plans (GSPs) in the San Joaquin Valley, where problems such as groundwater overdraft, subsidence, and dry wells are perhaps most severe, are planning a dramatic increase in construction of dedicated groundwater recharge basins, as well as programs promoting on-farm recharge. In addition to establishing incentives for recharge, SGMA also rectified the long-standing lack of connection between the regulation of surface and groundwater, calling out the depletion of interconnected surface water through excessive pumping as an “undesirable result” that must be avoided and mitigated (Cantor et al., 2018, p. 2). GSAs are now undertaking efforts to better understand where and how surface and groundwater are connected, knowledge that will be critical for promoting greater recharge.

However, numerous hurdles remain embedded in the riparian water rights legacy that remains unchanged under SGMA. First, under law, groundwater basin recharge must be accomplished by using water held under appropriative rather than riparian rights. This is because appropriative rights were designed to accommodate the need to divert and sometimes store water — especially important in a region where rainfall does not always fall in the time and place where it is most needed. Riparian rights, on the other hand, are a holdover from how water rights are conceived and practiced in the eastern and southern parts of the country where rainfall is greater and more evenly distributed year-round. Under riparian rights, landowners adjacent to the river have a right to use its flow, whatever it might be at a given time. It does not entail the right to store the water for later or divert it to another location. This takes groundwater recharge off the table for water obtained through a riparian right.

Land adjacent to rivers has high potential for recharge. The only way to accomplish recharge under a riparian right would be for landowners to dedicate their land in a way that allows for natural recharge to occur, such as by dedicating it to ecosystem restoration or flood control. SGMA provides some motivation for such land transfers, and new partnerships between landowners, water agencies and environmental nonprofits focused on restoration are emerging. Many of these arrangements involve the transfer of lands to state or nonprofit entities who manage them for ecosystem, flood control or recreational benefits (Pottinger, 2018, River Partners, 2025).

Another way in which California’s legacy of water rights constrains aquifer recharge is the view that recharge is not a beneficial use of water on its own. This means that for water to be stored in a groundwater basin, there must be plans for its use. That beneficial use could involve extracting the water later for municipal, agricultural, or environmental use, or it could entail keeping the water in the groundwater basin to help avert or reduce problems such as seawater intrusion or land subsidence. However, the latter set of uses, in which benefits derive from keeping water stored in the basin, are not yet clearly defined in the law. This slows down the process for the growing number of groundwater users and managers seeking to use surface water for groundwater recharge (Miller et al., 2018, pp. 4-5).

While in some groundwater basins under SGMA, landowners are not required to have a permit to draw water out of the ground, they do need one to put water into the ground. This is due to the need to establish and document the “beneficial use” of that stored water as well as to ensure that water diverted and stored underground does not impinge on the surface water rights of others. Each appropriative surface water right specifies the amount, location, and timing of water diversion, and its beneficial uses. When recharge is introduced, very often the timing of diversion might change — for example, shifting to more diversions in winter months when water is plentiful — and sometimes the location might change to an area best suited for recharge. Although these changes may sound simple, it may take months to years for the State Water Resources Control Board to approve them, largely due to antiquated systems of tracking water rights usage and the need to study the specific implications of such changes for other water users.

When legacy rights impinge on modern need, people seek out workarounds. Hence, there are nascent efforts underway to capture flood waters for groundwater recharge. Beginning in 2022, under executive orders from the Governor the Board has established methods for “temporary” permits, of 6 months and 5 years in length, for recharge of groundwater using “excess” water in wet years. In very wet years, the amount of water flowing toward the ocean exceeds the amount allocated to surface water users during the wetter months. Putting this water underground holds the potential for considerable public benefit, by reducing flooding risks and storing water for future use in dry years.

However, to take advantage of this opportunity, landowners and water districts need to act fast, implementing recharge strategies during the time when flood waters are present. Acting fast has not been the State Board’s strong suit. The temporary permits are an effort to speed up the process. The Board has sought to speed up their review of the “90–20” rule of thumb, under which applications that request permission to recharge water when flows are above 90% of normal while taking only 20% of this amount are more likely to receive approval (because diverting these waters for recharge is very unlikely to impact other water claims). There is now even a way to receive approval to recharge flood waters without a permit under certain conditions (State Water Resources Control Board, 2025b). However, only small amounts of water have been recharged to date through these temporary permits, and approval still takes several months, which means that one needs to begin the process — including submitting permit application fees — prior to the rainy season when it is still uncertain whether excess flows will arrive.

The evolution of groundwater governance in California described here highlights the sticky nature of institutional norms around water use in the state. Although the groundwater laws enacted in 2014 represent a substantial change in the policies and regulations guiding groundwater management in the state, they are, as Leahy puts it, “an example of how what occurs ‘overnight’ can be a century in the making,” (5). Climate change is bringing more extreme weather to the state, which may sharpen the focus of policymakers in enacting needed legislative change. However, the less visible, incremental steps between major policy shifts are also crucial. The recent steps to establish temporary permits to enable the capture of flood waters for groundwater recharge only scratch the surface of this challenge but may be the first step toward larger reforms.

Providing adequate water in the arid and semi-arid West is partly a climate challenge, but also a population question. How much water is needed is determined by how many people and businesses need it. But how many people need it can depend on whether it is available. Some crops can be fallowed in drought years, but other crops and people cannot. If you build a water supply sufficient to promote growth, and new population and businesses move in, the demand for water hardens, i.e., taps cannot be turned off. And if droughts lengthen, new supplies and storage must be built or greater efficiencies in water usage must be achieved to accommodate new growth or water-stressed existing growth. Such is the story of California’s San Francisco Bay Area.

When plans were brought forward to create the Hetch-Hetchy Regional Water System [HH RWS], the planners originally only had residents of the City of San Francisco in mind. At the time, residents obtained their water from the Spring Valley Water Company which privately managed local storage and supply. Officials realized, though, that the company could not meet the needs of city functions and a growing population. This was evident during the 1908 earthquake in the City of SF that caused a large fire, which became difficult to extinguish due to insufficient water resources. City public officials sought to identify and capitalize upon a stable water supply for the city’s present and future development in the early twentieth century. This moment would be pivotal for the future of municipal water supply in the Bay Area because it represented a shift from private to public management. The Hetch-Hetchy reservoir inundates the Hetch-Hetchy Valley in Yosemite National Park and serves as water storage for the system. Protections enacted by the National Park Service on this federal land preserve this water as a pristine source with a limited extent of impact from human activities. The City of San Francisco was authorized to capitalize on this water source with development in the park by the Raker Act of 1913, named after John Raker who was a congressman of the Central Valley.

This authorization resulted in the construction of O’Shaughnessy Dam in 1923, with a unique design that would accommodate two later additions made to its height. The original project was under the purview of the San Francisco Mayor and the Army Corps of Engineers. After the city purchased all infrastructure previously owned by the Spring Valley Water Company in 1930, the San Francisco Public Utilities Commission [SFPUC] was created in 1932 and enacted these additions, to provide a total 360,000 acre-ft of storage (Chatman et al., 2003, p. 6). The dam is 300 ft tall with 117 ft of that height being a structural base located underground. The SFPUC purposefully built a dam that far exceeded their necessary storage in order to meet anticipated need in the coming decades. While the commission likely could not have predicted that San Francisco would be the major city of today in a state that is currently the fourth-largest world economy, the gold rush era and construction of the transcontinental railroad as well as other major development on the West Coast meant San Francisco at that time was already becoming a major metropolitan hub for both commerce and population. What makes this dam unique is that it was constructed as an investment by the end user, independent of federal and state project funding. San Francisco also has the right-of-way for use of all of its pipes and tunnels. The question then lingers for many as to why the City of San Francisco opted to share this water with other municipalities across the Bay Area that had no role in the construction.

By the 1960s, San Francisco expanded the water system with the construction of Don Pedro Reservoir, in collaboration with the Turlock Irrigation District [TID] [Bay Area Water Supply and Conservation Agency (BAWSCA), 2025a]. The creation of this separate reservoir increased overall water availability because it enabled the SFPUC to satisfy its water obligations to TID while also having plentiful supply for themselves. The SFPUC has obligations to supply water to TID because Modesto and Turlock are the senior water rights holders on the Tuolumne River. San Francisco holds 570,000 acre-ft of storage in the Don Pedro Reservoir of a total 2,010,000 acre-ft and leaves the remainder to Modesto and Turlock. San Francisco also has an obligation to redirect 51.71% of its own share toward the streams managed by Modesto and Turlock to maintain appropriate stream flow for fish populations (Kastama, 2024).

Since the Don Pedro and O’Shaughnessy dams were large upfront investments into the system, San Francisco hoped to recover those costs. Hence, it was incentivized to find outside wholesale customers that could be physically interconnected to the city’s water system. These customers were other cities throughout the Bay Area, who advocated early on behalf of San Francisco for the original passage of the Raker Act and who now could contribute meaningfully to overall revenue for the SFPUC from this project. In addition, the SFPUC was gaining revenue from hydroelectricity that, by terms of the Raker Act, could only be used for public benefit, meaning sold to provide power as well for the Bay Area region. The inclusion of hydroelectric infrastructure in the overall project design was a provision required by the Raker Act (San Francisco Public Utilities Commission and Hanson, 2005, 27). For most of its costs, the SFPUC was heavily reliant upon tax bond measures voted on by San Francisco residents. Conflict arose when San Francisco officials proposed a water rate for outside customers that was significantly higher than what they charged their own retail customers.

The other Bay Area cities formed the Bay Area Water Users Association (BAWUA) to directly negotiate with the SFPUC on rates. The support of BAWUA enabled the City of Palo Alto to lead litigation against the SFPUC and renegotiate these rates (United States Court of Appeals, Ninth Circuit, 1977, 1). BAWUA then evolved through Assembly Bill 2058 into the Bay Area Water Supply and Conservation Agency [BAWSCA] (California State Assembly, 2002, 1). The purpose of this special district was to address the persistent concern that outside customers did not have an adequate voice in the SFPUC’s decisions. BAWSCA collaborates closely with the SFPUC and drafts guidance that has consistently influenced management since the agency’s founding in 2003 (BAWSCA, 2025). The system’s wholesale customer base has expanded to 24 municipalities as of 2025.

The SFPUC has faced challenges with both water supply and funding, confounded by the diversity of communities it must deal with. To put this into perspective, it is important to first understand the timeline of interconnection and the realities of unequal water usage. Roughly a quarter of the municipalities that became wholesale customers [Hillsborough, Daly City, Burlingame, Atherton, San Bruno] were incorporated within a decade of the system’s construction and lacked an established water supply. Other municipalities throughout the twentieth century were gradually connected to the system as population expanded across the region. Certain wholesale customers such as the Purisima Hills Water District, which serves Los Altos Hills, report a per-capita water consumption that is 200% to 300% higher than that of other municipalities (Wang and EKI Environment & Water, Inc., 2025, p. 47).

The SFPUC and BAWSCA both recognized that this would make the task of maintaining a sustainable supply more difficult. SFPUC’s Water System Improvement Program, which was drafted and approved in the year 2008, was created as a response to this challenge (BAWSCA, 2010). As outlined in an Interim Water Supply Limitation, the purchases of wholesale customers for water could not total more than 184 mgd [million gallons per day] and retail customers could purchase no more than 81 mgd. The wholesale portion was divided into allocations and a negative incentive framework of enforcement fined municipalities for exceeding their allotted totals. However, BAWSCA members were unable to agree on how to divide these allocations and failed to recommend a uniform approach to SFPUC, and the allocation system ultimately implemented may not have adequately reflected member needs. Water allocations have real implications for the future of Bay Area residential and commercial development. For example, a review of plans to expand housing in 20 jurisdictions in San Mateo County revealed that all but one has inadequate supplies to support planned housing in the fourth and fifth years of a drought (Lowell, 2025).

At the other end of the socioeconomic spectrum, East Palo Alto was a primarily nonwhite community at its time of incorporation in 1983 and was the lowest per-capita water user on the system. However, a surge of economic growth and new population outpaced the city’s ability to support with its exclusive reliance on its Hetch-Hetchy allocation. Hence, the city issued a moratorium on new development in 2016 which curbed significant progress this community had made. East Palo Alto City Council simultaneously began to plan for shifting toward alternative water sources, such as desalination as well as the Palo Alto Park Municipal Water Company and other potential local groundwater suppliers (Hernandez, 2021, p. 19). However in 2018, the cities of Palo Alto and Mountain View both agreed to share a portion of their allotment with East Palo Alto (Davila, 2018). This collaboration mechanism is made possible by the 2018 Water Supply Agreement [WSA], which allowed customers to share their allotment, also referred to as an Individual Supply Guarantee [ISG], with other customers (Alameda County Water District and SFPUC, “Alameda County Water District and SFPUC, 2018”, 14). The policy demonstrates that even though it was difficult for municipalities to reach consensus, the framework still had the ability to evolve toward enhancing collective compromise and addressing newer needs. Even so, East Palo Alto is not alone in its motivation to explore alternative water sources. Other municipalities have developed their own storage and tapped into groundwater supplies to account for their seasonal variability in demand and increase overall supply reliability.

Supply reliability will be key in conversation when discussing how infrastructure, funding, and drought are now interacting with each other. In 2025, most of the Hetch-Hetchy infrastructure will have reached or surpassed its 100-year age mark, meaning it is soon in need of repair. These repairs could total billions of dollars but are crucial to making sure that the system can continue to meet demand standards (Bulloch et al., 2017, p. 3). The City of San Francisco and various wholesale customers have developed a dependency over time on Hetch-Hetchy’s water supply and are thus particularly vulnerable. Additionally, there are certain areas of concern in the design, including the crossing of conveyance piping across major fault lines and water discharge outlets encased in concrete, that need to be carefully managed. Over the past several years, the State of California has issued regulations, such as California Water Code Section 10632 involving Water Shortage Contingency Plans for urban water providers, and State of Emergency declarations to decrease overall usage in drought conditions (California Public Law, 2025, Water Code Section 10632). For SFPUC, this decrease in demand raised concerns about water rate revenues and being able to cover major imminent repairs in the years ahead.

Added to all of this are new uncertainties surrounding potential new regulatory constraints on how much water SFPUC can take from the Tuolumne River system. A 2018 update to the Bay Delta Plan — intended to ensure that the needs of all water users, including the environment, are satisfied — requires that SFPUC and the irrigation districts on the Tuolumne leave more water in the river to meet environmental needs. Climate change is among several drivers creating the need for more water availability in the river system during dry periods. The 2018 amendments have not yet been implemented as negotiations around potential alternatives proceed, but if fully implemented, SFPUC could lose as much as 40% of its supply during dry periods (Alternative Water Supply Plan, 2024, p. 36).

At the same time, in a July 2025 BAWSCA meeting, the city of Mountain View raised concern over a specific policy in the WSA called a Minimum Purchase Requirement [MPR] (BAWSCA, 2025b, 00:51:45). Initially applied to certain municipalities to discourage shifting toward imported alternative supplies, this provision requires the city of Mountain View to purchase a set amount of water at minimum annually. When Mountain View acted in accordance with drought restrictions, water usage dropped to below this amount and the city had to pay the difference but was instructed by the state that legally they could not use this extra water. In the 2021 Amended WSA, the SFPUC attempted to accommodate by enabling the choice for customers with an MPR to transfer both their ISG and financial responsibility among other customers (City and County of San Francisco and BAWSCA, 2021, 18–19). However, BAWSCA is further proposing renegotiation of the MPR itself, but this would further decrease base revenues for the SFPUC (BAWSCA, 2025b, 1:15:20). The public comment of BAWSCA’s July Board Meeting additionally mentioned concern regarding water rates in May of 2025 that were 10.5% higher than what they were originally projected to be back in the year 2022. It demonstrates the broader understanding that a decrease in demand may motivate an increase in rates at the detriment of wholesale customers. In this case and others of decreased water usage, with a mean 9% reduction across all wholesale customers between years 2020 and 2023, it becomes more difficult for the SFPUC to balance the priorities of fair water rates with the obligations of system maintenance and repair (Wang and EKI Environment & Water, Inc., 2025, p. 47). Only further worsening the issue, water that sits unused in reservoirs for longer periods of time loses its residual chlorine and deteriorates in quality. The remedy for this issue further drives up operation and maintenance costs in what is already a strained financial situation.

The evolution of the Hetch-Hetchy system illustrates that water infrastructure can encourage population growth and harden demand for water at a time when water may become harder to store and scarcer in the future. The path dependency of how cities solved their water problems in the past has created the institutional structures that must resolve future water allocations, rates and new projects.

Much of the American West was settled and developed in the nineteenth century when the primary industry was agriculture. Farming required water, but water in the Western region was abundant in some places and scarce in many others. This gave rise to large gray infrastructure to store and convey water from where it was abundant to where it was needed. Irrigation systems enabled farmers who owned property that was not adjacent to bodies of water to transport surface water to their lands. Appropriative rights secured the usage of the water and groundwater rights enabled them to sustain their operations during drought periods. Over time, the environmental movement and the growth of urban residential and commercial sectors put a strain on a water supply system that was rooted in economic and legal assumptions of the past. Climate change not only exacerbated these tensions, but at critical times prompted the need to adapt water supplies within the system of existing infrastructure. How these legacy features will change to resolve these tensions is murky at best and reflects the overall characterization of water resources management as being a continually evolving landscape.