The "era of pools"

If you want to understand the state of play on the Colorado River, you have to understand three very important letters.

The letters are ICS, shorthand for Intentionally Created Surplus.

Among water nerds, it is an infamously arcane, controversial and yet critically important program that basically allows certain (typically large) water users in the Lower Basin to bank volumes of unused water in Lake Mead.

The ICS program allows them to save water for a rainy — er, dry — day while providing some operational flexibility to the overstressed reservoir and its hydroelectric turbines.

The program was introduced in 2007, but it’s set to expire unless renewed in the next plan for the Colorado River.

The tentative Colorado River deal we covered last month between the Lower Colorado River Basin states of Arizona, California and Nevada proposes extending the ICS policy, as well as making some changes designed to secure buy-in from California and address some of the program’s criticisms, which we’ll get into in a moment.

But ICS is more than a bargaining chip.

In the eyes of some river analysts and stakeholders, the conservation pool idea at the center of the program could be the future of Colorado River management.

I am not an expert, but I did get an improbable A- in a water law class (intended for law students, no less!) while I was studying for my master’s at the University of Montana. So I’ll do my best.

The first thing to understand is that in most western states, water rights are (usually) distributed according to a system of prior appropriation. That system is governed by the principle of “first in time, first in right.” The first entity to establish a use of a water source has priority access to that water in times of shortage. Meaning, a user downriver can have priority over a user upriver, so long as they established their right first.

But if you fail to use your water, you lose it.

In other words, when entities with contracts for Colorado River water in the Lower Basin use less than their allocation, that water typically flows to the next user in the priority line.

In theory, this helps resolve some of the tensions inherent in the system of prior appropriation.

In practice, it incentivizes overuse.

ICS offers a way around this problem, allowing certain users to store water they don’t immediately need in a “pool” — a sort of conceptual reservoir that sits atop Lake Mead — that they can withdraw later or use to offset other reductions.

What’s crucial to understand is this water generally sits outside of the priority system.

That’s especially important for users with low-priority access to the Colorado River who stand to lose the most in times of shortage. One of these low priority users is the Metropolitan Water District of Southern California, which pushed to develop the program in the early 2000s to manage a volatile, shrinking water supply in light of its low priority relative to other users in the state. Metropolitan continues to be the program’s main user.

History buffs may be interested to know that Metropolitan has the unique right under its almost 100-year-old contract with the federal government to bank water behind what was then called Boulder Canyon Dam (now known as the Hoover Dam, which creates Lake Mead). The district didn’t need to use this right until the 2000s, by which time drought had struck the basin and Arizona had fully grown into its entitlement — its water had previously flowed unused to California.

But Arizona, for reasons grounded elsewhere in the complicated legal history of the river, objected to Metropolitan’s exercise of its storage right. (Remember that for all the spirit of collaboration among the Lower Basin states in recent months, Arizona and California are historic riverine rivals.)

In the resulting dustup and discussion, the states created an early form of the ICS program as part of 2007 interim guidelines on the river. The program was then expanded with the 2019 Drought Contingency Plan.

Now, the holders of ICS water include Metropolitan, California’s Imperial Irrigation District, the Central Arizona Water Conservation District (or CAP), the Southern Nevada Water Authority, and two tribal nations: the Colorado River Indian Tribes and the Gila River Indian Community. Metropolitan holds the greatest volume of ICS water by far.

In practice, there are multiple types of ICS water. It can be created by conserving water that would otherwise be used — like when farmers fallow fields — by investing in infrastructural improvements to minimize waste, by bringing on new supplies, even by funding conservation programs in Mexico, which also has a Colorado River entitlement and is considered part of the Lower Basin.

One important restriction to keep in mind is that, under the DCP, those with ICS water stored in Lake Mead can’t withdraw it if the reservoir drops below 1,025 feet. As the future of the program is debated, the question of how low is too low for ICS withdrawals is central. (The lake sits at 1,044 feet as of June 30.)

The program has exploded in popularity, and for good reason.

It’s so popular, in fact, that about half of the water stored in Lake Mead is ICS water.

On one level, that’s great — more people are leaving water in the reservoir.

But there’s a catch. Definitionally, ICS water belongs to someone who, at some point, will want to take it out and use it.

And, as you might hear if you follow discussions about the program, it “counts” for the purposes of determining shortages while also being protected from shortage declaration. When the federal government releases its 24-month projections for Lake Mead elevations, it includes ICS water in that calculus, thus forestalling shortage declarations.

To put a finer point on that idea: if ICS water wasn’t counted for shortage determinations, the Lower Basin would be subject to much harsher cuts than it is otherwise.

And if the entities that banked water in the river decide that they want to use it, reservoir levels drop.

The above idea is at the center of the debate around ICS water’s operational neutrality.

Because ICS water counts for the purposes of determining shortage, it is not “operationally neutral” — it adds to the reservoir elevation, but it does not lift all boats.

It provides some short-term operational flexibility for the reservoir — particularly for hydropower generation and recreational uses like boating and fishing, which benefit from more water regardless of how it got there — but the real boon is for individual contract holders.

What the Colorado River needs, though, is permanent, systemwide conservation.

ICS has thus been the subject of some criticism in academic and policymaking circles.

Some of that criticism has even been a bit of a sore spot for those who have helped put the program together. A 2024 Kyl Center report that gave the program some credit for providing operational flexibility but dinged it for failing to secure long-term conservation generated a blog post from the Arizona Department of Water Resources that lambasted the report as full of “baseless accusations and little substance.” (Read more about that war of words in a previous edition here.)

Proponents of ICS argue that the program has led to some systemwide conservation because assessments levied on stored water and other mechanisms have created several hundred thousand acre-feet of systemwide water (admittedly at a time when the basin’s needs are measured in the millions). And they say that consumptive use in the Lower Basin has dropped during the life of the program. A 2025 report from Metropolitan estimates that more water is stored now — about 2 million-acre feet — than would be if the program didn’t exist.

The report concedes, though, that if all ICS water were withdrawn, reservoir levels would be lower than if the water were never banked at all.

And therein lies the rub.

Another line of criticism is that the program circumvents the priority system.

If you represent an irrigation district in Yuma with secure, high-priority rights, you don’t get the same benefit that a low-priority user like the Central Arizona Project does. Or, take on-river users like cities in Yuma or Mohave counties that don’t get to participate in the program.

But the ability to temporarily use less water without fear of losing it to the next person in line is central to the program.

In May, Arizona, California and Nevada announced a tentative deal to share cuts of the river’s water over the next two years. It was important news. The states of the basin have long been unable to reach consensus on how to manage the river’s water into the medium-term future. A seven-state, basin-wide deal continues to be elusive, but a three-state deal is certainly a step in the right direction.

Still, just about everybody we spoke to for this story cautioned that the three-state deal isn’t final. The Lower Basin states among themselves and all seven states in general are still actively negotiating. And what vision for the river the federal government adopts when it rolls out its final plan later this summer is anyone’s guess. (Recent statements from Bureau of Reclamation head Scott Cameron suggest that the Lower Basin proposal has some purchase, though that doesn’t mean Arizona won’t feel the pain of cuts.)

The deal is tricky because its success relies on California agreeing to share in cuts that would otherwise fall almost exclusively to Arizona, and that’s “not an easy sell in California,” Bill Hasencamp the manager of Colorado River resources for Metropolitan, told us.

The three-state deal also sweetens the pot for California — again, remember that Metropolitan is the biggest user of ICS — by extending the program to 2038 and expanding accumulation limits.

And, perhaps most importantly, the withdrawal limit would be moved from 1,025 feet elevation at Lake Mead to 1,000 feet, with each state subjected to its own elevation-dependent delivery limits. Collectively, this increased ICS capacity can be used to meet at least some of the cuts called for in the three-state deal.

Not everyone, however, is on the same page. Gabriel Lopez, the chairman of the Ak-Chin Indian Community, made that much clear during the ARC meeting in May, echoing an argument common among some senior users on the river.

The deal does increase assessments on ICS water in order to contribute to systemwide conservation. And it allows entities with ICS water to meet required cuts by converting that stored, earmarked water to systemwide water. Along those lines, it expresses a general goal of achieving ICS operational neutrality.

But the deal doesn’t really say how broad-based operational neutrality might be reached, other than that “the Lower Division States will consult to determine when and how to convert ICS to operational neutrality at higher elevations in Lake Mead.”

That “higher elevations” phrase is important. Because of Lake Mead’s critically low levels, if all ICS were converted to operational neutrality overnight, the Lower Basin would be subject to sweeping shortages beyond what the three Lower Basin states have already agreed to take under the proposal/proposed deal.

But we don’t know if or when the reservoir’s levels will recover. Suffice it to say, achieving operational neutrality isn’t painless or an immediate priority given the elevation of the reservoir.

Hasencamp said one idea would be to convert a certain portion of new ICS water — say, 10% — each year. But low reservoir levels mean the details may have to wait.