Initial forecast: Lakes Mead and Powell headed for record low in 2018

Posted by on

With an underwhelming snowpack right now, the U.S. Bureau of Reclamation’s initial 2018 forecast (pdf here) projects combined storage in Lake Mead and Lake Powell, the two primary reservoirs on the Colorado River, will drop to 21.7 million acre feet by the end of 2018. That would be the lowest Mead/Powell combined year end storage since Powell was first filled in the 1960s.

WARNING: Just this afternoon, I was discussing reservoir storage data with some folks working on Colorado River policy analysis and I strongly discouraged using the 2018 forecast yet. It’s early. It could snow a lot. It also could not snow a lot. The error bars on a forecast made in January are huge, it will almost certainly change in one direction or the other. But for what it’s worth, here’s my updated combined storage graph with the 2018 forecast added.

Storage in Lake Mead, Powell

There’s still enough water in storage to prevent a 2019 Lower Basin shortage declaration. But the risk for 2020 is rising.

It finally rained in Albuquerque

Posted by on

It finally rained yesterday morning in Albuquerque, a bit after 8 a.m., ending a 96-day dry streak.

Breaking Albuquerque’s dry streak, which ended Jan. 10 at 96 days

The water in the bottom of my gauge looked like strong tea as the rain washed out three months’ dust, and the relief was not measurable, but large. I ended up with 0.09 inch (2.3 mm) at my house, and the Albuquerque airport gauge run by the National Weather Service got 0.03 inch (0.8 mm).

But….

My skin is cracking, and conversation at the office this week turned to moisturizing techniques. Since Oct. 1, the “water year”, 2017-18 is the third driest year in more than a century of records here, a ranking that only changed slightly among years that, for all practical purposes, saw no meaningful rain on the landscape this far into the winter season.

In my morning paper Kerry Jones (NWS meteorologist, graduate of UNM’s Water Resources Program, skilled communicator of such things) gave these “yeah buts”:

“We would need unprecedented wetness, almost equivalent to the dryness we have experienced, to make up the ground we have lost,” he said.

 

On the need for federal legislation to implement Colorado River drought plans

Posted by on

Eric Kuhn* of the Colorado River District wrote an interesting memo (pdf here) for his board’s meeting next week that lays out the options and reasoning behind current discussions about whether federal legislation will be needed to implement Colorado River Basin drought plans.

The “Law of the River”, which governs allocation, distribution, and management of the Colorado’s water is an interlocking body of statute, compact, treaty, court decision, and executive branch actions that makes tweaks tricky. If you want to do something in one part of the law – say, for example, adjust allocations to respond to drought – you have to be mindful of the impact it has on other areas of the law, even if everyone’s in agreement on the steps to be taken.

At last month’s Colorado River Water Users Association meeting, there was an interesting discussion (during a panel moderated by Eric) of whether federal legislation is needed to implement the various institutional widgets being developed under the rubric of the “Drought Contingency Plan(s)”.

There is a general consensus that legislation will be needed for the Lower Basin part of the DCP, which sets new guidelines for reducing water deliveries from Lake Mead to California, Nevada, and Arizona under conditions of Lake Mead emptiness. Here’s Eric:

Eric Kuhn on the need for federal legislation for Colorado River Drought Contingency Plan

This piece is relatively straightforward, and those favoring this approach say Congress can probably do this in a relatively straightforward, non-controversial way despite its current dysfunction. Simple stuff can still get done.

But there are other layers of complexity, including the question of whether, once we unlock the Congressional Action on the Colorado River box, we should take the opportunity to put other stuff in it. Maybe, for example, the states of the Upper Basin should ask for legislation creating a more flexible framework for operating Upper Basin reservoirs to ensure we keep enough water in Lake Powell to avoid compact delivery problems. Eric, in his board member, argues for caution in this regard:

Eric Kuhn on the risks of broader federal legislation for Colorado River Drought Contingency Plan

* Disclosure: Eric and I are collaborating on a book.

What happened in the Colorado River Basin in the winter of 1976-77?

Posted by on

At yesterday’s monthly Colorado Basin River Forecast Center briefing, Greg Smith noted, by way of analogy, the winter of 1976-77. Smith explained that he wasn’t forecasting – the fact that the evolution of this year’s forecast is similar to 1976-77 doesn’t mean that the rest of this year will be like that year, or that this year’s runoff will be like 1976-77. But picking analog years is a great communication tool, to give us a sense of what actually happened, historically, in conditions similar to those we might see today.

So let’s look at 1976-77.

  • Naturalized inflow from the Upper Colorado River Basin at Lee Ferry was 5.4 million acre feet, the lowest in the USBR’s Natural Flow Database (which goes back to 1906)
  • Lake Powell dropped 3.4 million acre feet, the fourth largest one-year drop since Glen Canyon Dam was built (1990, 2002, and 2013 had bigger drops)
  • I graduated from high school

None of these are encouraging analogs.

 

Overcoming “use it or lose it” on the Colorado – an example

Posted by on

Yesterday I pointed out how much water is being stashed in Lake Mead as an example of how folks on the Colorado River are overcoming the old “use it or lose it” problem in western water.

Here’s another example, this time with water taken off of the river and stored underground, in this case excess water in Phoenix’s allocation being stored via a collaborative relationship with Tucson, which has big spreading basins and aquifer storage capability:

“This is Colorado River water that they can’t use today. But if they don’t use it, they don’t have it later in the future,” Molina said. “We worked out an agreement with them where they will store extra water in our recharge facilities at no cost to us.”

The key here is the creation of a new institutional arrangement – a Phoenix-Tucson water banking deal – to overcome a shortcoming in the existing institutional arrangement. Here’s the issue: Phoenix doesn’t currently use its full Colorado River allocation. Four years ago, it toyed with the idea of simply storing its unused allocation in Lake Mead. But that can’t happen, because rules. The rules proved difficult to change, so Phoenix and Tucson developed a side deal, inventing a new institutional widget that allowed them to accomplishment something quite similar under the existing rules.

Overcoming “use it or lose it” on the Colorado River

Posted by on

The “use it or lose it” problem in western water happens when water users who conserve are penalized by having the saved water simply go to another user.

A series of policy innovations over the last decade to overcome this problem are showing up right now in a big way in Lake Mead. In all, through these various mechanisms, more than 700,000 acre feet of water are being left in Lake Mead this year.

Without this water, Mead would probably be under elevation 1,075 right now, and we’d be having our first formal shortage declaration.

Not to alarm you further, but the Jan. 1 runoff forecast for New Mexico is really really bad

Posted by on

As I mentioned, this is the driest start to a water year in a century in Albuquerque. The preliminary Jan. 1 runoff forecast from the federal Natural Resources Conservation Service bears this out. The forecast, based on snow measurements, is stark. NRCS has 40 years of snow records, and for many sites, this is the driest start to a water year on record. In other words, it’s not just down here in the city. It’s up in the mountains of northern New Mexico and southern Colorado, where snowpack feeds New Mexico’s rivers.

With months of snow season yet to come, this should be considered very preliminary in nature. But the statistical probabilities suggest that, for many of New Mexico’s rivers, the chances of a record low year are higher than the chances of an average year.

Some numbers (all of these are median forecast numbers):

  • Rio Grande near Lobatos, near the Colorado-New Mexico border: 15 percent of average
  • Embudo Creek at Dixon: 16 percent of average
  • Rio Grande at Embudo:
  • Rio Grande at Otowi: 24 percent of average
  • Rio Grande at San Marcial, at the head of Elephant Butte reservoir: median forecast of essentially nothing, which is not plausible but the snowpack is so lousy that the model kinda breaks here
  • Pecos at Santa Rosa: 18 percent of average

A few things to remember.

First, it could snow a lot. These numbers could come up. The forecasts for the next month are “meh“, but this could happen.

Second, it could get worse. See previous link to “meh” forecast for the next month, and also recall that both the recent weather and the long term climate have trended on the warm side, which means that for a given amount of snow, paltry as it is, less water is ending up in our rivers.

Third, as I have written before, this is happening in the context of some very important changes in the approach to water management in New Mexico over the last few decades, in terms of conservation and diversification of supply, that leave our human water systems in a very resilient position. That resilience is likely to be very seriously tested this year.

A brief tutorial in how bad the Colorado River Basin snowpack is right now

Posted by on

Let’s look at the new Colorado Basin River Forecast Center graphic of projected runoff into Lake Powell, shall we?

Jan. 1, 2018 Colorado Runoff projection for Lake Powell inflows, courtesy CBRFC

The folks at CBRFC has done a lovely update of their graphical presentation. The story it’s telling right now – not so lovely.

Let’s take this step by step.

First, the green lines in the middle of the graph. Those are “average” for the April-July runoff, where the solid line is the mean, and the dotted line is the median. (UNM WRP students, there’ll probably be a question at some point on the quiz about the difference, and why you’d choose one or the other for policy discussions. Prof. Fleck is just kidding, right?)

The dark blue line is the automated forecast, based on current snowpack and some mathemagic the CBRFC uses that we all trust. The light blue band is essentially the error bounds, where the top of the bar says there’s a one in ten chance it’ll be that high, and the bottom a one in ten chance it’ll be that low.

Then, reading left to right, the X axis is the date of the forecast. So the fact that the blue line is below 4 million acre feet means that when they did their mathemagical incantations (I believe that’s the technical term) on Jan. 1, they concluded that there’s a wide range of possible runoffs, spread from just above 6 million acre feet to just above 2 million acre feet, with a midpoint a bit under 4 million acre feet of inflow into Lake Powell this runoff season.

But another way, this is really, really bad. Or, as one of the smart managers said in an email this morning, “sobering”.

Here’s the link, so you can join us in clicking obsessively while we worry.

Lake Mead ends 2017 not in shortage

Posted by on

Lake Mead ends 2017 at elevation 1,082.5, almost two feet above last year at this time. Lake Powell ends the year at 3,623, up more than 20 feet from a year ago. Combined storage in the two primary Colorado River reservoirs ends the year up more than 2 million acre feet.

Willow Beach on the Colorado River downstream from Hoover Dam, May 2017

This is in part the result of a good snowpack in the winter of 2016-17, but is more than that. Excess runoff into Powell this year from that snowpack was 1.14 million acre feet. The only way you get from there to an increase in storage of 2 million acre feet is by using less water. And that is perhaps the most remarkable piece of Colorado River news as we end 2017.

In the lower basin states of Arizona, Nevada, and California, this year’s preliminary estimate of total use – 6.77 million acre feet – is the lowest since 1987.

Consider:

  • Southern Nevada this year has taken just 81 percent of its full Colorado River allotment.
    • Its use of Colorado River water has dropped 25 percent since its peak in 2000, at time during which its population has risen 55 percent. This is not the result of a shift to groundwater, this is straight up conservation success.
  • Southern California (bolstered by a wet snowpack in the Sierra Nevada), is taking just 4 million acre feet of water.
    • That is its lowest Colorado River water use in the history of modern record-keeping on the river, which dates to the U.S. Supreme Court’s 1963 Arizona v. California decision.
    • Southern California’s use of Colorado River water this year is down 25 percent from its 2002 peak.
  • Arizona will use 2.5 million acre feet of Colorado River water this year, its lowest since 2005.

Part of this, of course, is the big Sierra Nevada snowpack. But it’s also driven by the remarkable decoupling of water use from population and economic growth in the arid southwestern United States. According to a new dataset of municipal water use published this year by the U.S. Geological Survey, total water use in the municipalities served with Colorado River water declined 7 percent from 2010 to 2015, even as population rose 10 percent.

Population is going up. Water use is going down. When people have less water, as I wrote in my book, they use less water.

If we were ending 2017 with Lake Mead’s elevation below 1,075, we’d be in shortage, there would be mandatory cutbacks, and we’d all be writing hand-wringing pieces about the looming apocalypse. The fact that it’s ending at 1,082.5 has largely gone uncommented upon, but if shortage matters, then the lack of shortage must be equally important.

Notes on sources: