Resolves yes if the U.S. Bureau of Reclamation reports Lake Mead's elevation to be below 895 feet anytime before 2025 ends.
1,000
1.00The USBR 2-year predictions put the lowest modeled elevation of Lake Mead at the end of 2025 as 1035 feet. With 2024 snowpack looking reasonable, it would take a drought of monumental proportions in 2025 to get anywhere close to deadpool.
The US Bureau of Reclamation predicts the worst probable outcome for Lake Mead at the end of 2025 as pool elevation of 1034 feet. It would be very surprising for their modeling to be off by ~140 feet.
To reach deadpool, Lake Mead would have to drop 152 feet. The largest single-year change in Mead's elevation during the drought has been 47 feet (in 2011). So we'd need three consecutive years of record-setting drops to hit deadpool. This market resolves in two years. Even if USBR takes no effort to conserve Mead's water, I can't imagine reaching deadpool before 2025. On top of that, Colorado snow pack is above normal in 6 of its 8 river basins so far in 2023. I think the probability of deadpool by 2025 is <1%.
I don't see any plausible scenario by which Lake Mead can reach deadpool before the end of 2025.
If the next three years are as dry as 2000, 2001, and 2004 (3 of the 4 driest years), we'd still only reach 913 ft. The USBR estimates the probability of Lake Mead dropping below 900 feet by May 2026 as 0%[1]. The minimum estimate for September 2024 is 993 feet[2], leaving 100 feet for the final year (4x bigger drop than the largest on record). The most recent 24-month study puts September 2024 closer to 1012 feet[3]. My own simulations show Lake Mead above 949 feet in 99% of scenarios. I didn't make any optimistic adjustments based on "heavy shortage" conditions that trigger at 1050 feet or "extreme shortage" conditions at 1025 (based on the interim guidelines).
1: https://www.usbr.gov/lc/region/g4000/riverops/crss-5year-projections.html
2: https://www.usbr.gov/uc/water/hydrodata/crmms/current/12_2022/921/dashboard.html#pool_elevation/
3: https://www.usbr.gov/uc/water/crsp/studies/24Month_12.pdf
Federal officials warn "it could", seek action to prevent this from happening: https://www.latimes.com/environment/story/2022-12-16/risk-of-dead-pool-looms-at-colorado-river-meeting
There are two questions here. "If nothing happens, will it read dead pool?" and then "conditional on YES to the above, will some successful intervention be performed".
Historical water levels: https://mead.uslakes.info/level.asp
The last year's drought was extreme, and resulted in a drop of ~25 feet (maybe 20, maybe 30, depending how the year ends, I guess). A drop of >150 feet is needed to reach "dead pool" status. Three summers remain before this market closes. So, each summer would need to have a drop twice as large as that of this last year, which was already unusually large.
@ScottLawrence because the lake is shaped like a V, declines become steeper as the reservoir empties. So past numbers aren't quite as relevant as they look
@mndrix Is there somewhere to see that actual profile of the lake (area as a function of height)? If it's actually a V, this is a difference of around 15% (30% if the lake is a cone) as the level drops from ~1050 to ~900. That's not going to make up for the fact that every one of the next 3 years would need to have a factor-of-2-larger drop.
I also worry about things like changing upstream conditions. It could be, for instance, that an upstream dam is getting rapidly depleted, and once it stops discharging, the level in lake Mead will fall far faster. But I haven't read anything that hints that that would be the case (there aren't any upstream dams so large, I thought).
@ScottLawrence I've failed several times to find an accurate (or approximate) 3D model of the lake to calculate height-volume relations. At this point, I only know that it's vaguely V or U shaped on the short axis, and roughly descending on the long axis. If you find anything, please share.
I agree with your sentiment and reasons for No, as well as your concerns about the broader Colorado watershed/storage. I don't yet own shares in this market because my models aren't yet helpful enough to calculate a probability from which to derive a Kelly bet.
@mndrix There are maps like this, FWIW:
@Duncan This site is useful for seeing some potential bottlenecks: https://www.nps.gov/lake/learn/nature/storage-capacity-of-lake-mead.htm
@Duncan this doesn't seem to show the topography beneath the water's surface. Am I reading it wrong?
@Duncan The table titled "Area and Capacity of Lake Mead" in the storage capacity link is excellent. That's almost exactly what I've been looking for. Thank you.
@ScottLawrence based on the link from @Duncan, here's a quick best fit curve from Wolfram Alpha[1]. x is the water elevation in feet and y is the capacity in acre-feet:
182.737 x^2 - 306326. x + 1.30451×10^8
1: https://www.wolframalpha.com/input?i=best+fit+curve+for+%7B%7B1229%2C30167000%7D%2C%7B1221.4%2C28946000%7D%2C%7B1219.6%2C28667000%7D%2C%7B1205.4%2C26483000%7D%2C%7B1050%2C10230000%7D%2C%7B950%2C4553000%7D%2C%7B895%2C2547000%7D%7D