On the shortcomings of the way we measure water “use”

When I first got serious many years ago about the project of writing (in the newspaper at the time) about New Mexico water, I went looking for the numbers. How much do we have? Who uses what?

It’s a task that became central to my work. Eric Kuhn and I spent three years writing an entire book about the importance of having good numbers and using them to make good decisions.

As I prepare for another fall semester with first year water resources grad students, the question is fresh upon my mind. We spend a lot of time helping students up the bottom slopes of the “water numbers” mountain.

It’s doubly fresh because I’m in the midst of working up, with my Utton Center colleague Kate Tara, input in response to the U.S. Department of Interior’s request for comment on the agency’s post-2026 Colorado River management guidelines. A call for good data, used well, is key to our comments.

In preparing our comments, I had occasion this morning to re-read a really useful paper published earlier this year by Amy McCoy and colleagues about Colorado River data. The paper’s focus is a deep dive into the Bureau of Reclamation’s Lower Colorado River Decree Accounting reports, which to a water data nerd like me, writing about the Colorado River, have long served as a mineable lode of water policy ore. Which they are. But McCoy et al make a crucial point that I tend to forget: In my search for answers to the questions of “how much do we have” and “who uses what”, we can miss things that are incredibly important, but that fall through the cracks of this kind of an epistemology:

Water accounting in river systems endeavor to monitor and track diversions, deliveries, inefficiencies, and savings. Theoretically, water accounting creates transparency for the public, and can be a tool to improve river and water management, particularly as demands grow and supplies are nearly or fully allocated. However, accounting also reflects the historic cultural conditions that were in place when water laws, policies, and infrastructure were initially developed in the modern era. Rivers are complex systems, and accounting often takes a focused lens on elements that directly relate to the economy, such as consumptive use for agriculture and cities. This focus excludes complex elements that are difficult to track, that are not a direct part of the economic system, that are nonconsumptive uses, or that do not have legal allocations or entitlements. In the Colorado River system, elements outside of the historically constructed legal and accounting systems have included environmental uses, tribal water, and in many cases, groundwater. Because these water uses have not been accounted for, any degradation or changes can more easily go unnoticed.

This epistemology biases us toward ignoring important things.

The paper is McCoy, Amy L., et al. “A Survey of the Bureau of Reclamation’s Decree Accounting Reports in the Lower Colorado River Basin.” Journal of Water Resources Planning and Management 149.3 (2023): 04022085. (I think it’s here and not paywalled, though I sometimes can’t tell because I’m writing this on a computer on the university network.)


  1. It’s not “nearly or fully allocated” in many cases as “over allocated”, but I think the other major issue is that things not seen are hard to verify and track and ground water and acquifers are not always well understood. There can be layers with all the geological complexity, salinity issues, over pumping, outright pollution and corruption and with each year as we learn more then there’s also the historical legacy. The questions then become how much can you afford to clean up to use and then reuse as much as possible. If you build your system well you can be sustainable within your population and precipitation and then also supplement with some imports but as time goes on there will be less and less available for imports. Sooner or later someone has to decide “No more room at this inn…”

  2. Having been involved in the data collection process in both river management and water accounting on the Colorado River it would have been interesting to read the paywalled paper.

    I can only relate my own experiences and viewpoints here. I’ll start with this. The key to knowing what goes on in water management is in the integrity of the data that was gathered. Typically the USGS standard is a reading every 15 minutes. This is dictated primarily by the telemetry limitations in the GOES satellite system. Be it from a stream, river or canal. Reclamation’s data collection on the lower Colorado is through Line Of Sight LOS radios and can collect more data than the GOES system. However, there are still constraints on the amount of data that can be transmitted due to bandwidth limitations.

    This brings us to the question. How can we collect detailed data? The type of data that is needed for water accounting purposes. Water that is diverted from the Colorado River needs to be measured multiple times every minute. The results of that minute recorded in a Data Logger and the bare minimum amount of important data is transmitted through the telemetry system and looked at by water accounting. The detailed logs from the data logging equipment dumped into the hydro tech’s laptop when they made site visits. It was a very detailed collection as we could see just exactly how and when water was appropriated. You could also see site traits (I referred to them as personalities). There wasn’t a thing that was missed. I personally looked at water accounting data needing to be of the highest integrity. It went into the system design and collection. Very few people actually understood the detail work involved in the collection process. For the sites entrusted to us, the amount of individual measurements exceeded 44 million. That’s more than the amount of people that depended on the water.

    It would have been interesting to me to have read that paywalled paper. Always interesting to see what their take was.

Leave a Reply

Your email address will not be published. Required fields are marked *