Let’s start thinking about water!

A hydro-economic approach for quantifying well performance thresholds and recoverable groundwater yields in Texas

Thompson and Young’s paper shows up at a good time with increasing water-level declines affecting more and more wells in the Carrizo-Wilcox Aquifer in Central Texas [and you ain’t seen nothing yet…]. In fact, they tested their methodology in the Carrizo-Wilcox with the first large water-level declines associated with the Vista Ridge Project, where groundwater is produced and piped 140 miles to San Antonio. They call their tool HELPER, the Hydro-Economic welL PERformance model, which considers the “cost of energy required to lift water to land surface, (2) the cost of pump equipment, and (3) the cost of well drilling and installation.”

Their analysis shows groundwater production in the unconfined zone may be most limited by operational thresholds such as decreased transmissivities. In confined settings, performance may be most limited by affordability thresholds associated with production costs. The lead author may be relieved to hear that I won’t be writing a comment on the paper [although we need to get together to drink beers and discuss the usefulness of specific capacity tests].

Revisiting Gunnar Brune’s “Major and Historical Springs of Texas” with an analysis on the fractal character of springflow

A few years ago, I helped write a short entry into a book with worldwide coverage on the status of Texas springs. I knew exactly where to go to get an estimate on how many springs had gone dry: Gunnar Brune’s report on springs for the Texas Water Development Board. However, as I included Gunnar’s estimate in our writeup, I realized that it had been 50 years since someone had updated that number (Gunnar’s report was published in 1975, with most of the fieldwork completed by 1971). So Galaviz, a talented undergraduate student, and I started using satellite imagery and other means (including a dog) to “revisit” all the springs in the report and update the number that had gone dry.

The first thing we found was that the data in the report was “messy” with inconsistencies that we could address with Brune’s later book, The Springs of Texas, Vol. 1. After correcting the flow status of springs in the 1975 report with those in Brune’s 1981 book, we found that 14 percent of springs had gone dry by 1981 with 23 percent of springs dry today, an increase of 64 percent. Because Brune’s book did not include the entire state and we found the book’s information much more reliable, we refined our analysis to include only those springs shared between the report and the book. This refined analysis showed that 11 percent of springs had gone dry by 1981 compared to 30 percent today, an increase of 173 percent, or 2.7 times more. We also found that springflow volumes follow a fractal distribution. Using the fractal relationship, we estimated total springflow in the state at 2.1 million acre-feet per year.

Evaluating future water availability in Texas through the lens of a data-driven approach leveraged with CMIP6 general circulation models

The climate modeling community puts out a new batch of climate models every so often, with the latest called CMIP6 (Coupled Model Intercomparison Project Phase 6). In their assessment of “water availability,” Li and others first compared CMIP6 to CMIP5 to note any changes in model outputs for Texas. They found no change in precipitation trends between the two sets of models but found CMIP6 (SSP2–4.5 and SSP5–8.5) to run hotter than CMIP5 (RCP4.5 and RCP8.5) (CMIP6 uses shared socioeconomic pathways [SSPs] while CMIP5 uses representative concentration pathways [RCPs]; both are about the same but not exact).

For Li and others, “net water availability” is defined as the difference between precipitation and actual evapotranspiration, which differs from water availability used for water planning. Therefore, their conclusions are based on what’s happening within the region at the land surface. A decrease in “net water availability” suggests less runoff to rivers and streams (and thus less water in reservoirs) and less recharge. One issue with the study is that it assumes that a region’s water supply is contained within the region. While in many cases that is true, it’s not true in all cases, for example, Regions E and M and, to a lesser degree, Regions C and H. Furthermore, changes in effective recharge (recharge that meets the needs of increased pumping) are not dependent on changes in total recharge. Regardless, given these caveats, some interesting things come out of the study.

There’s a substantial difference between the response of net water availability between SSP-4.5 and SSP-8.5 (the latter being the warmer of the two). For example, Region F (where net evapotranspiration is important to local water supplies) shows a 10 to 30% increase in net water availability under SSP-4.5 with a -10 to 10% change under SSP-8.5 over the next 50 years. On balance, Li and others show increased net water availability for SSP-4.5 with lower levels in several regions for SSP-8.5, but nothing to lose sleep over. Beware their conclusions about water supplies in the regions because they don’t fully understand the planning process (their analysis is focused on averages, not the droughts of record that Texas water planners use, and they use projected needs to make conclusions without considering water management strategies in the plan).