Texas Data Center Water Supply: What Developers Are Getting Wrong Before They Break Ground

The Short Answer

Many large rural parcels in Central Texas do not sit on a productive aquifer. The groundwater is real and proven nearby, but the geology beneath your specific site may be Cretaceous clay more than 66 million years old. Drilling into it will not change that fact. The developers who find this out after signing are the ones who needed this article three months earlier.

Table of Contents

1. Why Water Is Now a Fatal Flaw in Texas Data Center Deals
2. The Geology Problem No One Explains Before the Land Decision
3. What a Proper Water Due Diligence Actually Looks Like
4. A Real Project: What MES Found on a 500 Acre Central Texas Site
5. The Four Source Water Strategy That Made the Project Viable
6. The Cooling Architecture Decision That Changes Everything
7. Rainwater Harvesting: The Supply Source Texas Law Actively Supports
8. What MES Delivers Before the Deal Closes
9. Frequently Asked Questions
10. Work With MES on Your Texas Site

Why Water Is Now a Fatal Flaw in Texas Data Center Deals

Texas is one of the most active data center markets in North America. Land costs, power access, and favorable tax structures have accelerated investment across the state. But one variable is routinely underweighted until it threatens the deal: water.

A large data center can consume 150,000 gallons per day or more. Most of that demand is usually non potable cooling makeup, not drinking water. On a large rural Texas parcel, the assumption is almost always the same: drill a well and solve the problem.

That assumption is breaking deals.

State policy on data center water use is moving fast. As of October 2026:

• The Texas House Natural Resources Committee took testimony on data center water use on June 23, 2026.
• On August 3, 2026, the Governor directed the PUC and ERCOT to audit data centers in ERCOT's interconnection process. The audit asks for each project's projected water use, water sources, and cooling technology, and whether it brings and reuses its own water.
• On September 21, 2026, the Governor directed TCEQ to pause the issuance of all permits related to data center projects until ERCOT completes its review.
• The Governor has said he will work with the Legislature on data center safeguards. The next regular session starts in January 2027.

Any TCEQ permit your project needs may wait on that review. Developers who treat water as a late-stage detail are building timeline risk into the foundation of every deal.

The Geology Problem No One Explains Before the Land Decision

Why Big Rural Parcels and Good Groundwater Often Sit Apart

In parts of Central and South Texas, productive shallow groundwater is not distributed evenly beneath the land. The water-bearing formations are thin Pleistocene river terrace gravel deposits that track tightly along ancient river corridors. In the study area described below, the producing wells draw from only 14 to 32 feet of saturated gravel. A mile or two in the wrong direction and the geology transitions to Cretaceous Navarro Taylor marl: dense clay and shale that the Geologic Atlas of Texas describes as "strikingly uniform throughout." Wells drilled into it rarely produce usable water.

In this setting, large, flat, developable tracts tend to sit on the upland terrain, which is marl country. The aquifer exists and is excellent. It is just somewhere else.

What the Geologic Atlas of Texas Tells You Before You Spend a Dollar

The Geologic Atlas of Texas, available through the USGS Texas Geology Web Map Viewer, maps the two surface geology units that matter in this setting: Qt (Pleistocene fluviatile terrace deposits, the aquifer) and Kknm (Navarro Group and Marlbrook Marl, the aquitard). A direct query of the atlas at any parcel coordinates returns which unit sits at that location. This is the first check MES runs on every Texas water engagement, and it takes minutes. Developers who skip it sometimes spend months finding out the hard way.

What a Proper Water Due Diligence Actually Looks Like

Desktop Hydrogeologic Screening: What It Is and Why It Comes First

A desktop hydrogeologic screening is a structured analysis of publicly available well records, geologic mapping, and elevation data performed before any drilling program begins. It is not a field investigation. It does not require a drill rig. It answers the question that should be answered before the land decision is conditioned on groundwater: is there any realistic path to on-site water, and if not, what sources are actually available?

MES conducts this screening using the Texas Water Development Board's full Submitted Drillers Report database, including lithology logs, well test records (particularly 36 hour constant rate pump tests), water level data, and borehole construction records. Ground elevations are measured at reference wells and cross validated. The Geologic Atlas is queried directly at the site and at every producing well in the study area.

The output is a complete, defensible picture of the water supply situation, produced at a fraction of the cost of a dry drill hole.

Why Pump Test Data Is the Most Honest Number in the Analysis

Reported well yields vary widely in quality. Driller estimates are unreliable. Jetted yields reflect the equipment, not aquifer limits. The only number that tells you what the aquifer actually delivers is a constant-rate pump test of 24 hours or longer. MES identifies and anchors every analysis to verified pump test data, and flags estimated figures for what they are.

A Real Project: What MES Found on a 500-Acre Central Texas Site

The following is an anonymized summary of a recent MES engagement. Client name and site location are withheld.

The Setup

A large proposed data center campus in Central Texas needed a water supply plan to support up to 150,000 gallons per day of demand, approximately 104 gallons per minute continuous. The development team assumed on site groundwater would cover most of the non potable load. MES was engaged to screen that assumption before the drilling program was authorized.

What the Screening Found

MES pulled records for all 54 wells within approximately five miles of the site, covering 206 lithology intervals, 23 well test records including three 36 hour pump tests, and 28 water level measurements. The geologic atlas was queried at the parcel and at every producing well.

The result was unambiguous. Every productive well in the study area, including public supply wells verified at 100, 153, and 200 gallons per minute, sat on Pleistocene terrace gravel (Qt). The project parcel sat on Cretaceous Navarro-Taylor marl (Kknm). The nearest correctly located deep boring, drilled 1.6 miles from the site, logged 100 straight feet of clay and shale without a single foot of gravel.

Four independent lines of evidence confirmed the finding:

1. Mapped geology. Direct query of the Geologic Atlas of Texas returned Kknm (clay aquitard) at the project parcel and Qt (terrace gravel) at the pump-tested public supply wells it checked.

2. The nearest deep boring. One hundred feet of clay and shale, no gravel, exactly consistent with a marl-upland location.

3. Gravel hit rate by distance. Across all 54 wells analyzed, gravel hit rates drop from 91 to 100 percent within one mile of the river corridor, to 33 percent in the site's distance band, to zero percent beyond three miles. The site sits 2.1 miles out.

4. Nearby borings by a regional water authority. Four borings drilled 0.7 to 1.2 miles from the site logged clay and sandy clay with only trace gravel fragments. The upland surface between the site and the productive terrace is marl.

The One Exception Worth Testing

The mapped Qt terrace margin runs along a drainage feature that reaches the parcel's southwestern boundary. At that location only, MES estimates a 20 to 25 percent probability of a modest productive completion. One targeted test boring at that corner, at an MES-estimated cost of $10,000 to $20,000 (less if combined with the building geotechnical program), is worth drilling. The parcel interior is not.

MES's Conclusion to the Client

"The region has water. The terrace aquifer two miles west is excellent and proven by 36-hour pump tests. The project's demand is modest. The problem is a two-mile gap between the demand and the resource. This is a pipeline problem, not a hydrogeology problem."

The Four Source Water Strategy That Made the Project Viable

Rather than abandoning the project, MES designed a tiered water supply strategy around sources already within reach. Each source covers a defined share of demand.

Source 1: Municipal Water Supply (Potable and Fire Suppression)

An existing tap at the property corner connects to a local water supply corporation whose own terrace wells carry verified 36-hour pump test results. The ask is sized to the true potable, domestic, and fire suppression load, not the full planning volume. Sizing the first request to that load, rather than the full planning figure, keeps the negotiation productive from the first stakeholder meeting.

Source 2: Non Potable Reuse from an Adjacent TCEQ Authorized Facility

The adjacent property holds a TCEQ reuse authorization tied to a 0.95 MGD wastewater reclamation facility currently in permitting. MES is the engineer of record on that facility. Coordinating delivery of reclaimed water to the data center campus covers toilets (which need Type I reclaimed water), irrigation, and a significant share of cooling makeup demand. At a time when data center draw on potable supply is under legislative scrutiny in Texas, sourcing non potable demand from a reclamation facility is also the strongest public perception play available to the project team.

Source 3: Rooftop Rainwater Harvesting

Texas law directly supports rainwater harvesting:

• Tax Code Section 151.355 exempts rainwater harvesting equipment and supplies from state sales tax
• Tax Code Section 11.32 lets a local taxing unit exempt property where approved water conservation initiatives are in place, if it chooses to

Rainwater also carries near-zero total dissolved solids, which supports higher cycles of concentration in cooling towers. Each harvested gallon can displace more than one supplied gallon in cooling makeup. Under a closed loop cooling design, a large single story data hall plus hardscape catchment area can cover the entire non potable load on an annual average basis.

Source 4: A Contingent On Property Test Boring

One targeted test boring at the southwestern drainage corner. Non gating. Low cost. Clear decision rule: ten or more feet of saturated gravel with its base below the projected water table elevation completes as a test well and triggers pump testing. Anything less closes the groundwater question permanently, without spending much money on the wrong answer.

The Cooling Architecture Decision That Changes Everything

In a Texas data center water plan, the largest variable is often not the geology. It is a mechanical design decision made early in the project: closed loop versus hybrid cooling tower.

Cooling Design	Total Water Demand	Non-Potable Strategy
Closed-loop (base design)	~25,000 GPD	Rainwater + reuse covers essentially all of it
Hybrid cooling tower	~150,000 GPD	Full coordination with municipal and reuse sources required

Under the closed loop design, the municipal ask shrinks to domestic and fire service only. The reuse supply and rainwater harvesting cover non potable demand. A modest on-property well, if it completes, carries the entire cooling load independently.

Under the hybrid design, every supply source has to be sized to the full planning figure. The reuse supply conversation, the municipal negotiations, and the groundwater question all get significantly harder.

Locking down the cooling architecture with the mechanical designer before advancing water supply negotiations is the cheapest risk reduction move available to the project team.

Rainwater Harvesting: The Supply Source Texas Law Actively Supports

Rainwater harvesting belongs in every Texas data center water plan, regardless of which primary supply source is selected.

Why It Works

• Rooftop capture does not pump an aquifer or divert a stream, which is valuable where Edwards Aquifer Authority, TCEQ South Texas Watermaster, and groundwater conservation district jurisdictions overlap
• Near zero TDS supports higher cooling tower cycles of concentration
• Texas Tax Code Section 151.355 exempts rainwater harvesting equipment and supplies from state sales tax
• Tax Code Section 11.32 lets local taxing units exempt property with approved water conservation initiatives
• Captured volume may offset stormwater detention requirements

What It Yields

At a design basis of 34 inches of rain per year and 85 percent capture:

• 300,000 square feet of catchment yields approximately 14,800 GPD
• 500,000 square feet yields approximately 24,700 GPD

Under a closed loop cooling design at approximately 25,000 GPD total demand, a large single story data hall plus hardscape can cover the entire non potable load on an annual average basis.

What MES Delivers Before the Deal Closes

MES produced the following for this engagement before a single drill bit turned:

• A complete desktop hydrogeologic screening of 54 wells and the project parcel
• A four source water supply strategy sized to both cooling architecture scenarios
• A defined decision rule for the one recommended test boring
• Identification of the adjacent reuse supply as a coordinated non potable source
• A rainwater harvesting yield analysis with Texas Tax Code documentation
• A recommended next step sequence that gates nothing on the groundwater outcome

The client received a defensible, buildable water plan at a fraction of the cost of a failed drilling program.

Frequently Asked Questions

Does a large Texas parcel in a rural area always have groundwater available?

No. Productive aquifer access in Central and South Texas depends first on which geologic unit sits beneath the parcel. In the study area above, Pleistocene river terrace gravel yielded 100 to 200 gallons per minute to properly built wells, but the terraces track narrow river corridors. Parcels situated on Cretaceous clay uplands are unlikely to produce usable groundwater from the clay, which the Geologic Atlas of Texas shows as hundreds of feet thick in this part of the state.

What does groundwater due diligence cost compared to a dry drill hole?

A desktop hydrogeologic screening of the type MES performs typically costs a small fraction of a single exploratory well. A well that comes up dry in Cretaceous marl is money spent on the wrong answer, and the cost grows with depth and mobilization. The screening identifies whether drilling is warranted before that money is committed.

How does TCEQ reuse authorization work for data center cooling?

TCEQ authorizes reclaimed water use under 30 TAC Chapter 210. The rules set water quality by type of use. Type I covers uses where people may contact the water, such as toilet flushing, fire protection, and irrigation of public parks. Type II covers uses where contact is unlikely, including cooling tower makeup water. Reuse of domestic wastewater falls under Subchapters A to D, and the treatment plant needs its own wastewater permit or authorization. Industrial or commingled wastewater can use Subchapter E, known as 210E. All-domestic wastewater does not qualify for 210E. Either path requires an engineering submittal and TCEQ review. See TCEQ's reclaimed water page. MES has active experience as engineer of record on reuse permitting in Texas and can coordinate both the permitting and the downstream supply strategy.

Does Texas regulate data center water use specifically?

As of October 2026, Texas has no statute written only for data center water use, but the state is acting. Data centers that receive the Texas Water Development Board's annual water use survey must return it under Water Code Section 16.012(m), and on September 14, 2026 the Governor directed TWDB to enforce that duty. The September 21, 2026 TCEQ permit pause and the June 23, 2026 House committee testimony are described above. Changes are anticipated in the 2027 legislative session. Findings and strategies built now should account for that regulatory trajectory.

Can rainwater harvesting realistically meet data center non potable demand?

Under a closed loop cooling design, yes, on an annual average basis. At approximately 25,000 GPD total demand, a large single story data hall plus surrounding hardscape provides sufficient catchment area at Central Texas rainfall averages. Under a hybrid cooling tower design at 150,000 GPD, rainwater serves as a useful supplement but not a primary supply. Storage sizing requires a daily water-balance model on local rainfall records before design.

Work With MES on Your Texas Site

Modern Engineering Solutions provides water and groundwater due diligence, TCEQ permitting, reuse facility engineering, and multi source water supply strategy for private developers, data center operators, and infrastructure investors.

We engage before the land decision closes. We produce clarity before money is committed in the wrong direction.

Licensed in Texas, Colorado, California, Oklahoma, Nevada, Arizona, Florida, Kansas, and New Mexico.

Schedule a water due diligence consultation

Learn more about MES Water Engineering Services

Prepared by Michael Groselle, P.E., Modern Engineering Solutions. Texas P.E. No. 125681. TBPE Firm Registration No. F-20221. mod-eng.com. McKinney, TX and Golden, CO.