February 2022 TAGD Business Meeting
Study Results: Delineating Injection Well Buffer Zones in Brackish Aquifers
Juan Acevedo, BRACS Hydrologist, TWDB Jack Sharp, Professor Emeritus in Geology, UT- Austin
5 Wondrous Places You Should Visit at Least Once in Your Lifetime (1).pdf
Study Results: Delineating Injection Well Buffer Zones in Brackish Aquifers
1. Study Results: Delineating Areas Designated or
Used for Class II Well Wastewater Injectate
TAGD Winter Business Meeting
Juan P. Acevedo
TWDB – BRACS
February 08, 2022
4. HB 30: Requirements
— TWDB-BRACS Identifies and designates Brackish
Groundwater Production Zones (BGPZs)
— BRACS = Brackish Resources Aquifer Characterization System
— Determine the amount of brackish groundwater
production from BGPZs over:
— a 30-year period,
— a 50-year period; and
— without causing significant impact to water availability or
water quality
— Make recommendations regarding reasonable
monitoring
— to observe the effects of brackish groundwater production
within the BGPZs
5. Statutory Requirements for BGPZ Designation
Must have brackish water In areas of the state with moderate to high availability and
productivity
Must have hydrogeologic
barriers
Sufficient to prevent significant impacts to freshwater
availability or quality
Cannot be within these
boundaries
Edwards Aquifer within the Edwards Aquifer Authority, Barton
Springs-Edwards Aquifer Conservation District, Harris-
Galveston Subsidence District, Fort Bend Subsidence District,
and Dockum Aquifer
Cannot be already in use Brackish water already serving as a significant source of water
supply for municipal, domestic or agricultural
Cannot be used for
wastewater injection
Permitted under Title 2 of Texas Water Code, Chapter 27
[Class II Underground Injection Control (UIC) Wells used for
saltwater disposal (SWD)]
6. Statutory Requirements & Criteria for BGPZ
Class II wells are
potentially
injecting
wastewater
into the
brackish
portions of
Texas Aquifers
7. Aquifer BGPZ Example
- Previous BRACS studies
applied 15-mile buffer to
all class II wells injecting
into aquifer study areas.
- Stakeholders pushed back
on the 15-mile buffer
designation
- too conservative
- scientifically
defensible?
12. Potential Mapping Techniques
1. Analytical solutions (Preferred method):
— Stable
— Easy-to-use
— Simplifying assumptions but exact solutions
— EPA (1994), Bear & Jacobs (1965)
— Applicable on a regional scale
2. Numerical solutions :
— Accommodate complex systems
— Intensive data requirements
— potentially unstable, require advanced users
— Modflow 6
— Applicable on a small/local scale
Current EPA procedure is to
calculate
“Zone of Endangering Influence”
13. — Tier 1 Analysis (no flow direction)
— EPA (1994)
— Bear and Jacobs (1965)
— Compute maximum migration extent
— Tier 2 Analysis (flow direction
available)
— Both gradient and direction of flow
— Bear and Jacobs (1965)
Proposed Methods:
Two-tiered – Analytical Solutions
Injection Well
Potential
Injectate
Migration without
gradient
Potential
Injectate
Migration extent
with gradient
Injection Well
Potential
Extent of
Injectate
Migration
Tier 1
Tier 2
x
x
Ambient Flow Direction
18. Tool Testing: Nacatoch Aquifer
— Original designation (2019)
— 525 Class II wells
—84 SWD wells
—441 EOR wells
— Updated Method (2021 Study)
— 435 Class II wells
—60 SWD wells
—375 EOR wells
— Largest injectate radius
—6 miles
19. Next Steps
BRACS team to apply tools and analyze outputs of injectate
mapping tool and complete QA/QC
1. Apply tools and procedures for Class II injection wells to aquifers
ready for zone evaluation.
2. Provide list of Class II injection wells injecting to formation of
interest to RRC for review and feedback.
3. Conduct the whole aquifer zone evaluation for all other statute
requirements
4. Select buffer distances for Class II injection wells
— Injection mapping tool provides model for injectate migration but
not buffer distance.
— BRACS is developing guidance and will seek stakeholder feedback.
Please contact us to be added to the stakeholder list.
20. —Contact us with comments/questions or to be
added to the stakeholder list
— Juan P. Acevedo, Juan.Acevedo@twdb.texas.gov
—Or visit the study website:
www.twdb.texas.gov/innovativewater/bracs/projects/Injection/index.asp
-
Thank you!
Caverns of Sonora, TX
21. CONCLUSIONS
• Brackish groundwater desalination is being and will continue to be
implemented to meet Texas’ growing water needs.
• There are abundant Class II injection wells that could be used to dispose
of the residual concentrates from the desalination plants – 30,000-100,000
mg/L.
• We propose tier I and tier II analytical models to assess the areas of
review (zones of endangering influence).
• In some cases, numerical models may have to be used.
• Class II wells offer a viable potential solution to disposal of concentrates.
Suggestion
• Underpressured, depleted petroleum reservoirs offer another attractive
potential for disposal of desalination waste fluids.
(Class II EOR wells become Class I wells) JMS
22. Desalination - One Future Solution
• Texas has abundant brackish and saline water.
• Desalination technology is improving; costs of
desalination are decreasing. Reverse osmosis currently
the preferred/most promising process.
• Concerns:
• Location of waters for desalination;
• Effects of fluid production; &
• Disposal of residual fluids
JMS
26. Class II Wells
• There are ~ 180,000 Class II wells in the USA.
• Most are in Texas, California, Oklahoma, and
Kansas.
• Texas has the most (over 53,000, although only about
31,000 are active).
• Produced waters are generally “brines” with a TDS of
~45,000 ppm (DOE, 2006). This can vary from ~30,000
to >100,000 ppm depending upon the source of waters being
desalinated.
JMS
28. Injectate Mapping Techniques
– EPA (1994) assumes no ambient hydraulic gradient
– Bear and Jacobs (1965) includes an ambient hydraulic gradient
Tier 1
Screening level
evaluation
Tier 2
Used when
refining model
32. Summary: Dispersion
• 50% isocontour represents average injectate migration
• Analytical solutions match average injectate migration
• Analytical solutions must consider radial flow from the
injection well and regional flow.
33. Fully penetrating well -
Variable density fluids
Comparing Bear and Jacobs (1965) with numerical model
Denser into lighter Lighter into denser
Same densities
34. Partially penetrating well –
Variable density fluids
Comparing Bear and Jacobs (1965) with numerical model
Denser into lighter Lighter into denser
Same densities
35. • Density may vertically stratify injectate
• Anisotropy plays an important role in the spread of
injectate, needs site-specific information
• Migration underestimated in fully penetrating wells
• Migration overestimated in partially penetrating wells
• More Class II wells are partially penetrating
• Modeling variable-density is complex and is not suitable
for regional-scale studies
• Suitable numerical solutions include:
SEAWAT; USG-Transport; MODFLOW 6, FEFLOW,
SUTRA, HST3D
Summary: Partial Penetration & Density
36. CONCLUSIONS
• Brackish groundwater desalination is being and will continue to
be implemented to meet Texas’ growing water needs.
• There are abundant Class II injection wells that could be used to
dispose of the residual concentrates from the desalination plants
– 30,000-100,000 mg/L.
• We propose tier I and tier II analytical models to assess the
areas of review (zones of endangering influence).
• In some cases, numerical models may have to be used.
• Class II wells offer a viable potential solution to disposal of
concentrates.
Suggestion
• Underpressured, depleted petroleum reservoirs offer another
attractive potential for disposal of desalination waste fluids.
(Class II EOR wells become Class I wells)
JMS
Increasing gap between anticipated demand and existing supply
State water planning process identifies various sources of future water supplies – onto next slide
Brackish groundwater resources are an important component of TX water supplies.
BRACS group at TWDB is mandated to identify brackish groundwater resources in TX
The legislature has provided more guidance on this task with HB 30 – onto next slide
Juan is the expert on this
Juan is the expert on this
Juan is the expert on this – note the large 15 mile radius for all the Class II injection wells
TWDB solicited SOQs to work on developing tools and processes to estimate the extent of injectate mapping
WSP team including SS Papadopulos & Associates and Dr. Sharp was selected for the project
To begin the project, TWDB identified 18 aquifers with potential Brackish Groundwater resources for WSP to assess:
Aquifer parameters
Presence of Class II injection wells
Injection well parameters
Injection well data available from RRC and aquifer data available from TWDB was used to check for:
Presence of Class II injection wells
Potential for presence of Class II injection wells
In this example, no Class II injection wells were found to intersect with the Blossom aquifer
For certain other aquifers such as the Gulf Coast, many Class II injection wells were found
Aquifer parameters relevant for injectate mapping were obtained from TWDB GAMs and other available sources and a default aquifer parameters table was developed. This was later used in the injectate mapping exercise. The graphic is just a snippet from the default table and the entire table will be made available when the report is made public
EPA ZOEI calculations not applicable for this project since we are interested in modeling the full transport of injectate migration and not changes in hydraulic head from mounding
Numerical models were also considered for this project. However, the project team and the workgroup reached a consensus that analytical models are the appropriate path forward. The project team conducted various numerical modeling exercises to confirm that the analytical models perform just as well as the numerical models under various scenarios.
A two-tiered approach was finalized
1. in the first tier the EPA model and Bear and Jacobs model is used to estimate a maximum radius of injectate
2. For certain wells, a flow direction might also be assumed to obtain a more advanced estimate of injectate migration
Some limitations of the analytical modeling approach were identified such as this one:
When multiple Class II injection wells are injecting relatively close to each, they may have an impact on the ambient groundwater flow field and thereby on the migration of injectate from the nearby wells.
In such cases, the migration estimates from the analytical wells might be slightly different from that obtained from numerical models
The workflows discussed earlier to process the RRC databases was automated on an online platform and it provide an output in the form that can be directly edited by TWDB staff and used to check for wells that intersect the BRACS formations
These online tools are available on the FME web server. It contains both data processing tools required in order to download the RRC class II well data base, and to process its information into a usable input for the injectate mapping tool.
Key steps necessary for the well intersection tool is to have a completed BRACS study aquifer surfaces (in raster format) since this tool uses the aquifer surfaces as an input to intersect the wells that are potentially injecting into our aquifers of interest.
*Please note that all these online tools are meant for internal use by TWDB staff and not available for general public use
The injectate mapping tool is also hosted online and requires the input from the two FME tools and aquifer parameters
Default aquifer parameters are available and can be edited by TWDB staff where more recent or better data is available
Here’s a snapshot of the online injectate mapping tool
*Please note that all these online tools are meant for internal use by TWDB staff and not available for general public use
If you remember the initial slide from the 15 mile buffer example, here are the results from the injectate mapping tool – note the vast difference in the estimated extent of injectate migration
Please note that the tool only estimates injectate migration, not the actual buffers
TWDB staff conducts a more detailed analysis to come up with the actual buffers – next slide
Juan is the expert on this
Dr. Sharp has used the results from this study to propose a novel idea and I’ll be handing it over to him to discuss it
These techniques map the injection transport in the subsurface, they are not the actual buffer distances that the TWDB will apply for class 2 injection wells.