Engineering Systems Analysis of Shale Gas Water Management
The objective of this modeling effort is to simulate the characterize the chemical and hydraulic character of flowback water generated during the fracture of shale formations, and how these may influence the planning of gas generation facilities over a forty five year life cycle of individual wells, fields of wells, and plays of multiple fields.
Document Type
Report
Report Type
Topical Report
Report Period
March 2012
Author(s)
Thomas D.Hayes, Ph.D.; Blaine F. Severin, Ph.D., P.E.
Corporate Source
Gas Technology Institute (GTI)
Sponsor
Research Partnership to Secure Energy for America (RPSEA)
Engineering Systems Analysis of Shale Gas Water Management
Barnett and Appalachian Shale Water Management and Reuse Technologies
RPSEA Report No: 08122-05.10
Data collected at 19 wells in the Marcellus shale were analyzed for total salt production, flowback rates, and extrapolations for long term water recovery. The median Marcellus event was used to interpret water recovery during the expected life of a projected play based on the particular needs of gas generation in the Marcellus.
A Microsoft Excel Visual-Basic systems model was developed to incorporate a range of operating criteria on the projected life-cycle of the play. The core model tracks up to 32 wells per field at up to 350 fields per play. Concentrations and flow may be simulated by any reasonable set of parameters, such as the median Marcellus event. The projections are made on a daily basis for up to 30 years from the first completion. The variables include, but are not limited to, individual closure dates, number of wells per field, number of wells per field, number of refractures, installation rate, mobilization rate, days between refracture, number of drilling rigs mobilized, and recovered water rate.
The analyses show that there are three distinct periods in the life of a play; young age, middle age, and old age. The water flow and salt generation during each period will likely dictate a different water management strategy for each period. The magnitude of water and salt generation is exceedingly high and early planning for each period in the life of the play is recommended.
Extrapolations for the life cycle in Barnett are also presented based on the assumption that water and salt generation is similar to the median Marcellus flowback event. Differences in geology and aggressiveness in the development of each play, however, yield very different projections for the Barnett compared to the Marcellus. It is expected that each resource will need a water handling and management strategy tailored to some extent for that resource. Without sound strategies in place, water management has the potential to determine the pace of drilling, the number of refractures implemented, the periodicity of refractures, the life span of water reuse options, and the shut-in dates of individual wells.
