Investigation of Water Coning Phenomena in a Fractured Reservoir Using the Embedded Discrete Fracture Model (EDFM)

Daniel Wong, Florian Doster, Sebastian Geiger, Eddie Francot, Francois Gouth

Research output: Contribution to conferencePaperpeer-review

5 Citations (Scopus)
1 Downloads (Pure)


Naturally fractured reservoirs hold significant reserves but are highly heterogeneous and are challenging to simulate flow in. Dual Porosity (DP) methods, although widely used, require fine tuning using production data and thus lack predictive capability in green field applications. The Embedded Discrete Fracture Model (EDFM), which explicitly represents complex fracture network geometries at the cost of computational efficiency, is an excellent tool that can complement the DP method. Using EDFM, we study water coning phenomena in a green fractured Latin American gas field. We do this by simulating gas flow in a stochastically generated sector model representative of the field. EDFM simulations performed using different well rates revealed three possible flow regimes that the field may experience: stable flooding at low flow rates, coning with matrix production at moderate flow rates, and coning without matrix production at high flow rates. These results have implications in field development plans and can also be used for validating any DP models that may be used for full field simulations in the future. Through this study, we demonstrated how EDFM can be used, before any field production, to gain insights into the flow behaviour in a green field.
Original languageEnglish
Publication statusPublished - 6 Jun 2019
Event81st EAGE Conference and Exhibition - Excel Centre, London, United Kingdom
Duration: 3 Jun 20196 Jun 2019


Conference81st EAGE Conference and Exhibition
Country/TerritoryUnited Kingdom
Internet address

ASJC Scopus subject areas

  • Geochemistry and Petrology
  • Geophysics


Dive into the research topics of 'Investigation of Water Coning Phenomena in a Fractured Reservoir Using the Embedded Discrete Fracture Model (EDFM)'. Together they form a unique fingerprint.

Cite this