1 Innovative CO 2 injections in carbonates and advanced modelling for numerical investigation

Innovative CO2 injection strategies are being carried out in ENOS Project (EU H2020 Programme, http://www.enos-project.eu). First results based on field tests conducted at Hontomín, and the advanced modelling developed so far will be analyzed and discussed in this article, as well as, the description of future works. The evolution of operating parameters such as flow rate, pressure and recovery term during the tests confirm the CO2 migration through the fractures.


INTRODUCTION
Most of experiences on CO 2 geological storage worldwide have been conducted in rock formations with high permeability in the pore matrix, mainly in sandstones [Michael et al., 2011] [Krevor et al., 2012] [Torp and Gale 2004] and in some cases in carbonates such as AEP Mountaineer Project [Gupta 2008] [ Mishra et al., 2013], Michigan and Williston Basin [Finley et al., 2013] [ Worth et al., 2014] in USA.There are also CO 2 -EOR projects which injects CO 2 in carbonate formations such as the IEAGHG Weyburn-Midale CO 2 project [Wilson and Monea, 2004] in Canada and the Uthmaniyah CO 2 -EOR demonstration project in Saudi Arabia [Liu et al., 2012]   The design of safe CO 2 injection strategies and the understanding of trapping mechanisms in carbonates with poor matrix porosity and fluid transmissivity through the fractures are challenging matters so far.To give a proper solution, the study of hydrodynamic and mechanical effects induced by the CO 2 plume migration in the fractures, and those ones produced by the geochemical reactivity due to the acidification of reservoir water, is needed to increase the knowledge on the behavior of these reservoirs for CO 2 geological storage and later industrial deployment [de Dios et al., 2017].
Hontomín Pilot Plant [Neele et al., 2014], operated by Fundación Ciudad de la Energía (CIUDEN), is the only current onshore injection site in Europe for CO 2 geological storage, recognized by the European Parliament [EP resolution 2014] as key test facility for CCS technology development.The pilot is located close to Burgos in the north of Spain, and its reservoir is comprised of fractured carbonates with poor matrix porosity [Campos et al., 2014].
To demonstrate innovative injection strategies and history matching approaches for increased confidence of operators in safely managing sites is a priority within ENOS Project (EU H2020 Programme, http://www.enos-project.eu).It is expected to increase the understanding on CO 2 injection in fractured carbonates with low primary permeability, and to develop safe and efficient operational procedures using real-life experience from running the Hontomin pilot [Gastine et al., 2017].CO 2 injection in rock formations with main fluid transmissivity through fractures usually requires high values of pressure, which means a risk for the pair seal-reservoir integrity [Vilarasa et al., 2014].On the other hand, as mentioned above, the geochemical reactivity due to reservoir water acidification impacts on the carbonate permeability [Gaus et al., 2015].These matters must be considered to design safe and efficient injection strategies in ENOS project to improve the hydrodynamic stability and control of storage integrity.First injections conducted at Hontomín using synthetic brine and CO 2 will be described in the article, analyzing the evolution of operational parameters and discussing the results.
To model the CO 2 migration through the fractures and predict the injection effects in the carbonates is a challenge as well.An advanced modeling workflow with FracaFlow™ used to elaborate a Digital Fracture Network (DFN) [Bourbiaux et al., 2005] around the injection well and characterize the main properties of the fracture network will be described in the article.The dynamic characterization of fracture properties was then performed using an advanced automated history matching with CMOST™ to model the pressure behavior around the injection well based upon a previous modelling work [ Le Gallo et al., 2017].
Taking into account the first results from injections and the modelling developed so far, the authors will describe the planned works to be conducted in ENOS project in order to find solutions based on real life experiences.
The results from first injections performed at Hontomín site confirmed the singularity of this reservoir where CO 2 migration is through the carbonate fractures.The evolution of main operational parameters such as well head pressure (WHP), flow rate, bottom-hole pressure (BHP) and distributed temperature along the well tubing confirm the injection of CO 2 in liquid phase.Taking into account the information provided by the first results, it is necessary to determine the long term evolution of BHP and flow regarding the injection strategy used and particularly the pressure recovery period during the fall-off phase according to the cumulative amount of CO 2 injected on site.

DESCRIPTION OF PILOT PLANT
Hontomín site represents a structural dome where the pair seal-reservoir is located within Jurassic Formations (Marly Lias and Sopeña respectively).Overburden is formed of Dogger, Purbeck and Weald and the underlying seal is located at Triassic Keuper [Rubio et al., 2014].Figure 1 shows the lithological column of Hontomín site and the geological cartography of the area.The main challenge faced during Hontomín hydraulic characterization was the low injectivity existing on site.The injection of brine and CO 2 to characterize the pair seal-reservoir produced geomechanical changes and geochemical reactivity effects that improved the permeability in the fracture network while the matrix does not appear to significantly contribute to the storage capacity for the time being [de Dios et al., 2017].
Hontomín is at the early injection phase, thus, all long term effects that condition the safe and efficient CO 2 geological storage must be determined and analyzed.

INNOVATIVE CO 2 INJECTION STRATEGIES
It is planned to inject on site up to 10 000 metric tonnes of CO 2 during the period 2016-2020 in ENOS project, with the purpose of better knowing the behavior of this tight fractured reservoir, mainly what concerns the improvement of hydrodynamic stability and control of storage integrity, for finding safe and efficient operation conditions.
Therefore, the design of CO 2 injection strategies to be conducted in the project must be based on criteria of efficiency and safety, for later up-scaling to industrial deployment.The operating procedures must ensure efficient energy consumption, maximizing reservoir capacity and preserving seal integrity [Gale et al., 2001].
As mentioned before, CO 2 injection in carbonate reservoirs with low matrix permeability shows specific features that are different from injection in porous media.Considering the CO 2 migration is dominated by the fracture network, the following gaps need to be studied in order to define proper strategies for the injection: • Bottom-hole pressure (BHP) evolution and its influence in the cap-rock integrity and reservoir behavior • Bottom-hole temperature (BHT) evolution and the analysis of thermal effects due to injection • Monitoring CO 2 evolution along the well tubing and the fluid density reached at the bottom hole.• Energy consumption and operation performance for each planned injection Most of existing experiences worldwide were based on continuous flow of injected CO 2 [Nordbotten et al., 2004], but taking into account that in future storage operations at industrial scale unexpected effects may occur, such as high pressure values reached at the reservoir that put in risk the seal integrity and/or low injectivity that conditions the process efficiency, it is needed to design and test injections adapted to these scenarios.
First injections conducted at Hontomín in ENOS were developed in discontinuous process, what supposed to inject brine or CO 2 in periods of 8, 24 or 48 hours in order to assess the evolution of pressure and flow, and to determine the term for BHP recovery during the fall-off periods.These discontinuous strategies aim at studying the improvement of hydrodynamic stability in the fractured reservoir.
The use of brine and CO 2 seeks to compare the results with each fluid, with different hydraulic properties as density and viscosity.Usually, first tests are conducted with brine in order to refine the design and performance of CO 2 injections according to the prior results.
High pressure values are necessary to inject CO 2 in Hontomín site, with longer pressure recovery periods than media with transmissivity dominated by matrix porosity.Unexpected reservoir behavior could take place during the injections planned in ENOS project, conditioned to the cumulative amount of CO 2 injected on site and the bottom hole pressure evolution.That means it is needed to analyze the evolution of the injected flow maintaining a safe BHP value and also the evolution of the pressure values while injecting CO 2 at constant flow rate.
In order to address these goals two types of tests were designed: • Injection tests in pressure control mode • Injection tests in flow control mode In the injections monitored by pressure, the setpoint corresponds to the well head pressure (WHP) value in the range of 60-80 bar.In CO 2 injection case, the setpoint is always equal or higher than 75 bar to ensure the liquid phase of injected fluid, taking into account the temperature at the well head (WHT) is usually set in the range of 10-30º C.
On the other hand, in the tests that use flow rate as control parameter, the setpoint is in the ranges of 1-2 kg/sec using CO 2 and 1-3 kg/sec for brine, which are consistent with the existing capabilities of CO 2 injection and water conditioning facilities.
Figure 3 shows the evolution of BHP data from two P/T sensors (see figure 2) located in the open hole of HI well, WHP in the injection well, BHP from the P/T sensor located in the seal of HA well (see figure 2) and flow rate of brine and CO 2 during 8 hours of CO 2 injection in flow control mode (setpoint 2 kg/sec).Despite being a short injection, since the operation period was 8 hours, as CO 2 permeates and push the reservoir water through the fractures it was needed a final incremental of well head pressure close to 30 % from initial value to maintain the flow rate value of 2 kg/sec along the test.Therefore, future injections must prove if pressure evolution follows the trend showed in the short term or not, being necessary to conduct tests during longer periods in similar conditions.
The main goal is to assess if the final BHP of 180 bar, considered a safe value to preserve the pair seal-reservoir integrity, is reached, and to determine additionally the period of time needed, the operational conditions and the cumulative amount of CO 2 existing on site when it occurs.
BHP evolution shown in figure 3   On the other hand, this test was conducted once several injections were carried out using firstly brine and subsequently CO 2 as mentioned above, with recovery periods during the fall-off phase in the range of 48-60 hours for a pressure recovery range of 95-100%.New tests aim to analyze how BHP recovery will be regarding the final pressure value reached in the reservoir and the cumulative amount of CO 2 injected on site.
Figure 4 shows the evolution of the operating parameters for 24 hours of CO 2 injection in pressure control mode (set point 80 bar).These data are: BHP from HI P/T sensors, WHP in the injection well, BHP from P/T sensor located in the seal of the HA well and the flow rate of brine and CO 2 .analyzed above, hydrodynamic effects induced by the injection in the fractures play a key role to understand the reservoir behavior.It is also needed to conduct tests during longer periods in similar conditions for checking if the flow evolution follows the trend showed in the short term or not, and if safe BHP value of 180 bar is reached and when it happens.
This test was the last one of eleven injections, six of them performed with 430 m 3 of brine and the rest with 490 metric tonnes of CO 2 .A period of 15 days was needed to recover 98% of initial BHP value, while in similar tests conducted with brine the recovery period was close to 10 days.This fact may be due to supercritical CO 2 is a compressible fluid affected by the closure of fractures during fall-off periods when the injection stops, what conditions the biphasic migration of CO 2 and brine through the fracture networks.On the other hand, as much CO 2 is injected on site BHP increases in the reservoir and longer periods for pressure recovery are needed.
As regards the thermodynamic parameters along the injection well and open-hole that determine the operation performance and thermal effects into the reservoir, figures 5 and 6 show respectively the thermal profile along HI tubing and BHT evolution during the 24 hours injection test in pressure control mode.The graphic at the top of figure 5 shows the temperature values along the HI well tubing previously the injection.The graphic in the bottom corresponds to values recorded at the final phase of CO 2 injection.The data were recorded by a Distributed Temperature Sensing System (DTS) [Hilgersom et al., 2016] anchored to HI well tubing (see figure 2).
In both graphics, the range of temperatures corresponding to the distances between 0 and 200 m belongs to the part of DTS optic fiber buried close to surface to connect the control room with HI well head, and therefore, these data correspond to environmental temperature and do not have to be taken in consideration.
Graphic at the top of figure 5 shows homogeneous thermal gradient of the existing brine in the annular space between the tubing and the casing/liner, which could be quite similar to geothermal gradient.Hontomín reservoir is a "cold" formation taking into account the BHT value was 44º C prior to injection.
On the other hand, the graphic at the bottom of figure 5 shows the temperature evolution along the tubing once 22 hours of CO 2 injection were conducted on site.The heating and subsequent cooling for the distance of 1200 m (1000 m MD depth in HI well), correspond to effects induced by the choke [van der Zante and van der Broek, 1998] located in this position of well tubing mentioned above.Initially CO 2 heating is produced due to friction between fluid and the wall of choke and finally the cooling due to the Joule-Thomson effect produced by the CO 2 expansion at the choke exit [de Dios et al., 2017].
The BHP value of 16O bar (figure 4) and temperature along the tubing (bottom of figure 5) reveal that the injection was performed with liquid CO 2 , reaching a fluid density at HI bottom hole of 0,828 t/m 3 at the end of the test.
As regards the thermal effect produced in the saline aquifer due to CO 2 injection, figure 6 shows the evolution of BHT data from P/T sensors of HI well, WHT in the injection well, BHT in the sensor located into the seal of HA well and flow rates of brine and CO 2 for the 24 hours of CO 2 injection test in pressure control mode.The gradual cooling of reservoir water happened for the 24 injecting hours from 44º C, that was the BHT initial value, to 33º C taking into account the wellhead maintained an averaged temperature value of 10º C during CO 2 injection, as figure 6 shows.Relevant temperature decrease at HI bottom hole occurred due to the injection of brine previously to CO 2 injection in order to pressurize the well head to 75 bar, and subsequently to clean up the tubing when CO 2 injection finalized, as mentioned above, while BHT in HA remains constant at the seal what proves its integrity.
The injection tests are being analyzed from the hydraulic characterization phase with a compositional dual media model which accounts for both temperature effects and multiphase flow hysteresis [Le Gallo et al., 2017], to effectively simulate the alternating brine and CO 2 injection tests that were and will be conducted on site.Next point adresses the modelling developed so far to characterize the fractured carbonates at Hontomín site for CO 2 geological storage.

MODELLING OF CO 2 MIGRATION THROUGH FRACTURES
The The DFN is then upscaled into the reservoir grid inducing an anisotropy of fracture permeabilities due to the orientation of the two sets of fractures.In this approach, there are significant uncertainties regarding the various parameters of the fractures such as aperture and permeability which strongly influence the model response.
Therefore, an advanced uncertainty workflow is used to best tune the model to the well conditions.The modelling followed a sequential approach: first matching the single phase parameters such as fracture permeability during the brine injection periods and then matching the two-phase parameters such as fracture relative permeability during the brine and CO 2 injection periods.
In order to minimize the number of model parameters, e.g.avoiding modelling the flow behavior within the well, the well flow rates are modeled at reservoir conditions aiming at matching the measured bottom-hole pressure.The well parameters are acquired every minute and would consequently limit the simulation time step, thus the well parameters are averaged on an hour basis while ensuring consistency of mass balance and synchronicity of model with well operations.

MODELLING SINGLE PHASE BRINE INJECTIONS PERIODS
Based upon the previous modeling work [ Le Gallo et al., 2017], several single phase parameters are considered within the CMOST™ uncertainty modelling workflow to best enable pressure matching: • Multipliers for fracture permeabilities in each of the 3 directions of the grid which vary several orders of magnitude with uniform distributions • Multiplier for matrix permeabilities (identical in each of the 3 directions of the grid) which varies one order of magnitude with an uniform distribution since the matrix permeability was previous matched [Le Gallo et al., 2017 The uncertainty model CMOST™ investigates the parameter domain based upon a Latin Hypercube Sampling approach and converges towards its optimum after about 80 different simulations as shown in figure 7. Beyond the initial 100 simulations, no significant improvement of the match (cumulative difference between simulated and measured BHP for HI well) is obtained and the residual relative error is about 7% as shown in figure 8 which mainly corresponds to the slight mismatches during the fall-off periods.The single phase match is quite satisfactory and the most influential parameters, shown in figure 9, are: 1. Multiplier for fracture permeability in the I direction (north-south) 2. Multiplier for fracture permeability in the J direction (east-west) 3. Multiplier for fracture porosity The brine single phase flow in the carbonate reservoir is mainly influenced by the fracture characteristics (permeability and porosity) and to a lesser extent by the matrix ones.The optimal match shown in figures 7 and 8 is then used as base case to model the two phase CO 2brine injection periods.

MODELLING TWO PHASE BRINE & CO 2 INJECTION PERIOD
As flow in this naturally fractured carbonate reservoir is controlled by fracture characteristics, the two-phase parameters selected are related to the fracture relative permeability for the CO 2brine as shown in figure 10 The two-phase parameters adjusted within the CMOST™ uncertainty modelling workflow are the following end-points (see figure 10) of the relative permeabilities for the fractures: • The maximum of the CO 2 relative permeability, KRGRL, which may vary between 0.01 and 1 with a uniform distribution • The maximum of the brine relative permeability, Krlmax, which may vary between 0.75 and 1 with a uniform distribution • The CO 2 critical saturation, OneminusSGcrit, which may vary between 0 and 0.5 with a uniform distribution • The brine critical saturation, SLrg, which may vary between 0.1 and 0.25 with a uniform distribution The ranges of variation of these parameters are determined based upon the previous modelling work [ Le Gallo et al., 2017].Using the best results from the single phase modelling (base case in this workflow), the uncertainty model CMOST™ investigates the parameter domains based upon a Latin Hypercube Sampling approach and converges towards its optimum after about 30 different simulations as shown in figure 11.Beyond the initial 50 simulations no significant improvement of the match is obtained and the residual relative error is about 4.5% as shown in figure 12, which mainly corresponds to the mismatches during the fall-off periods.The twophase match is quite satisfactory and the most influential parameters, shown in figure 13, are: 1.The maximum of the CO 2 relative permeability, KRGRL, which optimum is at about 0.035 2. The CO 2 critical saturation, OneminusSGcrit, which optimum is at about 0.49 3. The maximum of the brine relative permeability, Krlmax, which optimum is at about 0.769 The optimal solution after the history matching is shown in figure 11.The main mismatches between the field and simulation occur during the fall-off period as the simulated pressure always returns faster than the measured pressure to its base level.An attempt to mitigate this discrepancy was to account for the uncertainty to rock compressibility and to numerical dispersion due to grid size.The current results significantly improve the previous modelling results [ Le Gallo et al., 2017].However, the influence of the compressibility of the various rock facies is not significant as shown in figure 9. -Optimal history-matched bottom-hole pressure at the injection well HI during the single phase brine and two phase CO 2 -brine injection periods

ANALYSIS OF RESULTS
During the injection tests, the CO 2 migrates about 80 meters away from the injection well as shown in figures 15 and 17.The impact in temperature is limited to the near wellbore region as shown in figures 16 and 17 due to thermal inertia of the reservoir rock.The pressure disturbance is extending further away from the injection well as shown in figure 17.The local pressure increase only extends as far as the CO 2 and relaxes quickly in the model.The pressure relaxation at the well (see figure 14) is always quicker than in the pilot.Consequently, there is still on-going work to match the current injection tests.

DISCUSSION AND FUTURE WORKS
As mentioned above, the injection tests conducted so far in Hontomín for ENOS project aim to better understand the fractured reservoir behavior in order to improve its hydrodynamic stability.Those ones conducted in pressure control mode revealed the decrease of injected flow to maintain constant the WHP value.On the other hand, tests conducted in flow control mode shown the pressure increases at the well head to hold a constant value of flow rate.The matter to discuss is if these results correspond to wellbore effects in the short term or they set a trend on long term behavior of pair seal-reservoir.
Something similar happens with the pressure recovery period during the reservoir fall-off phase, which plays a key role to determine the reservoir stability [Eiken O. et al., 2011].The terms of BHP recovery depend on the injected fluid (brine/CO 2 ), due to different hydraulic properties of each one, and on the cumulative CO 2 injected on site.As happened in the study cases for the tests conducted in pressure and flow control modes described above, it is necessary to demonstrate if this recovery trend is due to short term effects occurred in the well vicinity or corresponds to the long term behavior of reservoir.context, the model was set up as a dual-permeability and porosity tacking into account the temperature effects and relative permeability hysteresis.The bottom-hole pressure match is quite satisfactory as shown in figure 14.
It is planned to perform continuous injections at Hontomín site during several days hereinafter to address all mentioned gaps, for gaining knowledge on the management of operational parameters to control the storage integrity for long term operations, and particularly, for setting of pressure recovery periods to ensure a safe re-start of injection.On the other hand, alternative strategies using cold CO 2 (temperature below 10ºC) will be designed and tested, with the aim of finding efficient operational parameters that are consistent with the safety of the process.As mentioned above, there is still on-going work to refine the modelling developed so far to investigate how the CO 2 plume migration and trapping mechanism will be.

CONCLUSIONS
Results from first injection tests conducted at Hontomín site within ENOS project confirmed the singularity of this reservoir where CO 2 migration is through the carbonate fractures.Thus, when pressure remains constant at well head with a value equal or higher than 75 bar for ensuring liquid injection, flow rate decreases considerably as CO 2 is expanding within the fracture network.On the other hand, when flow value is constant during injection the well head pressure highly increases as much CO 2 is injected on site.
As regards the period of time necessary for pressure recovery on the bottom hole during the fall-off phase, it depends on the injected fluid due to different hydraulic properties, and the cumulative amount of CO 2 existing on site.
Regarding the thermal profiles corresponding to injections, liquid CO 2 phase is ensured along the tubing which corresponds with an efficient operation, reaching fluid density values close to 0,83 t/m3 at the bottom hole.
The modelling performed so far shows a small lateral extension of the CO 2 away from the injection well during the injection tests with an even smaller extension of the temperature disturbance.As for most of the CO 2 storage projects in aquifers, the main impact is the extension of the pressure disturbance which drives the CO 2 .The future modelling work shall focus upon the pressure fall-off periods which appears faster in the model than in the field.

ACKNOWLEDGEMENTS
ENOS project has received funding from the European Union´s Horizon 2020 research and innovation programme under Grant Agreement Nº 653718-ENOS.
Authors acknowledge the Innovation and Networks Executive Agency (INEA) for the funding, as the role played by the coordinator Bureau de Recherches Geologiques et Minieres (BRGM) and the rest of partners of the consortium.
Sole responsibility lies with the authors, so the European Commission is not responsible for any use that may be made of the information contained in this article.

AUTHOR CONTRIBUTIONS
Dr. J.Carlos de Dios is Director of the Research Program for CO 2 Geological Storage in Fundación Ciudad de la Energía (CIUDEN) and member of the Management Board of ENOS project.He is

Fig 1 .
Fig 1.-Lithological column and geological map of Hontomín area

Fig 2 .
Fig 2.-Schemes of injection well HI (left side) observation well HA (right side) and panoramic view of the pilot

Fig 3 .
Fig 3.-Evolution of operating parameters for 8 hours of CO 2 injection in flow control mode reveals a final overpressure of 15 bar from the hydrostatic value existing previously to the injection start.At the beginning and end of the test WHP reached the values of 75 and 98 bar respectively with BHP values of 144 and 159 bar.Pressure incremental at the well head is not proportional to the overpressure at the bottom hole, which is due to a choke was installed in the HI well tubing to ensure the integrity of the pair seal-reservoir[de  Dios et al., 2017]  what produced a pressure drop at the bottom hole during the injection.

Fig 5 .
Fig 5.-Thermal profile of HI well tubing previously (top) and during CO 2 injection (bottom)

Fig 6 .
Fig 6.-Evolution of BHT and WHT during 24 hours of CO 2 injection in pressure control mode

Fig 7 .Fig 8 .
Fig 7.-Evolution of objective function characterizing the error between model and well measurements with the number of simulations

Fig 9 .
Fig 9.-Most influential parameters on objective function characterizing the error between model and well measurements

Fig 10 .
Fig 10.-CO 2 -brine relative permeability in the fracture [Le Gallo et al., 2017]and end-points used in history-matching

Fig 11 .
Fig 11.-Evolution of objective function characterizing the error between model and well measurements with the number of simulations

Fig 17 .
Fig 17.-Evolution of pressure (top), temperature (middle) and CO 2 saturation (bottom) profiles as a function of the distance from the injection well at the top of Sopeña formation

CO 2
injections were performed in liquid phase that is the most efficient operation.However, the injected fluids have cooled the pair seal-reservoir that could produce impacts that would put in risk the integrity of rock formations [Gor andPrévost 2013].The heat transfer between the reservoir rock and the biphasic fluid needs to be investigated in order to test the viability of pair seal-reservoir for CO 2 storage[Somaye et al., 2010].Particularly, the changes induced in geomechanics of rock massif and fractures must be studied to better understand the behavior of these reservoirs during the injection, preserving the seal integrity.The modelling work was performed as an initial characterization work in order to tune the model to injection tests conducted during the hydraulic characterization phase [de Dios et al., 2017].The model was based upon previous work [Le Gallo et al., 2017] and updated to account for a recent characterization of geological heterogeneities which was performed within ENOS project [Le Gallo and de Dios, 2018].In order to account for the various operations and the geological

Table 1
Operating parameters for 8 hours of CO 2 injection in flow control mode (2 kg/sec)

Table 2
Operating parameters for 24 hours of CO 2 injection in pressure control mode The reservoir behavior is different depending on each injection phase.Thus, initially a flow of 2.2 kg/sec injected into the reservoir was needed to hold the WHP value of 80 bar.As the fluid fills the fractures, this value decreases up to reach 1.7 kg/sec at the end of the test, what suppose 23% less than the initial flow value.As happened in the injection conducted in flow control mode ] • Multiplier for fracture porosity which varies 50% with an uniform distribution since the base model was previous matched [Le Gallo et al., 2017] • Multiplier for matrix porosity which varies 50% with an uniform distribution since the base model was previous matched [Le Gallo et al., 2017] • Compressibility of matrix and fracture medium which vary several orders of magnitude with uniform distributions to best improve the pressure response during fall-off periods