GIN method applied to the consolidation of cracked rocks: case study of the Memve’ele hydroelectric dam (southern Cameroon)

The empirical GIN (grouting intensity number) method is one of the most popular methods for structural stabilization and is increasingly used in many projects. The concept of this method is to limit the combination of pressure and volume injected to a specific grouting intensity number in order to control the energy induced in the rock fractures and to avoid heaving. In order to ensure the water tightness of the foundation of the Memve’ele hydroelectric dam, the treatment by the GIN method was implemented. The protocol implemented consisted to carry out a veil of injection drilling, consolidation, and control of effective treatment on each geotechnical hazard zone. Thus, 1240 injection boreholes were opened over a total length of 1600 m, i.e.: 613 pre-treatment boreholes and 627 consolidation boreholes. The results of the monitoring tests show that the 0.7 cement grouting treatment contributed to a reduction in crack permeability to less than 5 Lu (Lugeon units), a reduction of over 94%. This confinement of the rock was confirmed by geophysical imagery whose signature at the fractured interval varied after treatment. Geophysical and geohydrological characterization through permeability testing and electrical tomography revealed the existence of highly permeable discontinuity zones (over 90 Lu) below the bedrock surface, up to 15 m wide. Notwithstanding the volume of drilling carried out, the results obtained support the use of the GIN method in the stabilization of foundations of structures in the granite-gneissic basement zone.


Introduction
Controlling grouting pressure is vital to the success of any grouting operation, high pressure can result in grout spreading into areas beyond any possible usefulness or damage to structures by displacing rock in their foundation, although deeper penetration into any opening that can be grouted is beneficial (Rafi et al. 2015;Bisso et al. 2020). Thus, pumping pressure control is recommended, especially for near-surface fracture injection. However, according to authors (Willson 2012;Tsuji et al. 2012), the US practice has routinely injected at a lower pressure compared to other methods. Thus, as a precaution, its injection pressure increased was suggested so that the inspector can recognize the onset of uplift.
In practice, the pressure was sometimes adjusted conservatively due to project limitations. According to Steyn and Mouton (2012), in a dam construction project in South Africa under variable geological conditions, strict control of leakage beneath the dam was necessary due to the limited size of the catchment. In order to allow the use of high pressures to inject the desired volume of grout into the rock, the pressure applied was limited to half the pressure of the cover while avoiding damage to the rock structure. Different methods are used in the construction of dams, among them the GIN method which has been used in many projects.
The grouting intensity number (GIN), defined as the combination of pressure and volume, has been used in many projects (Lombardi and Deere 1993;Ladiges et al. 2010;Kettle et al. 2012;Rafi et al. 2015;Usman et al. 2017;Shahzad et al. 2017;Nakamura et al. 2021;Ntomba et al. 2021). By applying this method, they found that it is very effective in improving the injection operations by gradually reducing the pressure as the cumulative volume increases. However, the GIN must initially be selected based on experience, literature, and understanding of geology.
As part of the development of its energy sector, the Government of the Republic of Cameroon is committed to developing a number of hydroelectric projects, such as the Memve'ele hydroelectric development on the Ntem River. According to Schleiss and Pougatsch (2011), the construction of a dam requires specific studies and treatment of the foundations to enable the structure to cope with the considerable forces to which it may be subjected over time. This work requires, on the one hand, a good knowledge of the deformation of the supporting terrain and, on the other hand, the choice of methods to ensure the waterproofing and stability of the structure Milanović 2021).
Applications of the GIN method on soils and formations are widespread. In order to determine if this method can be applied to all geological environments, GIN treatments were applied to the rock foundation of the Memve'éle hydroelectric dam. The main dam of this structure was built on gneissic bedrock with a ring structure, with anastomosing joints. The treatment for the reduction of the permeability and the increase of the cohesion of this massif consisted in works of injection of cement slurry by the GIN method. In order to monitor and assess the effectiveness of this treatment, the geological conditions of the site were evaluated through permeability tests and geophysical measurements by tomography before and after injection. This article presents the results obtained from the treatment of the Memve'ele hydroelectric dam and proposes to the industrialists, as well as to the scientific community, a useful information basis for the design of stability studies of civil engineering structures.

Geological setting
The Memve'ele dam is located in Central Africa, between latitudes N 02° 15′ and N 02° 31′ and longitudes E 10° 15′ and E 10° 30′, in the lower Ntem basin (Fig. 1). Its catchment area is 26,350 km 2 and includes numerous waterfalls, including the one named after the locality. The Memve'ele waterfalls, with a height of about 35 m, offering favorable conditions for the development of a hydroelectric power plant. The regional geology consists mainly of Archean Paleoproterozoic and plutonic rocks, the oldest of the Precambrian basement, and which, in southwestern Cameroon, form the northern edge of the Congo Craton (Owona et al. 2011). Previous work shows that the regional geology consists in detail of three litho-tectonic units, namely the Nyong, Ayina, and Ntem. The rocks are well differentiated but are similar in petrology, metamorphism, tectonics, and geochronology (Owona et al. 2011;Ntomba et al. 2020;Chougong et al. 2021). The geological context suggests that the formations encountered are largely composed of gneisses and pyroxene-hornblende granitic rocks that derive from metamorphosed Precambrian sedimentary formations. They are characterized by the development and distribution of faults and schistosity (Owona et al. 2011;Bisso et al. 2020).
The foundation of the main Memve'ele dam is located in a valley about 1200 to 1500 m wide with low hills on both sides. The highest hills are on the right bank at an altitude of 400 m while the highest parts on the left bank rise to about 420 m. The local geology consists of granulite gneiss (migmatite) dating from the Precambrian era with diabase intercalations and loose overburden of Quaternary terrain (Fig. 1). The rock is hard but crossed by several families of joints whose most important replicas are of direction N 135°, N 45° Ntomba et al. 2020;Ntomba et al. 2021). Boreholes drilled show the presence of sound   granitic gneiss on the left bank at an altitude of 393 m, and falls on the right bank at 375 m, i.e., 6 to 8 m below the present riverbed level. The riverbed is generally covered with a medium layer of sand (Fig. 1).
Both banks are covered with thick laterite that is fairly dense and residual yellow to brownish clay with some detritus, ranging from 8 to 14 m deep on the right bank and reaching up to 20 m on the left bank. Marsh deposits exist mainly upstream of the left bank and in the shallows. They consist of deposits of clayey silt, organic clay, and fine sand Ntomba et al. 2020).

The GIN method
In general, with the GIN method, the energy acting in an injected rock is approximately proportional to the product of the final injection pressure and the injection volume (Lombardi and Deere 1993). This energy can cause hydrofracturing which must be limited. The product of pressure and volume should not exceed a given GIN to avoid certain problems. According to the pressure-volume graph presented by Lombardi and Deere (1993), the trajectory of the injection will touch the GIN curve at different points depending on the quality and tightness of the rock mass, i.e., the size of the fracture opening. It can be used to choose the distance between boreholes in the application of the split spacing method.
Although the GIN method has been a significant advance towards more efficient and economical injection, various limitations and ambiguities have been introduced. The uncertainty associated with GIN selection makes it difficult to establish stopping criteria based on it. According to Lombardi and Deere (1993), injection should be stopped when one of the limits is reached, i.e., maximum pressure, maximum volume, or hyperbole. However, El Tani (2012) suggested that injection should continue after reaching the hyperbola, using decreasing pressure until the point where there is no more grout flow. This point is on the so-called ZFP (zero flow paths) curve, which is the combination of pressure and volume at which grout flow is zero (refusal). Thus, grouting is completed at the intersection of the ZFP curve and the GIN hyperbola (Fig. 2). In this study, the GIN was described based on the grout propagation radius.

Collection of data and treatments
The structural analysis allowed us to identify 2076 closed and open joints with centimetric to metric lengths along the route of the main dike. This analysis consisted of a systematic characterization of the joints, including their description and measurements of direction, dip, azimuth, length, and opening. For this, it was based on visual assessments, photographs, and compass measurements.
The Lugeon test (NF P94-131) is a test put in place to evaluate the possibility of water circulation in the soil and to detect heterogeneities or fissures. It consists of injecting water under pressure into a cavity consisting of a portion of the borehole of known dimensions and measuring the injection rate for different pressure levels, for a given time. The equipment used consists of drilling machines; a sealing system; a water injection device; and a measuring system. The techniques used consisted to (i) carry out inside the rock, by extraction, a cavity of diameter 76 mm, then to connect this cavity to the surface of the ground by an injection tube. The cavity consists of a portion of drilling between the bottom and an obturator which limits it at the top; (ii) produce and maintain a constant hydraulic load inside the cavity by injecting water under pressures varying between 0.3 and 0.5 MPa, or even 0.7 MPa; and (iii) measure every minute the volume of water injected in the cavity as a function of time, i.e., 5 min for each pressure level.
The geophysical survey allowed to visualize the deep structure and the potential anomalies of the dam foundation before and after treatment with 0.7 cement grout. For this study, a 2D electrical resistivity tomography was implemented. The measurements were carried out along 12  (2012) and Zero flow path. ξ is the advance ratio which is equivalent to relative penetration proposed in Gustafson and Stille (2005). Advance ratio of 1 indicates ZFP curve where grout is at rest straight longitudinal profiles arranged along the axis of the dam or at 90°. The measuring device consisted of a flute with 48 electrodes. The electrodes were spaced laterally by 2 m in order to cover the area of interest evenly. Each profile was measured separately using the Wenner equipment. Also, more than 600 measurements were made for dipoles from 2 to 20 m long in nine levels. A total of 4200 dipoledipole measurements were performed on all profiles. The location of the electrodes was carried out at the total station, on each profile, and with an uncertainty lower than 10 cm. The resistivity measured on each profile was edited in Prosys II to eliminate outliers and jerks due to variations in resistivity of the surface coating. For further processing, the RES2DINV inversion software was used. Also, to consider the distribution of resistivity at the right of each profile, the "Robust" configuration, which minimizes the absolute value of the deviation of the measurements, appeared to be the most suitable for the data inversion.
The purpose of the injections is to seal or consolidate a cracked rock mass. The voids are filled by grout introduced by pressure and brought by drillings which cross the cracks and the discontinuities in the rock mass. In order to optimize the treatment of the main dam foundation, 1240 grouting holes were opened as shown in Table 1.
These injections were carried out in two phases, namely, the injections of the axial sealing veil of the dam and those of consolidation. The injections of the veil make it possible either to carry out veils intended to reduce the flows of percolation through the foundation of the dam, or screens under the foundations against the risk of pressure of heave or foxing, or to improve the impermeability of the bottom of the reservoirs and to increase the resistance of the rock mass. Those of the foundation of the main dike of Memve'ele were carried out by alternating primary, secondary, tertiary, and even quaternary drilling, slightly inclined to the north of 10° from the vertical (towards the secondary spillway) while remaining always in the plane of the veil. These injections into the rock were carried out at a spacing of 8 m for the primary holes and 4 m for the secondary, tertiary, and quaternary holes with depths varying between 15 and 25 m.
Consolidation injections were carried out in the highly fractured zones following a 3-m grid, in depths ranging from 3 to 5 m. These boreholes, all vertical, were injected in a single up-hole.
The aim of this method is to fill the voids existing in a rocky or loose mass by introducing, by pressure through perforations, a "cement slurry" product that will set; the properties of the injected rocky mass must be modified in the desired direction.
The main idea is to use, during the work, a constant injection intensity defined by the product. The formula used is that of Lombardi and Degree (1993).
where P is the final grouting pressure (in bar) and V is the final volume of grout injected (in l/m). The GIN value is calculated at zero flow rate, i.e., with the grouting pumps switched off.
The grouting pressure should be as high as possible, in order to increase the range of action, but at the same time low enough to avoid unnecessary hydrofracturing, hence, the concept of "injection intensity." The treatment zone is defined in advance by the average distance reached by the grout, called the radius of action or radius of the grouted zone, noted R with The GIN injection method is defined by limit envelope curves that take into account the following three factors: maximum pressure (Pmax), maximum absorption (Vmax), and limit intensity (Fig. 3). As the pressure varies with depth, grouting is stopped when the product P×V reaches a fixed limiting intensity.  The type of grout used is a stable, dense, limited-settling cement grout with high final mechanical properties and resistance to washout. It is formed by a mixture of cement, water, and admixture. Following the results of the suitability tests, in relation with the requirements of the CCTP (technical and particular specifications) of the works, the choice was made for the grout E/C = 0.7.

Results and interpretation
Estimation of permeability using Lugeon tests before cement slurry treatment Figure 4 above summarizes the permeability tests carried out before the treatment of the foundation with cement slurry, particularly at the exploratory stage. These values are mostly lower than 20 Lu with a small proportion between 20 and 90 Lu. These tests reveal high permeability values up to 90 Lu, reflecting the intense fracturing of the rock.

Veil injections
The results shown in Fig. 5 sufficiently demonstrate the effectiveness of the injections. Thus, the grout absorption rate decreases considerably, overall, for approximately equal depths from primary to secondary holes and from secondary to tertiary, and becomes very low for quaternary and control holes. Figure 6, which corresponds to the most critical zone (fault zone), highly fractured and therefore the cement grout absorption is the highest, further confirms the good confinement of the foundation. Thus, the average grout absorption is 63.77% in the primary holes, 19.97% in the secondary holes, 8.54 l/m in the tertiary holes, 5.8% in the quaternary holes, and only 1.93% for the control holes. This corresponds to average absorptions of 125.2 l/m, 39.21 l/m, 16.77 l/m, 11.38 l/m, and 3.79 l/m for the primary, secondary, tertiary, quaternary, and control holes respectively.

Consolidation injections
The histogram presenting the results of the consolidation injections carried out upstream and downstream of the axial wall of the dam (Fig. 7) shows that the average grout absorption of the upstream line injected first is 111.2 l/m at the level of the primary holes and 43 l/m at the level of the secondary holes, i.e., three times less. The same phenomenon of confinement is observed on the downstream line since the average absorption at the primary holes of 51.4 l/m was reduced to 19.9 l/m only for the secondary holes.

Lugeon test
The permeability tests carried out at the end of the treatment campaign with cement slurry 0.7 in depths of 15 to 25 m in successive increments of about 5 m, give almost zero values below 5 UL (Fig. 8). Figure 9, which is extracted from the injection veil of the Dike-phase 2 (MD0+720.00~MD0+730.00), gives more idea on the extent of the modification of the waterproofing confirming the effectiveness of the treatment. The high average permeability of 68.4 Lu in the initial phase is reduced to 2 Lu in the final test, which is a permeability decrease of 97.07%.
The comparison of the results of the permeability tests and the injections before, during, and at the end of the injection campaign shows a reduction in permeability, and  therefore an increase in the tightness of the rock by sealing its cracks with the hardened grout, as shown by the disk and the traces of grout illustrated in Fig. 10 in the cores of a control hole at the end of the campaign. Based on the above observations, the critical disadvantages of conventional grouting are the complexity of the grouting process and operation which leads to grout segregation, excessive grout loss, causes frequent incidents, and damage in terms of hydrofracturing and hydrojacking of the foundation and high project cost based on the grouting volume. Therefore, it is preferable to adopt the GIN method because it is more efficient, controls quantities, saves time, reduces project cost, and improves productivity while ensuring the quality of grouting work.

Geophysics
The resistivity sections, resulting from the inversion and corresponding to the nine rows of blocks of the internal structure of the dike are presented in Fig. 11. The first (S1) corresponds to the untreated bedrock. The second (S2) presents the image of the bedrock after consolidation with 0.7 cement grout. On section S1, the measured resistivity values show a great diversity of bedrock facies. From this section, a discontinuous zone of bedrock appears between the abscissas 35 and 51 m. At section S2, the bedrock sequence appears to be quite homogeneous over its entire extent. The geophysical aspect of this study allowed to obtain a diversified lithological identification and to characterize the groundwater flow conditions of the studied area. This result also suggests a consolidation of the foundation of the dam after treatment.

Conclusion
The following conclusions were drawn from this study: 1. Structural, geophysical, and geohydrological characterization through permeability tests and electrical tomogra- Grout disc Traces of grout phy revealed the existence of highly permeable discontinuity zones (of more than 90 Lu) below the surface of this massif, which can reach 15 m in width. 2. The treatment of this foundation with 0.7 cement grout injections has considerably reduced this permeability with almost zero control values below 5 Lu, i.e., a reduction in permeability of about 94.4%. 3. The effectiveness of the said treatment is attested by a considerable decrease overall for approximately equal depths from primary to secondary holes and from secondary to tertiary, and becomes very low for quaternary and control holes. 4. The results of the consolidation injections carried out upstream and downstream of the axial wall of the dam show that the average grout absorption of the upstream line injected first is 111.2 l/m at the level of the primary holes and 43 l/m at the level of the secondary holes, i.e., three times less. 5. The confinement of the rock was confirmed by geophysical imaging whose signature at the fractured interval varied from conductive to highly resistive after treatment.
However, we recommend that geophysical characterization should be always added to the geological procedures to better refine the geological treatment model, and to confirm the effectiveness of the method at the end of the treatment.
Acknowledgements As this paper prepared in the COVID-19 pandemic situation, the authors dedicate this paper and to express their heartfelt gratitude and appreciation to all the healthcare workers and other frontline workers in the world for their challenges and sacrifices since the start of this pandemic.

Funding Open access funding provided by University of Lausanne
Data availability Datasets for this research are included in this paper or point to where the references are compiled.

Declarations
Ethics approval I, hereby, Armel Zacharie Ekoa Bessa, consciously assure that for the manuscript "GIN method applied to the consolidation of cracked rocks: case study of the Memve'ele hydroelectric dam (southern Cameroon)," the following is fulfilled: (1) This material is the authors' original work, which has not been previously published elsewhere.
(2) The paper is not currently being considered for publication elsewhere. (3) The paper reflects the authors' research and analysis truthfully and completely. (4) The paper properly credits the meaningful contributions of co-authors and co-researchers. (5) The results are appropriately placed in the context of prior and existing research. (6) All sources used are properly disclosed (correct citation). Copying of text must be indicated as such using quotation marks and giving proper reference. (7) All authors have been personally and actively involved in substantial work leading to the paper and will take public responsibility for its content. The violation of the Ethical Statement rules may result in severe consequences. I agree with the above statements and declare that this submission follows the policies Arabian Journal of Geosciences as outlined in the Guide for Authors and the Ethical Statement.
Consent to participate By signing below, I am indicating our consent to participate in the research. I understand that the data collected from my participation will be used primarily for a Ph.D. work of the first author, and will also be used in summary form for journal publication, and we consent for it to be used in that manner.

Low deformed bedrock
Crack area

Fig. 11
Evidence of crack sealing by cement grout observed by tomography