Empirical design for the excavation method and support system at Budong-Budong Dam Diversion Tunnel, Central Mamuju, West Sulawesi Province

Budong-Budong Dam, located in the West Sulawesi Province of Indonesia, was designed to use a tunnel to divert river flow during the dam construction. The Rock Mass Rating was used to examine the tunnel design with a less detailed drill core investigation. The rock strength parameter is only based on the field estimation. The entire diversion tunnel will pass through Breccia with weak strength. Evaluation is required using integrated and precise data to guarantee safety during and post-construction. The RMR classification was used to determine the excavation and preliminary tunnel support, while the Q-System was used to compare the support of the tunnel. Those classifications were used to anticipate the geological conditions uncertainty. Geological data were collected by geological mapping, evaluation of subsurface drilling, and laboratory analysis. The tunnel will pass through a Tuffaceous Breccia and Tuffaceous Sandstone insert of Tuffaceous Siltstone. In addition, a poor-quality rock was found around the inlet and outlet, and a fair-quality rock in the middle of the tunnel. Hence, drilling and blasting were recommended for the excavation. The tunnel’s proposed support system included shotcrete, wire mesh, and rock bolts in the middle of the tunnel. Steel sets are added around the inlet and outlet.


Introduction
One of the challenges in tunneling work is being unable to choose tunnel conditions.In contrast to surface construction, where general conditions can be seen, tunnels are built under conditions that can only be assumed with limited data [1].It will also be a challenge to construct a diversion tunnel, which will be used to change river flow during the construction of the Budong-Budong Dam in Salulebo Village, Central Mamuju Regency, West Sulawesi Province.From technical planning data, the diversion tunnel, which is 320 meters long, is planned to have an excavation width of 7 meters and will be constructed on the east-south side of the main dam, as shown in Figure 1.Rock bolts, steel sets, shotcrete, and wire mesh are used as support systems in tunnels.The tunnel inlet is planned to be at an elevation of +36.00 meters above sea level, and the tunnel outlet is at an elevation of +34.70 meters above sea level.The investigation also found that the groundwater elevation was above the tunnel alignment.According to geological investigations and drilling outcomes at three points during the planning stage, Rock Mass Rating (RMR) was used empirically to determine the excavation technique and the diversion tunnel's support.According to the investigation results, the tunnel will pass through fair-quality rock from the inlet to the outlet [2].However, the rock strength parameters in the design are still based on estimated field values [3].
Additionally, the tunnel design only considers the RMR classification.Detailed evaluation and other rock mass classification are required to anticipate geological uncertainties [4].In most cases, using at least two classifications is suggested to improve and produce more accurate results when designing and engineering rocks [5].Several rock mass quality assessments are also recommended for cross-checking [6].This study utilizes multiple rock mass characteristics, especially RMR [7] and Q-System [8].The diversion tunnel's quality rock mass is determined using four drilling data (BH-08, BD-13, BH-03, and BH-09).By considering the geological characteristics in the research area, this study aims to obtain more precise information for determining the kind of excavation and preliminary tunnel support.

Geological Investigation
The geological investigation includes surface geological mapping and a description of core drilling.Surface geological mapping is conducted over 1 km 2 around the study area.Surface geological mapping involves observing lithology, investigating geological structures, and collecting data on discontinuity orientations.Meanwhile, observing subsurface core drilling results includes determining lithology, weathering degree, measuring Rock Quality Designation (RQD), discontinuity conditions, discontinuity spacing, and other factors to assess rock mass quality.Uniaxial Compressive Strength (UCS) testing in the laboratory was carried out on rock samples to obtain mechanical rock properties.This research uses boreholes BD-13, BH-08, BH-03, and BH-09 for subsurface investigation.BH-08 and BH-09 are located around the inlet and outlet of the tunnel, while BD-13 and BH-03 are located in the midsection of the tunnel.The location of the drill points can be seen in Figure 2. BH-08 has a drilling depth of 25 meters, and BH-09 has a drilling depth of 15 meters.For the midsection of the tunnel, BD-13 and BH-03 both have a drilling depth of 40 meters.

Rock Mass Rating
Rock Mass Rating (RMR) was determined based on six parameters [7], including the strength of rock (P1), RQD (Deere, 1989) (P2), discontinuity spacing (P3), discontinuity conditions (P4), groundwater conditions (P5), and discontinuity orientations (P6).Each parameter's rating can be adjusted based on the observation conditions.A rock's strength parameter may be assessed using laboratory tests such as the PLI (Point Load Index) or UCS (Uniaxial Compressive Strength).RQD is measured by comparing the total length of rock greater than 10 cm with an entire drill length.The distance between discontinuities is considered discontinuity spacing.Discontinuity conditions are determined based on surface, aperture, persistence, and weathering conditions.Equation ( 1) is used to determine the RMR value, which is based on the sum of each rating parameter.
RMR is classified into five classes, depending on the total rating of all parameters.Classes in Table 1 are used as a guideline for deciding the type of excavation and preliminary tunnel support.Empirical excavation and tunnel support types can be assessed by utilizing the RMR values and classes through the provisions referred to in Table 2, which shows the details of the excavation work methods for each rock classification class in the tunnel.Excavation work methods from a tunnel can be full face, benching, or multiple drifts, with excavation progress adjusted based on the class of the rock.Table 3 shows the detailed specifications of the support system used based on class classification or RMR value.
In Table 2, it is explained that an advance in digging will be directly proportional to the class of rock mass.The rate of excavation progress increases with higher rock mass class.Likewise, with the type of excavation, the better class has a bigger area that needs to be dug in the tunnel face.In contrast, Table 3 shows the detailed specifications of the support system used based on class classification or RMR value.Table 3 indicates that fewer forms of reinforcement are required for rocks with greater rock mass classes, including shorter rock bolt lengths, larger spacing requirements, and thinner shotcrete.

Q-System
The Q-System is a rock mass quality method initially presented in 1974 [10] and is currently being developed to evaluate tunnel support systems by the NGI (Norwegian Geotechnical Institute).In general, three main factors are used in Q-System classification: jointing degree (size of block), joint friction, and active stress.The degree of jointing expressed by RQD [9] and Joint set number (Jn), joint friction was  defined by comparison between the number of joint roughness (Jr) [11] and joint alteration (Ja), while active stress was determined from joint water reduction number (Jw) and stress reduction factor (SRF).These parameters from the three main factors were used to calculate the Q-value using Equation (2).

Conditions of Surface Geological
Refers to the Map of Regional Geology sheet of Mamuju [12] and Pasangkayu [13], the diversion tunnel is located in the Lamasi Volcano Formation (Toml).The formation is composed of Tuff, Lava, and Volcanic Breccia composed of local Andesite-Dacite with intercalated Limestone Sandstones and Shales of Oligocene to early Miocene age.The geological mapping results show that surface conditions in 1 km 2 of the study area comprise three lithologies: Tuffaceous Breccia, Tuffaceous Sandstone, and Barley-Silt Deposits, as shown in Figure 2. Structure, such as fault and fold, was not found at the tunnel location, but joints were found at some observation stations.Based on observations on the surface, 1-2 joint sets were found.Beddings were found in some observation locations around the tunnel.The measurement results show that the rock strike at the study site is dominated in the northeast-southwest direction.The average dipping of the rock layers is about 20-25°.In comparison, the diversion tunnel has a path in the northeast-southwest direction.The excavation is assumed to be in the same direction as the rock strike, referring to the rock's strike and dip measurement.Megascopic description of observations on the surface of Tuffaceous Breccia rocks has a bright gray color, fragments measuring 2-6 cm, a sand-sized matrix, matrix-supported, and poor sorting, as shown in Figure 4.The fragment of rock has angular to subangular roundness degrees.Figure 5 is Tuffaceous Sandstone, which has a megascopic description of blackish brown in weathered condition and blackish gray in fresh condition, medium to fine sand grain size, closed packing, and good sorting.Figure 3 shows the deposits found along the Salulebo River.The sediment is silt to barley or cobble size (<256 mm).These deposits come from rocks in the downstream area of the river, which are transported by the river flow and deposited around the research location.

Conditions of Subsurface Geological
Observations of subsurface geological conditions were carried out through the results of four drilling points located around the diversion tunnel location.Figure 6 is an example of drill core analysis carried out at a depth of 30-35 meters below the surface.From this drill core observation, the type of lithology, weathering level, rock strength, RQD measurements, conditions and spacing of discontinuity, and 7 groundwater condition are observed, which are used as parameters for assessing the quality of the rock mass.

Figure 6. Examples of Rock Core Drill Analyzed at Diversion Tunnel
Based on four core drill observations along the tunnel, there were Tuffaceous Sandstone and Tuffaceous Breccia.Tuffaceous Siltstone was also found on drill observation.A cross-section of lithology along the tunnel can be seen in Figure 7.The tunnel inlet is planned to be at an elevation of +36.00 meters above sea level, and the tunnel outlet is at an elevation of +34.70 meters above sea level.From the secondary data as bore log data, the investigation also found that the groundwater elevation was above the tunnel alignment.

Figure 7. Cross-Section Lithology of Diversion Tunnel Based on Drill Observations
The characteristics of the rocks in the tunnel were divided into four zones.Table 4 displays the zones' division and the tunnel depths for each zone.Rating parameters and the calculation of rock mass quality for each zone are displayed in Table 5 and Table 6.Rock mass quality zoning based on RMR is presented in Table 5, while zoning based on Q value is shown in Table 6.

Determination for Tunnel Excavation Method
Refers to the calculation in Table 5 that shows RMR rock mass quality, diversion tunnels will pass through of poor to fair rock quality.The decision to choose the kind of excavation to be applied is based on the rock class or quality.Top heading and bench excavation are suggested for poor and fair-class rock masses.The suggested excavation advance for rock mass of poor quality is one up to five meters, and supports are installed simultaneously with excavation.For fair rock mass quality, a 1.50-3.00m excavation advance is suggested in the top heading, and support should be installed after the blasting is done.

Determination for Tunnel Support System
Besides being used to analyze the type of excavation, RMR can be used to establish support systems.Refers to the RMR value in Table 5, the tunnel consists of poor and fair quality rock class.According to the RMR classification, preliminary tunnel support comprises three components: steel set, rock bolt, and shotcrete.Table 7 describes the details of each support tunnel component.Fair-quality zones do not require steel sets and also need thinner shotcrete with wider rock bolt spacing if compared to poorquality zones.Apart from that, in the fair quality zone, wire mesh is required only on the tunnel crown, whereas in the poor quality zone, wire mesh is needed on the walls and crown of the tunnel.
Otherwise, the determination of tunnel support is also obtained by utilizing Q-System classification.Q value in Table 6 and equivalent dimension (De) values are necessary to decide on a support system using the graph in Figure 8 [14].The equivalent dimension is obtained by comparing height or span in meters and ESR (Excavation Support Ratio) [10].Based on data from detailed engineering design, it is known that the planned excavation height is 7 m [15].While the ESR value for this study, a value of 1 was used.The plotting results in Figure 7 are used to determine the support system category.The plotting results show that the tunnel rock requires a category 3 support system in the third zone, category 5 in the first until the second zone, and category 4 in the fourth zone.Details of each category are explained in Table 8.From Table 8, it can be seen that the lower the rock class, the shotcrete thickness will be thicker, and the rock bolt spacing will be tighter.The right side of the graph in Figure 7 has recommendations for bolt lengths, but more analysis is required.Using longer bolts than the diagram suggests will be essential, and extending bolts by lowering the Q value will generally be necessary.

Conclusion
Based on the evaluation of the core drill, it is known that the diversion tunnel elevation consists of Tuffaceous Breccia and Tuffaceous Sandstone insert of Tuffaceous Siltstone.Assessment of rock mass quality by RMR classification, same as the tunnel outlet, shows that around the tunnel inlet has poor rock quality.In contrast, the midsection of the tunnel has fair rock quality.On the other hand, when assessing the quality of the rock mass with the Q-System classification, the midsection of the tunnel has poor quality, and it produces very poor rock quality at the tunnel inlet and outlet.The tunnel is recommended to be excavated with top heading and bench with advancing is 1.5-3 m for the midsection and 1.0-1.5 m for the around the inlet and outlet of the tunnel.Support is suggested to be installed after each blast in the midsection, while for other than the midsection, support installation is simultaneous with the excavation, erect support 10 meters from the face.Combining steel set, rock bolt, wire mesh, and shotcrete is suggested for the tunnel support system.Detailed specifications of those combinations are adjusted following the rock mass quality in each tunnel zone.

Figure 2 .
Figure 2. Geological Map Around the Diversion Tunnel.

Figure 3 .
Figure 3.The appearance of deposits around the Salulebo River.

Figure 8 .
Figure 8. Category of Support Based on the Relationship between Q and De Values[14]

Table 5 .
Classification of each Zone using RMR

Table 6 .
Classification of each Zone using Q-System

Table 7 .
Detail Support System Pattern Based on RMR Class