Characterization of Reservoir Rocks of X Gas Field, Surma Basin, Bangladesh

- This paper deals with the rock physical analysis of primary reservoir rocks in the X gas field, which includes identifying the transition zone between mechanical and chemical compaction zones and the amount of cement in the reservoir rock. The analysis is performed by plotting different log data (Gamma-ray, density, neutron porosity, and sonic log) against depth, constructing the cross plot, and plotting of sonic and porosity logs with cement model. The transition zone, which is the boundary between mechanical and chemical compactions, indicates a sharp increase in density and a decrease in porosity with no lithological change at 2,576 m depth (Well-2). Among three gas sands, the upper gas sand and upper part of the middle gas sand are located within the mechanical compaction zone. The lower parts of the middle gas sand and lower gas sand are within the chemical compaction zone. Cement model analysis showed that the upper gas sand and upper part of the middle gas are composed of unconsolidated sand. In contrast, the lower part of middle gas sand and lower gas sand comprise of consolidated sand, having almost 2% contact cement.


Introduction
The Surma Basin is the Subbasin of Bengal Basin located in the NE part of Bangladesh, the central gas-rich province. Among the twenty-six gas fields discovered in Bangladesh, eighteen gas fields are located in the Surma Basin. The total recoverable proven and probable gas reserve of twenty-six gas fields have been estimated to be at 27.12 TCF, out of which 12.45 TCF was produced, leaving only 14.55 TCF (Petrobangla, 2015). The country gas demand has already surpassed about 3,200 million cubic feet per day, whereas the average supply of gas is around 2,700 MMCFD (Petrobangla Annual Report, 2016). The demand-supply mismatch has been a longstanding problem in the country energy sector. In this situation, sustainable management of the gas field is essential, along with discovering new gas fields. Therefore, proper reservoir characterization in any gas field is necessary to do the proper reservoir management. The reservoir properties, i.e. stiffness, porosity, permeability, ongoing diagenetic process etc. depend on the depositional and diagenetic process of a basin. When sediments were first deposited in the basin, they were in a liquid-like state, loose, porous, and unstable. Then these sediments underwent a series of changes with time, depth, and temperature. The most important change is the loss/ gradual decrease of porosity by mechanical and chemical compaction where mechanical compaction (MC) is stress-dependent, and chemical compaction (CC) depends on the temperature. Reservoir rock initially experiences mechanical compaction (MC) with depth, which ceases at a specific temperature when chemical compaction (CC) started. This process changes the reservoir rock properties, i.e. decreasing porosity and increasing velocity and density through progressive burial at greater depth. The zone with this specific temperature (about 69 -77ºC) where MC changes to CC is, namely, transition zone (TZ). When a reservoir rock lies within the mechanical compaction zone, it is considered to be unconsolidated and vulnerable to subsidence. However, a small amount of cement at the grain contacts significantly increases the rock stiffness (Vernik and Nur, 1992;Dvorkin and Nur, 1996), and the rock becomes stiffer and changes the rock properties significantly. Therefore, it is crucial to know the regime (i.e. MC or CC) beause the reservoir rock might experience (maximum temperature and pressure).
This study identified the transition zone between mechanical and chemical compaction of three gas sands (Upper, Middle, and Lower gas sands) by tracking the changes in geophysical property in wire line logs from Well-2 of the X Gas Field (Figure 1). Moreover, quantitative cementation of primary reservoirs rocks was evaluated using the rock physic template.

Geological Setting
The Bengal Basin covers approximately 11,000 km 2 , roughly half of which record deposition within offshore settings (Reimann, 1993). The basin contains approximately 20 km of Cenozoic siliciclastic sediments. The Ganges-Brahmaputra and associated or ancestral rivers have been transporting clastic sediments to the Bengal Basin (Uddin and Lundberg, 1999). The basin has two broad tectonic provinces: (1) stable shelf, where thin sedimentary successions overlie the rocks of the Indian craton in the northwestern part of Bangladesh and (2) thick basin fill that overlies the basement of undetermined origin in the south and east (Bakhtine, 1966;Khandoker, 1989  deeper part of the Bengal Basin is bounded by the hinge zone to the west and the fold belt to the east ( Figure 1a). Compared to those in the fold belt area, strata in the deeper basin have experienced limited tectonic deformation. The Surma Basin is located in the northeastern part of the Bengal Basin, which was the main depocentre during Miocene. A series of N-S trending anticlinal folds are located in the eastern part of the Surma Basin, which possibly developed due to the Early to Middle Miocene collision between the Indian Plate and Burman Plate (Hiller and Elahi, 1988).
The studied X gas field is one of this N-S trending anticline ( Figure 1b). The Surma Basin has been experiencing its evolutionary history when the Bengal Basin took the shape of a remnant ocean basin (Ingersoll et al., 1995) during Oligocene. By the Early Miocene, the Himalayas and Tibet started to uplift very rapidly, and a major upthrust movement of the Shillong Plateau along the Dauki Fault separated the Surma Basin from the stable shelf. Therefore, this basin became an active sediment depocentre. In the Eocene time, the paleo-depositional shelf was located on the Surma Basin when delta plain was farther north. The delta shelf edge shifted south in the Early Miocene, and the studied area became an upper shoreface to prodelta setting (Alam, 1995). The shelf edge continuously shifted southward, the Surma Basin became dominated by the intertidal to delta plain setting, which continued until Early Pliocene, Silty shale with fine-grained sanstone.
Alternating grey to dark grey, well laminated shale and light grey sandstone, hard and well consolidated, clay pebbless also present' Alternation of grey to dark grey well well laminated shale and sandstone of light grey, hard and well consolidated clay pebbles also presents. The Miocene Surma Group is divided into two units: 1) Bhuban and 2) Bokabil Formations (Holtrop and Keizer, 1970;Hiller and Elahi, 1984;Khan et al., 1988), both of which extend throughout the Bengal Basin. Johnson and Alam (1991) have interpreted the Lower Surma Group as prodelta to delta front deposits of a mud-rich delta system, while the upper Surma Group represents shallow marine deposits to a tide-dominated coastal setting within a cycle of a transgressiveregressive regime (Sultana and Alam, 2001). The thickness of this group varied significantly from 2,700 m to over 3,900 m (Parvin et al., 2019). The overall stratigraphy of the Surma Basin is given in Table 1.
The Surma Group reservoirs are divided into three main gas zones, namely upper gas sand (UGS), middle gas sand (MGS), and lower gas sand (LGS). Table 2 shows the top and base of the UGS, MGS, and LGS sands of Well-4 in the X gas field.

Methods and Analytical Techniques
The transition zone can be identified by the interpretation of the different geophysical logs. A sudden and distinct change in log properties can indicate changes from MC to CC. Moreover, the cement model can validate the depth range by evaluating cementation quantitively. A cement model is a rock-physical template where elastic properties (i.e. P wave velocity Vp, S wave velocity Vs, bulk density, etc.) of rock is plotted against porosity. In this study, Vp versus average porosity template is used to validate our transition zone analysis.
Wireline logs that are used in this study include Gamma-ray (GR), density (RHOB), neutron porosity (NPHI), and P-sonic log (DT). The density porosity was calculated from the bulk density log (RHOB) using the following equation: where the matrix and the fluid density are assumed as quartz and brine density as 2.638 g/ cc and 1.1 g/cc, respectively.
The neutron porosity log was corrected before being averaged using the following equation: Finally, the average porosity was calculated using the following equation: where: is the neutron log porosity, is the corrected neutron porosity, is the saturation of mud filtrate, and is the hydrogen index of the gas of this formation.

421
The analysis was performed by plotting these logs in several ways. Plotting includes logs against depth, constructing cross plots, and plotting sonic and porosity logs with cement models.
Moreover, compressional velocity (Vp) is calculated from the P-sonic (DT) log using the following equation: where Vp is in m/s, DT is in µs/ft.
The rock physic template used in this study is adapted from the previous study where the friable sand model is adapted from Rahman and Mondol (2017), and contact and constant cement (2%) models are adapted from Avseth et al. (2005). The constant cement model is a combination of the contact cement model and the friable cement model. The constant cement model assumes that the clastic rock has reduced porosity to a certain degree due to contact cement, whereas further porosity reduction is due to the noncontact-cemented particles located inside the pores (Grude et al., 2015). (soft) sand model. Intermediate constant cement curve represents 2% contact-cemented line. These models were used to evaluate qualitative cementation within different reservoirs.

Result Transition Zone Identification
Gamma-ray was used as a proxy for facies (sands or shales) as no core and cutting samples are available in this study. No distinct changes were observed in gamma-ray values except few local low and highs where the values range between 30 to 57 API (Figure 3a). There is a sharp increase in density, and a decrease in neutron porosity was identified at a depth of 2,576 m. The density above and below this depth shows the comparatively constant value with 2.2 and 2.65 g/cm 3 , respectively (Figure 3b). A similar trend was also found in the neutron porosity plot where porosity ranges between 52% to 22% above and below the depth, respectively (Figure 3c).
Cross plot of NPHI versus RHOB of UGS, MGS, and LGS from Well-2 also reveals two distinct clusters with significantly different neutron and density values (Figure 4). The upper cluster shows higher porosity and lower density than the lower cluster, where porosity decreases with increasing density significantly.

Rock Physical Analysis
For the quantitative analysis of this three-gassand, data from Well-4 was plotted with cement models. The average porosity was plotted against velocity to identify the quantitative cement percentage of different sands ( Figure 5). The lower gas sand and lower part of the middle gas sand have higher sonic velocity and lower average porosity compared to the upper gas sand and the upper part of the middle gas sand. The UGS followed the friable sand model, where porosity loses through compaction. Most of the MGS also followed the UGS trend. However, a few MGS and LGS points sharply increased the velocity with minimal porosity change and followed the 2% constant cement model. The data distribution illustrated the diagenetic differences within sandstones.

Discussion
Rock properties such as velocity, bulk density, and porosity continuously changes after deposition due to several compaction processes. Rock initially undergoes mechanical compaction (MC) up to a certain depth with a specific temperature (about 69 -77ºC). This depth is called transition zone (TZ), as discussed earlier. In the MC regime, rocks are loose and uncemented and show higher porosity, lower density, and lower velocity than the rocks in chemical compaction zone. These changes are easily visible in wireline logs, and UGS and the upper part of MGS are located in mechanical compaction zone showing higher porosity and lower density and velocity (Figures 3 and 4).The transition zone starts at 2,576 m below which lower part of MGS and LGS has experienced chemical compaction zone.
As mechanical compaction is mainly controlled by the effective stress, it should be expected that the development of fractures and deformed grains in thin section of studied rock. The chemical compaction is associated with dissolution and precipitation of minerals within sedimentary rocks (reservoir rock), which are easily visible in thin section. Unfortunately, there was no core data for this study. However, two core analyses (2,420m and 3,420m) were adapted from Hossain (2009) who used core samples to identify over pressure and temperature at the Well-2 in the same X gas field. First core is located in identified mechanical compaction zone and second core  Figure 4. Cross plot of neutron porosity and density of UGS, MGS, and LGS from Well-2. The data clusters represent the variation of reservoir properties. from chemical compaction zone. If two samples are compared, there is a significant decrease in porosity with increasing depth from 2,420 m to 3,420 m where porosity in the rocks from MC zone is significantly higher than the rocks in the CC zone ( Figure 6).
The core sample analysis from 2,420 m shows the development of fractures in plagioclase feldspar along weak planes, which is probably because of mechanical compaction of the rock. The sample at 2,420 m depth also shows well-preserved primary interparticle porosity (blue stain) and grains are loosely packed which indicates that these rocks undergo mechanical compaction (Figure 7). The temperature in this zone is approximately 58ºC that is not high enough to initiate chemical compaction in this zone (Figure 8).    LGS TZ = 2.576 km~6 9 0 C The sample from 3,420 m depth shows a complete loss of primary interparticle porosity by matrix infilling (Figure 6b). Secondary porosity in sandstone develops through feldspar dissolution during deep burial (Hossain, 2009). The presence of corroded quartz grains, oversized pores, porosity channels between grains and cement, and partially dissolved rock fragments satisfies the criteria for secondary porosity (Schmidt and MacDonald, 1976) which indicates that these rocks are located in a chemical compaction zone. Moreover, the quartz and lithic fragments percentage significantly change between the two depths. The quartz fraction increases by 20%, while lithic fragments decrease two-third ( Figure 6). These lithic grains might be precipitated as quartz cementation, explaining the significant differences in mineralogy between them. In addition, the transition depth temperature in this zone is high enough (69º C) to initiate chemical compaction of the rock in this zone. Therefore, these analyses validate the findings of TZ depth and sandstones compaction history As UGS and upper part of MGS are located above transition zone (TZ), they are loose, porous, and unstable. Below TZ, at the initial phase of chemical compaction, a small amount of cement at the grain contacts will significantly increase the stiffness (Vernik and Nur, 1992;Dvorkin and Nur, 1996) and cease the stress-dependent mechanical compaction even though the cement is relatively soft (Dvorkin et al., 1994). Cemented rock has a higher strength than an uncemented rock of the same porosity and mineralogy. Besides, the cemented rock permeability might be higher than that of uncemented rock at the same porosity. Because of loose pore-filling material or noncontact cement increases the specific area, it decreases permeability (Avseth et al., 2010). Therefore, during gas production, it should be kept in mind that these three reservoirs are vertically located in different compaction regimes, and the reservoir properties of these sands have variation. UGS and upper part of MGS are unconsolidated, which may lead to sending issues due to the overproduction. This also may trigger the overburden rock subsidence in the studied area as these rocks are loose and have no sufficient strength to hold overburden when gas is depleted. On the contrary, during any fluid injection for enhancing recovery in the lower part of MGS and LGS, the fluid type should carefully be chosen, because these zones are in the CC regime and have almost 2% cement. Improper fluids may react with the cement and seal the porosity instead of increasing them. However, reservoir quality depends on many other factors, i.e. temperature, pressure, etc. that need to be included during the reservoir management plan for sustainable development.

Conclusions
Identifying the cement model transition zone, analysis has revealed the qualitative and quantitative information about reservoir properties. The UGS and upper part of MGS are located in the mechanical compaction zone and vulnerable if overproduced. However, the lower part of MGS and entire LGS are experiencing chemical compaction. These reservoir rocks have sufficient strength, with almost 2% contact cement. However, their properties may degrade during improper chemical injections for enhanced recovery. Yet, very few works in rock physical have been done in Surma Basin. Therefore, it is recommended that every gas field should run this type of work for better reservoir management.