Basin Modelling in The Mahakam Delta Based On The [PDF]

Citation preview

© IPA, 2006 - 21st Annual Convention Proceedings, 1992



Disc Contents



Contents



IPA 92-11.04 Search



PROCEEDINGS INDONESIAN PETROLEUM ASSOCIATION Twenty First Annual Convention, October 1992 BASIN MODELLING IN THE MAHAKAM DELTA BASED ON THE INTEGRATED 2D MODEL TEMISPACK Jean Burrus * Etienne Brosse * Ghislain Choppin de Janvry Yves Grosjean ** Jean-Louis Oudin **



ABSTRACT The petroleum system of the Mahakam delta is investigated using a two-dimensional reconstruction of the history of HC generation and migration along a 80 km long regional section. We find that the classical ’perascensum’ model, in which gaseous HC are generated in the deep overpressured, overmature shales, and sweep liquid HC during their vertical migration, needs to be revised. The main conclusion of this study is that the coal-rich, sand-rich and normally pressured delta-plain facies, located in the synclines, need to be considered as the most effective source rock, rather than the deep overpressured marine shales. Also, migration appears to take place mostly parallel to bedding up-dip along structure flanks rather than vertically across bedding. Our model explains the distribution of gaseous HC at Tunu structure, and of oil at Tambora structure. It is consistent with distribution of present-day subsurface temperatures, shown to be affected by recent meteoric water circulation, with coal maturity and with overpressures distribution. INTRODUCTION The Mahakam delta (Figure 1) is Indonesia’s second hydrocarbon province. oil and gas accumulations are found in complex stacked deltaic sandy reservoirs generally encountered between 2 and 4 km depth. The principal fields (Figure 1) are : Handil (oil and gas), Nilam and Radak (gas with minor oil rim), Attaka (mainly oil), Bekapai (light oil and gas). More recent discoveries include Tunu (gas) and Sisi (gas with minor oil). The petroleum system of the Mahakam delta has been studied by various authors over the past decade.



* Institute Franais du PL-trole *’ TOTAL - Francc



**



Most previous work focused on the organic geochemistry of the regional potential source, a rather uniform type 111 series with interbedded coals. While most authors have described the nature, distribution and variability of the organic matter, few authors have attempted to address the geochemical evaluation of the delta together with its geological evolution, including structural history, thermal history, and history of overpressures development and subsequent fluid flow. Recently appeared numerical basin modelling techniques provide a useful way to address simultaneously the geochemical history and the geological history of petroleum provinces through an ’integrated’ approach. This paper presents some results obtained when applying IFP’s integrated numerical model TEMISPACK to study the petroleum history along a 80 km east-west regional section located in the center of the Mahakam delta. This section crosses the South Tambora, Tunu and Sisi structure. South Tambora, where some oil shows were found, is equivalent to the giant oil field of Handil (Figure I ) , but has no significant closure. Tunu is an asymmetrical gas accumulation. with a steeper flank along the more shaly facies found eastward. Sisi was only recently drilled. Gas and minor quantities of oil were found, and appraisal is still going on. The purpose of this study is to discuss a regional scenario of HC generation and migration. Geological, geophysical and geochemical data used in this study were collected and synthesized by TOTAL TNDONESIE in 1988-1990. The numerical simulations were carried out at IFP on a CRAY computer in 19901991. This modelling study is part of a regional synthesis carried out by TOTAL. A summary of data and main conclusions of this synthesis were already given by Duval et al. (1992). This paper includes four sections. In the first section, we present the numerical model used for the simulations.



24



In the second section, we briefly summarize the principal geological and geochemical characteristics of the studied section. In the third section, we show model outputs pertaining to (a) distribution of overpressures, (b) distribution of subsurface temperatures and source rock maturity, (d) H C migration trends. The last section is a discussion and conclusion section. PRINCIPLES OF TEMISPACK MODELLING TEMISPACK is a finite-volume model which reconstructs the history of petroleum generation and migration along a 2D evolutive mesh representing a regional cross section. Equations used in TEMISPACK can be found in Ungerer et al. (1990), Burrus et al. (1992a). The program contains five different modules. The backstripping module simulates the sedimentation rate along the section, by decompaction of the layers using normal compaction curves. 'Normal' means that neither undercompaction nor mineral diagenesis are considered. The mechanical compaction module uses the sedimentation rates computed previously to simulate the development of overpressures, the direction and magnitude of water flow and the porosity. The following parameters need to be inputs: the permeability of layers (a function of porosity through Koseny-Carman law), their surface porosity, and their mechanical strength (through effective stress/porosity relation, a classical concept in overpressure modelling: Smith, 1971). All these parameters are lithology dependent. In this study, effective stress/porosity relations were reconstructed from wireline logs and pressure data (RFT, mud density) following the details given elsewhere (Burrus et al., 1992b). The free parameters were therefore the permeabilities, which were adjusted against the observed overpressures. The thermal module computes the transient history of subsurface temperatures. In this study, as convective flow was considered, the heat equation was solved in conduction and convection. The inputs to the module were thermal conductivities and heat capacities estimated from the lithology, in accordance with Table 1. A variable basement heat flow was imposed at 15 km below sedimentbasement interface, and the surface temperature was kept constant at 25"G, the present day mean surface temperature at the top of the sediments. The main free parameter is the subcrustal heat flow, which is adjusted against the observed subsurface temperatures and thermal maturity indicators. The HC generations module computes the rate of kerogen cracking (or TR: transformation ratio) by



solving a system of parallel kinetic reactions (Tissot and Espitalie, 1975). Kinetic parameters for the source rock were adjusted against experimental pyrolysis yield curves (Table 2). Theoretical TRs were compared to observed maturity indicators like H I (hydrogen index) using an approximate relation TR=HIo-HI/HIo. The H C migration module computes the mass of H C displaced across the mesh, in response to the driving forces: buoyancy, gradients of overpressures, capillary forces. It is assumed that petroleum migrates as a separate phase flow through the pore network, and that other mechanisms (diffusion, adsorption, etc) are relatively negligible. The results are displayed as the volumetric H C saturations in each mesh cell. This is obtained by solving a two-phase (water+HC) Darcy equation, together with the compaction equation. Additional parameters are: the initial potential of the source rocks, estimated from~Rock Eva1 experiments (see Table l), the relative permeability functions used in multi-phase Darcy equation (degree two functions of saturations), the capillary pressures, which represent the difference of pressure between the water phase and the H C phase, and the petroleum fluid properties (density and viscosity). The density and viscosity of the H C phase are theoretically affected by PVT and H C chemical composition changes. These were not explored in this study. We assumed the H C generated from kerogen cracking behave like a 'compressed gas' from a density (350 kg/m ) and viscosity Pa.s) standpoint. Capillary pressures were assumed to be around 10 MPa in marine or transgressive shales, and neglected elsewhere. Practically, a TEMISPACK modelling study is stepwise organized, starting from the most simple reconstructions (backstripping), and ending with complete coupled reconstructions (burial / overpressures / thermal / HC generation and migration). A litho-chrono-stratigraphic model of the studied section was established first, and all well controls gathered: lithofacies. overpressure, porosities, temperatures. source rock potential and maturity, etc. Then the section was decompacted using the backstripping module. Consistency between palaeo-bathymetries and stratigraphy was checked. Then the past and presentday pressure regime was simulated and the pressure history calibrated against observed pressure distribution. Permeabilitics were first estimated from the lithological model, then refined to fit the observed pressure. Due to the presence of interbedded shale and sands, considerable Permeability anisotrogy was introduced (up to five orders of magnitude; see details in Burrus et a]., 1992b). The next step was to model thermal and maturity histories, while adjusting a consistent subcrustal heat flow reconstruction against



25 observed temperatures, organic maturity and other thermal indicates. Since a convective component was recognized in the thermal field, some calibration of horizontal permeabilities was also achieved while matching temperatures. Finally, the coupling of all modules lead to discuss the history of HC expulsion and migration. This last step must be viewed as qualitative rather than quantitative. In particular, comparison of quantities of HC computed by a given model and real accumulations is not relevant since computations are 2D, while real migration pathways are 3D. GEOLOGICAL AND GEOCHEMICAL CHARACTERISTICS: AN OVERVIEW. The Mahakam delta basin (see location in Figure 1 and section in Figure 2) contains an accumulation of more than 9 km of post-mid Miocene sediments. They are overlying older (Upper Eocene ? to Oligocene Lower Miocene) sediments not involved in this study. Geodynamics-Structures. The Mahakam basin forms the eastern part of the wider Kutai basin (Figure 1). Its development, between Upper Eocene and mid-Miocene, is roughly coeval with the opening, 150 km to the north-east, of the North Makassar basin, a marginal extensional basin linked to the subduction of the Indian-Australian plate under Sulawesi (Situmorang, 1982;Letouzey et al., 1990). Subsidence is considerably reduced to the south of the delta across the Pater Noster platform. A major east-west compression appeared during the Pliocene. It is still active at present-day time. It was responsible for the inversion of the whole Kutai basin and for the formation of the Meratus range 200 km to the southeast of the delta (Figure 1). In the Mahakam delta region, the structuration is attenuated when compared to the Kutai inversion structures, which are accompanied by more than 2 km of erosion, and major thrusting. To the west of the present-day delta, the structuration consists in the NlO-N20 folds and thrusted folds of the Samarinda anticline (Figure 1). Within the present-day delta, the structuration consist in three parallel folded axes, not affected by thrusting: the Handil-Badak axis, the Bekapai-Attaka axis and the Pemarung-Sisi axis. Erosion, estimated from log and seismic data, is relatively high at handil (700-800 m) and Badak (500-600 m ) , small (