Quantifying the uncertainty in the ultimate recoverable oil reserves using the Monte Carlo simulation techniques from ‘OWA’ Marginal Field, Onshore Niger Delta, Nigeria
Journal Title: Geology, Geophysics & Environment - Year 2018, Vol 44, Issue 4
Abstract
A review on the development of marginal oil fields in Nigeria has now become an important strategic issue if it is to remain amongst the top producers in the global market, and these fields are vast, available all over the Niger Delta. One of the factors that makes a field marginal is the size of its reserves. Stochastic estimation gives a certainty in terms of the possible number of outcomes within the range of input parameters. In this work, four (4) deviated wells and 3D seismic volume (362 inlines and 401 traces) were interpreted for the evaluation of the field. The petrophysical evaluations were interpreted using the Power Log software and the Seismic, Geographix and Petrel softwares. Stochastic reserve estimation was done using Monte Carlo sampling techniques and subjected to uncertainty quantification using the Crystal Ball software by varying distributions and measuring sensitivity impact on the overall reserves. The production profile was predicted based on some assumptions and history matching which result in the overall Expected Ultimate Recovery (EUR). The petrophysical analysis shows the reservoirs to be within the unconsolidated continental Benin Formation denoted as ‘Intra-Benin’ sands, an unconventional reservoir as supposed the normal reservoir rocks within the Agbada Formation. This indicated high porosity (0.28), water resistivity (7 Ω∙m), and water saturation and also inferred Heavy Oil (low API). Nine hydrocarbon sands were identified but only three (B1, D and E), representing shallow, mid and deep reservoirs were further evaluated. 1P and 2P reserve estimates were 4.8 MMBO and 5.7 MMBO for B1; 15.2 MMMscf and 16.4 MMMscf for D; 8.4 MMMscf and 8.8 MMMscf for E respectively. The Monte Carlo simulation of 1,000,000 trials with mainly triangular distribution assumption generated P10, P50, P90 were 6.5 MMBO, 5.6 MMBO and 4.4 MMBO for B1; 17.5 MMMscf, 13.7 MMMscf and 10.8 MMMscf for D; 10.4 MMMscf, 8 MMMscf and 6.1 MMMscf for E respectively. The sensitivity impact of the input parameters were estimated and ranked, and the coefficient of variability ranges within 15% to 20% for the reservoirs indicating that there is a very low level uncertainty of reserve estimation around the P10, P50 and P90 percentiles which could be positive for investment decisions. ‘OWA’ marginal field reflects a typical low reserve (EUR) category found within the Niger Delta basin.
Authors and Affiliations
Olubunmi C. Adeigbe, Isaac Folorunso Odedere, Omowunmi Idera Amodu
Geomechanical modellingl of Paleozoic Shale Gas Formation: a case study from the Baltic Basin, northern Poland
Geomechanical modeling has become a powerful tool for optimization of well placement and trajectories, designing drilling processes and stimulation of reservoir formation as well as effective reservoir management. In thi...
Coal mining waste dumps as secondary deposits – examples from the Upper Silesian Coal Basin and the Lublin Coal Basin
Polish coal mining industry generates approximately 30 million Mg of mining wastes per year, which represents the largest amount of industrial waste in Poland. This results in the disposal of a huge volume of wastes at w...
Department of Environmental Protection
Department of Environmental Protection
3-dimensional surface magnetic susceptibility distributions with depth to investigate anthropogenic pollutants boundary in Abuja Metropolis
Changes in the magnetic susceptibility of topsoil, especially in urban areas, are greatly affected by the deposition of anthropogenic metallic dust arising from industrial activities, the burning of fossil fuels, car wea...
Basic unit size in the analysis of the distribution of spatial landscape elements on the basis of the lithostratigraphic geodiversity of the Ojców National Park (Poland)
Spatial analyses of landscape element diversity are always carried out on the basis of a previously established analytical grid. It divides the research area into ‘n’ smaller areas – basic units before calculating variou...