Prediction of thermal maturity by indirect methods using seismic attributes in the central part of the Persian Gulf

Abstract

AbstractIn this paper, a method is proposed for the prediction of thermal maturity in the source rock using indirect methods. The applied data are well logs (neutron, density, resistance, and acoustic) in 13 wells and seismic data in six oil and gas fields in the central part of the Persian Gulf. Well-logs and seismic data are much more abundant than geochemical data and cover an extensive area in the oil and gas fields. These properties compensate for the lack of geochemical data that are scattered and limited to a few wells. This study is carried out in two steps. First, the amount of thermal maturity in the Kazhdumi Formation is calculated from well logs and is presented as an index in each well. Data obtained from organic thermal evaluation analyses are used to validate the results of thermal maturity prediction. These data include Rock-Eval pyrolysis in two wells. Then, seismic data are processed and studied in two-dimensional sections at the location of the target fields. In this step, seismic attributes are extracted from the seismic data using the multi-attribute regression analysis method, and thermal maturity is calculated using these attributes. Prediction is performed by probabilistic neural network analysis, and a seismic section is extracted indicating variations in thermal maturity in the Kazhdumi Formation.Keywords: Source rock, Thermal maturity, Well log, Seismic attribute Introduction The Kazhdumi Formation is an important source rock in the Persian Gulf basin, in the south of Iran (Bordenave Burwood, 1990). Thermal maturity of the Kazhdumi Formation is low in the central Persian Gulf compared to the eastern and western sectors (Rabbani 2008; Ghasemi-Nejad et al. 2009; Rezaie Kavanrudi et al. 2015; Rabbani et al. 2014; Baniasad et al. 2019). This formation is over-mature in proximity to the Hormoz Strait. The TOC content of this formation depends on the variety of depositional environments across the basin, increasing to the northwest with a maximum of 6 wt% in the proximity of the Hormoz strait (Rezaie Kavanrudi et al. 2015; Noori et al. 2016). The kerogen type is mainly ⅡS and type Ⅲ in different areas (Ghasemi-Nejad et al. 2009). The stratigraphic equivalents of the Kazhdumi Formation are the Burgan Formation in the west and south of the Persian Gulf (Kuwait), producing hydrocarbons from the second-largest hydrocarbon oilfield in the world, and the Nahr-Umr Formation in Qatar and Iraq. The Burgan and Nahr-Umr formations consist of fluvial sandstone in the south and east of the Persian Gulf compared to the shale-dominated volume in the north and central part (Ghasemi-Nejad et al. 2009; Noori et al. 2016).The source rock potential of this formation is largely unknown and there is a lack of published reports in the central part of the Persian Gulf. the aim of this study is to use log and seismic data for indirect estimation of thermal maturity in this area. Materials & MethodsIn this study, neutron, density and sonic as well as gamma-ray logs are used to predict thermal maturity. The maturity index (MI) by Zhao et al. (2007) was used to describe the level of thermal maturity based on well logs. The response of the neutron and density logs is affected by the fraction of water in a formation. The hydrocarbon density also decreases due to thermal maturation. The maturity index is calculated based on the equation from Zhao et al. (2007)(Eq. 1) N: the number of log readings or the number of samples.Øn9i: neutron porosity of rock samples with 9% density porosity or higher. The 9% is a cut-off for porosity in the calculations. Values lower than 9% are indicative of very dense formations such as anhydrite or dense dolomite known as non-source shales. The cumulative value of the calculated MI is presented as a maturity index in the Kazhdumi Formation in each well. In the second step, seismic data analysis, inversion of seismic data and log prediction are carried out. Inversion analysis starts with seismic data processing using well logs and 2D post-stack seismic data. An acoustic impedance log is created by the combination of density and sonic logs. Check-shot data are used for depth-to-time conversion resulting to the correlation of well and seismic data. Following this, an initial strata or impedance model is constructed. This model is produced using the seismic volume, available well data and defined horizons. Seismic data, the initial strata model as well as available wells are applied as the input data for inversion analysis, and then the inversion method is selected. Discussion of Results & ConclusionsGenerally, thermal maturity status in the Kazhdumi Formation is between immature to mature in the study area in the central part of the Persian Gulf. According to the presented data in this study, this formation is mainly immature in 2, 3, 5 and 7 fields. Field number 4 is early mature and number 1 is mature. Field 6 shows an immature to mature level indicating that the maturity varies in this field.In the next step, seismic attributes are selected by regression analysis for MI prediction. Attributes selection is a process for the extraction of seismic attributes from the raw seismic data for modeling the target log. Multi-attribute analysis is an automatic procedure for the selection of the most relevant seismic attributes to the target log.Finally, extracted attributes are used for the log prediction by Probabilistic Neural Network analysis (PNN). It is trained based on the selected attributes in the previous step and then predicts the target parameters.Acoustic impedance is recognized as the most important seismic attribute reflecting the geological properties (Chopra and Marfurt, 2005). The optimum number of attributes for MI prediction is four which shows the lowest prediction error, although the training error continuously decreases by adding more attributes. The validation plot of the target log is estimated by excluding the data step by step from the calculation. The reliability of the regression model is tested by comparing the prediction with the actual log values.Thereafter, the 2D seismic section is converted to an MI volume. The variation in MI is continuous laterally and vertically, therefore, can be tracked throughout the basin.A comparison of the computed MI with geochemical data indicated that this method is applicable for thermal maturity prediction in the study area. The increase in MI corresponds to an increase in the Tmax values, thus providing a good indicator of thermal maturity variation.The last point to consider is the significance of the selected seismic attributes and their relationship with the target parameter. Results indicated that acoustic impedance is the most important seismic attribute in MI prediction. Acoustic impedance contains information about the velocity and formation density which are both affected by the formation fluids (Broadhead et al. 2016; Atarita et al. 2017). It is inversely related to the organic matter content (Harris et al. 2019). The computed thermal maturity is inversely related to the neutron porosity and water saturation which are both controlling parameters of the acoustic impedance in a formation.Other attributes that have been used in the log prediction are time and amplitude-weighted frequency. These attributes are related to different geological properties of source/reservoir formations (Taner et al. 1994; Chen and Sidney 1997; Chopra and Marfurt 2005). Amplitude envelope (reflection strength) mainly represents acoustic impedance and is useful for identifying porosity, hydrocarbon and gas accumulation, sequence boundaries, and lithological/depositional environment variations (Chen and Sidney 1997; Hart 2002). Average frequency is defined as a signature of events and is useful for correlation, often reflecting oil and gas reservoirs by seismic attenuation (Taner et al. 1994). Amplitude-weighted frequency is a product of the amplitude envelope and the instantaneous frequency, providing a smooth estimation of instantaneous frequency by removing spikes and noises (Chen and Sidney 1997).To sum up, well logs and seismic attributes are successfully applied to predict thermal maturity in the Kazhdumi Formation in the central part of the Persian Gulf.The analysis shows that the Kazhdumi Formation is mainly immature to early mature in the central part of the Persian Gulf.Seismic data are spatially continuous which is an advantage in source rock evaluation, resulting in continuous predictions of thermal maturity in hydrocarbon fields.Using seismic data is also cost-effective and less time-consuming than geochemical testing.

Keywords

• source rock
• thermal maturity
• well log
• seismic attribute