Saturation Calculation for Low-Resistivity Reservoirs Caused by Pyrite Conductivity

Abstract

In the current petroleum industry, as oil and gas resources are continuously developed, easily accessible reservoirs are gradually diminishing, making complex reservoirs, especially low-resistivity reservoirs, key exploration targets. By analyzing the main geological factors contributing to the formation of low-resistivity reservoirs in the LN oilfield clastic rock formations and through the integrated use of logging and petrophysical data for property modeling, we considered the impact of pyrite content on low-resistivity reservoirs and reconstructed a low-resistivity reservoir density-volume model. Using a parallel conductivity model, a saturation calculation method was established to account for the conductivity induced by pyrite in low-resistivity reservoirs, which was able to accurately identify oil layers in practical applications. This pyrite-conductivity-based method for calculating saturation in low-resistivity reservoirs effectively addresses key issues in the identification of such reservoirs, significantly improving identification accuracy. It provides a new technical approach for the evaluation and development of complex reservoirs and holds important theoretical and practical significance for enhancing the efficiency of unconventional oil and gas resource development.

1. Introduction

In oil and gas field exploration and development, accurately calculating saturation is crucial for improving efficiency and reducing risks. Low-resistivity reservoirs often exhibit characteristics such as highly saline formation water, thin interbedded sandstone and mudstone layers, high-bound water saturation, and the presence of conductive minerals [1,2,3,4,5]. These factors make it difficult for conventional methods to accurately evaluate the hydrocarbon content, increasing the complexity of reservoir evaluation. Consequently, traditional methods fail to precisely calculate hydrocarbon saturation, thereby affecting fluid identification and reservoir characterization.

To address this challenge, researchers have proposed various improved methods, including parameterized Archie’s formula [6,7,8,9], DeWitte’s shaly-sand conductive model [10], Alger’s parallel theory [11], and Poupon’s laminated medium model [12]. Many studies have explored more accurate saturation calculation methods by improving conductivity models, applying the J-function concept, and utilizing resistivity cross-plot models [13,14,15,16].

The main difficulty in dealing with low-resistivity reservoirs lies in identifying their origins, including quantitative and qualitative analyses of the rock mineral content and saturation calculations [17,18]. The complexity of lithology and petrophysical properties, combined with the low contrast in resistivity between the oil and water layers, makes it difficult for resistivity logs to effectively distinguish fluid types [19]. Currently, big data, artificial intelligence, and machine learning methods are widely used to calculate reservoir parameters based on logging curves [20]. However, the accuracy of saturation prediction may decline across different regions [21]. Some studies have attempted to calculate saturation using nuclear magnetic resonance (NMR) porosity combined with resistivity (While the results are comparable to traditional models, they fail to accurately reflect reservoir hydrocarbon saturation) [22]. Non-electrical methods, such as NMR logging techniques, including spectrum differentiation, shift methods, and time-domain difference techniques, can qualitatively identify hydrocarbons [23], but still cannot provide quantitative saturation calculations.

These studies collectively indicate that evaluating low-resistivity reservoirs requires the integration of geological, mud logging, well logging, and petrophysical data to establish highly adaptive saturation calculation methods for different regions, thereby improving the accuracy of low-resistivity reservoir identification and saturation estimation.

In this study, by analyzing the causes of low resistivity in the LN block, it was determined that the reservoir’s low resistivity is caused by the conductivity of pyrite. Consequently, a saturation calculation formula that considers pyrite conductivity was developed based on the dual-water model, and a mineral framework model incorporating pyrite was established. In contrast, traditional resistivity-based saturation calculation methods fail to account for the influence of conductive minerals, treating them in the same category as sandstone. This leads to significant errors in the saturation calculation. Furthermore, non-electrical saturation calculation methods mainly rely on capillary pressure, which imposes high experimental requirements and lacks computational stability.

2. Geological Characteristics of Low-Resistivity Reservoirs

Low-resistivity reservoirs in different geological settings may exhibit varying resistivity and formation resistivity factors. Currently, low-resistivity reservoirs are typically defined as those where the resistivity ratio between the oil-bearing layer and a pure water layer within the same oil-water system is less than 2 [24]. Key factors influencing resistivity include high irreducible water saturation, formation water salinity, and the presence of conductive minerals.

When the pore space contains a significant amount of immobile water and the formation water has a high salinity, the reservoir’s conductivity is enhanced, resulting in a low-resistivity reservoir [25]. In the Jurassic well areas of the LN oilfield, the lithology is predominantly siltstone and fine sandstone. The rocks in the formation are generally water-wet, as indicated by the core data from the wells in this area shown in

Table 1. Hydrophilicity identification analysis table of rock samples.

Detection Number	Wetting Index		Relative Wettability Index	Wetting Type
	Water Wetting	Oil Wet
1	1	0	1	Strong hydrophilic
2	0.87	0	0.87	Strong hydrophilic
3	0.8	0	0.8	Strong hydrophilic
4	0.62	0	0.62	Hydrophilic
5	0.37	0	0.37	Hydrophilic
6	0.78	0	0.78	Strong hydrophilic
7	0.83	0	0.83	Strong hydrophilic
8	0.68	0	0.68	Hydrophilic
9	0.79	0	0.79	Strong hydrophilic

. The strong water-wet nature of rocks leads to the adsorption of formation water on rock surfaces. Combined with water analysis data showing high salinity in the formation of water, the coupling of these two factors increases the bound water content in the rocks, thereby reducing resistivity.

During the early exploration of the Jurassic well areas in the LN oilfield, analyses indicated that most reservoirs were formed in a reducing environment, which is highly conducive to the development of pyrite. During diagenesis, pyrite was widely distributed [26,27]. As shown in

Table 2. Heavy mineral identification: Statistical table of pyrite content.

Serial Number	Well Number	Lithology	Total Number of Particles	Pyrite, %
1	#1	Light grayish−white fine sandstone	199	79.4
2	#2	Green−gray medium sandstone	693	79.65
3	#3	Brown coarse sandstone	544	91.54
4	#4	Brown coarse sandstone	720	95.28
5	#5	Brown coarse sandstone	593	96.96
6	#6	Brown coarse sandstone	825	87.52
7	#7	Brown fine conglomerate	1306	71.36
8	#8	Light gray−white conglomerate	546	58.29
9	#9	Light green−gray fine sandstone	368	96.47

, heavy mineral identification (minerals with a density greater than 2.9) was performed for the wells in this block, revealing a high pyrite content among the heavy minerals. During resistivity logging, the distribution of pyrite can form conductive pathways, resulting in lower measured resistivity values, consequently leading to low-resistivity reservoirs. From this, it can be concluded that the presence of pyrite reduces resistivity measurements, making pyrite one of the primary causes of low resistivity in LN oil reservoirs.

3. Physical Property Modeling of Low-Resistivity Reservoirs

3.1. Establishing the Porosity Model

Using core experimental data, modeling was conducted to fit various parameters of the well logging data in the LN oilfield, focusing on the porosity modeling of this block. As shown in Figure 1, the porosity-depth data from the core samples in the region were fitted against the corresponding density log values, establishing the relationship between porosity (POR) and density (DEN). The fitted curve demonstrates a strong correlation between porosity and density.

3.2. Establishing the Permeability Model

The core physical property data of this block were analyzed for the porosity-permeability relationship. As shown in Figure 2, the permeability fitting results indicate a moderate positive correlation between porosity (POR) and permeability (PERM). The correlation trend follows an exponential form, which was used to establish the permeability interpretation model for this region.

The modeling results were applied to wells not involved in the initial modeling for validation purposes. As shown in Figure 3, the comparison between the physical property modeling curves and core data points demonstrates that the fitted curves align closely with the trends of the core analysis data. The fitted porosity and permeability curves are consistent with the core data, providing fundamental data for subsequent saturation calculations.

4. Establishing the Irreducible Water Saturation Model

By fitting the nuclear magnetic resonance (NMR) bound water saturation data with porosity from rock samples in this block, a calculation formula for bound water saturation was established. As shown in Figure 4, the fitting results reveal a strong exponential relationship between the bound water saturation (Swi) and porosity (POR) for this block.

5. Principle of the Low-Resistivity Reservoir Saturation Calculation Model

5.1. Archie Formula

In clean sandstone, which is free of clay and fully saturated with formation water, resistivity is directly proportional to the resistivity of the formation water. The proportionality constant is referred to as the formation factor F, and F follows a power-law relationship with porosity φ. In clean sandstone that is not 100% water-saturated, the resistivity is proportional to the resistivity of the fully saturated sandstone, and the proportionality constant is referred to as the resistivity index I. By combining F and I, the formula for calculating water saturation can be derived (Equation (1)):

$$Sw={({\displaystyle \frac{abRw}{{\phi }^{m}Rt}})}^{{\displaystyle \frac{1}{n}}}$$

(1)

In the formula, the rock electrical parameters are a = 2.3392, b = 1.0291, n = 1.8186, m = 1.2643; R_w = 0.015 ohm.m is the formation water resistivity; R_t is the resistivity of undisturbed formation, ohm.m; φ is porosity, v/v; and S_w is the formation water saturation, v/v.

5.2. Dual-Water Model

The dual-water model is used to calculate the saturation of irreducible water and shaly reservoirs, taking into account the conductivity of both bound water and wet clays. This model assumes that apart from the difference in formation water conductivity due to salinity, the conductivity of shaly formations is the same as that of clean formations with equivalent porosity and saturation. The conductivity of formation water is determined by the parallel conductivity of free water and bound water, with bound water treated as a solution with distinct conductive properties [28,29,30]. Let the conductivity of the free and bound water mixture be C_wm and the formation conductivity be C_t. Based on Archie’s formula for sandstone, Equation (2) can be derived as follows [31]:

$$Ct={\displaystyle \frac{S{w}^n{\phi }^{m}Cwm}{ab}}$$

(2)

According to the concept of the two-water model, the conductivity of the mixture is determined by the parallel conduction of free water and bound water [32]; thus, Equations (3) and (4) can be obtained:

$$C_{wm}({\phi }_f+{\phi }_b)={\displaystyle \frac{{\phi }_f}{Rwf}}+{\displaystyle \frac{{\phi }_b}{Rcw}}$$

(3)

$$Cwm={\displaystyle \frac{[\frac{(Sw-Swb)}{Rwf}+\frac{Swb}{Rcw}]}{Sw}}$$

(4)

Substituting Equations (3) and (4) into Equation (2), we obtain Equation (5):

$$Ct={\displaystyle \frac{[\frac{S{w}^n}{Rwf}+S{w}^{n-1}(\frac{1}{Rcw}-\frac{1}{Rwf})Swb]{\phi }^m}{ab}}$$

(5)

where R_wf is the free water resistivity, ohm.m; R_cw is the resistivity of bound water, ohm.m; and S_wb is the bound water saturation, v/v.

5.3. Method for Constructing the Low-Resistivity Reservoir Saturation Calculation Model

In the calculation and evaluation of saturation in low-resistivity reservoirs, traditional well-logging-based saturation calculation methods are often limited by their adaptability to complex geological conditions, resulting in inaccurate saturation estimations. A new saturation model was established for the low-resistivity reservoirs in this block using a parallel conductivity model to calculate the formation water saturation [33]. The feasibility of this model was validated through comparison with other models.

As shown in Figure 5, the volume model for this region was constructed primarily by considering the sandstone, shale, pyrite, and fluid components. A density-volume model was developed based on the volumetric content of these components.

The relationship between the components can be obtained using the density skeleton volume model (6):

$$(1-\phi -Vsh-Vp){\rho }_{sand}+Vsh{\rho }_{sh}+Vp{\rho }_p+\phi S_{xo}{\rho }_{mf}+\phi (1-Sxo){\rho }_o=DEN$$

(6)

where V_sh is the clay content, v/v; ρsand is the sandstone density, g/cm³; ρ_sh is mudstone density, g/cm³; ρmf is slurry density, g/cm³; ρ_o is oil density, g/cm³; ρ_p is pyrite density, g/cm³; S_xo is the water saturation of the flushing zone, v/v; V_p is the pyrite content, v/v.

Through the establishment of the density skeleton volume model, the relationship between the pyrite content and the water saturation of the flushing zone can be obtained, as shown in Equation (7):

$$Vp=ASxo+B$$

(7)

Of which:

$$A=-{\displaystyle \frac{\phi ({\rho }_{mf}-{\rho }_o)}{{\rho }_p-{\rho }_{sand}}}$$

(8)

$$B=-{\displaystyle \frac{\phi {\rho }_o+(1-\phi -Vsh){\rho }_{sand}+Vsh{\rho }_{sh}-DEN}{{\rho }_p-{\rho }_{sand}}}$$

(9)

For each parameter in the density skeleton volume model, the shale content is calculated by the GR curve ($V_{sh}={\displaystyle \frac{2^{GCUR\cdot SH}-1}{2^{GCUR}-1}}$, $SH={\displaystyle \frac{GR-G{R}_{\min }}{G{R}_{\max }-G{R}_{\min }}}$, the GR curve is the natural gamma curve, which is the second red solid line in Figure 3, GCUR is an empirical coefficient, with a value of 2 for older strata and 3.7 for Tertiary strata), and the porosity is calculated by the physical property modeling relationship. As shown in Figure 6, a new rock conductivity model tailored to the characteristics of this block was designed. In this model, R* represents the resistivity of the mineral framework model excluding pyrite, while R_t denotes the resistivity of the entire mineral framework model, corresponding to the formation resistivity measured by resistivity logging.

The two-water model and pyrite conduction are combined by parallel conduction. The components other than pyrite are regarded as a whole by the two-water model, which is equivalent to the equivalent model in Figure 7, and Equation (10) is obtained:

$$Cxo=CpVp+C^{\ast }{V}^{\ast }=CpVp+C^{\ast }(1-V_p)$$

(10)

In the formula, V* = 1 − V_p is the volume of other components in the formation except pyrite, v/v; C* is the conductivity of other components of the formation except pyrite; and C_p is the conductivity of pyrite.

The sandstone, mud, mud, and bound water are regarded as a whole. Since pyrite is removed, the porosity in this whole is expressed as Equation (11), and the equivalent conductivity formula of the whole two-water model is obtained, namely Equation (12):

$${\phi }^{\ast }={\displaystyle \frac{\phi }{1-Vp}}$$

(11)

$$C^{\ast }={\displaystyle \frac{[\frac{Sx{o}^n}{Rmf}+Sx{o}^{n-1}(\frac{1}{Rcw}-\frac{1}{Rmf})Swb]{\phi \ast }^m}{ab}}$$

(12)

In the Equation, R_mf is the mud resistivity, ohm.m; S_xo is the water saturation of the flushing zone, v/v; and φ* is the porosity of other components except pyrite, v/v.

By substituting Equations (11) and (12) into Equation (10), the calculation formula for the flushing zone saturation is obtained, that is, Equation (13):

$$C_{xo}=C_p{V}_p+{\displaystyle \frac{[\frac{{S_{xo}}^n}{Rwf}+{S_{xo}}^{n-1}(\frac{1}{Rcw}-\frac{1}{Rwf})Swb]{\phi \ast }^m(1-V_p)}{ab}}$$

(13)

Substituting Equation (7) into Equation (13), the final calculation formula of flushing zone saturation can be obtained, that is, Equation (14):

$$\begin{array}{l}{C}_{xo}=C_P(ASxo+B)+\\ {\displaystyle \frac{[\frac{Sx{o}^n}{Rmf}+Sx{o}^{n–1}(\frac{1}{Rcw}–\frac{1}{Rmf})Swb]{(\frac{\phi }{1-ASxo-B})}^m(1-ASxo-B)}{ab}}\end{array}$$

(14)

Similarly, according to Figure 8 and Figure 9, the water saturation formula of the undisturbed formation can be obtained, that is, Equation (15):

$$C_t=C_p{V}_p+{\displaystyle \frac{[\frac{S{w}^n}{Rwf}+S{w}^{n–1}(\frac{1}{Rcw}–\frac{1}{Rwf})Swb]{(\frac{\phi }{1-V_p})}^m(1-V_p)}{ab}}$$

(15)

Since the detection depth of the triple-porosity logging corresponds to the flushed zone, the flushed zone water saturation is first calculated using shallow resistivity through Equation (14) based on the aforementioned saturation calculation equations. The calculated flushed zone water saturation is then substituted into Equation (7) to obtain the pyrite content. Finally, the pyrite content is substituted into Equation (15) to determine the original formation water saturation.

5.4. Selection and Optimization of Model Parameters

For the low-resistivity characteristics of the LN oilfield, this model emphasizes the influence of irreducible water saturation and the presence of conductive minerals on the formation resistivity. These factors are not isolated but are coupled together, collectively shaping the properties of the low-resistivity reservoirs.

For the parameters in the equation, the mud density ρ_mf is set to 1.35 g/cm³, the mud resistivity R_mf is set to 0.168 ohm.m, the water density ρ_w is set to 1.0 g/cm³, the oil density ρ_o is set to 0.8 g/cm³, the sandstone density ρ_sand is set to 2.65 g/cm³, the mudstone density ρ_sh is set to 2.45 g/cm³, and the pyrite density ρ_p is set to 4.99 g/cm³. Since the apparent resistivity of pyrite has no accurate value [34], the pyrite resistivity R_p is set to 0.01 ohm.m. The resistivity R_sh and φ_sh porosity of mudstone are selected by the pure mudstone section to calculate the immovable water resistivity R_cw using Equation (16):

$$Rcw={\displaystyle \frac{Rsh{{\phi }_{sh}}^m}{a}}$$

(16)

Except for engineering parameters such as mud density and mud resistivity, all other mineral physical parameters are referenced from Schlumberger’s mineral logging response parameters [35,36]. After determining the parameter values required for the equations, the low-resistivity model can be used to calculate the formation water saturation. It is particularly emphasized here that the equations are solved iteratively, where saturation is calculated to estimate the resistivity. This process continues until the error in the true resistivity is minimized, at which point the corresponding saturation is determined. The results are then compared with those obtained using the conventional dual-water model and Archie’s formula to validate the feasibility of this model.

6. Analysis of Geological Characteristics of Low-Resistivity Reservoir

Considering the existence of conductive minerals and the influence of irreducible water in this block, the saturation calculation equation is improved by establishing the corresponding formation rock conductivity model to capture the influence of complex factors on resistivity to accurately calculate the saturation. As shown in Figure 10, the analysis of the low-resistivity layers in the area shows that the well section 4565–4570.5 m is the test oil layer; the fourth channel in the figure is resistivity logging, and the resistivity of the test reservoir is less than 2 ohm.m; the ninth is the saturation calculation and comparison. The water saturation calculated by this method is significantly lower than that of the Archie formula and double water model, and the daily oil production of this layer is 296.75 square oil layer. The saturation calculated by this method is compared with the eighth irreducible water saturation. It can be seen that the movable water saturation is low, which is in line with the conclusion of the oil test.

7. Conclusions

By analyzing the main geological factors contributing to the formation of low-resistivity reservoirs in the clastic rocks of the LN oilfield and through the integrated use of well logging and petrophysical data for property modeling, the impact of pyrite content on low-resistivity reservoirs was considered. A density-volume model was used to derive the expression for calculating the pyrite content and flushed zone saturation. Based on the parallel conductivity theory, a saturation model was developed, incorporating the conductive effect of pyrite. This model first calculates the flushed zone water saturation and pyrite content using shallow resistivity, and then determines the original formation water saturation using deep resistivity and the calculated pyrite content.

The model demonstrated significant advantages in the practical application within the LN oilfield, showing high consistency with the oil testing results, which validates its effectiveness. It provides strong support for decision-making in oilfield development.