Sedimentary Microfacies Analysis and Reservoir Prediction of Braided River Delta Reservoirs in Central Asia’s S Gas Field

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This article analyzes sedimentary microfacies and predicts reservoir distribution in a river delta. This work is relevant for geological exploration, as deep layered deposits are often highly heterogeneous and challenging to interpret using traditional seismic methods.

 

The authors propose an integrated approach combining core, well log, and seismic analysis with enhanced seismic data.

 

Comments and Recommendations:

1. Section 3. An explicit mathematical formulation of the problem with equations and boundary conditions for seismic inversion is missing. Formulas (5)–(7) describe the inversions, but how the regularization parameters λ and constraints for predicting thin sand intervals are determined is not shown.
2. An important element of numerical model analysis—an assessment of the convergence of the problem solution as the dimensions of the system elements change (analysis of the degree of system discretization)—is missing. This, in turn, may lead to a lack of scalability of the results. This can be discussed within the framework of relation (2).

Author Response

Comment 1 Section 3. An explicit mathematical formulation of the problem with equations and boundary conditions for seismic inversion is missing. Formulas (5)–(7) describe the inversions, but how the regularization parameters λ and constraints for predicting thin sand intervals are determined is not shown.
Response: We agree. Section 3.5 (Stochastic Optimization Seismic Inversion) has been restructured into three subsections (marked in red):
(1) Section 3.5.1 Forward model and boundary conditions: Matrix formulation d = G·AI + n with G = WD; reflection-coefficient equation; boundary conditions at top/base, soft well constraints (±5%), and fairway-based penalties outside MF1–MF2 belts.
(2) Section 3.5.2 Objective function and regularization: PDF-regularized objective retained. λ = 0.8, selected by leave-one-well-out cross-validation (0.1–2.0, step 0.1), giving mean well AI correlation 0.86 and RMSE 420 (g·cm⁻³)(m·s⁻¹).
(3) Thin-sand constraints: Inverted AI below 25th percentile of channel-sand PDF + GR < 50 API (core-validated) + occurrence within MF1–MF2 fairways (Table 2).
Location: Section 3.5.1–3.5.2 (red text); Section 3.1 for GR validation.

Comment 2 An assessment of the convergence of the problem solution as the dimensions of the system elements change (discretization analysis) is missing. This can be discussed within the framework of relation (2).
Response: We added Section 3.5.3 Discretization sensitivity and Table 3, based on independent inversions of the original 2 ms and re-sampled 4 ms seismic volumes (supplementary experiments). At four calibration wells, mean AI correlation decreased from 0.86 (2 ms) to 0.83 (4 ms); mean net sand thickness deviation was 6.8%; lateral outlines shifted by less than one trace (~25 m). This demonstrates weak sensitivity to temporal discretization within the tested range.
Location: Section 3.5.3; Table 3 (red); Section 5.4 Limitations.

Reviewer 2 Report

Comments and Suggestions for Authors

This study proposes an integrated workflow for the Middle-Lower Jurassic of Block S that parallels geologically constrained thin-sand prediction strategies in continental rift basins. The structure of the study is clear, the figures are informative, and the workflow is described in detail. The topic is timely and has direct practical significance for gas exploration on the right bank of the Amu Darya. The research design is appropriate. The methods and results are adequately described. The results are clearly presented. The conclusions are supported by the results. The data are presented in tables, figures and equations correctly valued and interpreted in the paper. This research has important value. However, before publication, there are still some areas in the article that need to be improved. Below is my comment.
The sand ratio cutoffs are given without explanation of how they were derived or validated against core measurements. Please check this issue!
Authors must explicitly identify and number all nine microfacies and ensure consistency between the text, figures, and tables.
The study concludes that active rift formation during the Middle–Lower Jurassic sedimentation resulted in a graben–horst paleotopography. The study does not include a structural map or cross-section illustrating the graben–horst geometry. Since this paleotopography is said to determine the formation of the delta and the distribution of facies, a schematic structural map for Block S showing the major faults as well as the paleotopographic elevations and depressions would significantly strengthen the geological context.
The equations for the steerable pyramid are presented, but the number of decomposition scales, the number of orientations per scale, and the frequency bands of the radial filters are not specified. These are the parameters that determine the performance of the method, and they must be specified to ensure reproducibility.
The regularization weight λ in Equation 6 is defined solely as a regularization weight, without any information provided regarding its determination. Please specify the value(s) used and the selection criterion.
In Figure 16, the scale legend indicating north is missing next to the thickness values marked with color codes—although the color scale and unit notation are visible on the left, the unit should be clearly indicated.
In line 96: All Figures, Schemes and Tables should be inserted into the main text close to their first citation (e.g. Fig. 2-4.) Please check this issue elsewhere in the article.
References should be numbered in the order of their citation.
In general, the article makes a good impression.

Author Response

Comment 1 The sand ratio cutoffs are given without explanation of how they were derived or validated against core measurements.
Response: Section 3.1 now reports core–log cross-validation at 0.5 m intervals in five cored wells (n = 412 contacts). GR < 50 API agrees with core sand–mud contacts in 87% of cases. Sand-ratio cutoffs (>0.4, 0.2–0.4, <0.2) bracket porosity (~8%, ~5%) and permeability (~2 mD, ~0.5 mD) thresholds. Table 1 caption updated (red).
Location: Section 3.1; Table 1.

Comment 2 Authors must explicitly identify and number all nine microfacies and ensure consistency between the text, figures, and tables.
Response: We added Table 2 (MF1–MF9) and rewrote Section 4.2.4 with numbered microfacies codes. Table 1 references MF1, MF2, MF5. Abstract and Conclusions updated.
No.  Code  Microfacies 1    MF1   Underwater distributary channel 2    MF2   Mouth bar 3    MF3   Interdistributary bay 4    MF4   Subaqueous natural levee 5    MF5   Beach bar 6    MF6   Littoral mud 7    MF7   Coastal swamp 8    MF8   Sand flat 9    MF9   Mud flat
Location: Table 2 (red); Section 4.2.4 (red); Table 1; Conclusions (red).

Comment 3 No structural map or cross-section illustrating graben–horst geometry.
Response: We added supplementary structural maps from 3D seismic interpretation as Figure 1(d): four English-labeled panels at Upper Member top, Middle Member top, Lower Member top, and Lower Member base (depth in m; faults in red). Figure 1(a,b) retains the tectonic framework and stratigraphic column only.
Section 2 expanded to link graben–horst paleotopography to provenance and sand fairways (red text).
Location: Figure 1(d) (red); Section 2 (red).

Comment 4 Steerable pyramid: decomposition scales, orientations per scale, and frequency bands not specified.
Response: Parameter paragraph added after Equation (4) (red): K = 4 scales; N = 6 orientations (every 30°); centre frequencies 10, 20, 40, 80 Hz (one octave); 2 ms sampling; 5° orientation grid.
Location: Section 3.4 (red paragraph).

Comment 5 λ in Equation 6: value and selection criterion not provided.
Response: λ = 0.8 via leave-one-well-out cross-validation (see Reviewer 1, Comment 1).
Location: Section 3.5.2 (red).

Comment 6 Figure 16: north arrow and thickness units missing.
Response: We updated the reservoir-prediction figure using supplementary experimental outputs (now Figure 17 in the revised manuscript because Figure 1(d) was added):
(1) Figure 17(a): Cross-well stochastic inversion profile A–A′ across wells XE-1, XX-23, and XX-24 (figures/inversion_profile.png). (2) Figure 17(b): Planar predicted reservoir thickness map with north arrow, distance scale, and caption stating thickness in m (0–14 m; figures/reservoir_thickness_map.png).
The former four-panel acoustic-impedance maps (original Figure 18) contained Chinese labels and were removed; only the English-labeled panels above are retained.
Location: Figure 17 (red section); Section 4.6 (red).

Comment 7 Figures should be inserted close to first citation.
Response: Figures 2 and 4 placed immediately after first citation in Section 3.2. Figure 3 remains in Section 3.4. New Figure 1(d) placed in Section 2; reservoir-prediction results in Figure 17 (Section 4.6). The supplementary four-panel impedance maps with Chinese labels were not included.
Location: Section 3.2, Section 3.4, Section 2, Section 4.6.

Comment 8 References should be numbered in citation order.
Response: Reference list reordered to follow first appearance in text.
Location: References.

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

All my comments were taken into account and necessary corrections were made. The paper can be accepted.

Author Response

Reviewer Comment: All my comments were taken into account and necessary corrections were made. The paper can be accepted.   Response: We thank the reviewer for this positive assessment.