The manuscript investigates the influence of turbulent inflow conditions on large eddy simulation (LES) of fire plume development near the ground. The authors compared simulations with uniform inflow and turbulent inflow generated by a divergence-free spectral representation (DFSR) method under two wind conditions. The topic is relevant for fire–atmosphere interaction modeling. The manuscript provides useful numerical experiments and shows that the effect of inflow turbulence becomes more significant under stronger wind conditions. It addresses an interesting problem related to LES modeling of fire plumes. The reviewer recommends minor revision, but the paper can be improved by addressing the following comments before been considered for publication.

1 The combustion process is modeled using a one-step global methane reaction combined with an eddy dissipation model. While this simplified approach is common in LES fire simulations, the manuscript does not discuss its potential limitations. The authors should briefly discuss the implications of this simplified combustion model and whether it could influence the comparison between turbulent and uniform inflow cases.

2 The mesh resolution near the flame is stated to be 0.25 m × 0.25 m × 0.06 m, but the manuscript does not discuss whether this resolution is sufficient for LES of the fire plume. The authors should clarify whether any grid sensitivity test has been performed.

3 About the averaging time, the averaging period used for statistical analysis is 240 s, which corresponds to approximately 3–8 flow-through times depending on the wind speed. This averaging window may be relatively short for obtaining statistically converged mean fields in LES of atmospheric flows. The authors mention that the averaging time is sufficient for plume statistics, but no quantitative evidence is provided.

Minor comments:

• The captions should specify whether the plotted quantities are time-averaged or instantaneous.
• “Form Figure 4”, should be “From Figure 4”
• Temperature unit in Figure 13 is missing
• Legends in Figure 15 are not consistent, and some are two small.

This manuscript addresses a critical research gap in fire plume LES modeling by investigating inlet turbulence effects under weak/moderate wind conditions, and the research topic has practical academic value for fire plume modeling. The numerical methodology is described in detail, the results are adequately analyzed and supported by sufficient data. However, the numerical methodology verification is not sufficient, and it requires more comparison and analysis. 

(1) Why is the DFSR method is used? The authors should justify the choice and discuss its validity in generating turbulent ABL.

(2) Add mesh sensitivity analysis (at least 3 resolutions) with quantitative comparison of key plume parameters (centerline T/velocity, TKE, etc.), and clarify grid convergence and selection rationale.

(3) Provide quantitative time-averaging convergence verification (convergence curves of key statistics with averaging time) and explain the 250 s averaging start time.

(4) Justify the models used in the work (SGS model, combustion model, radiation model, etc.) and provide the missing key model parameters (EDM empirical constants, SGS model coefficients, and so on.).

(5) Calculate non-dimensional numbers (Fr/Ri) to quantify the buoyancy-inertial force balance and explain the differential effect of inlet turbulence under 2/5 m·s⁻¹ wind speeds.

(6) The authors should provide more information about the turbulent fluctuations in Figure 4. Why are the fluctuations much larger for 5ms-1 case?