This paper discusses the topic of wind turbine extreme load statistical extrapolation methods. By utilizing the Gaussian mixture model (GMM) as a multi-modal statistical model to fit the long-term distribution of load response under uncertain wind condition, the data aggregation first and fitting afterward (AFFA) method can be used to estimate the characteristic load with 50-year return period. The performance of the proposed method is validated against the existing methods through four cases: the maximum out-of-plane blade tip deflection, the maximum blade root out-of-plane bending moment, the maximum blade root in-plane bending moment, and the maximum tower base side-to-side bending moment. A comparison of the fitting first and aggregation afterward (FFAA) and AFFA approach is conducted, and the advantages of GMM in terms of both accuracy and flexibility is demonstrated. The paper is concise and well organized with most of the reviewers’ comments properly addressed. I would recommend this paper to be accepted on condition that the following minor issues are further addressed to improve the clarity and completeness of the paper:
1. The discussion regarding Figure 5 is missing in the main body of the context.
2. Regarding the Comment G.1 of Reviewer #2, in Line 68-71 of the revised manuscript, the authors claim that the tail distribution of a mixture probability distribution is likely to be dominated by only one of the distributions. This implies that the tail distribution of GMM might also be dominated by a single Gaussian distribution component since GMM is also a kind of mixture probability distribution. I am not sure this explanation is the true reason that the GMM outperforms other mixture distribution. Consequently, I recommend the authors discuss more about the advantage of GMM over other mixture distribution models or compare the performance of other mixture distribution models in Sections 3-5.
3. The word “sample” is used throughout the paper with various meanings, which is a bit confusing.

This paper discusses the topic of wind turbine extreme load statistical extrapolation methods. By utilizing the Gaussian mixture model (GMM) as a multi-modal statistical model to fit the long-term distribution of load response under uncertain wind condition, the data aggregation first and fitting afterward (AFFA) method can be used to estimate the characteristic load with 50-year return period. The performance of the proposed method is validated against the existing methods through four cases: the maximum out-of-plane blade tip deflection, the maximum blade root out-of-plane bending moment, the maximum blade root in-plane bending moment, and the maximum tower base side-to-side bending moment. A comparison of the fitting first and aggregation afterward (FFAA) and AFFA approach is conducted, and the advantages of GMM in terms of both accuracy and flexibility is demonstrated. The paper is well organized and expressed. The authors have well addressed the reviewers' and the associate editor's comments, and revised the paper accordingly.

Thank you for your careful consideration and response to reviewer and editor recommendations. We appreciate all your hard work in bringing this work to completion and for selecting WES for publication.
Paul