This manuscript analyzes historical drought trends in Australia using multiple drought indicators and explainable machine learning to assess hydrometeorological drivers and variability influences. While well-structured with a robust methodology and significant findings, improvements in clarity, statistical transparency, and rigor are needed. The attached document evaluates the manuscript against HESS criteria with recommendations to enhance its impact and alignment.

The manuscript presents a statistical assessment of the historical trends of drought in Australia. The methodology and the analysis are well designed and contain elements of novelty, while the results are interesting, relevant and generally well discussed. There are, however, area where the manuscript could be improved, in particular with regards to the explanation of some of the methodologies and their contribution to the final results, as well as discussion of the implication of the statistical choices made.

Overall, the manuscript is of good quality and I recommend its publication in HESS with minor amendments.