The authors carried out downscaling simulation of the mid-Holocene Australian climate based on the WRF and compared it with the CESM simulation of the mid-Holocene Climate and obtained an added value.  The study is certainly very interesting. Therefore, I suggest a minor revision.

1. I see that the authors have also given almost identical distribution of vegetation in the WRF for both the Holocene and control experiments. This is similar to the fact that most PMIP experiments still use the vegetation distribution of present day. Actually, the reconstruction results suggest that the vegetation distribution in the mid-Holocene may have been quite different from the present day and may have impacted on the regional climate (Thompson et al., 2022; Sun et al., 2023). For this reason, I suggest that authors might try to drive vegetation models (Such as Biome et al.) with CESM outputs and WRF outputs in order to examine the regional vegetation response to mid-Holocene climate change.

Thompson, A.J., Zhu, J., Poulsen, C.J., Tierney, J.E., & Skinner, C.B. (2022). Northern Hemisphere vegetation change drives a Holocene thermal maximum. Science Advances, 8.

Sun, Y., Wu, H., Ramstein, G. et al. Revisiting the physical mechanisms of East Asian summer monsoon precipitation changes during the mid-Holocene: a data–model comparison. Clim Dyn 60, 1009–1022 (2023). https://doi.org/10.1007/s00382-022-06359-1

2. Besides, in the introduction somewhere, previous progress in downscaling studies on other regions of the mid-Holocene should be reviewed.

Huo, Y., Peltier, W. R., and Chandan, D.: Mid-Holocene monsoons in South and Southeast Asia: dynamically downscaled simulations and the influence of the Green Sahara, Clim. Past, 17, 1645–1664, https://doi.org/10.5194/cp-17-1645-2021, 2021.

Summary

This manuscript assessed the ability of the CESM and WRF to simulate the mid-Holocene climate of Australia and the equatorial tropics of the Indonesia-Australian monsoon regions with respect to bioclimatic modelled proxy data, in terms of temperature, precipitation and plant available water index. This study provides the first downscaled paleoclimate analysis of the mid-Holocene in Australia, and is well written. The followings need to be commented or addressed before it is publishable.

General comments

For Introduction

1. Add the description about the need for RCM in paleoclimate modeling and the progress in downscaling for the paleoclimate simulations.

For the Results   

1. It is inappropriate to use only one metric, the mean absolute error, to assess the model simulations. As is shown in Fig.5-6, the differences between MH and piControl over the northern Australian in WRF and CESM show a warm and dry conditions, which is totally disagree with the pollen proxy datasets, and those differences are even more pronounced in WRF. It seems that the downscaling results do not improve the simulations over the Australian.
2. Does the warm-dry biases over the northern Australian in WRF pre-industrial simulation have an impact on the simulation results for MH?
3. Change the 850 hPa divergence difference in Fig.9 to the Whole-layer water vapor flux divergence difference.
4. Give the possible reason for the differences in annual plant available water index over the Indonesia between WRF and CESM in Fig.10.
5. Add some discussion about the added value of WRF compared to the driving forces.

Specific comments

What does the dots mean in the Figure S5?