Based primarily on CESM1.2 large ensemble results, this manuscript investigates the evolution of near-surface wind speed (NSWS) in the Northern Hemisphere under varying CO₂ concentrations. It reveals the particular behavior of NSWS over Europe and explores the possible underlying causes. What interests me most is the finding that the Atlantic Meridional Overturning Circulation (AMOC) plays a key role in driving the evolution of NSWS over Europe in response to CO₂ changes, mainly by modifying the temperature gradient over the Atlantic. The analysis is sound, and the logical flow is clear. To effectively disseminate these new findings, I believe this manuscript is worthy of publication in this journal.

Some comments:

1. A large portion of the manuscript discusses NSWS changes across the entire Northern Hemisphere, while the title focuses specifically on Europe. It would be better to unify the scope; either adjust the title or concentrate the discussion more on Europe.

2. Both the CESM large ensemble and CDRMIP experiments have limitations. The former may suffer from model dependency, while the latter may be affected by internal variability due to the limited number of models involved. Therefore, it remains uncertain whether NSWS would behave exactly under CO2 removal scenario as presented in this manuscript. However, the results do robustly suggest that AMOC is the primary driver of the NSWS evolution.

This study investigates how near-surface wind speed (NSWS) across NH land areas responds to CO2 ramp-up and ramp-down. Increasing CO2 concentrations are shown to lead to an overall decrease in NSWS. When CO2 levels decrease, NSWS quickly recovers within the first several decades but then stall until the end of CO2 removal. NSWS then declines again, particularly over Europe, in the initial stabilization period, followed by a slow recovery in the mid to late stabilization period. These non-monotonic changes in NSWS are attributed to the AMOC changes and the associated changes in temperature gradients and westerly jets. This finding underscores the pivotal role of the AMOC in shaping the hysteresis of NSWS in CDR experiment.

Though short and simple, this study addresses an interesting subject not covered in previous studies. The results of this study are not surprising and can be inferred from previous studies (e.g., An et al. 2021). Nevertheless, it is beneficial to quantify it. In this regard, I appreciate this study. However, I found substantial room for improvement in the study. Among others, the possible mechanism(s) is too simple and not justified in quantity. I suggest that the authors address the following issues when revising the manuscript.

Mechanism

NWSW is explained for both NH and NA. However, only the changes over NH are presented in Fig. 1b. It would be helpful to show the temporal evolution of NA NSWS in a new panel below Fig. 1b. This would be particularly useful when discussing Fig. 3.

NSWS changes are mainly explained by SAT gradient changes. However, SAT gradient does not explain everything. An example is Fig. 3. While AMOC continues to weaken from the ramp-up to early ramp-down periods (Fig. 3a), NA SAT gradient increases during the ramp-up period and then rapidly decreases (Fig. 3d). This SAT gradient changes do not correspond to NA westerly jet  changes (and presumably NA NSWS changes as well), especially during the ramp-up period (Fig. 3e). Can you explain why?

Another example is Figs. 1b and 3. Fig. 1b shows a rapid weakening of NH NSWS during the initial stabilization period, followed by a steady decrease afterward. This evolution cannot be explained by NH SAT gradient change shown in Fig. 3b. This mismatch indicates that other factors also affect NSWS changes. Such factors should be discussed in detail.

Model biases

It is stated that NSWS climatology in each model is similar to ERA5 climatology. The spatial correlation over 60°S–70°N (not just NH land areas) ranges from 0.72 to 0.85 (L125). Is this also the case when considering only NH land areas? Beyond the spatial distribution, does the model reproduce the intensity of NSWS? I suggest presenting NWSW climatology for ERA5 and each model in the SI.

Comparison to CDRMIP

Figs. 1b and S2 reveal a notable difference in NSWS changes among the models. These differences are qualitatively related to the different AMOC changes. Could you quantify it by establishing NSWS-AMOC relationship among the models? Since NSWS changes resemble 500-hPa wind changes, 500-hPa winds could be also used to increase the sample size.

Cross-member correlation

I am not sure what the purpose of this analysis is. Ensemble spread of AMOC is rather small during the ramp-up period and is unlikely to affect ensemble spreads of NA SAT gradients and westerly jets. As ensemble spread increases during the ramp-down period, AMOC more effectively explains ensemble spread of NA climate properties. This does not suggest “the crucial effect of AMOC recovery on weakening the NA SAT gradient” in L248-L249. Rather it suggests that the internal variability (not trend!) of NA climate properties is closely associated with AMOC variability during the ramp-down period.

Bi-regression analysis

I had a difficult time understanding L253-L257: “During the ramp-up period (2001–2140), GMST and AMOC explain the variance of NH extratropical NSWS for 98% and 1.2%, respectively. During the early ramp-down period (2141–2220), GMST and AMOC explain 95.3% and 4.2% of the variance, respectively. While during the late ramp-down period (2221–2280), GMST and AMOC explain 24.5% and 73.2% of the variance, respectively”. Does this mean that AMOC is not important for NSWS changes during the ramp-up and early ramp-down periods? If so, this contradicts the key conclusion of the paper – the critical role of AMOC hysteresis in NSWS hysteresis.

Minor points

Fig. 4 is not critical and could be moved to the SI.

Fig. 5 should be extended to the year 2500 (i.e., the end of stabilization). The same applies to Fig. 7.