Please find my review attached.

The manuscript represents objective global climatology of moisture transport axes. The paper is well-written and the figures are well designed. The method effectively captures moisture transport outside the tropics, including high latitudes.

My major concern is the moisture transport axes (MTAs) within the deep tropics. Specifically, Fig. 3 shows a high frequency of MTAs in the ITCZ region, which I interpret as moisture transport along the ITCZ from east to west. As far as I understand, the ITCZ signifies moisture convergence driven by surface trade winds, typically not organized into distinct filaments. In this instance, the method identifies convergence into the ITCZ as a peak in transport and creates an impression of moisture transport along the axis, not into the axis.



In Supplements, the authors show an alternative method based on normalised water vapour transport. This method gives similar results in extratropics but does not detect ITCZ as an MTA; therefore, in my view, it is better suited for the identification of MTAs. Alternatively, separating MTAs into two types, e.g., dominated by transport along the axis vs transport into the axis or dominated by advection vs water content, might help avoid confusion.

Furthermore, the authors assert that the method is well-suited for monsoons based on good performance in India and West Africa. However, it appears that, e.g., the Australian monsoon, which brings moisture toward the north of the Australian continent in January-March/April, is not explicitly represented. It may be that moisture transport in this region is not organised in filaments. If this is the case, discussing this nuance in the manuscript could provide a more comprehensive view of the method's applicability across different monsoon systems.

Finally, it would be good to cite the recently published paper (Konstali et al. 2024) that utilises MTAs described in this study. Even though Konstali et al. (2024) cite the pre-print of the manuscript under review submitted to another journal, it would be worthwhile adding a citation of Konstali et al. (2024) in this manuscript highlighting how MTAs contribute to rainfall in different parts of the globe. Following the approach outlined in Konstali et al. (2024), it may be valuable to investigate how frequently MTAs are linked with cyclones and/or fronts, especially in light of findings by Spensberger and Spengler (2018), which demonstrated that heat and moisture transport can be effectively utilised for classification of fronts. Integrating this analysis into Section 5 of the manuscript could significantly enhance the discussion.

Specific comments:

l. 38-42: I would disagree with the statement that ‘it is unclear to what extent the concept [of AR] can and should be extended to the subtropics.’ Subtropical latitudes are indeed affected by extratropical weather systems, as you suggested in that paragraph, more often in winter, but also warmer months. Catto et al (2015) show that the highest percentage of cold fronts associated with WCB are found between 20-30degS since the majority of WCBs in their dataset are found equatorward of 40degS. Some papers use the term AR in relation to Australian rainfall (e.g., Rauber et al 2022 (10.1029/2020JD032513), Reid et al. 2020, 2021, 2022). I would argue that subtropics are regions of active ARs as high moisture sources are nearby (Gimeno et al 2021). I think Fig4b of the submitted manuscript supports the idea that moisture is actively transported in the subtropical regions (15-25 deg), depite a relatively low total column water (Fig. 4a). Even though the distribution of axes frequencies at the top of Fig.4 shows a dip in the subtropics, these MTAs are important and might be responsible for extreme rainfall in subtropics and low midlatitudes.

l.93: A single threshold based on the global average is potentially biased towards low latitudes with higher moisture content and, therefore, contradicts the idea of avoiding thresholds that you postulated at the very beginning.

L.130: In fig. 2d, is the transport away from a cyclone associated with a warm front? It seems to be too far from the cyclone centre to be a part of a warm front but maybe it still is.

l.135: The moisture transport around a TC does not represent the actual moisture transport which has not been identified as AR by any of the algorithms in ARTMIP-ERA5. Since TCs are relatively rare, their contributions to climatology are small. However, is there a way to remove or improve those axes?

l. 147: (raised in my general comments) To what extent does ITCZ advect moisture in the zonal direction? My understanding of ITCZ is that it represents the convergence of moisture advected from subtropics. In my view, this advection mostly happens not in filaments.  Fig.4 (c,d), tropical moisture axes advect moisture eastward and mainly equatorward.

Perhaps in Fig.3, you can find a way to indicate areas with predominantly poleward and equatorward advection, that might be useful to understand tropical MTAs. However, as I said, the normalised moisture transport from Supplement better represents moisture transport in the tropics, in my opinion.

l. 165, Section 5: This section focuses on Fig 4 and, therefore, mixes tropical and extratropical MTAs. Maybe Fig. 4 can be discussed separately in a different section and the discussion in Section 5 can be confined more to regions where ARs are observed highlighting their relationships with fronts and cyclones and their role in tropical-extratropical interactions.

l. 200: The peak around 60S in Fig. 4c is weak but can be defined. Can eastward flux at 60 deg lat be associated more with eateries around Antarctica than circulation around the cyclone centre, which usually happens over short distances? I am not sure how this can be objectively measured other than separating the moisture transport into transient and stationary components (which is an interesting avenue, perhaps).

l.205: I’d recommend showing distributions for tropical and extratropical MTAs separately as they may represent different processes.  Also, as far as I understand, this plot shows moisture transport only along the axes, it does not tell you how much moisture is advected within atmospheric rivers as a whole. Therefore, the distribution of moisture transport within atmospheric rivers might be more poleward.

l.218: You mention that high latitude ARs/moisture transport is similar to AR in the midlatitudes. In the two Antarctic cases shown in the paper, AR/MTA stretches from the subtropics to the Antarctic coastline. What do you mean by saying that those high-latitude events are ‘similar’ to their ‘midlatitude counterparts’? I read that they are similar but still different (maybe that’s not what you meant). In my view, they are created by the same process, i.e., moisture advection associated with a cyclone/frontal circulation that must be the leading cause of ARs formation.

l.252: My reading of Papritz et al is they explore moisture transport into the Arctic. Can you mention their Figure that shows the extrusion of the warm moist air from the polar region? Also, the statement that the transport around the cyclone centre advects air that is still relatively warm should be supported by temperature analysis. As a side comment, it would be interesting if a similar approach could be applied for heat advection.

l.264: Subtropical climate is very variable. In winter, extratropical weather systems are often observed in subtropics, especially in the area below the subtropical jet that increases baroclinicity. In warmer months, extratropical circulation shifts further poleward but fronts are still frequently intruding subtropics. Either frontal circulation or cutoff lows can create a strong moisture advection in the subtropics, particularly in late summer-early autumn. ARs are important for rainfall in South Africa (e.g., Blamey et al 2018), Australia (e.g., Reid et al 2022), subtropical South America (Reis et al. 2022), the Middle East and North Africa (Massoud et al. 2020), Southeast China (Xu et al. 2020 https://doi.org/10.1071/ES19027). The subtropical climate varies a lot from one region to another but there are many subtropical areas for which moisture transport in the form of ARs is critically important as they create extreme rainfall events.

l. 292: Can you explain why rainfall associated with this moisture transport remained weak given a close proximity to the ITCZ? (Though it could have been strong in a relative sense in this arid region)

l. 305: It is a good comment on moisture recycling, perhaps, it could be mentioned earlier in the manuscript to avoid misinterpretation (apologies, if I missed a mention).

l.313: In Fig. 4c, the occurrence of poleward moisture transport in high latitudes(>60deg) is higher than the equatorward transport, suggesting that more moisture is advected into the polar regions than outside.

l.321: The moisture transport ‘along the straits of the Maritime continent’, also mentioned earlier in the manuscript, is interesting. Looking at seasonal rainfall (e.g., Fig 1 in Bukowski et al. 2017, 10.5194/acp-17-4611-2017), I cannot see why the moisture transport axis lies exactly between islands. If it is not an artefact, can it be explained?

Supplement, l. 14: Following Wille et al. and Gorodetskaya et al., I think, moisture transport into high latitudes is very important for polar regions. You say that a lower threshold leads to spurious detections over Antarctica. Did you check if axes that would be identified with a lower threshold were not associated with precipitation over Antarctica and, therefore, could be suspected as spurious?

Technical comments:

FIg. 1: Please define the blue shading.

Fig. 3: I barely see continents, can you please make them more visible?

Fig. 5,6,7,9: Could you please mark the location of the target area for moisture axes with a symbol? Are Fig.5 (c,d) and subsequent plots shown for a particular season?