Review of "Lightning-intense deep convective transport of water vapour into the UTLS over the Third Pole region", by Prashant Singh and Bodo Ahrens, submitted to Atmospheric Chemistry and Physics (ACP)

This paper investigates the role of lightning-associated convection in transporting water vapour into the upper troposphere and lower stratosphere (UTLS) over the Himalayas and Tibetan Plateau ("Third Pole") region. The authors use lightning data from the Tropical Rainfall Measuring Mission (TRMM-LIS), along with forward trajectories derived from ERA5 reanalysis and high-resolution ICON-CLM simulations, to track moist air masses. The goal of the authors is to assess their contribution to the well-documented water vapour enhancement observed by MLS and ACE-FTS (which are more appropriate than AIRS) within the Asian Summer Monsoon (ASM) anticyclone.

It is well known that, in the tropical lower stratosphere, and also within monsoon systems, the stratospheric water vapour entry values are primarily controlled by the freeze-drying of moist tropospheric air at the cold point tropopause (CPT) (Brewer, 1949; Randel and Park, 2019; Smith et al., 2021; see also the introductions of the Ploeger et al. or Clemens et al. papers cited in your paper). Deep convection that directly crosses the tropopause is also under debate but remains much more difficult to quantify. As a result, all statements related to the stratosphere in this paper remain very qualitative; even the position of the WMO tropopause or the cold point tropopause is completely ignored. In my view, no robust conclusions can be drawn from this study regarding any impact on stratospheric water vapour.

Unfortunately, even regarding the upper troposphere, the findings are quite weak, especially when compared to earlier studies such as Price et al. or Singh and Ahrens (2023). The correlation between enhanced upper tropospheric water vapour and lightning counts is not new, and actually appears more clearly in daily data than in strongly averaged climatologies such as Fig. 1. Even the domain-averaged daily time series (Fig. 2) show large inconsistencies, with unexplained spikes between 15 March and 15 June. The correlation coefficients are actually weakest during the monsoon time (Table 3), the time period when intense thunderstorm activity is expected.  

The only truly new contribution, in my view, is the comparison of trajectory behavior between ERA5 and ICON-CLM, as you also highlight in your abstract. While the coarser-meshed (~30 km) convection-parameterized ERA5 data show slow ascent, with air parcels crossing the Himalayas and reaching the upper troposphere over the Tibetan Plateau, the convection-permitting km-scale ICON-CLM model reveals faster vertical and more direct transport for the same events (Figs. 3, 4, and 5). 

However, there is significant potential to improve the presentation and interpretation of these results. For example, the color bar in Fig. 5 is not readable, and the visual contrast makes interpretation difficult. Moreover, interpreting the highest trajectory points in Figs. 3d and 3g as being in the stratosphere seems, at best, an overinterpretation. Without proper reference to the cold point or WMO tropopause, such a claim cannot be supported with confidence.

In your conclusions, you attempt to link your findings to the recently identified significant wet bias in the lowermost stratosphere in climate models (Charlesworth et al., 2023; Ploeger et al., 2024). However, your results are strongly confined to the upper troposphere. Furthermore, the wet bias in ERA5 upper tropospheric water vapour, diagnosed in your paper by comparison with MLS and AIRS data, is also present in the high-resolution ICON-CLM model, which you otherwise describe as more physically realistic in terms of vertical transport along trajectories. This is really confusing.

Given these concerns, I can only recommend rejection of the current version. The manuscript would need to be fundamentally rewritten. Possibly, Figs. 3, 4, and 5 could serve as a starting point for a completely new and more focused version.

A few other important points:

**Introduction**  

If you want to make claims about the stratosphere, large parts of the introduction would need to be rewritten to reflect the relevant processes and literature more accurately.

**Singh 2015**  

This reference appears to be grey literature and, in my view, should not be used in a peer-reviewed journal submission.

**Lagrangian Tracking**  

There is no proper citation of the Lagrangian trajectory tool used in the study. I strongly recommend using a well-established and widely cited tool such as FLEXPART or MPTRAC for this type of analysis.

The study focuses on water vapour transport mechanisms driven by lightning-induced deep convection in the Third Pole region, a global lightning hotspot where UTLS (upper troposphere and lower stratosphere) water vapour exchange significantly impacts climate. The research addresses critical regional and global climatic implications.The combination of multi-source data (TRMM-LIS, ERA5, AIRS, MLS) with km-scale ICON-CLM modeling and Lagrangian tracking creates a robust "observation-reanalysis-model" validation framework, demonstrating methodological advancement.The logical and the results are reasonable. The structure is clear, and the writing is well. However, there are some concerns about this study.

1.Lightning datasets (TRMM-LIS, ISS-LIS) span long periods, but the impact of temporal discrepancies (TRMM: 1995–2015; ISS-LIS: 2019–2020) on result consistency requires clarification.

2.Incorporate recent (post-2023) observational studies on deep convection in the Third Pole (e.g., ground-based radar or satellite retrievals) to enhance timeliness.

3. Correlation analysis (e.g., Table 2) shows the highest lightning-UT water vapour correlation over the Tibetan Plateau (r = 0.78). Whether the region’s low lightning frequency (<4 flashes/km²/year) affects statistical significance?