General Comments

This manuscript describes the D-KULT project, a national scale initiative aimed at demonstrating the operational feasibility of eco-efficient flight routing by building an end-to-end information chain connecting numerical weather prediction, climate response modeling, flight planning, and air traffic control. The topic is timely and societally important. Furthermore, I commend the authors for leading such a complex project across multiple domains.

I find several of the results interesting but many of the details on why such results are observed or sensitivity analyses are deferred to future forthcoming publications.

My high-level concern is that this paper in its current structure ends up downplaying the scientific insights that were gleaned from the project. I therefore recommend the authors to revise the paper with the primary goal being a structural reorientation of the paper to make the new scientific nuggets be highlighted and perhaps sections that primarily describe the consortium workflow (e.g., section 3.1) can be moved to an appendix or supplementary material.

Specific comments

This paper covers a breadth of topics and I realize that it both serves as a "preview" of other publications in preparation and a summary of prior published articles so I keep my specific comments focused on only a few of the issues that I think struck me the most. I also realize that this paper is already fairly long and some of my suggestions ask for more quantitative results, but perhaps moving some of the programatic descriptions of the activities and coordination to a supplementary material might help free space to present some quantitative results. I leave this to the authors.

• The two-moment cloud ice microphysics scheme

In my initial reading I thought this was one of the most important scientific results in the paper, but I realized that Figure 8a is actually a subset of lines (RHi > 100%, 110%, 120%) from Hanst et al 2025 (which additionally has RHi > 105%). Any new insight on this figure beyond Hanst et al 2025, should be really highlighted in this section. Additionally, the paper does not also show the precision curve from the earlier Hanst et al paper which suggests that even in the "3/10 ensemble members agree" case the precision was ~40% so, of the predicted ISSR regions less than half were actually confirmed by radiosondes. Which I think is the central challenge given the <10% prevalance rate. Is this correct? If so I think this needs to be highlighted and further discussion on this would be warranted, especially if new insights since the publication of Hanst et al 2025 have been gleaned.

It will also be useful to discuss what the spatial coverage of the radiosonde data was in this comparison.

The paper also states that the 10-member ensemble was initialized from the operational one-moment analysis which "has practically no supersaturation". This results in the spin-up period. I would be interested to know what the resulting ISSR distribution and its evolution look like compared and if it reflects the measured atmospheric state. Additionally did the spin-up period present any real operational hurdles or was this largely immaterial?

The forecast skill analysis (Section 4.1.2) shows ETS declining from 0.7 at 6-hour lead time to 0.4 at 24–30 hours. Crucially, the paper states that "the forecast skill is insensitive to the choice of microphysics scheme". This I think is a result worth expanding on. If the primary source of uncertainty is the underlying linearities in the atmosphere rather than the microphysics parametrization, this has important implications - improving the micorphysics scheme alone cannot substantially extend the forecast horizon for PPC prediction. I encourage the authors to discuss this more explicitly and connect it to the practical planning timelines discussed in the other sections.

• The operational tests

Section 4.2.3 reprots that 9 flights were selected for active contrail avoidance and 16 used the tool in shadow mode. However essentially no quantitative information on these flights are presented - fuel burn penalties, comparison of predicted vs observed contrails etc. The authors state that "evaluation of the changing climate effect of the test flights is ongoing and the results will be published elsewhere". This presents a real problem from a review perspective because it makes it difficult to evaluate the claims that the program accomplished the implementation of eco-efficient flights without referring to future publications. In the current state these sections describe the activites and workflow. I think it will benefit the paper if at least some minimum quantitative results are presented.

In lines 589 presents an interesting finding about the average thickness of the avoidance regions being 4000ft while the assumed reference value was 2000ft. This is an important finding that I wish had more details on the characteristics of the PPC regions. Was this an artifact of the ICON 2-moment model and it's spin up period? What implications does this have for the operational constraints of contrail avoidance?

• Climate response function comparisons

The finding that contrail aCCF peak values in Clima-3s are up to 10–20 times higher than in Clima-1(s) in some regions despite identical underlying meteorology is an interesting result and potentially represents a major source of uncertainty in the eco-efficient routing optimization. The authors attribute this primarily to differences in "calculation steps". It unclear what exactly this means from physically, and there is no justification provided. Is this a numerical artifact of a regridding step described as a best-practice recommendation in using the aCCFs (lines 511-512)? Or is this because of the differences between the aCCFs and the CoCiP representation. The paper acknowledges the impact this might have on the optimized trajectories but further analysis or discussion would be useful.

It is unclear if the results in Figure 9, where a trajectory optimized using Clima-1s fields yields higher contrail climate estimates when evaluated with Clima-1 is a result of the same issue with differences in calculation steps or if it is due to other reasons. Unfortunately the readers are directed to forthcoming paper for an explanation.

Technical corrections

Title -  Would it be more accurate to replace "routine" with "operational prototype of" or similar perhaps.

Abstract - The abstract is structured more as an executive summary of a project report than as a scientific abstract. It does not state the main quantitative results of the study.

Figure 3 caption and Section 3.2.3 - The term "masked" (as in "Clima-3s masked by ensemble mean PPC") appears in Figure 3 without definition in the main text

Section 4.2.1 numbering - There are two subsections both labelled "4.2.1" (one on sensitivity to weighting, one on strategic flight planning).

Line 501 - There is no section 2.2.3 in the paper, this must be a typo.

F-ATR vs ATR - I believe this terminology is follows from Megill et al 2024 but the "F" is never defined or explained in this work.

Figure 9 - is this per flight distance/ flight or for all the flights that were optimized?

Lines 621 - 622 - "54% of the planned aircraft were kept out of the PPC area" - worth clarifying if this is 54% of all the aircraft considered in the simulation or 54% of those whose planned routes intersected the PPC positive regions.

Please find attached my review.