Reply on RC1

The language has been changed to appropriately describe the isotopic tie to Standard Light Antarctic Precipitation in Section 2.1. The IAEA Certification report No. 63 states: “For the elements hydrogen and oxygen, the two international measurement standards used to calibrate all relative stable isotope ratio measurements are named VSMOW2 (Vienna Standard Mean Ocean Water 2) and SLAP2 (Standard Light Antarctic Precipitation 2) [2, 3]. The isotopic ratios of these two materials span almost the total range of isotopic compositions of natural water samples on earth. All stable isotope ratio measurements for hydrogen and oxygen performed worldwide are thus directly or indirectly calibrated versus these two international measurement standards, which have replaced the previously available water reference materials VSMOW and SLAP in the year 2006.”

The language has been changed to appropriately describe the isotopic tie to Standard Light Antarctic Precipitation in Section 2.1. The IAEA Certification report No. 63 states: "For the elements hydrogen and oxygen, the two international measurement standards used to calibrate all relative stable isotope ratio measurements are named VSMOW2 (Vienna Standard Mean Ocean Water 2) and SLAP2 (Standard Light Antarctic Precipitation 2) [2,3]. The isotopic ratios of these two materials span almost the total range of isotopic compositions of natural water samples on earth. All stable isotope ratio measurements for hydrogen and oxygen performed worldwide are thus directly or indirectly calibrated versus these two international measurement standards, which have replaced the previously available water reference materials VSMOW and SLAP in the year 2006." Based on the above report, we have revised the text: "The data consist of measurements of hydrogen and oxygen isotopes in water vapor, where the ratio of heavy to light water isotopes in a sample is expressed in δ notation (Epstein et al. 1953, Mook 2000 relative to internationally recognized primary reference materials Vienna Standard Mean Ocean Water (VSMOW) and normalized to Standard Light Antarctic Precipitation (SLAP) in accordance with IAEA reference material (2017):

EQUATION
where R is the isotopic ratio 18O/16O or D/H (i.e., 2H/1H). The δD and δ18O symbols refer to fractional deviations from VSMOW, normally expressed in parts per thousand (per mille or ‰). In practice, we maintain a suite of secondary reference waters that are rigorously calibrated to the primary reference materials (VSMOW and SLAP). Storage of our secondary reference waters is in accordance with methods described in IAEA Technical Note No, 43, (Newman et al. 2009 fig. 1 is a general presentation of the EastGRIP project then please at the reference. This is an original figure to this publication, but yes, it derives inspiration from another figure from H. C. Steen-Larsen, a co-author on this paper. Figure 1b is a map obtained from Wolfram Research and a citation is now given in the caption.
Line 153 to 157: This is a typical but not helpful assessment of an airborne system, as the mentioned problems can easily be solved without substantial additional cost.
Choosing the right platform is a complex process taking many aspects into account, often non-scientific reasons like funding and access to knowledge. Addressing the relevant aspects is necessary to establish better rational based approaches rather then today often seen intuitive choices.
Finding a platform that satisfies constraints imposed by ease of use, range of flight, payload capacity, adaptability for user fabricated modifications and budget can be challenging. We found the S2 platform met our needs and had the advantage of being locally produced.
The chosen platform should be described with all components, normal procedures and limitations as the paper title focusses on the technical part of the overall system. Please clear some inconsistencies in numbers (text vs. app. C vs. Black Swift Technologies homepage, e.g. payload mass.).
We do not identify any discrepancies between the text and app. C, however it appears that there are differences between the designs given to us by manufacturer and those on the website. Ours are the current numbers. We have added important payload and platform specifications within the text at line 164 originally found in app. C.
Revised text: "The aircraft can carry up to a 3.5 kg payload for up to 90 mins. At arctic temperatures with the payload used in this study, we found 45 mins of flight time typical and apt for climbing 1600m and including needed sampling time." 2.4.1: Can you please add a system diagram to fig. 3? This is now included.
2.4.1: Was the payload leak tested in low temperature conditions and mechanical vibrations (inflight conditions)?
Testing was performed in an arctic freezer prior to the field campaign to confirm material decisions. The intercomparison with the separate system described in Section 2.3 was in part motivated to address concerns of the system performance in conditions on-site. Vibration was not independently tested but motivated taking samples in pairs.
Revised text at line 364: "Paired sampling was motivated primarily by the inability to test the low temperatures, the 12G forces exerted on the flasks during launch, and inflight vibration forces in a "benchtop" setting." 2.4.1: Did the choice of components and materials take into account a potential corruption of the air samples?
We explored the use of Teflon, Tedlar, and stainless steel bags. We observed memory effects in all three of those options. We were initially apprehensive to use glass due to the potential percussive forces a flask could experience during takeoff and landing. This proved to not be true after test flights.
Revised text at line 181: "We explored and tested the efficacy of holding water vapor within Teflon, Tedlar, and stainless-steel bags and we observed memory effects in all three of those options. Glass was the only material found where sample carry over was minimal." Line 186: I don´t understand the half sentence "yielding appr. 50 flasks …". I guess a time reference is missing? This has been addressed by adding "over the 5-minute flush-fill process for each sample" to line 193.
Line 199ff: I guess the temperature and humidity of the undisturbed air at the position of the UAV is meant? Please add a description of the sensor installation, as this is essential for a later discussion.
The temperature and humidity sensor is attached approximately halfway down the right wing and is part of the combined sensor package used by the autopilot. It is installed pointing forward. Text to mention this has been added to line 209.
Revised text: "Both sensors are included as part of the forward pointing package to assist in autopilot flight on the right wing of the aircraft." Line 207ff: "Flasks … opening a single port on the flask." Does this mean that the Picarro-System sucks air out of the flasks, which reduces the pressure inside? If yes, is there an influence on the isotopic result because of condensation?
There is an influence because of condensation of the water vapor, but only at the point when there is a strong pressure gradient between the flask and the cavity. For the initial sampling period of a few minutes, the measured isotope value is effectively constant. We will report the stability of the isotope value over the course of the integrated measurement period in the data product. We have added text to address this in line 219.
Revised text: "In this manner the Picarro analyzer is pulling the sample air from the dead end of the flask, reducing the pressure slowly over time."… "Additionally, to address any issues associated with any reductions in flask pressure near the end, the last 3 minutes are also cropped." Line 209ff: I guess you have experiences in appropriate flushing and filling times. Can you please explain this a bit more detailed or cite a proper reference?
We have added the appropriate explanation to line 225 Revised text: "These timings were empirically derived from consistent plateaus of both isotopes and water concentration between the beginning and ending tails." 2.5: Are pressure differences between flasks and Picarro system an issue?
The Picarro instrument maintains a controlled pressure within its measurement cavity. Text has been added to line 221 to mention this.
Revised text: "Pressure within the analyzer cavity is carefully controlled at 50 Torr by the instrument with high speed PID controlled valves on both ends of the cavity." 2.6: The trials described in this section make the most of the possible, which is much more then often seen. I suggest discussing the difference between airborne and stationary samplings (e.g. fluxes?). Standard deviation and root mean square lead to the same result, but have other constrains. So I suggest to be careful presenting 1 sigma values.
We apologize, but we aren't exactly sure what is meant by this comment. As for the difference between airborne and stationary samples, stationary samples were used to establish an empirical uncertainty by sampling all flasks from all three pods over the course of 90 mins. Built into that is an implicit assumption that we are sampling the "same air" over that time period. This may not be true and sampling is subject to diurnal variation at that time scale. This is why we report our uncertainty as the worst-case scenario as water vapor from standard water would likely outperform outside air.
Make sure, that sec 2.4 and 2.8 do not double each other.
In section 2.4, we clarify that a typical analysis day will be explained in 2.8, to alert the reader that additional information will be forthcoming in a later section. Section 2.4 lays out each individual process with appropriate technical details. Section 2.8 lists those methods and shows how those methods are used during a sampling day. With that said, we feel that each section is worth keeping to ensure clarity on measurement scheme vs. typical day.
Revised text: "A typical flight day including sampling is found in Section 2.8." Line 347: As altitude maintaining power setting you normally do not use "idle".
Language has moved from "idling" to "holding altitude".
Fixed. Fig 7.: The overlaid periodic changes in temperature and humidity correspond to the heading of the UAV. As the installation is not described (comment X.) nor it is clear if the airspeed or the ground speed is commanded by the autopilot it cannot be excluded that an improper installation and/or data correction leads to this result. At this point I do not agree with the options described in lin 434ff as better sensors cannot heal improper installations.
The airspeed and the ground speed are not determined by the pilot. The autopilot solves for the desired path and climb/descent rate. Conceptually, there exists a tradeoff between controlling climb rate and air speed but that decision is outside the purview of the pilot during flight. Wind speed is measured and used as part of the autopilot but not of appropriate reportable quality for making a correction in this study. Since submitting this document for review, we have obtained the ability for our system to determine reportable 2D wind speeds and we have included text within Conclusions and Outlook at line 480 to mention the new sensor.
Revised text: "We plan to leverage an existing anemometer used by the autopilot in order to assist in the correction as well as produce an additional 2D wind speed for the flight." Line 582: Acknowledgements Fixed.
Additional comment: In the time since submission, we have determined that the first two flights of our six total do not have sufficiently useful calibrations. This was discovered from referencing lab notebooks from which it was found that calibration protocols were not correctly followed at the beginning of the field campaign. The midair isotope values for both the June 3rd and June 6th flights are precise and the midair isotope gradients real, but the values are not accurate to the standard necessary for reporting in this paper. To remain conservative in this pilot study, we are choosing to omit the data from the text and data product. Please note, this does not change any conclusions in the paper and only necessitates minor explanatory changes in the discussion, which has already focused on the June 12 th flight onwards.