Review of "A Novel Simplified Ground-Based TIR System for Volcanic Plume Geometry, SO2 Columnar Abundance, and Flux Retrievals"

This paper presents a comprehensive methodology for retrieving SO2 column density using a ground-based TIR system (2 IR cameras and one visible camera). The study is well-structured, detailing the three key steps: computing the instrument view geometry and the preparation of the data from both IR cameras (one broad band and one narrow band with a filter at 8.7µm), calibration (using the radiative transfer model called MODTRAN), and SO2 retrieval (using look-up tables from MODTRAN). The authors provide an analysis of error propagation and uncertainty quantification, making the findings valuable for the atmospheric and volcanology communities. Overall, this is a well-executed study that presents a useful methodology for SO2 flux retrievals, with a balanced discussion of its advantages and limitations.

Suggestions of technical corrections:

Line 37: "but are punctual" → "but are limited to specific locations"

Lines 50-51: "and as part of a continuous and real-time volcanic monitoring system." → ", and as part of a continuous, real-time volcanic monitoring system."

Lines 77-78: "manufacturer supplied" → "manufacturer-supplied"

Line 80: Why is the conversion into brightness temperature performed at 10.02 µm?

Figure 2a: The quality of the text box could be improved.

Line 145: In the legend of Figure 4, replace "m2" with "m²".

Lines 162-166: Equations 9, 10, 11, and 12 are not easy to conceptualise. To help the reader, a new illustration could be added to clearly show x(j) and explain mx and qx. Alternatively, additional text could be included to clarify these equations.

Line 169: In Equation 3.8, does the reference correspond to Equation 8?

General question regarding Section 3.1: How is the wind direction (ω) determined? Is it estimated during the field campaign, or is it derived from GEO (SEVIRI) or LEO (TROPOMI) data? If this is the case, it should be explicitly stated.

Lines 180-184: SEVIRI retrievals estimate a plume top altitude of 6.9 km at 00:55 UTC on April 1, 2021. You state that this is in good agreement with the VIRSO2 nighttime measurements. Could you specify the volcanic plume top altitude obtained using VIRSO2’s field of view and geometric considerations, both with and without wind correction?

Lines 190-193: The error increases for ω > 45°. Could you provide an explanation for this? Additionally, what conclusions can be drawn from the comparison between this study and the tool provided by Snee et al. (2023)?

Lines 200-202: The text mentions "the next section," but the order of presentation is Sections 4.1, 4.3, and 4.2. Consider reordering the text to follow the sequence 4.1, 4.2, and 4.3, or, if this is not logical, switching the order of Sections 4.3 and 4.2.

Lines 200-202: "at first, the non-perfect transmissivity of the 8.7 µm filter produces a 'ghost image.' Then, the filter temperature affects the NB measurements, and finally, an adjustment is necessary for the BB camera, considering that the clear sky temperature often does not match the MODTRAN simulations." The mention of "non-perfect transmissivity" and "filter temperature effects" for the NB measurements appears closely related. Additionally, some introductory information about the use of MODTRAN would be beneficial. Improving the clarity of these three lines would help the reader grasp the concept more easily.

Line 206: The phrase "do not frame exactly the same scene" describes a common issue known as an X/Y shift between the two cameras. This terminology could be included for clarity.

Line 213: "For the BB camera, this effect is small enough that a correction is not required (Prata et al., 2024)." Given that the vignetting effect results in a range of 4 K for the NB and only 0.4 K for the BB, it seems reasonable to assume that vignetting correction is not necessary for the BB camera.

Lines 230-232: "A 'correction image' (Fig. 7c) is simply obtained as the difference between the black target temperature image Tn,NB(i, j) and its mean value MEAN(Tb,NB) over the 320 × 240 array." It is unclear why MEAN(Tb,NB) is subtracted from Tn,NB(i, j). Could you provide an explanation? Is this a way to normalise the ghosting effect? What is the purpose of applying this offset? Clarifying this would be helpful.

Line 287: In Equation 15, it appears that the term B is not explicitly defined.

Line 306: In Figures 8a and 8b, using the same Y-axis range (e.g., 210–290 K) could improve visual consistency.

Line 313: In Section 4.3, you wrote, "as described in Sect. 4." Did you mean "As described in the introduction of Section 4"?

Lines 322 & 331: To fully understand Equations 16 and 17, it is crucial to define the term B.

Line 351: In Section 4.3, you wrote, "as already reported in Sect. 4." Please specify whether this refers to the introduction of Section 4, Section 4.1, or Section 4.2.

Lines 350-355: It would be helpful to explicitly define T_BB(sky), T_BB(ground), T_'NB(sky), and T'_NB(ground).

Line 374: In Figure 10a, there still appears to be some vignetting effect. If confirmed, this should be mentioned.

Line 401: Equation 22 does not seem to match Figure 11. The equation states:

BH = D₀ tan(θ)

However, if B_MP represents the mean altitude of the MODTRAN plume layer above sea level, the correct expression should be:

B_MP = (D₀ + HT/2) tan(θ) + h₀

B_MP should also be added to Figure 11.

Line 415: It may be useful to remind the reader that T_S represents the clear sky temperature.

Line 429: In Figure 13a, there still appears to be some vignetting effect. If confirmed, this should be mentioned.

Please see the attached comments and suggestions