Overall evaluation:

  Moon et al. adds the aerosol iodine recycling processes into the GEOS-Chem model, simulates global iodine species distribution in 2022, and performs observation-simulation comparison. By doing so, the authors demonstrate that the chemical conversion from aerosol iodide back to gas-phase reactive iodine species is critical to global iodine chemistry. This work represents an important step in the development of GEOS-Chem, especially for studying halogen chemistry. The simulation results are acceptable, considering that this is the pioneering step in incorporating the aerosol iodine recycling module. However, there is a major concern about the treatment of heterogeneous reactions in this iodine module. It looks like the authors are merging the treatment of heterogeneous and aqueous reactions, which is unusual and confusing and may lead to unreasonable modeling results. This work has the potential to become a great paper if the above major comment could be properly settled. I also have some minor comments mainly about improving clarity and adding discussion. Overall, considering the existence of a major comment, I would suggest a major revision.

Major comment:

1. The expression of rate constants for heterogeneous reactions is very confusing. In Table B1, taking the reaction B1 as an example, k = kchem*[HOBr]*[Cl-]*[H+], where this k is noted as khet. However, in formula B1 (line 772), khet is calculated in another way where uptake coefficient is applied. Somehow these two calculations of khet are inconsistent. It seems that the calculation of khet in Table B1 and B2 is aqueous-phase reaction rates rather than heterogeneous reaction (meaning gas-particle interaction) rates.

  I also have concern about the unit of kchem (M s^-1). Assuming that [HOBr], [Cl-], and [H+] all have a unit of M, then the unit of k in Table B1 becomes M s^-1 * M * M * M = M^4 s^-1, which is strange since the authors say the unit of khet is s^-1.

  Table A3 also has similar issues. Within the Table, kchem has a unit of M s^-1, while in note c, the unit of kchem becomes M atm^-1, which is inconsistent. What’s more, since khet = kchem*[HOI]*[DOM]*[H+] and the value of kchem is given, it is not clear how Alpha, Dg, Henry’s K0 and Henry’s CR contribute to the calculation of khet. If the above parameters are indeed needed, why is the uptake coefficient of HOI not needed here?

Minor comments:

1. Line 69: although ocean emissions are undoubtedly the largest initial iodine source, the possible continental iodine source should not be ignored. Using nitrate-Tof-CIMS, the signal of HIO3 gas can be observed in continental urban areas, showing possible continental iodine source. Some brief discussion on continental iodine sources could be mentioned in the introduction part, although it is discussed in line 606-616 as remaining uncertainties.

2. Line 82: the concept of iodide dehalogenation is confusing to me. Since the authors mention that aerosol halides (which are supposed to refer to Cl-, Br-, and I-) are involved, it seems that this process should be generalized as dehalogenation rather than specifically iodide dehalogenation.

3. Line 86-87: the concepts of Bry and Cly, and dihalogen species are mentioned for the first time here. To avoid misunderstanding, please define them specifically. What is the composition of Bry and Cly, respectively? A definition can be given similar to that in line 93 for Iy.

4. Line 99-100: it is inspiring to know that there will be a follow-up paper that addresses the impact of this newly incorporated iodine chemistry on oxidants. It would be very interesting to know the impact on not only traditional oxidants, such as HOx, O3, and NO3 but also Cl and Br that can oxidize VOCs.

5. Line 105: does the HETP module also estimate sulfate?

6. Line 231: since some reaction rates are dependent on aerosol acidity, the authors should evaluate the accuracy or uncertainty regarding pH simulation.

7. Line 252-253: the authors mention tuning of aerosol iodine interconversion rates. This is an important statement, in which enough details should be given: (1) Exactly speaking, what reactions are tuned? (2) What kinds of observations are referred to?

8. Line 286: how shall we understand “out-of-the-box”?

9. Line 290-302: the simulation was performed for the year 2022, while observations used to compare with simulation were not necessarily performed in the same year. A brief discussion should be provided to state possible uncertainty caused.

10. Line 452-454: the authors mention a hypothesis that aerosol organics can combine with iodide and thus slow down iodide dehalogenation. Are there any laboratory experiments that support this hypothesis?

11. Line 692 Table A2: for note a, in principle the uptake should happen on the surface of all kinds of aerosols or surfaces, although with different uptake coefficients. For note b, the authors increase the uptake coefficient of I2Ox from 0.02 reported by Sherwen et al. to 0.10 shown in this table. Is there any fundamental evidence that supports this kind of tuning? At a minimum, some discussion should be provided. By the way, in line 695, “or 0.02” should be changed to “of 0.02”.

12. Line 774: A should be referred to as aerosol surface area density rather than aerosol surface area.

13. Line 860: In Table B3, khet only increases several percent when uptake coefficient is increased 100 times. Is this really true? If so, why is the khet so insensitive to changes in uptake coefficient?

14. The authors are suggested to briefly mention how well GEOS-Chem can simulate O3, since O3 + iodide reaction is very important for iodine activation.

15. How are iodine oxides (I2O2, I2O3, I2O4, and I2O5) produced in the model? Do they interconvert with each other? 

This study introduces speciated aerosol iodine and aerosol iodide recycling to the global chemical transport model. They find that aerosol recycling of iodide to form Iy is more than twice as fast as the other Iy sources combined, which is one potential contribution to atmospheric halogen chemistry. I have some comments, as below.

1. Line 203-208: The lifetime of iodate against conversion to iodide is estimated to be 1h in the fine mode and 24h in the coarse mode in the model. Are these lifetimes and the associated rate constants supported by experimental measurements or theoretical calculations? Additionally, to represent the enhancing effect of aerosol alkalinity on HIO₃ uptake, the reactive uptake coefficient for HIO₃ onto coarse aerosol is assumed to be 10 times higher than in the fine mode. What is the basis for the scale factor applied in the model? The quantitative uncertainty analysis about these assumptions and their effects on the results are needed.
2. Section 2.2.4 provides extensive background on SOI and existing observations, but it lacks clear methodological details. It is not sufficiently explained how these mechanisms are implemented in the model, how the formation and conversion rates are calculated, or how fine- and coarse-mode aerosols are treated in the simulations. More details should be clarified clearly to ensure reproducibility.

3. Line 273-276: The manuscript states that the same rate constants for bromide are applied to iodide in the HOX and XNO₃ dehalogenation reactions. It is unclear what the justification is for this assumption, given that the chemical reactivity of iodide may differ significantly from bromide.

4. Section 2.4 lacks sufficient details regarding the spatial and temporal coverage and quality control of the compiled observations.
5. Line 347-358: The manuscript discusses differences in sea surface iodine emission schemes (MacDonald et al., 2014 vs. Sherwen et al., 2019) and notes that the new scheme increases iodide concentrations, particularly in polar regions. However, it is unclear whether this updated emission scheme has been tested in the current model version with the full new iodine chemistry. In addition, the attribution of the large differences between modeled and observed IO at coastal sites solely to coarse model resolution may be oversimplified; the potential contributions of uncertainties in chemical mechanisms and emission inventories should also be considered and discussed.
6. Line 375-389: The simultaneous presence of positive and negative biases (underestimation of SOI and iodide, overestimation of iodate at high latitudes) indicates systematic errors in the model, yet the authors do not clearly explain the underlying physical or chemical causes. Additionally, the normalized mean biases are  large (e.g., -83.3%), and it is recommended that the authors clarify whether such substantial deviations are within an acceptable range for the model.
7. Line 532-533: Why does the new iodine chemistry scheme lead to a simulated increase in surface ozone (up to +11%) over the Southern Ocean and Antarctica, which contradicts the well-established role of iodine as an ozone-depleting agent? What is the specific chemical mechanism responsible for this net ozone production in the model?
8. Line 634-647: The authors note that the performance of iodine chemistry in GEOS-Chem varies significantly across model versions due to differences in oxidant fields and other factors. Given that the model's performance is so strongly dependent on the overall chemical environment, how can the specific effectiveness of the new iodine chemistry mechanisms introduced in this study be objectively evaluated, rather than merely assessing their integrated performance within one particular model version (v14.4)?