Minor comment:

1. Abstract line 21: add be ‘can be served’
2. ‘mineral dust being predominant’. Give a number percentage to ‘predominant’
3. Usually dust is coarse…how can it serve as a key point for the entire size distribution?
4. This line is redundant “ Saharan dust exhibited smaller and less absorbing particles than Arabian dust.” Given you already mentioned their angstrom coefficient.
5. Introduction line 16: “ Larger and less absorbing particles are expected in Asian dust. “ previously you said Saharan dust are smaller and less absorbing. So shouldn’t Asian dust be larger and more absorbing? (give references)
6. Section 3: “ Predominance of fine particles is observed in 18% of total instantaneous data with AE > 1.4, meanwhile aerosol mixtures occur in almost half of the database with AE values between 0.6 and 1.4. “ give reference
7. Section 3 lie 19 “All the days showing instantaneous values higher than 0.2…” are you talking about AOD? At what wavelength? Why this specific wavelength
8. Figure 1: what does different wavelength say? Is it just because the instrument measured at these many wavelengths or is there any rationale behind using all?
9. Does Figure 2 have data from both the sites? Make it clear
10. Can you please describe in methods section how you derived volume distribution used in Figure 3? How does the volume distribution of F look like? Include that
11. Can you also propose some clustering techniques given your criteria to automatically classify the dust type instead of meticulous human review? And compare to your classification

Comments and Technical Corrections:

Line 21, page 1: The term "near-infrared range" can be vague. Please specify the exact wavelengths used (e.g., 1020 nm and 1064 nm) for better reproducibility.

Lines 16-17, Page 2: The statement "Larger and less absorbing particles are expected in Asian dust" seems to contradict the Abstract, which indicates less absorbing particles for African dust. Please clarify this apparent inconsistency.

Lines 22-23, Page 2: The discussion of aerosol-cloud interactions could include the “cloud invigoration effect”. Aerosol-cloud interactions depend not only on chemical composition but also on atmospheric conditions such as convection, atmospheric instability, and water vapor availability. Please consider expanding this discussion.

Line 13, Page 3: "several wavelengths," please specify all the exact wavelengths used by the CE318 sun-photometers in this study (e.g., 340, 380, 440, 500, 675, 870, 1020, 1640 nm).

Line 16, page 3:  Regarding "Ångström exponent (AE) by fitting AOD between 440 and 870 nm" - is the Ångström exponent estimated by fitting multiple wavelengths or calculated directly using the standard two-wavelength equation: AE = -ln(AOD₄₄₀/AOD₈₇₀)/ln(440/870)? Please clarify the methodology.

Line 23, Page 3:  Regarding "reject inversion data if AOD values are less than 0.4" - please specify if this threshold applies to a specific wavelength (e.g., AOD at 440 nm < 0.4) or to all spectral AOD values.

Line 25, Page 3:  Regarding “additional quality control is implemented to level 1.5 inversion data: the AOD must meet level 2.0 quality”. Please clarify the quality control statement. Are you using level 1.5 inversion data with level 2.0 AOD as an additional filter, or should both inversion and AOD be at the same quality level (either both 1.5 or both 2.0)?

Line 27, Page 3:  Regarding “removed the threshold regarding the AOD value”. Please clarify: was the AOD threshold completely "removed" or was it "decreased" from 0.4 to 0.2? The sentence "removed the threshold" can be ambiguous.

Line 33, Page 3: Regarding "Ångström exponent (AE)", please avoid repeating the definition as AE has already been declared in Line 16.

Figure 1: The contrast between AE and spectral AODs is insufficient for grayscale viewing. Please enhance the plot by using different line styles, markers, or increased line thickness to improve clarity.

Line 19, Page 4: Regarding "showing instantaneous values higher than 0.2" - please specify what parameter this refers to. Is this spectral AOD at a specific wavelength? If so, which wavelength?

Lines 22 (Page 4) to 2 (Page 5): Would it be possible to prepare a table summarizing the AOD and AE thresholds for each aerosol scenario? This would improve clarity and allow for more clear definition.

Lines 4 to 16, Page 5: The classification approach, which includes not only aerosol properties but also synergy between models and the evolution of these properties, is very important. Would it be possible to prepare a flowchart or similar diagram to explain this methodology more clearly?

Section 4.1: What is the specific importance of aerosol classification using only sun-photometry? Does the classification in this section improve upon the classification presented in Section 3? Please explain more clearly the value and rationale for analyzing classification using only sun-photometry. I suggest including a brief introduction at the beginning of this section to explain why the classification of aerosol scenarios using solely  sun-photometry  is important.

Line 9 , Page 6: Regarding “aerosol typing” do you mean “aerosol types”?

Line 12, Page 6: Please clarify the expression used for the Ångström exponent calculation. Consider using more precise wording such as "AE calculated from AOD at 440 and 870 nm" or "AE estimated from the wavelength pair 440-870 nm" instead of "AE values obtained from the fitting AOD between 440-870," which implies a multi-wavelength fit rather than a two-wavelength calculation.

Line 32, Page 6: Regarding "To quantify this effect, we introduce a new Ångström Exponent parameter, AENIR" - since the Ångström exponent has already been defined earlier in the text, it is not necessary to introduce it as a "new parameter." Instead, simply explain that AE is computed using the wavelength pair 1020-1640 nm and denote it as AENIR or AE(1020-1640) to distinguish it from AE calculated at other wavelength ranges.

Equation 1: This equation should either be presented earlier in the text when the Ångström exponent is first introduced, or it should be removed here and simply state that AENIR is estimated using the wavelength pair 1020-1640 nm (following the same methodology already defined for AE).

Figure 2: Please specify which wavelength pair the AE refers to (e.g., AE(440-870) or another pair of wavelengths).

Line 11, Page 7: Regarding "Ångström Exponent (AE)" - this parameter has already been defined in Line 16, Page 3. Please avoid repeating the definition. Consider checking throughout the text to remove redundant definitions of previously introduced parameters.

Line 7, Page 8:  Regarding “coarse mode”. Please explain how coarse and fine modes are defined. Are these based on a specific size threshold from the AERONET inversion algorithm, or should readers refer to Figure 3 for this information? Clarification of the size separation criterion would improve understanding.

Line 11, Page 8: Please check the formatting consistency throughout the text for aerosol type abbreviations (e.g., "MD" vs 'MD'). Ensure the same format is used consistently for all aerosol categories.

Line 11, Page 11: How is the AAE (Absorption Ångström Exponent) estimated using AOD? Please clarify the methodology, as AOD represents total extinction (scattering + absorption), not absorption alone. Are you using AAOD (Absorption Aerosol Optical Depth) from AERONET inversions, or is there another method to isolate the absorption component?

Line 12, Page 11: Regarding "Single Scattering Albedo (SSA)" - once defined for the first time, you do not need to repeat the full term. Simply use "SSA" in subsequent mentions.

Lines 16 to 18, Page 12: Please consider moving this text to serve as the definition of AAE at the beginning of the section, rather than placing it at the end. This would improve the logical flow of the presentation.