The manuscript addresses a topic of scientific interest during the last 2 years, such as the variation in air pollution during the COVID-19 lockdown. The study is interested in analyzing the variation of the chemical composition of PM2.5 (not only its concentration) obtained in an area with particular geographical conditions such as a semi-arid city of northern China. Several studies have reported changes in the concentrations of atmospheric pollutants such as PM, O3 and NO2 during the lockdown measures, but few studies have delved into the variation in the chemical composition of PM. This approach allows carrying out more detailed analyzes of atmospheric chemistry by relating the fluctuation of emission sources and the implications on the chemical composition of PM.

I appreciate if the authors can offer a response/discussion to each of the following comments:

1. The authors define as objective of the study "identify the long-term chemical characteristics of PM2.5 in a semi-arid city". However, can one year of study be considered a long-term study?
2. The manuscript suggests that the results obtained "can provide a new insight for the formulation of effective policies to improve aerosol pollution in semi-arid regions". The authors should go beyond the generality and could suggest concrete measures to improve public policies based on the results achieved.
3. It would be interesting to present a comparative analysis of the variation in the composition of PM2.5 (not only concentrations) between the year of study and an average of previous years (to be possible). This is a good way to identify PM2.5 chemical composition anomalies during the COVID-19 lockdown measures.
4. The authors calculated and reported two indicators related to secondary aerosols, the sulfur oxidation ratio (SOR) and the nitrogen oxidation ratio (NOR). What is the usefulness of these indicators and how are the results interpreted? What additional information do the indicators provide regarding the concentrations of SO2 and SO4?
5. Please check in the title "3.2 Factors influencing PM2.5" the word "metrological" since it should be "meteorological".
6. The source apportionment of PM2.5 was carried out for each of the four seasons. But how did the COVID-19 lockdown measures impact on sources of PM2.5?

This manuscript analyses the composition and sources of ambient PM2.5 in the Honhot region in China before and during the COVID-19 lockdown. The information presented in the study is relevant because, unlike other existing studies, data were collected well before beginning and after the lockdown period which allowed capturing business-as-usual and lockdown PM2.5 samples. Results are well presented and structured, and the discussion goes straight to the relevant findings. By applying the PMF model, it can be proposed those sources that contribute significantly to the ambient levels observed with statistical confidence. Nevertheless, some details must be addressed before it is accepted for publication at ACP.

1. Some sentences are repetitive in the abstract, e.g. L25-27, and within the entire document.

2. One aim is to identify the long-term characteristics of PM2.5 in the studied region, however, analysing one year is not sufficient to understand long-term variations unless their results are discussed and compared with those in existing studies, which are not reported.

3. Introduction includes studies from most regions of the world, but Latin America was not included where also interesting studies have been made. I recommend you revise and include the following studies where appropriate:

-Mendez-Espinosa, J. F., Rojas, N. Y., Vargas, J., Pachón, J. E., Belalcazar, L. C., & Ramírez, O. (2020). Air quality variations in Northern South America during the COVID-19 lockdown. Science of the Total Environment, 749, 141621.

-Hernández-Paniagua, I. Y., Valdez, S. I., Almanza, V., Rivera-Cárdenas, C., Grutter, M., Stremme, W., García-Reynoso, A. & Ruiz-Suárez, L. G. (2021). Impact of the COVID-19 lockdown on air quality and resulting public health benefits in the Mexico City Metropolitan Area. Frontiers in public health, 9, 642630.

-Nakada, L. Y. K., & Urban, R. C. (2020). COVID-19 pandemic: Impacts on the air quality during the partial lockdown in São Paulo state, Brazil. Science of the Total Environment, 730, 139087.

4. In line 129: the authors did not define what a strict analytical procedure is.

5. In several sections, calculations and results are reported for a year time-scale, but it is not defined if this refer to a calendar or sampling year.

6. It would be convenient if the authors propose a hypothesis in introduction and then discuss their findings in light of it, e.g. L196-199.

7. In most sections, PM2.5 composition is claimed to be different from other Chinese regions but the reason behind this is not discussed. This issue is critical and must be addressed.

8. L230-235: Statistical tests must be conducted to identify if the changes observed for each component between periods were significant.

9. L236-238: The percentage of SNA decreased during and post lockdown, but the reason behind this behaviour is not discussed.

10. Since the main objective of the study was identifying changes in emissions during the lockdown, why PMF was not applied to conduct an additional analysis of sources prior and during lockdown?

11. How did the apportionment to PM2.5 change during the lockdown? This is not reported.

12. Overall, the text is clear and understandable but there are some sentences that require re-writing and re-wording (L95-inhabitants?, L114-analysis, L181, L222-contributed by X % to total PM2.5…, 278-benefical?, L283-easier, you meant faster?... ).