Review of Bernet, Brem, Bukowiecki, Henne et al.

Long-term trace gas and black carbon measurements at the high-altitude station Mount Kenya: tropical atmospheric variability and the influence of African emissions

OVERVIEW & RECOMMENDATION:  This paper follows a pair of papers that reported the installation (Henne et al., 2008a, JAMC) and first 5 years of CO/CO2/CH4/O3 data at Mt Kenya (2002-2006, Henne et al., 2008b, ACP) by adding roughly 5 more years of data (2020-2024). Seasonality of the data and precipitation & temperature cycles are displayed. Flexpart analysis of air parcel origins at the free tropospheric (FT) Mt Kenya site are described using the same regional partitioning as in Henne2008b. Comparisons of data with CAMS are included for the more recent data attempting to learn more about source apportionment; interpretation is somewhat limited because CAMS and data agreement are not very good. Black carbon (BC) data for the more recent period are displayed.

    As a Data Report, the new data are useful and should be published. As a scientific analysis, the paper is uneven in quality and not as well organized as the 2008b paper. Mis-statements about uniqueness of the network are made (easy to fix). However, more importantly, the paper is not as well organized as an ACP Research Article should be. The paper would have greater value to the scientific community if the authors focused less on the CAMS comparisons and more on trajectories for source apportionment, updating some of the most useful figures in Henne2008b. Even more, the obvious analysis to add is a comparison of the newer and first 5-year period concentrations. In view of two new papers that computed FT ozone trends (~2000-2023) from the Nairobi sonde FT record (Thompson et al., 2025; Van Malderen et al., 2025), comparing the earlier and more recent Mt Kenya records is an important result. Although Henne2008b and your Fig 3c show some offsets between the sondes at Mt Kenya level and your GAW data, the correlations are very good. It is recommended that the paper analyses be augmented with a trends calculation and ordered as follows: Climatology, seasonality, diurnal patterns, trajectory/sector analysis (like Fig 8 in Henne2008b), trends for the 4 species in Fig C1, finally source apportionment. Additional recommendations and references follow.

Abstract Comment. Also applies to the Introduction. (1) It is important to acknowledge the other two African WMO/GAW stations on the continent and then point out as you do that the Mt Kenya is the only equatorial station in Africa…sample wording follows.

Abstract. Long-term observations of atmospheric composition are essential for understanding regional and global climate impacts. Although the Global Atmosphere Watch (GAW) programme provides a network of worldwide measurements, continuous atmospheric measurements across Africa are limited to three stations. This study presents in-situ measurements of trace gases and black carbon from the Mount Kenya GAW station (MKN) from 2020 to 2024, offering a unique dataset from equatorial Africa. Its location exposes MKN to air masses from both hemispheres, enabling detection of emissions and providing insights into tropical variability such as seasonal and diurnal cycles. We present carbon dioxide (CO2), methane (CH4), carbon monoxide (CO), ozone (O3), and black carbon (BC) measurements and describe seasonal and diurnal characteristics. Trajectory calculations with emissions estimates for CO and methane give African vs non-African contributions to greenhouse gases and air pollution. CO and BC were mainly linked to household fuel use and industrial energy, with biomass burning contributing during dry seasons. Methane variability was driven by agriculture and seasonal wetlands, but large uncertainties remain in all emission estimates. We also compare the 2020-2024 measurements to Mt Kenya trace-gas data from 2002-2012. (give % change). Comparison of the observations with Copernicus Atmospheric Monitoring Service (CAMS) model products shows CAMS fails to capture O3 and BC dynamics during rainy seasons. Our findings confirm the value of MKN observations for evaluating atmospheric models and emission inventories, highlighting the need to expand African measurement infrastructure. 235 words

(2) The wording above reflects a strong Recommendation to compute trends or “% changes” from means over the first 5-10 years and the more recent 5-yr period. Because your attributions (line 13-14) have such large uncertainties, better to put concrete results on the changes in the abstract and to call attention to the changes in the Introduction and Conclusions sections! That will be of more value to readers and likely enhance citations to the paper.

Introduction Comment: Not enough credit and context to other studies. The section about Lamto and Rwanda measurements sounds rather dismissive.

Suggested Introduction Edits-

*The World Meteorological Organization’s Global Atmosphere Watch (GAW) Programme is one of only a few internationally coordinated initiatives (NDACC, de Maziere et al. [2018] and its affiliated networks, e.g., AGAGE, GML’s HATS, SHADOZ), dedicated to long-term, systematic observations of atmospheric composition on a global scale (World Meteorological Organization (WMO), 2014). *Through a network of hundreds of stations, GAW delivers high-quality data from spatially representative sites that monitor atmospheric conditions, with an emphasis on sites with minimal local influence. This global framework is essential for understanding large-scale patterns and long-term trends in atmospheric composition. However, despite its wide reach, significant observational gaps remain—particularly across tropical regions and the Global South (consider omitting this expression; it can be controversial). Africa, though one of the most climate-vulnerable continents, is typically under-represented in atmospheric monitoring networks, including for greenhouse gases (GHGs), primarily due to different national priorities, limited resources and infrastructure in emerging economies.

Continuous GHG observations are essential for verifying and reducing uncertainties in bottom-up emission estimates, as demonstrated in Europe (Henne et al., 2016; Saboya et al., 2024) or other regions (Bukosa et al., 2025). Despite a few monitoring stations (e.g. Morgan et al. (2015); Labuschagne et al. (2018); Tiemoko et al. (2023)), much of Africa still lacks the comprehensive GHG monitoring needed for robust emission assessments. This study explores the lessons learnt from the longterm data measured at the remote, high-altitude GAW station on Mount Kenya (give latitude/longitude, alt). Besides an extensive data analysis to investigate atmospheric variability, we simulate air masses with an atmospheric transport model and combine them with emissions from bottom-up emission inventories. This integrated approach allows us to explore not only temporal variability but also the spatial and sectoral origins of the observed species. This assessment is made with full recognition that a denser GHG observation network across the continent is ultimately needed for robust verification and constraint of emission estimates.

The Mount Kenya station (MKN), operational since 1999, represents a unique monitoring site in tropical Africa. The recurrent meridional migration of the Intertropical Convergence Zone (ITCZ) that oscillates between approximately 20 °N and 5–8 °S depending on boreal and austral seasons (Henne et al., 2008a; Lashkari and Jafari, 2021; Hu et al., 2007), exposes the station to fundamentally different advection regimes throughout the year. These include continental air from the northeast during boreal winter and marine tropical air from the southeast in boreal summer. Moreover, more local anthropogenic processes

and biomass burning emissions also influence the station, particularly during daytime………Summarize the Henne et al, 2008b findings here. More recently a global tropospheric ozone study included evaluation of east African ozone trends for more than 20 years of Nairobi ozone soundings. Total tropospheric ozone changes (surface to tropopause) were estimated at ~(1.5-3.5)%/dec for the period ~2000-2023 (Thompson et al., 2025; Van Malderen et al., 2025) with the lower value representative of a trend at the altitude corresponding to Mt Kenya. Increases near the surface were closer to + 5%/decade because Nairobi is a polluted city of ~4 million.

The continuous and comprehensive greenhouse gas and air pollution datasets at MKN are unprecedented in the tropical African region and underline the importance of the MKN measurement site. While early carbon monoxide (CO) and surface ozone (O3) data were reported previously (Henne et al., 2008b), the renewal of the power line has enabled largely gap-free, continuous aerosol measurements since 2015, and measurements of carbon dioxide (CO2), methane (CH4), CO, and surface O3 since December 2019. Multiple trace gases were measured with flask samples at MKN by the Global Monitoring Lab (GML) of the National Oceanic and Atmospheric Administration (NOAA) from 2003 to 2011 (e.g. Lan et al. (2025)), but were not continued afterwards. Kirago et al. (2023) investigated MKN CO in-situ and flask measurements up until 2022, but more recent years and other species have not yet been explored. Indeed, this study presents the first comprehensive analysis of the recent continuous datasets, focussing on the period 2020 to 2024.

Few studies have investigated similar compounds in the region. DeWitt et al. (2019) studied GHGs and air pollutants at the Rwanda Climate Observatory (Mt. Mugogo, 2590m a.s.l), but data were limited to 2015-2017. Earlier black carbon (BC) and aerosol measurements in East Africa have been analysed at urban and rural sites (Kirago et al., 2022; Gatari and Boman, 2003; Makokha et al., 2017; Khamala et al., 2018), but the recent multi-year BC data from MKN were not included. In addition to the continental East African site MKN, the Maido station (2155m a.s.l) on Reunion Island (Callewaert et al., 2022) and the Lamto station (155m a.s.l.) in Cote d’Ivoire (Tiemoko et al., 2021, 2023) provide tropical data on GHGs and air pollutants. However, the former cover only 20 months of Maido data, and Lamto is strongly influenced by local sources due to its low altitude. The continuous, remote MKN measurements therefore fill a critical gap in the tropical observation network.

In this study, we (i) analyse in-situ trace gas and aerosol measurements at MKN, (ii) evaluate the performance of Copernicus Atmospheric Monitoring 60 Service (CAMS) model products for several species (suggest putting this last), (iii) compare surface ozone with vertical profiles from ozonesondes launched in Nairobi, (iv) simulate atmospheric transport using the particle dispersion model FLEXPART, and (v) combine transport simulations with bottom-up emission inventories for fires, wetlands, and anthropogenic sources ……

REFERENCES:

In addition to the SHADOZ websites in *Sec 2.2.3,* a primary reference that displays FT Nairobi seasonality that is similar to that at Mt Kenya is Fig 9c is the following. It is appropriate to reference one of the ‘archival’ SHADOZ papers (2003, 2012 or 2017, the latter with the v06 data) in the data section.

Thompson, A. M., Witte, J. C., Oltmans, S. J., Schmidlin, F. J., Logan, J. A., Fujiwara, M., Kirchhoff, V. W. J. H., Posny, F., Coetzee, G. J. R., Hoegger, B., Kawakami, S. Ogawa, T., Fortuin, J. P. F., and Kelder, H. M.: Southern Hemisphere ADditional Ozonesondes (SHADOZ) 1998-2000 tropical ozone climatology. 2. Tropospheric Variability and the Zonal Wave-One, J. Geophys. Res., 108, 8241, doi: 10.1029/2002JD002241, 2003

Reference for NDACC:

DeMazière, M., Thompson, A. M., Kurylo, M. J., Wild, J., Bernhard, G., Blumenstock, T., Hannigan, J., Lambert, J-C., Leblanc, T., McGee, T. J., Nedoluha, G., Petropavlovskikh, I., Seckmeyer, G., Simon, P. C., Steinbrecht, W., Strahan, S., and Sullivan, J. T.,: The Network for the Detection of Atmospheric Composition Change (NDACC): History, status and perspectives, Atmos. Chem. Phys., https://doi.org/10.5194/acp-2017-402, 2018

Trend References:

Thompson, A. M., Stauffer, R. M., Kollonige, D. E., Ziemke, J. R., Johnson, B. J., Morris, G. A., Cullis P., Cazorla, M., Diaz, J. A., Piters, A., Nedeljkovic, I.,Warsidikromo, T., Silva, F. R., Northam, E. T., Benjamin, P., Mkololo, T., Machinini, T., Félix, C., Romanens, G., Nyadida, S., Brioude, J., Evan, S., Metzger, J.-M., Dindang, A., Mahat, Y. B., Sammathuria, M. K., Zakaria, N. B., Komala, N., Ogino, S.-Y., Quyen, N. T., Mani, F. S., Vuiyasawa, M., Nardini, D., Martinsen, M., Kuniyuki, D. T., Müller, K., Wolff, P., and Sauvage, B.: Tropical tropospheric ozone trends (1998 to 2023): New perspectives from SHADOZ, IAGOS and OMI/MLS observations, Atmos. Chem. Phys., 25, 18475–18507, 2025  https://doi.org/10.5194/acp-25-18475-2025

Van Malderen, R., Thompson, A. M., Kollonige, D. E., Stauffer, R. M., Smit, H. G. J., Maillard Barras, E., Vigouroux, C., Petropavlovskikh, I., Leblanc, T., Thouret, V., Wolff, P., Effertz, P., Tarasick, D. W., Poyraz, D., Ancellet, G., De Backer, M.-R., Evan, S., Flood, V., Frey, M. M., Hannigan, J. W., Hernandez, J. L., Iarlori, M., Johnson, B. J., Jones, N., Kivi, R., Mahieu, E., McConville, G., Müller, K., Nagahama, T., Notholt, J., Piters, A., Prats, N., Querel, R., Smale, D., Steinbrecht, W., Strong, K., and Sussmann, R.: Global ground-based tropospheric ozone measurements: reference data and individual site trends (2000–2022) from the TOAR-II/HEGIFTOM project, Atmos. Chem. Phys., 25, 7187–7225, https://doi.org/10.5194/acp-25-7187-2025, 2025

This study of Bernet et al, presenting data from a renewed effort for continuous of multi-year in-situ measurements of trace gases and black carbon from the Mount Kenya GAW station (MKN) is well articulated and presented. In-situ observations at MKN have been conducted since the early 2000s, albeit with lengthy data interruptions, with a more recent focus on continuous data from 2020 to 2024. The authors aimed to highlight the importance of ground-based data for a better understanding of atmospheric variability and emissions in an otherwise sparsely sampled tropical African region. The trace gases measured include carbon dioxide (CO2), methane (CH4), carbon monoxide (CO), ozone (O3), and one aerosol component, namely black carbon (BC).

The study identifies various emissions impacting the site and demonstrated why ground-based measurements are so crucial for the validation of atmospheric models. In this specific study, the CAMS model was used, and its limitations were noted and discussed. CAMS data often underestimate diurnal variability (for example O3 i this case) and fail to capture dynamics during rainy seasons (for BC in this instance). Overall, the paper is well structured and presented in a format that flows well and systematically.

The introduction section explained precisely what the authors had set out to do …

“In this study, we (i) analyse in-situ trace gas and aerosol measurements at MKN, (ii) evaluate the performance of Copernicus Atmospheric Monitoring Service (CAMS) model products for several species, (iii) compare surface ozone with vertical profiles from ozonesondes launched in Nairobi, (iv) simulate atmospheric transport using the particle dispersion model FLEXPART, and (v) combine transport simulations with bottom-up emission inventories for fires, wetlands, and anthropogenic sources. “

In this regard, the various objectives of the paper were achieved.

The authors also successfully demonstrated that Mount Kenya station is uniquely positioned for atmospheric observations, due to its altitude of 3678 m a.s.l.- having minimal and clearly defined local emissions. Furthermore, it was demonstrated that the local climate is strongly influenced by the Intertropical Convergence Zone (ITCZ), leading to distinct rainy and dry seasons, each with their own unique trace gas characteristics.

It is noted that in-situ observations at MKN have been conducted since the early 2000s, with a focus in this work on the continuous data from the more recent 2020 to 2024 time period. It would be useful to extend the discussion a bit further to also include the first period of measurements, but due to the large data gaps it might not be possible and will introduce more bias to the overall data set. One simply cannot extend trend calculations over periods where such large data interruptions have occurred. I think the authors should use this argument to explain / elaborate on this, in order to clarify the reason(s) for not extending the discussion to include the entire data set since inception.

The study utilized an atmospheric model to understand the origin and variability of the trace gases and aerosols observed at MKN. FLEXPART, a Lagrangian particle dispersion model, is well known and documented and simulated air mass transport to MKN quite well, providing insights into various emission sources and pathways. Furthermore, the use of FLEXPART simulations revealed the significant impact of seasonal air mass movements on atmospheric conditions, specifically at this station - Mt. Kenya (MKN). It was clearly demonstrated that the ITCZ creates two distinct hemispheric atmospheric regimes. The “take home message” of this study highlights the need for further improvement of emission inventories and monitoring networks across Africa to enhance the overall understanding of atmospheric processes.

With regards to the long-term greenhouse gas measurements at MKN, the authors managed to show increasing trends consistent with global patterns. They should perhaps expand that aspect of the discussion bit more, and mention & discuss the tropical derived values against the global background observations. This will further highlight the need for tropical and equatorial observations. 

Some specific edits required:

Figure 1 – please enlarge (similar sizing as Fig. 2) and add a placemark for the MKN station.

Line 283: correct typo “ground”

Line 344: Please rephrase – I did not see the 2x minima for CO2 clearly, however, the general remark of NE – SE seasonal displacement of the ITCZ still holds.

 Line 397: Please check the statement related to CO2 diurnal amplitudes – I could not easily discern the 417PPM CO2 and stated ~4PPM reduction during nighttime.

Line 444: correct typo CH4 (subscript)