Earthdata Forum

Dear all,

I hope this message finds you well.

Recently, I’ve been analyzing the relationship between atmospheric carbon dioxide and vegetation greenness. To do this, I plan to use the global time-series AIRS CO₂ product (SNDRAQIL3CMCCP), which I found on Earthdata:
https://disc.gsfc.nasa.gov/datasets/SNDRAQIL3CMCCP_2/summary

After downloading the data, I encountered a question:
In the variable "CO2 upper troposphere VMR", there are two matrices with dimensions 360×180. Could anyone clarify which one represents the actual atmospheric CO₂ concentration?

Also, if you know of any other global time-series CO₂ concentration products suitable for large-scale ecological analysis, I would greatly appreciate your recommendations.

Best regards,
Duan

Hello,

These are both the CO2 upper troposphere volume mixing ratios. Since AIRS is in a Sun synchronous orbit it crosses the equator at ~ 1:30 PM and 1:30 AM local time every orbit. The variable includes both orbit passes. The first orbit pass is ~ 1:30 PM ("daytime" at the equator) and the second orbit pass is ~1:30 AM ("nighttime" at the equator). You can read more about it in the following user guide:

https://docserver.gesdisc.eosdis.nasa.gov/public/project/Sounder/CLIMCAPS.L3.V2.README.pdf

The data you cited is a monthly mean product. There is also a daily version:

Chris Barnet (2019), Sounder SIPS: AQUA AIRS IR-only Level 3 CLIMCAPS: Specific Quality Control Gridded Daily V2, Greenbelt, MD, USA, Goddard Earth Sciences Data and Information Services Center (GES DISC), Accessed: [Data Access Date], https://doi.org/10.5067/47IB56XWPHB3

There is also a gridded product derived from OCO-2 observations:

Hai Nguyen, Junjie Liu, Susan Kulawik, David Baker, Jonathan Hobbs, Amy Braverman, Matthias Katzfuss, and Vineet Yadav (2024), OCO-2 Gridded bias-corrected XCO2 and other select fields aggregated as Level 3 daily files, Greenbelt, MD, USA, Goddard Earth Sciences Data and Information Services Center (GES DISC), Accessed: [Data Access Date], https://doi.org/10.5067/GRVGCTOKA8TL

A fused product that uses data from OCO-2 and GOSAT:

Hai Nguyen, Junjie Liu, Susan, Kulawik, David Baker, Jonathan Hobbs, Amy Braverman, Matthias Katzfuss, and Vineet Yadav (2022), Multi-Instrument Fused bias-corrected XCO2 and other select fields aggregated as Level 4 daily files, Greenbelt, MD, USA, Goddard Earth Sciences Data and Information Services Center (GES DISC), Accessed: [Data Access Date], https://doi.org/10.5067/ZS346LH1NTIS

and gridded OCO-2 GEOS products:

Brad Weir, Lesley Ott and OCO-2 Science Team (2022), OCO-2 GEOS Level 3 daily, 0.5x0.625 assimilated CO2 V10r, Greenbelt, MD, USA, Goddard Earth Sciences Data and Information Services Center (GES DISC), Accessed: [Data Access Date], https://doi.org/10.5067/Y9M4NM9MPCGH

Brad Weir, Lesley Ott and OCO-2 Science Team (2022), OCO-2 GEOS Level 3 monthly, 0.5x0.625 assimilated CO2 V10r, Greenbelt, MD, USA, Goddard Earth Sciences Data and Information Services Center (GES DISC), Accessed: [Data Access Date], https://doi.org/10.5067/BGFIODET3HZ8

All of the above products are derived from Level 2 observations. Please let me know if your are interested in Level 2 observations as well.

Thomas

Dear Thomas,

Many thanks for your prompt reply and valuable comments—I really appreciate your help!

I have a few follow-up questions regarding the AIRS CO₂ product:

For the variable "CO2 upper troposphere VMR," the description states: "Carbon Monoxide volume mixing ratio to dry air between 20,000–50,000 Pa, near the peak of sensitivity." Should this be "Carbon Dioxide" instead of "Carbon Monoxide"? Is this possibly a typo?

The unit of this variable is listed as m³/m³. To convert it to ppm, should I simply multiply the values by 10⁶?

I examined the spatial distribution of the CO2 upper troposphere VMR variable for January 2003. I noticed that CO₂ concentrations appear significantly higher over the South Pole, North Pole, and the Qinghai–Tibet Plateau compared to other regions. Is this pattern accurate?

Thank you again for your support. I also appreciate the additional CO₂ data resources you provided. Since I’m particularly interested in long-term global time-series data (2001–2022), I find the AIRS CO₂ product especially suitable for my research.

Wishing you a great day!

Best regards,
Duan

Hello,

Yes. That is definitely a mistake in the description for this variable.

Yes. That is how you can convert the mixing ratio to ppm.

I am not certain about the accuracy of the mixing ratios in those regions since there are limited validation datasets at those locations. Before drawing conclusions about the CO2 concentrations in those regions you may want to perform your own validation by comparing several different datasets.

You might also want to examine the following test report for this particular product:

https://docserver.gesdisc.eosdis.nasa.gov/public/project/Sounder/CLIMCAPS-Aqua_V2_AIRS_V7_L3_Composition_Products_Test_Report.pdf

You might also want to look at the following two datasets that I mentioned above. However, these are assimilated products, thus, the data over the poles are not direct observations.

Brad Weir, Lesley Ott and OCO-2 Science Team (2022), OCO-2 GEOS Level 3 daily, 0.5x0.625 assimilated CO2 V10r, Greenbelt, MD, USA, Goddard Earth Sciences Data and Information Services Center (GES DISC), Accessed: [Data Access Date], https://doi.org/10.5067/Y9M4NM9MPCGH

Brad Weir, Lesley Ott and OCO-2 Science Team (2022), OCO-2 GEOS Level 3 monthly, 0.5x0.625 assimilated CO2 V10r, Greenbelt, MD, USA, Goddard Earth Sciences Data and Information Services Center (GES DISC), Accessed: [Data Access Date], https://doi.org/10.5067/BGFIODET3HZ8

Thomas

Dear Thomas,
Thank you so much for your reply! Your suggestion is very helpful!
I will convert the CO2 and compare it with the other CO2 products.
Wishing you a good weekend.
Best,
Duan