Subsets of geostationary satellite data over international observing network sites for studying the diurnal dynamics of energy, carbon, and water cycles

Hashimoto, Hirofumi; Wang, Weile; Park, Taejin; Khajehei, Sepideh; Ichii, Kazuhito; Michaelis, Andrew; Guzman, Alberto; Nemani, Ramakrishna; Torn, Margaret; Yi, Koong; Brosnan, Ian

doi:https://doi.org/10.5194/essd-2025-33

Preprints

https://doi.org/10.5194/essd-2025-33

Preprints

17 Mar 2025

| 17 Mar 2025

Status: this preprint is currently under review for the journal ESSD.

Subsets of geostationary satellite data over international observing network sites for studying the diurnal dynamics of energy, carbon, and water cycles

Hirofumi Hashimoto, Weile Wang, Taejin Park, Sepideh Khajehei, Kazuhito Ichii, Andrew Michaelis, Alberto Guzman, Ramakrishna Nemani, Margaret Torn, Koong Yi, and Ian Brosnan

Abstract. The latest generation of geostationary satellites provide Earth observations similar to widely used polar orbiting sensors but at intervals as frequently as every 5–10 minutes, making them ideal for studying diurnal dynamics of land-atmosphere interactions. The NASA Earth Exchange (NEX) group created the GeoNEX datasets by collating data from several geostationary platforms, including GOES-16/17/18, Himawari-8/9, and GK-2A, and placing them on a common grid to facilitate use by the Earth science community. Here, we document the GeoNEX Coincident Ground Observations dataset (GeCGO) for terrestrial ecosystems studies, and provide examples for its use. Currently, GeCGO provides GOES-16 Advanced Baseline Imager (ABI) data over a 10 x 10 km area surrounding 1586 network sites across Americas. GeCGO make it easy to compare the time series of geostationary data with the diurnal ground observations including carbon/water fluxes and aerosol optical depth, and is extensible to other regions. We also developed GeoNEXTools to facilitate analyses that require both GeoNEX data and other NASA satellite data. The objectives of this paper are to introduce GeCGO and GeoNEXTools, and demonstrate their applications. First, we describe the details of GeCGO and GeoNEXTools. Second, we explain how GeCGO can be integrated with other satellite data. Finally, we showcase comparisons between GeCGO and observations from three ground-based networks.

Received: 16 Jan 2025 – Discussion started: 17 Mar 2025

Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims made in the text, published maps, institutional affiliations, or any other geographical representation in this preprint. The responsibility to include appropriate place names lies with the authors.

Download & links

Hirofumi Hashimoto, Weile Wang, Taejin Park, Sepideh Khajehei, Kazuhito Ichii, Andrew Michaelis, Alberto Guzman, Ramakrishna Nemani, Margaret Torn, Koong Yi, and Ian Brosnan

Status: final response (author comments only)

RC1:
'Comment on essd-2025-33', Anonymous Referee #1, 02 Apr 2025

Much hope has been placed on the hyper-temporal observations from the new generation geostationary satellites to complete the polar-orbiting satellites for environmental remote sensing. Yet a major barrier for the users is the standard, high-level science products from the GEO sensors that can be readily used in the specific applications, hence saving significant amount of time in going through the sophisticated satellite data preprocessing, which is nearly impossible for scientists working in ecology or environmental sciences. In this context, this study by Hashimoto and colleagues made a good effort in eliminating the GEO data barrier and provide a well-organized dataset containing processed high-level variables such as surface reflectance, LST, and solar radiation for more than one thousand of ecological and environmental network sites. In my opinion, this dataset and the R toolbox provided, which is functionally similar to the widely used TESViS MODIS data subset tool maintained by ORNL, is much needed and will be well welcomed by the ecologists and the carbon and water cycle science community. I have no major concerns with regard to workflow employed in creating the dataset. I do have some minor comments that the authors can consider in working on the revised version. The manuscript is well written and was a joy to read. I believe it can be accepted for publication in ESSD with a minor revision.
Specific comments:
L100: I suggest to add figures to illustrate the pixel numbering scheme of the image stamps with different sizes, something similar to TESViS did for various MODIS products.
L210: just to be a bit more accurate, MODIS onboard Terra/Aqua can in together provide four instantaneous measurements per day;
L230: when presenting correlation coefficients, suggest to provide p-value for knowing the significance of the correlation;
Figure 5: the data points on L1G plot is much less than that of L2G, perhaps due to cloud screening during the MAAIAC. Just to be fair with L1g, I wonder if the L2G is still superior than L1g when same subset of data (i.e., sites) is used.
L245: I am not sure what is the relevance of “MVC” here, as it is not recommended for generating multi-day VI composite due to its tendency of chose off-nadir observations (see, van Leeuwen et al. 1999);

* van Leeuwen, W.J.D., Huete, A.R., Laing, T.W. (1999) MODIS Vegetation Index Compositing Approach: A Prototype with AVHRR Data. Remote Sensing of Environment, 69, 264-280.
L260: “The aerosol has large diurnal variability in Aerosol Optical Depth (AOD).” suggest change to —> “The Aerosol Optical Depth (AOD) has large diurnal variability.”
L280: Just to be specific, PhenoCam is one kind of phenology ground observation network focusing on using time-lapse cameras, with another one is the USA National Phenology Network.
L290: “harvardfarmnorth”, this sounds like a site-code not site name;
Fig. 7: please add legend on the plot. Besides, I suggest to have y-axis title for each sub plot to facilitate the interpretation;
L310: ‘You can download the We will add…’, please revise.
The title for Sub-section 3.2.1 “Which NDVI or NIRv can represents annual GPP is more representative of annual GPP?” is grammarly incorrect and can be revised;

Citation: https://doi.org/10.5194/essd-2025-33-RC1
- AC1: 'Reply on RC1', Hirofumi Hashimoto, 16 Jun 2025
  
  We appreciate your recommendation to publish the manuscript in ESSD. We revised the manuscript following your comments.
  Specific comments:
  L100: I suggest to add figures to illustrate the pixel numbering scheme of the image stamps with different sizes, something similar to TESViS did for various MODIS products.
  
  We clarified it with adding more explanations, as
  “For instance, the TOA band 1 reflectance data has 0.01° spatial resolution at nadir, and thus its GeCGO data consists of 9x9 pixel value data”
  and revising the Table 2 by adding a GeCGO grid-size column.
  L210: just to be a bit more accurate, MODIS onboard Terra/Aqua can in together provide four instantaneous measurements per day;
  We revised the sentence as
  “each MODIS on TERRA or AQUA provides only one observation at a specific time of the day in the daytime”
  L230: when presenting correlation coefficients, suggest to provide p-value for knowing the significance of the correlation;
  We agree that it is better to provide the exact value in most of the cases. However, the p-values in the figures were too small (a: 3.82e-18, b: 5.31e-13, c: too small to calculate (statistic library returned 0.0), d: 1.38e-7) to show the values in the figure. Therefore, we decided to write “p<0.001”.
  Figure 5: the data points on L1G plot is much less than that of L2G, perhaps due to cloud screening during the MAAIAC. Just to be fair with L1g, I wonder if the L2G is still superior than L1g when same subset of data (i.e., sites) is used.
  If we use the same sites for L1G analysis, the correlation coefficients are almost same. That means the cloud screening is important for this analysis.
  L245: I am not sure what is the relevance of “MVC” here, as it is not recommended for generating multi-day VI composite due to its tendency of chose off-nadir observations (see, van Leeuwen et al. 1999);
  
  * van Leeuwen, W.J.D., Huete, A.R., Laing, T.W. (1999) MODIS Vegetation Index Compositing Approach: A Prototype with AVHRR Data. Remote Sensing of Environment, 69, 264-280.
  We removed it.
  L260: “The aerosol has large diurnal variability in Aerosol Optical Depth (AOD).” suggest change to —> “The Aerosol Optical Depth (AOD) has large diurnal variability.”
  We changed the sentence as you suggested.
  L280: Just to be specific, PhenoCam is one kind of phenology ground observation network focusing on using time-lapse cameras, with another one is the USA National Phenology Network.
  We revised the sentence as:
  “PhenoCam is the phenology ground observation network using time-lapse cameras”
  L290: “harvardfarmnorth”, this sounds like a site-code not site name;
  We changed the site ID to actual site names.
  Fig. 7: please add legend on the plot. Besides, I suggest to have y-axis title for each sub plot to facilitate the interpretation;
  We added legend and y-axis title as you suggested.
  L310: ‘You can download the We will add…’, please revise.
  We revised the sentence.
  The title for Sub-section 3.2.1 “Which NDVI or NIRv can represents annual GPP is more representative of annual GPP?” is grammarly incorrect and can be revised;
  We revised the title.
  
  Citation: https://doi.org/10.5194/essd-2025-33-AC1
RC2:
'Comment on essd-2025-33', Anonymous Referee #2, 20 Apr 2025

General comment:
This is a well-written manuscript that presents an invaluable dataset and a practical tool. The authors clearly describe the development and application of the GeCGO and GeoNEXTools resources, which will greatly facilitate the use of GEO data. I found no major issues.

Line comments:
L39: Typo - comma is missing after “the Western United States”.
L41: Typo – can be removed “(Hashimoto et al., 2021).”.
L115: ‘PhenoCam’ is perfferd name.
L278: For a general introduction to the PhenoCam and its network, Richardson et al. (2018b) is more appropriate.
L283: Typo – phonological --> phenological
L285: For V2 data, the appropriate citation is: Seyednasrollah, B., Young, A.M., Hufkens, K., Milliman, T., Friedl, M.A., Frolking, S. and Richardson, A.D., 2019. Tracking vegetation phenology across diverse biomes using Version 2.0 of the PhenoCam Dataset. Scientific data, 6(1), p.222.
L295-296: About the issue of early spring peak in GCC time series, would be better to refer to the second paragraph in the discussion section of the following paper, which provides a quite different perspective on this: Keenan, T.F., Darby, B., Felts, E., Sonnentag, O., Friedl, M.A., Hufkens, K., O'Keefe, J., Klosterman, S., Munger, J.W., Toomey, M. and Richardson, A.D., 2014. Tracking forest phenology and seasonal physiology using digital repeat photography: a critical assessment. Ecological applications, 24(6), pp.1478-1489.

Citation: https://doi.org/10.5194/essd-2025-33-RC2
- AC3: 'Reply on RC2', Hirofumi Hashimoto, 16 Jun 2025
  
  We greatly appreciate your useful comments. The references you kindly provided was so helpful to revise the manuscript.
  Line comments:
  L39: Typo - comma is missing after “the Western United States”.
  We corrected it.
  L41: Typo – can be removed “(Hashimoto et al., 2021).”.
  We removed it.
  L115: ‘PhenoCam’ is perfferd name.
  We corrected it.
  L278: For a general introduction to the PhenoCam and its network, Richardson et al. (2018b) is more appropriate.
  We corrected it.
  L283: Typo – phonological --> phenological
  We corrected it.
  L285: For V2 data, the appropriate citation is: Seyednasrollah, B., Young, A.M., Hufkens, K., Milliman, T., Friedl, M.A., Frolking, S. and Richardson, A.D., 2019. Tracking vegetation phenology across diverse biomes using Version 2.0 of the PhenoCam Dataset. Scientific data, 6(1), p.222.
  We corrected it.
  L295-296: About the issue of early spring peak in GCC time series, would be better to refer to the second paragraph in the discussion section of the following paper, which provides a quite different perspective on this: Keenan, T.F., Darby, B., Felts, E., Sonnentag, O., Friedl, M.A., Hufkens, K., O'Keefe, J., Klosterman, S., Munger, J.W., Toomey, M. and Richardson, A.D., 2014. Tracking forest phenology and seasonal physiology using digital repeat photography: a critical assessment. Ecological applications, 24(6), pp.1478-1489.
  We added discussion as
  “The detailed observations and analysis at the Harvard Forest Environment Measurement Site (EMS) revealed that the GCC peak can be explained only by combination of change in leaf traits and canopy structure (Keenan et al., 2014).”
  
  Citation: https://doi.org/10.5194/essd-2025-33-AC3
RC3:
'Comment on essd-2025-33', Anonymous Referee #3, 16 May 2025
This is a well-written paper that provides a detailed introduction to the GeoNEX dataset, GeCGO, and GeoNEXTools, along with a clear comparison and relationship to existing work. It offers comprehensive demonstrations of how to utilize the data through several practical examples. The documentation and codebase are well structured and accessible. The only concerns are related to occasional instability of the data download page and a few recommendations for improving the data structure. I recommend minor revisions before publication.
Specific comments:

There is a missing period at line 39 after “United States” and a missing comma at line 152 after “LST”.

Line 311 contains an unfinished sentence that needs revision.

The website (https://data.nas.nasa.gov/gecgo/data.php) appears unstable; it often takes a long time to load or fails to open entirely. Please double-check its accessibility.

Although the naming convention follows ORNL’s Terrestrial Ecology Subsetting & Visualization Services (TESViS), I suggest including a site inventory file. Not all TESViS sites are available here, and such a file would help users filter and identify sites of interest more efficiently.

I did not find the Level 2 data mentioned in the manuscript. Please verify its availability or clarify.

It is unclear which files correspond to the 17×17, 9×9, and 5×5 spatial subsets. Please clarify this in the documentation or metadata.
Citation: https://doi.org/10.5194/essd-2025-33-RC3
- AC2:
  'Reply on RC3', Hirofumi Hashimoto, 16 Jun 2025
  Thank you for your time and consideration to review our paper. We corrected our manuscript as your comments.
  Specific comments:
  There is a missing period at line 39 after “United States” and a missing comma at line 152 after “LST”.
  
  We corrected it.
  
  Line 311 contains an unfinished sentence that needs revision.
  
  We revised it.
  
  The website (https://data.nas.nasa.gov/gecgo/data.php) appears unstable; it often takes a long time to load or fails to open entirely. Please double-check its accessibility.
  
  The stability issue of NAS data portal was caused by hardware. The NAS team is working to replace the hardware and expects to solve the problem soon. We will also look for alternative public platforms to host the data so that the user community can have multiple access options.
  
  Although the naming convention follows ORNL’s Terrestrial Ecology Subsetting & Visualization Services (TESViS), I suggest including a site inventory file. Not all TESViS sites are available here, and such a file would help users filter and identify sites of interest more efficiently.
  
  We created an inventory file: https://github.com/nasa/GeoNEXTools/blob/main/data/site_list.csv.
  
  I did not find the Level 2 data mentioned in the manuscript. Please verify its availability or clarify.
  
  The Level-2 data is available as Level-1b. However, LST data is available only for North America. We added the following sentence,
  “Currently, the Level-2 LST data is available only for North America (i.e., Canada, U.S., and Mexico).”
  
  It is unclear which files correspond to the 17×17, 9×9, and 5×5 spatial subsets. Please clarify this in the documentation or metadata.
  
  We clarified it by adding an example in section 2.2 and added a column in Table 2.
  
  Citation: https://doi.org/10.5194/essd-2025-33-AC2

Hirofumi Hashimoto, Weile Wang, Taejin Park, Sepideh Khajehei, Kazuhito Ichii, Andrew Michaelis, Alberto Guzman, Ramakrishna Nemani, Margaret Torn, Koong Yi, and Ian Brosnan

Data sets

GeoNEX Coincident Ground Observations (GeCGO) Hirofumi Hashimoto https://doi.org/10.25966/y5pe-xp41

Model code and software

GeoNEXTools Hirofumi Hashimoto https://github.com/nasa/GeoNEXTools

Hirofumi Hashimoto, Weile Wang, Taejin Park, Sepideh Khajehei, Kazuhito Ichii, Andrew Michaelis, Alberto Guzman, Ramakrishna Nemani, Margaret Torn, Koong Yi, and Ian Brosnan

Viewed

Total article views: 373 (including HTML, PDF, and XML)

HTML	PDF	XML	Total	BibTeX	EndNote
312	48	13	373	13	17

HTML: 312
PDF: 48
XML: 13
Total: 373
BibTeX: 13
EndNote: 17

Views and downloads (calculated since 17 Mar 2025)

Month	HTML	PDF	XML	Total
Mar 2025	115	17	1	133
Apr 2025	116	18	6	140
May 2025	62	11	5	78
Jun 2025	19	2	1	22

Cumulative views and downloads (calculated since 17 Mar 2025)

Month	HTML	PDF	XML	Total
Mar 2025	115	17	1	133
Apr 2025	116	18	6	140
May 2025	62	11	5	78
Jun 2025	19	2	1	22

Viewed (geographical distribution)

Total article views: 372 (including HTML, PDF, and XML) Thereof 372 with geography defined and 0 with unknown origin.

Country	#	Views	%

Latest update: 16 Jun 2025

Short summary

We create the GeoNEX Coincident Ground Observations dataset (GeCGO) by extracting point data at observational network sites across Americas from the gridded GeoNEX products. The GeoNEX dataset is the high temporal frequent dataset of the latest geostationary satellites observations. We also release the software, GeoNEXTools, that helps handling the GeCGO data. GeCGO data and GeoNEXTools could help scientists use geostationary satellite data at their interested ground observational sites.


Total:	0
HTML:	0
PDF:	0
XML:	0