Data-like File Formats: CSV and floating point GeoTIFFs

In addition to the standard file formats that we support in NEO—JPEG, PNG, GeoTIFF, and GoogleEarth—many (not all) datasets support two additional “data-like” formats: CSV (comma-separated values) and floating point GeoTIFF. When you choose one of these formats for download, there are a few details that should be taken into consideration.

  • The values that these files contain have been scaled and resampled for visualization purposes in NEO and should not be considered for rigorous scientific examination. At best they are useful for basic analysis and trend detection but if you are interested in conducting research-level science we recommend that you use the original source data (which are not hosted by NEO, but we can assist you in identifying the source).
  • CSV files can get quite large at full resolution. For example, a 3600×1800 CSV file can get to around 61MB. If your software has difficulty opening a file of that size then please select a smaller resolution (e.g., 1440×720).
  • There are two flavors of CSV available in NEO:
    1. “Regular” CSV which includes the text-only values at the resolution the user specifies. This format is suitable for Excel (2007 and later) and many other applications.
    2. “CSV for Excel” In Excel versions prior to 2007, worksheets could not support more than 256 columns. To remedy this, this format option is resized to 250×125. The first row contains the longitude values for the center of the cell and the first column contains the latitudes.
  • Floating point GeoTIFFs contain 32-bit numerical data along with the geolocation information that is standard for the Geo format. These files can also get large as they are not internally compressed—e.g., a 3600×1800 GeoTIFF can be around 25MB.

These formats are not available by default on our ftp site but if you are interested in obtaining a long time series of a dataset in one of these formats, please contact us and we can perform a customized export to the ftp site in the format you need.

33 responses to “Data-like File Formats: CSV and floating point GeoTIFFs”

  1. Kevin Ward says:

    Hello Douglas,

    The only source that includes data going back that far is the DMSP-OLS nighttime lights product from NOAA:

  2. Michael Cheeseman says:

    I am trying to conduct research on 0.5 Degree MODIS LAI data. Is there anywhere else I can get preprocessed 0.5D data that can be used for rigorous science?

    Thank you!

  3. Kevin Ward says:

    Hello Michael,

    The production data (HDF) can be found at

    You might also try using the AppEEARS online analysis tool which gives you the ability to work with the source data without needing to download the data. AppEEARS does focus more on regional/subset analysis but it is probably worth a perusal:

  4. Patricio Fernandez says:


    Is there a possible way to get the chlorophyll concentration data by day, instead of having it grouped by week/month. Can you refer me to the source?

    Many thanks,


  5. Kevin Ward says:

    Hello Patricio,

    The daily chlorophyll data can be found through the NASA OceanColor website (where you can also find the source weekly and monthly data we use for NEO).


  6. Andy Scholand says:

    Hi Kevin,

    Thanks to you and your group for making these data available in such conveniently accessible formats, I really appreciate it. I would like to ask about the daily AOD floating point geotiff data. The algorithm theoretical basis document shows AOD values up to 3 (for example, Figure 29), but values above 1 appear quite rare according to that figure. The maximum AOD in the geotiff format appears to be 1. Were values above 1 truncated? Or should the values in the geotiff be scaled up by a factor of 3? Or is there some other conversion formula available?


  7. Gabriel Evangelista says:


    I am doing a research Internship at the California Institute of Energy and Environment. My colleagues and I are very interested in obtaining the HOURLY data of Cloud Fraction in Mexico City. The daily data is available in the CSV documents and we were hoping to obtain this very same information on hourly basis, mainly for the year 2017. Is there any place we could ask for this data?


  8. Kevin Ward says:

    Hello Andy,

    In order to produce the best visual product we have constrained the maximum value we show for the data to 1.0, but the source data would have the full range of values.


  9. Kevin Ward says:

    Hello Gabriel,

    NASA cloud fraction data is only available daily, at best, due to the frequency of the satellite overpasses. More frequent data would have to come from a geosynchronous satellite such as GOES, but those data would most likely be limited to just over the continental United States.


  10. Gabriel Evangelista says:

    Hello again Kevin,

    It is sad for our research to know this; nevertheless, thank you very much for your response.
    By the way , you were right, GOES is limited to just over the continental US.


  11. Juan Acero says:

    How can I download the entire dataset for monthly values without doing it one image at a time?

  12. Kram says:

    Grateful, I am trying to do a statistical evaluation of GDP estimates derived from nightlights.
    But I have no idea of remote sensing. How do I get the compressed dmsp data into excel to enable me proceed, please

  13. Gabriel Martin says:

    I am looking for a GeoTiff image of active fires at 0.5 degrees with fire information for August 2019, to work in a combined analysis of South American marsupial species richness also at 0.5 degrees. Does the file in your download page have that resolution and is it accurate enough for scientific research? Could you point me to the website where the data can be downloaded? Many thanks, Gabriel

Leave a Reply

Your email address will not be published. Required fields are marked *

Contact Us

Need to get a hold of someone at NEO? Just fill out the form below.

Trouble with this form? Submit your comment here.