Skip to content. | Skip to navigation

Personal tools


12.10-14.2007 - GES DISC Google Earth topic at AGU 2007 Fall Meeting

The poster "NASA A-Train Vertical Data (Curtains) in Google Earth" was presented at AGU 2007 Fall Meeting in San Francisco, CA. The poster and the related demo were very successful and got positive responses from many participants. The followings are comments and responses from presenters and participants and contents of the poster.

Summary of coments and responses from presenters and participants
  1. Providing an ability to display height for inter-comparison with other data, at least, put labels. It also means to identify the vertical scale on orbit curtain.
  2. Bringing in air campaign data, e.g., TC4, with the same vertical presentation as in A-Train.
  3. Allowing GE to serve a single client, and present orbits and chunks of orbits as selectable items that can be then fetched from the server.
  4. A bug in GE that they acknowledged, and even called it “depth fightingâ€? – they are going to fix it soon. One of their suggestions for now was to use radio button to display one parameter, or another, or both together, to avoid this depth fighting.
  5. Formulated requirements for vertical overlays to provide a more optimal model for curtains and dynamically generated surfaces for pressure levels.
  6. Register our curtains with GE.
  7. Google folks suggested using photograph overlay on curving billboard instead of Collada. (We tried it, but it is hard to control it.)
  8. Journal of the Virtual Explorer might be a good place to put an article.
  9. A number of people asked about the vertical scale of the curtain plots. Was it to scale? If not, what was the scale? We should look into ways to present users with this information, somehow. Perhaps a pop-up “baseball cardâ€? when some anchor is clicked on?
  10. Provide MLS curtains for JPL.
  11. Suggestion of providing tool for user to convert HDF into KML directly.
  12. Currently, Giovanni/GoogleEarth is fine, but what about when user downloads result from Giovanni, does some of his/her own processing, then wants to display?
  13. Post some of our work to the GoogleEarth gallery.
  14. Finding a way of linking from GoogleEarth back to a data selection page and implement it.
  15. For best KML usability, 3-4-5 rules should be followed. It means that there are no more than 3 levels, 4 folders and 5 things in each folder.
  16. Shouldn't do something in Google Earth just because you can.

The poster consists of the following sections:


Google Earth can combine satellite imagery, aerial photography, map data, and socio-economic data to present an interactive 3D portrayal of the world. Aside from its popularity with the general public, it is also helping scientists discover, add, and share information about different geographic-related subjects in the world. NASA Goddard Earth Science (GES) Data and Information Service Center (DISC) has been working to integrate NASA imagery in Google Earth in order to facilitate scientific research, including a web portal to introduce and display the results.

The NASA imagery includes both two dimensional (2D) flat data and three dimensional (3D) vertical data. We introduce a new solution to integrate the vertical data from the A-Train constellation satellites CloudSat, CALIPSO (Cloud-Aerosol Lidar and Infrared Pathfinder Satellite Observation), and Aqua (mainly MODIS and AIRS products) into Google Earth to vividly expose cloud, aerosol, and H2O characteristics and atmospheric temperature profile in the form of curtain along the satellite orbit.

The vertical data are first processed by the GIOVANNI (GES-DISC Interactive Online Visualization ANd aNalysis Infrastructure) A-Train system based on user-selected spatial/temporal range and physical parameters. The resultant image is divided into small transparent image slices, with each image slice representing the fixed temporal internal orbit range. A generalized COLLADA (COLLAborative Design Activity) 3D model is designed to render the image slices in three dimensions. Based on the designed COLLADA models and satellite orbit coordinates, an orbit model is computed and the latitude/longitude results presented in KML (Keyhole Markup Language) format. The resultant orbit curtain makes vertical data viewable, transparently or opaquely, in Google Earth. Thus, three-dimensional science research data can be made available to scientists in an easy-to-use visualization tool. Also, simultaneous visualization and efficient exploration of the relationships among quantitative geospatial data (e.g. comparing the vertical data profiles with MODIS, AIRS data and TRMM precipitation data) becomes possible.

NASA A-Train Constellation Formation

The NASA Afternoon A-Train Satellite Constellation is a succession of six U.S. and international sun-synchronous orbit satellites, consisting of Orbiting Carbon Observatory (OCO) (not launched yet), Aqua, CloudSat, Cloud-Aerosol Lidar and Infrared Pathfinder Satellite Observations (CALIPSO), Polarization and Anisotropy of Reflectances for Atmospheric Science coupled with Observations from a Lidar (PARASOL) (A French mission) and Aura. It makes possible synergy of information from multiple resources, so that more informationabout the earth condition is obtained from the combined observations thanwould be possible form the sum of the observations taken independently [from NASA Facts].

Vertical data image curtain in Giovanni

Giovanni provides a useful platform for bridging the geospatial imagery data with the implied science and explicitly displaying the results to scientific communities.

Right: Image curtain of cloud/aerosol and ice/water vertical data from CALIPSO rendered via Giovanni v3.

Giovanni Version 3 (G3) adopts service- and workflow-oriented asynchronous architecture. Standard protocols, such as OPeNDAP, GrADS Data Server, help G3 transparently work with local and remote data. Service-Oriented Architecture (SOA) guarantees that all data processing and rendering modules are implemented through standard web services.

Left: Image curtain of cloud vertical data from CloudSat Satellite rendered via Giovanni v3.

Vertical Data Orbit Curtain in Google Earth

-- From Image Curtain to COLLADA Model to Orbit Curtain in Google Earth

Orbit Curtains for CloudSat, CALIPSO, and MODIS/Aqua Data in Google Earth

Cloud Reflectivity (dBZ) from 1B-CPR from CloudSat
(GMT 18:23 –18:53 Oct. 23 & 27, 2007)
Atmospheric Temperature Profile from MODIS/Aqua
(GMT 12:00pm – 1:30pm Feb. 23, 2007
Cloud/Aerosol Classification and Ice/ Water Phase Discrimination from CALIPO/CALIPSO (GMT 12:00pm – 13:00pm Feb. 23, 2007)
Cloud Reflectivityfrom 1B-CPR/CloudSat & Cloud/Aerosol Classifi-cation and Ice/ Water Phase Discrimination from CALIPO/CALIPSO
(GMT 12:00pm –13:00pm Feb. 23, 2007)

Specific Cases from A-Train Vertical Data

California Fire on Oct. 20, 22, 24, 25, 2007 Cloud/Aerosol Classification & IceWater Phase Discrimination from CALIOP/CALIPSO
Tropical Storm Andrea near SC/GA/FL Atlantic coast on May. 9, 2007 Cloud and Aerosol from CloudSat and CALIPSO
Heavy Tropical Rains over China’s southern coast on Jun. 9 and 11, 2007 cloud and aerosol data from CloudSat and CALIPSO Intense precipitation over the Arabian Sea near Pakistan on 25 June 2007 cloud and aerosol datafrom CloudSat and CALIPSO

Synergy of 2D and 3D data in Google Earth

GMT 18:38:18pm to 18:48:46pm Aug. 29, 2006 3-hour rainfall data for Hurricane Ernesto from TRMM satellite from GMT 9:37am to 8:23pm Aug. 29, 2006 AND Cloud Reflectivity (dBZ) from 1B-CPR/CloudSat GMT 12:00:00pm – 1:00:00pm Feb. 23, 2007 Cloud Reflectivity (dBZ) from 1B-CPR/CloudSat AND Cloud/Aerosol Classification & Ice Water Phase Discrimination from CALIOP/CALIPSO AND Atmospheric Temperature profile from MODIS/Aqua

Left: GMT 12:00pm – 1:30pm Feb. 23, 2007 3-hourly (10:30am – 1:30pm) .25°x.25° rainfall data from TRMM AND Cloud Reflectivity (dBZ) from 1B-CPR/CloudSat AND Cloud/Aerosol Classification and Ice/ Water Phase Discrimination from CALIPO/CALIPSO AND Atmospheric Temperature Profile from MODIS/Aqua.


This solution allows a researcher to combine vertical data together with other geospatial data for scientific research. This ability to visualize and compare diverse data from different providers provides researchers with a novel and valuable tool for data exploration.

All GES DISC Abstracts at AGU 2007 Fall Meeting in San Francisco, CA.

Document Actions
NASA Logo -
NASA Privacy Policy and Important Notices
Last updated: Sep 09, 2009 02:27 PM ET