Monthly Archives: January 2010

Google.org Builds Cloud-based Image Processing Platform

To coincide with the opening of the Copenhagen Climate Summit, Google.org announced a collaboration with the Carnegie Institution for Science to build an online version of the Carnegie Landsat Analysis System (CLAS).   The existing CLAS system is a desktop tool that supports conversion from the raw satellite imagery, calibration, atmospheric correction, cloud masking and spectral analysis to create maps of forest cover, deforestation, and forest disturbance that can be overlaid with other geographic data.  The new version of the software, called CLASLite, does all of this online.

The Google.org folks write:

What if we could offer scientists and tropical nations access to a high-performance satellite imagery-processing engine running online, in the “Google cloud”? And what if we could gather together all of the earth’s raw satellite imagery data — petabytes of historical, present and future data — and make it easily available on this platform? We decided to find out, by working with Greg and Carlos to re-implement their software online, on top of a prototype platform we’ve built that gives them easy access to terabytes of satellite imagery and thousands of computers in our data centers.

Geoprocessing in the cloud with petabytes of satellite imagery while reducing computation from days to seconds.  That’s a compelling vision. The prototype, Earth Engine, is not yet available to the public, but  Google has pledged to make it accessible for free to any tropical country.  And while the initial target of this effort is deforestation, it seems only logical that the Earth Engine could very well be extended to cover other types of geoprocessing.

Distributing geoprocessing has been on its way for a while. Wolfram Research has been offering the server version of its Mathematica product as a way to distribute mathematical and statistical processing across many machines in a network. Brian Flood has done a fair amount of work on cloud-based geoprocessing with his Arc2Earth Cloud Services.  At Azavea, we’ve designed our own DecisionTree raster processing framework to both distribute work across multiple machines/processors/cores as well as be able to run in the Amazon Web Services EC2 environment. Each of these examples is aiming at several benefits:

  • Speed: desktop processing can take many minutes and even hours to complete.  By distributing the work across dozens or hundreds of machines, we can get responses that are fast enough to display the results in “web time” – a second or two.
  • Lower Cost: If we can acquire processing power as we need it, rather than buying and maintaining hardware and disks ourselves, we can lower the cost of computing substantially.
  • Simpler UI: By complex processing to be performed on the web, we can create crafted user interfaces that focus on the needs of a particular workflow rather than requiring that someone learn the far more complex tools in a Desktop GIS.

I’m pretty excited the prospects for bringing analytical and statistical services to a much larger audience via cloud services.

GPS in 2010: Facing the Competition

In my last blog post, I reported on issues with the existing GPS constellation that may cause signal disruptions, performance degradation and a decline in positioning accuracy for GPS users worldwide.  A desire for independence, as well as concerns about United States control over system access and an aging satellite infrastructure have prompted other nations to develop their own Global Navigation Satellite Systems (GNSS) that could compete with and/or complement the existing GPS constellation. 

The Galileo constellation, sponsored by the European Union (EU) and European Space Agency (ESA), was originally slated for full operation by the end of 2009.  Though test satellites have been launched to verify orbits and time synchronization for the new constellation, delays in planning and lack of funds have postponed the first operational satellites from being launched until sometime in late 2010.  The projected number of satellites in the Galileo constellation has already been reduced from 28 to 22 initially, due to cost overruns.

Russia’s GLONASS was fully operational back in 1995, but lack of funding due to the collapse of the Soviet Union eventually caused the system to fall into disrepair.  A new commitment in 2001, including the announcement of a partnership with India, has put the program back on track, though still far behind its target date of restoring full world coverage by the end of 2009.  Even after placing 3 new satellites in orbit in December 2009, GLONASS has only 19 working satellites in its constellation, which guarantees coverage only within Russian territory.  A total of 24 satellites are needed to provide global coverage. 

After initially expressing interest in assisting the EU with its Galileo constellation, the Chinese government is planning its own Compass constellation that will expand the existing Beidou Navigation System from regional China-only coverage to worldwide coverage.  The target completion date is 2015.  However, there were positioning issues with the first 2 satellites in 2009 that may ultimately delay or even prevent the Compass constellation from becoming a reality.  

The Quasi-Zenith Satellite System (QZSS) is a proposed three-satellite system that would provide enhanced GPS coverage within Japan.  The first satellite is scheduled for launch sometime in mid-2010.  Full operation is anticipated by 2013.  However, funding for the second and third satellites in the system is not expected until 2011 at the earliest, since it is contingent on the successful launch and operation of the first satellite.  QZSS is intended to enhance rather than replace GPS, and is expected to provide improved reliability and usability to the entire South East Asian region. 

The Indian Regional Navigational Satellite System (IRNSS) is a regional satellite system being developed by the Indian Space Research Organisation.  A seven-satellite system is scheduled for completion by 2012, and will provide accurate coverage of India and a 2,000-kilometer perimeter around its borders.  All space, ground and user components will be made in India. 

As the United States moves to modernize its own aging GPS constellation in 2010 and beyond, interoperability with one or more of the new constellations will ultimately become possible.  GPS receivers with dual capabilities will be able to get more accurate readings, and the addition of new satellites to any of these alternate constellations will provide needed back-up to the United States as its older satellites fail.  I will be looking skyward in 2010 to monitor these conditions and providing periodic updates on each constellation as news becomes available.

Artist’s rendering of a GPS III-A satellite, part of the U.S. modernization of the existing GPS constellation.  (Public domain image courtesy of http://pnt.gov/public/images/.)

Artist’s rendering of a GPS III-A satellite, part of the U.S. modernization of the existing GPS constellation. (Public domain image courtesy of http://pnt.gov/public/images/.)

Netflix Rental Maps

The New York Times has an interesting collection of maps of Netflix rental popularity for major cities (unfortunately not for Philadelphia).   I wonder how much Netflix takes geography into account with their recommendation system.

A Peek Into Netflix Queues

Google Adds Spatial Search to Maps Data API

Google slipped out a new feature in its Maps Data API over the holidays that was quiet but I think was fairly substantial.  If you recall, back in April, Google released a new API, designed for geographic data.  This original release included only a few features: the ability to store and manage spatial data and the ability to search it. A major limitation was that, despite being designed for storing spatial data, it didn’t support spatial searches.   With the latest version of the API, that’s now changed.

This is still a fairly limited capability, with only bounding box (rectangle) and point-plus-radius (circle) searches currently supported.  However, the ability to sort based on distance in combination with the spatial data services would be enough for many of the most straightforward mapping applications.

2010: A Pivotal Year for GPS

GPS technology was in the news a lot during 2009.  It was number one on the British Science Association’s March 2009 list of “Top Ten Inventions that Changed the World,” and number seven on a travel-related list of “Top Ten Inventions that Changed the World of Tourism” six months later.  In October 2009, Google’s announcement about a free GPS navigation tool for smart phones sent shockwaves through the entire GPS navigation industry and caused stock prices to plummet for a number of existing GPS device manufacturers.   In November, GPS World reported that the Dutch government was initiating a distance-driven tax to replace the existing road tax on cars.  The new tax will be calculated by onboard GPS receivers and is intended to reduce carbon emissions and traffic congestion.  In December 2009, GPS was on the ABC News list of “The Top Ten Gadgets of the Decade,” second only to the iPod.

However, a report on GPS issued in May 2009 by the Government Accountability Office (GAO) was much less favorable.  The GAO concluded that if significant action was not taken by the Air Force to meet critical goals for modernizing the system, the older satellites in the GPS constellation will begin to fail in 2010, and it is likely that the total number of operational satellites will fall below the minimum number of 24 needed to provide the level of GPS service that the world has grown accustomed to.  Currently, about half of the existing GPS constellation is approaching “single thread” operation status, where a critical system error could render could render a satellite inoperative. 

It is possible this assessment is overly pessimistic, at least in the short term.  The constellation is currently at its maximum number of 30 active satellites and has another five deactivated spares as well as satellites that are “in the barn” on the ground, ready to launch on short notice.  However, the GPS III upgrade has been repeatedly delayed and further delays could endanger the integrity of the system.  Colonel Brad Parkinson, the original architect of the GPS, recently testified to substantial concerns regarding the risk of a GPS “brownout” that could arise from reduced satellite numbers. 

A reduction in operational satellites would cause performance degradations and a decline in positioning accuracy that would do more than simply reduce the accuracy of your car navigation systems or impact the ability of your GPS-enabled smart phone to find the nearest public transit station.  For example, many Enhanced-9-1-1 emergency calls, particularly from mobile carriers, rely on GPS technology to determine and transmit caller location data to emergency dispatch centers.  Power companies also use the precision timing of GPS to synchronize power plants and substations.  Without it, they will lose the vital ability to trace back failures in the power grid in the event of a major blackout, and restoration of power could be significantly hampered.  In addition, the airline industry may need to delay, cancel or reroute flights to allow for projected gaps in GPS coverage.  Military operations would be impacted as well. 

The United States controls the GPS constellation, and although civilian access has been allowed on a global scale since selective availability (SA) was discontinued in May 2000, there has been growing concern in other nations that some level of selective availability could be re-established in the event of another terrorist attack or hostile military actions against the United States.  While the GPS III system will not include any SA capabilities, these fears, along with a desire for independence and concerns about the aging satellite infrastructure, have spawned the development of alternate Global Navigation Satellite Systems (GNSS) that would compete with and/or complement the existing GPS constellation.  Unfortunately, these systems will not be operational in time to provide any back-up to the United States in the event of one or more satellite failures within the next couple of years.  

I’ll have more information about the alternate GNSS in my next blog post. 

constellation

A diagram of the existing GPS constellation showing 24 healthy satellites in orbit around the earth. Failure of any of these satellites could cause some level of performance degradation and a decline in positioning accuracy for GPS users worldwide. (Public domain image courtesy of http://pnt.gov/public/images/.)