Sauf mention contraire dans les contenus, l'ensemble de ce site relève de la législation française et internationale sur le droit d'auteur et la propriété intellectuelle.
Google Earth Blog
We were asked by a GEB reader whether the satellite imaging companies would be collecting and publishing satellite imagery of Vanuatu. DigitalGlobe has a subscription service called ‘First Look’ that provides access to timely satellite imagery for disaster zones. It is targeted at crisis response organizations. It does not provide the imagery directly to the public. DigitalGlobe does, however, have a publicly available map showing what imagery is made available via the First Look service and it appears that satellites World View 1 and World View 2 have captured a number of images over the last few days of the island of Efate where the capital of Vanuatu lies.
In addition, the New Zealand Herald has published sections of a few of the images showing comparisons of before and after imagery.
DigitalGlobe’s First Look coverage map.
Also on DigitalGlobe’s map you can see the locations of other recent events that you may not have heard of, such as the eruption Costa Rica’s Turrialba Volcano, the eruption of Chile’s Villarica volcano or flooding in Lobito, Angola.
We hope that DigitalGlobe makes the imagery of these locations available to the public at some point in the future, or even better, that they get into Google Earth.
Last year we had a look at Satellogic, a satellite imaging company that is planning to launch hundreds of imaging satellites. However, they are still in the very early stages of development and have only launched a few experimental satellites.
Today we are looking at Planet Labs a satellite imaging company that has already launched 71 imaging satellites and claim to have the largest such fleet in space. They are focusing on gathering global imagery on a regular basis at a relatively low cost. Their imagery is relatively low resolution, at around 3m, but is improving with each generation. For this reason we are unlikely to be seeing their imagery in Google Earth any time soon, as Google prefers high resolution imagery, preferably aerial imagery, and if that is not available then high resolution satellite imagery, such as Digital Globe’s recent 30cm offering.
In comparison, Google’s SkyBox Imaging has launched only 2 satellites of a planned constellation of 24, but they are capable of sub metre resolution and HD video.
For more see the following TED talk by Will Marshall, the CEO of Planet Labs.
This is the twelfth in our series showcasing the Google Earth plugin. This Friday, March 20th, 2015 there will be a total solar eclipse. We have looked at number of eclipses in the past and one of our favourite tools is the HeyWhatsThat website’s eclipse page that makes excellent use of the Google Earth plugin. It features two panes, one showing the expected path of the eclipse on the earth and the other, using Google Sky, shows the current position and path of the Moon across the sky. Remember that most browsers will require you to give the plugin permission to run.
The total eclipse will only be visible along a narrow band north of Scotland, but a partial eclipse should be visible from much of Europe, weather permitting.
Another useful site created by Xavier Jubier has a list of eclipses and corresponding KML files that you can download to view the path of the eclipse in Google Earth. The relevant KML for Friday’s eclipse can be found here
The post Google Earth plugin showcase: HeyWhatsThat eclipse appeared first on Google Earth Blog.
Google added a number of new 3D areas over the weekend, amongst which was Niagara Falls, which lies on the border of Ontario, Canada and New York State, USA. For a list of other newly discovered 3D areas, check the comments at the end of this post or download our KML file, keeping in mind that it takes us a day or two to get new locations into the KML. Also of note in the new finds are Zagreb, Croatia (a new country, and capital city) and Prague, Czech Republic (a capital city).
Although Niagara Falls does have the old type of 3D model which includes an impressive spray effect and a rainbow, we feel the new imagery gives a much better overall effect as the whole region is modelled and the colours match well. You can still view the old model if you wish by switching to ‘historical imagery’.
Google’s automatically generated 3D imagery usually does very poorly with moving objects such as aircraft, cars, and water. Google has clearly had to manually touch up the model as it usually does for bridges and some buildings. If you get too close the edge of the Horseshoe Falls looks a bit too angular and the trees on the islands are not in 3D. The rapids above the American Falls have some major flaws in the 3D if you look up close. However, if you don’t get too close, the overall quality of the imagery is impressive and worth having a look at.
We believe the 3D imagery was captured prior to May 2013.
We have in the past looked at the some of the California fault lines where earth quakes are likely to occur.
A recent story in the news is that a recent report by the Working Group on California Earthquake Probabilities (WGCEP) shows that the probability of an 8.0 magnitude earthquake happening in the next 30 years in the California region is higher than previously thought.
The interesting part from a Google Earth perspective is that WGCEP has provided a KML file showing the fault lines in the California region and the associated probability of a large earthquake happening in the next 30 years. You can download the KML here.
Google Earth has an Earthquakes layer in the Gallery, provided by the US Geological Society (USGS) which shows historical large earthquakes globally. In addition, the USGS provides various KMLs that show all the recent earthquakes over a magnitude of 1.0. You can find them here.
If you want a global view of the tectonic plates that cause the fault lines, we recommend the map from MyReadingMapped that we reviewed here.
We have also looked at the aftermaths of earthquakes, such as the August 2014 Napa Earthquake.
The post California fault lines and earthquake probabilities appeared first on Google Earth Blog.
While looking around Cape Town, South Africa in Google Earth recently, we noticed a 3D model of a large South African flag on Devil’s Peak. It is the old type of 3D model from the Sketchup 3D warehouse. It looks like Google’s quality checks missed it when it got added. There is actually no flag at that location, or at least not of that size. We measured it using Google Earth Pro’s measuring tools at approximately 123m tall. Although flags have been temporarily erected by hikers at that location in the past, as can be seen in the last image in this blog post, there is no permanent flag there. Also of interest is that the model was uploaded to the 3D warehouse on September 28th, 2013, just a few days before Google stopped accepting models from the 3D warehouse.
The flag on Devil’s Peak, Cape Town.
This is not the first 3D model to have incorrectly got into Google’s 3D buildings layer. Back in 2009, for example, a model of the Burj Dubai (later renamed the Burj Khalifa) was incorrectly placed in Melbourne, Australia. There are a few other models that do not reflect reality, but can be considered Easter Eggs rather than mistakes. These include the Blues Brothers Bridge Jump and the Tardis. The Tardis even includes Street View inside it, although it doesn’t seem to be possible to view it using Google Earth. Both these locations now have Google’s new 3D mesh, but if you switch to historical imagery, they can still be seen.
The Blues Brothers Bridge Jump
To find these locations in Google Earth download this KML file.
Ever since Google Earth was first released, users have been spotting aircraft in flight in the imagery.
When an aircraft is captured in flight in Google Earth it is not uncommon for it to have a second ghostly image next to it, or in some cases a rainbow effect. This is caused by the way satellite cameras are designed. Satellites have multiple cameras for capturing imagery in different wavelengths of light. A common setup is to have a high resolution monochrome camera and then a separate camera that takes photos with various colour filters in quick succession. The multiple images are then combined to form what you see in Google Earth. However, if there is a fast moving object in the scene such as an aircraft, it will have moved between exposures and the ghosting or rainbow effects can be seen, depending on what type of camera the satellite is using. In addition to the aircraft’s movement, the satellite itself is moving and due to parallax the aircraft will appear to have moved in relation to the ground. This often results in the multiple images being offset from the direction the aircraft is travelling in.
The satellite took images in quick succession with blue, green then red colour filters, and finally a higher resolution image in monochrome.
In this image, the satellite took the high resolution monochrome image first, and then red, green and blue colour filters. We don’t know how the double image of the aircraft tail happened.
This image shows an aircraft and its shadow both exhibiting ghosting.
Satellites are also capable of taking images in spectral ranges far outside the visible range such as in the infrared. These images have a variety of uses including mineral exploration, environmental monitoring, agriculture and military. Digital Globe’s World View 3 that was launched last year and whose imagery we recently looked at shows on its datasheet that it is capable of capturing a variety of bands in the infra-red range.
To find the aircraft featured in this post, download this KML file.
When you place a 3D model in Google Google Earth you can use KML to decide how close the viewer has to be in order for it to be visible. For technical details on how this works and how to implement it see the KML documentation. This effect doesn’t just apply to 3D models but works for other KML features, such as placemarks and polygons. Google Earth Pro uses this concept for the regionation tool we looked at last month.
Today we are looking at how a similar effect is used by Google Earth to display its 3D imagery. As you fly around in Google Earth, you will notice that the 3D imagery fades out in the distance. However, if you actually measure the distance that it fades out, it varies considerably from location to location. It also varies depending on what screen size you are viewing Google Earth at. At full screen we found that for parts of New York, the limit is about 8km, whereas in Tokyo it is up to 30km.
Clearly, Google sets the maximum view distance for each piece of 3D imagery and they are not consistent. In fact, in New York there is a curious effect where you can sometimes see more distant imagery but not the imagery in the mid-field.
Presumably the reason for implementing the maximum view distance is to avoid possible issues with too much 3D being displayed at once. However, we believe that the maximum distance displayed could safely be increased quite a lot and we hope Google will consider increasing it.
In the above view of New York the old type of 3D buildings (1) from the Sketchup 3D warehouse can be seen at great distances, far exceeding that of the automatically generated 3D. The bridge in the distance (2) and the neighbouring buildings become visible before some of the nearer 3D imagery (3) next to Central Park.
We recently noticed that if you look at a given place in both Google Earth and Google Maps’ ‘Earth’ view, they look a bit different, with distant objects looking closer in Google Maps than in Google Earth. The main reason for this is that Google Earth by default has a different Field of View from Google Maps.
Last year we showed you how to change the field of view in Google Earth using a tour KML. After some experimentation we reckon that Google Earth by default uses a horizontal field of view (HFOV) of about 60° and Google Maps uses a HFOV of about 35° for its ‘Earth’ view.
So to match up a given location you need to download this KML file, which adjusts Google Earth’s HFOV to 35 degrees. In addition, you may want to try turning on ‘Photorealistic Atmosphere’ in Google Earth’s settings to get the colours to match up a bit better.
Praia Grande, Brazil, with Google Earth’s default settings. Note how the marked building and hill in the distance look further away than in the Google Maps screen shot below.
Praia Grande, Brazil, in Google Maps’ ‘Earth’ view.
Praia Grande, Brazil, in Google Earth with an HFOV of 35° and ‘Photorealistic Atmosphere’.
Last year, soon after we created our map of 3D areas, we had fun using the KML to show you the areas covered by country and US state. Now that we have released a timeline version of our KML file it is is a good idea to revisit the concept and measure areas by release date rather than by country.
Although Google Earth Pro is able to measure the area of a KML polygon, you can only do this one polygon at a time. So to get the area of multiple polygons it is better to use online tools that offer the service. We used this tool from Zonum Solutions.
After manipulating the data a bit we were able to draw the following chart showing the progress Google is making over time.
As of this week, Google has captured in 3D approximately 315,000 sq km, which, incidentally, is almost exactly the same area as Poland, the newest country to receive 3D imagery.
Given that the total area of land on the earth is approximately 148,940,000 sq km, it would take Google close to a thousand years to finish the whole earth at the current rate. This explains why they are sticking to highly populated areas. So what we really need is an estimate of the area of all the cities in the world. If any of our readers knows of a KML file with global population densities done using KML regions and not an image overlay, please let us know in the comments.
One of the latest 3D imagery finds is Santos, Brazil. It is worth having a look at, as it clearly demonstrates the benefits of an automated system of generating 3D. Modelling that one city by hand would be practically impossible. To fly there in Google Earth use this KML file.
Google has not updated its imagery updates map since December 17th last year. It has, nevertheless been pushing out satellite imagery at a fairly rapid pace. We can only speculate as to why it is not updating the maps. Perhaps its staff have been too busy with all the reorganization that seems to be going on at Google. Or perhaps the maps are hosted on Google Maps Engine, which has recently been deprecated.
This post is an appeal to Google to resume releasing its imagery updates maps. If the updates are now too frequent to do a map for each update, then a cumulative map once a month would be much appreciated.
Quite a lot of people enjoy looking through new imagery to see what can be found, and the imagery update maps make it a lot easier. A few sharp-eyed GEB readers manage to find new imagery, either by the date or by comparing Google Maps to Google Earth or to historical imagery, as the various systems tend to get updated in a staggered fashion. However, the updates are so numerous that it is not really worth trying to find them all and list them here.
Occasionally GEB readers let us know about interesting finds, such as the Costa Concordia we showed you in January. And thank you to GEB reader ‘Munden’ who recently let us know about a brush fire he spotted in new imagery in the south of Mississippi. The only copyright notice for the location is Google’s, and based on the high resolution, the imagery is likely aerial imagery. So is Google now capturing its own aerial imagery?
A brush fire in Mississippi. To view in Google Earth use this KML file.
To view in Google Earth use this KML file.
When Street View started it was named ‘Street View’ because it was captured using cameras mounted on the roof of a car driving along streets. Since then Google has been shrinking the equipment, initially fitting it on a trike and later making it small enough to carry on your back known as the Street View trekker.
Click the image to explore in Google Maps’ Street View, or download this KML to find the location in Google Earth. The blue Street View outlines do not show in Google Earth, but if you drag the yellow man directly onto the pin, it should still to take you to the zip line images.
Navigation can be a little difficult and the view directly downwards is not good at all. The technique usually used on roads to erase the Street View car by using imagery from the next spot along the road doesn’t work so well in the air.
Last week DigitalGlobe announced that they are now selling the new 30cm imagery to customers. Until recently it was actually illegal for US companies to sell satellite imagery at this resolution. As we have noted in the past, aerial imagery is typically of similar or better resolution and is not subject to that restriction, but for global coverage and bulk image capturing satellites work out much more cost effective.
DigitalGlobe has available for download six sample images. For your convenience we have put them into Google Earth image overlays which you can view with this KML file. For best results turn off the ‘3D Buildings’ layer.
A few notes on the sample images:
- The image from Pudong, Shanghai, China is actually already in Google Earth and was captured on December 18, 2014.
The default imagery for Auckland, New Zealand is slightly better quality than the DigitalGlobe image, but that is because it is aerial imagery that was captured in 2012 as part of the Auckland 3D imagery.
The image from Kalgoorlie Mine, Australia, is of noticeably better quality than the default imagery in Google Earth which is from 2013.
The default imagery in Google Earth for Sao Paulo, Brazil is actually newer than the sample image although not quite as good quality as the sample.
Left: 30cm sample image dated September 10, 2014. Right: Default Google Earth image from October 17, 2014.
Another interesting story from DigitalGlobe is that on January 27, 2015 they retired their imaging satellite QuickBird after 13 years in orbit. For the full story and to see QuickBird’s last image see the DigitalGlobe blog article.
We have looked at MyReadingMapped many times in the past. It has recently changed its url from myreadingmapped.blogspot.com, to myreadingmapped.com.
A few of the recent maps to be found on the site that we have not covered in the past are:
- A map of the Nicaragua Grand Canal that enables users to zoom in close on the rivers and valleys involved in terrain and satellite mode in order to determine for themselves the impact on the ecology. It is related to this Scientific American story.
A map of the Privatization Boondoggles of Local, State and Federal Government that traces the costly boondoggles of privatization.
A map of U.S. Industrialization (1640-1880) that traces US industrialization in four important time periods, prior to the American Revolution when it was dependent on water power up to the War of 1812 with the introduction of refrigeration, the railroad and the wide spread use of steam power, up to the American Civil War, and as a result of the Civil War. You can even turn on and off layers that indicated the union states and confederate states to understand the economic differences and the impact of industrialization on the Civil War.
A map of Farms by Type. This documentary, in the form of a Google Map, provides examples of 50 different types of farms around the world.
A map of Freshwater Biomes. This documentary, in the form of a Google Map, provides examples of Freshwater Biomes.
A map documenting the battles of the War of 1812 – 1815 between the United States and Great Britain.
A map documenting the Mexican-American War of 1846-1848.
All the maps are displayed on the site using Google’s My Maps, and you can easily view them in Google Earth by downloading them as KML. Just click the ‘share’ button found at the top of the map and select ‘Download KML’.
The map of the proposed Nicaragua Grand Canal is our favourite. We had heard about it in the news and it is nice to be able to explore the locations in Google Earth.
We have spent a lot of this month exploring the features of the now free Google Earth Pro.
Google has continued to add to 3D imagery, with 48 additions so far in February. Oddly though, they have also recently been removing a number of areas. This has happened in the past and usually it is only for a short while, but occasionally it has been a long time before the imagery has been replaced. Apparently Nashville has been removed and added again a number of times. We have not yet made a decision on how to handle this in our KML file. Suggestions are welcome.
Google has continued to update satellite imagery but have not updated their map since December 17th. We hope this is just temporary.
We also had a look at the Monster Milktruck ported to Cesium and were quite impressed with the performance. Cesium is certainly looking to be a viable replacement for some uses of the deprecated Google Earth plugin.
We celebrated Valentine’s Day with a romantic trip to Venice which received new 3D imagery this month.
The regionate tool is a way to make large datasets easier to work with by limiting the number of items displayed in Google Earth at any given time. How it does this is explained in this YouTube video from the Google Earth help page:
We decided to try it out using a KML file that we found on the Geo-Wiki site that we looked at yesterday.
The file we tested with, the Geo-Wiki validations dataset from the downloads page, contains just over 58,000 features. When we opened it in Google Earth it made the application sluggish and it was quite difficult to navigate around to view the data. We used the regionate tool, and it split the KML into 6,589 separate files all linked together from a master KML file. All the data was still there, but you don’t see all the features until you zoom in, making Google Earth much more responsive. However, the regionation means that you loose the overall picture and it is no longer easy to see which areas have large clusters of points. In addition, although it would be possible to distribute the regionated files as a zip file, there appears to be no easy way to distribute the regionated data as a single file that can be loaded directly in Google Earth. So although this feature would clearly be very useful and practically necessary for very large data sets, it may not be the most suitable solution in all cases.
Before regionation, Google Earth was sluggish and the map is cluttered.
After regionation, a lot lest clutter, and Google Earth works smoothly, but large scale patterns are lost.
Although the site has an option to try it out as a guest we couldn’t get that to work, but registration is free so we went ahead and did that. Once registered, the site uses the Google Earth plugin to display a variety of maps, mostly relating to global agriculture patterns. There is a wealth of information, such as where various crops are grown and livestock production patterns. The Google Earth plugin is also used as a means to gather feedback from users, who can select an area and submit comments relating to the maps.
An interesting map of cropland field sizes.
The site includes a game called cropland capture, which shows you a series of satellite images or photos and asks you to identify whether the image includes crop land. The results are then used to create a global map of land under cultivation. There is also a more advanced version that enables you to give more detailed information about land use using the Google Earth plugin, but it seems they only make it available during competitions. To get an idea of how it works see the YouTube video below:
Last month Google removed the US$399 a year price tag from Google Earth Pro and released it for free. There is no longer any need to sign up for a licence key, simply login with your email address and the key GEPFREE. Recently, we have been having a look at some of the features unique to Google Earth Pro. Today we will have a look at the measuring tools in Google Earth Pro.
There are two ways to get measurements in Google Earth Pro. If you look at the properties of a path or polygon, there is a ‘measurements’ tab which displays length for paths or perimeter and area for polygons.
The area and perimeter of the newly released 3D along the southern coast of Portugal.
In addition, the ‘ruler’ tool has several tabs not present in the standard version of Google Earth. They allow you to measure the radius and area of circles, measure the length of 3D paths, and measure the area and perimeter of 3D polygons.
The Plaza de Toros de La Ribera in Logroño, Spain, has an area of nearly 10,000 square metres.
Using Google Earth Pro we could easily measure the height of the spires on Cologne Cathedral to be approximately 155m. Wikipedia says they are approximately 157m.
Measuring the approximate area of glass used on one of the facets of WTC 1 is easy.
In December last year, Google updated the 3D imagery in New York and San Francisco with newer, better quality imagery.
Since then Google has updated a number of other cities. Thank you to GEB reader Ryan for letting us know about Louisville (KY) and Munich (Germany). We don’t always notice the update, especially if, as in the case of Munich, the update is completely contained within the area covered by the older imagery. Other times, it is more obvious, such as Buffalo (NY) and Pittsburg (PA), where the new imagery extended beyond the old boundary.
We generally haven’t bothered keeping track of these changes in our 3D imagery KML, as it would often be difficult to identify the borders of the new imagery or know when it was added, and Google does not put dates on the 3D imagery. However, we did try tracing out the border of the Munich imagery so that you can see the extent of the update and compare it with the older imagery. To see the outline in Google Earth download the KML file here.
Google usually leaves an overlap between new and old imagery, and a curious effect of this is that sometimes different imagery can be seen, depending on the viewing altitude. So for example, in Louisville (KY) because the imagery was taken at different times of year, once when the trees were bare and once when they had all their leaves, the difference is quite striking. See the location in Google Earth using this KML.
The newer imagery tends to be characterized by richer colours, as can be seen in this border between the two, with older imagery to the left and newer imagery on the right.
Google has released some Street View imagery in Greenland. Thank you to GEB reader Chris for bringing this to our attention. There is also significant expansion to the coverage of Bangladesh that we looked at earlier this month.
The new Greenland imagery includes sections captured by boat
This is the Greenland Street View car.
A map of additions to Street View since we last looked