31

I can currently only think of two cases:

  • to visualise potential changes in a landscape, such as the effect of adding windfarms, or the addition of a new building to an urban area. However both these examples are often done using CAD packages.

  • to impress an audience - which while no doubt effective to get an interest by stakeholders in a GIS, may not provide analytical benefits.

    alt text

Whilst 3D data is often critical (for floodplain analysis, river profiles, geological surveys etc.) does a 3D viewer add anything?

1
  • 1
    I think you should change the title to "When is 3D visualistion useful" because 3D GIS includes the analyses you are mentioning as useful and critical. Nov 18, 2010 at 9:06

8 Answers 8

14

From someone who is part of the "GE Generation"

Atmospheric Data

When you are visualizing atmospheric features, it is important to see a vertical profile of the atmosphere.

alt text

Sub-surface data

When working sub-surface (bore holes, or earthquakes as shown below). In the image below, magnitude is mapped to pin size, but depth is mapped (inverted) to altitude. This image clearly shows the subduction zone below the Andes, and how earthquakes that occur east of the Andes are at a deeper depth.

alt text

Solar System or Planetary Scale Visualization

This might be grouped in with the first image as atmospheric, but if you want to see interactions between the Sun and Mars, or as shown below, magnetic field lines, 3D is helpful:

alt text

Other Cases

I use it (in)frequently for

  • oceanographic data (similar to atmospheric or sub-surface),
  • tracking airplane flights,
  • skydiving,
  • SCUBA diving,
  • GPS tracks of skiing,
  • etc.
4
  • 8
    Those are some pretty nice screen shots, but for the Sub-surface earthquake data I don't think the 3d is helping any. You could use color to represent depth, and the 3d perspective is not conducive to visualizing the size of the pins. The pins further from whatever perspective you are using are distorted and smaller.
    – Andy W
    Nov 19, 2010 at 5:06
  • 4
    The first image is actually 2D. Except that it is rendered along a line on the globe, and therefore harder to read and distribute. I would imagine a 2D chart to be more helpful here. -- Imagine the second chart without the numbers next to the pins. Would it still be readable? Would it be in 2D?
    – relet
    Nov 19, 2010 at 11:29
  • @relet First image is 2D and for quantitative analysis a simple 2D image is better. But it is also proof-of-concept. When you have a 2nd satellite with a different ground track, the two 2D planes become 3D. Multi-sensor integration benefits from a qualitative/exploration overview such as this. Of course, for publication, it eventually gets distilled down to a simple 2D image or just a line graph.
    – mankoff
    Nov 19, 2010 at 15:43
  • @relet As for the second image... You can remove the numbers. It depends what you want to show. As a qualitative example of the deeper earthquakes east of the Andes, this view works very well, IMO.
    – mankoff
    Nov 19, 2010 at 15:44
10

I think both your points sum up the A to the Q quite well.

Some examples where its useful:

I think a key problem with 3D is the 'Google Earth Generation' (I just made that up), think that displaying everything in 3D is a good idea.

This kind of belongs on the bad maps link, but take this example: alt text

IMO, census data like this, does not need to be in 3D. The extra dimension is too confusing for most people to identify patterns in the data and IMO this kind of data would be better off in 2D. - you could argue you get the extra dimension to show a different value (as the colour could show a different attribute), but in this case = no need.

Saying that, I am sure people may laugh at this post in years to come, as the move to augmented reality, is kind of a new breed in 3D GIS anyway?

2
9

Environmental investigations are inherently three dimensional. As a simple illustration of what some environmental data might look like, here is an image from a simple 3D GIS I created (using VRML) in the '90s.

alt text

The box-like structures are buildings in an office park. The multi-colored "straws" display subsurface geologic readings taken every one to five centimeters down to 10 meters below ground surface. (There are over 400 of them here, collectively displaying over 80,000 values.) The colors distinguish soil properties. Visualizations like this help us understand where contamination might be, how it got there, and where it might go: all of them crucial questions. A better visualization might include semi-transparent "clouds" to show plumes of groundwater contamination, arrows to show the 3D vector field of groundwater flow velocities, and subsurface structures like sewers. (At the time I had to use AVS to do things like that, a 3D visualization platform that uses graphical modeling to process and display 3D data.)

An easy-to-use, powerful, ubiquitous 3D platform to support the easy creation of and direct interaction with these kinds of data could revolutionize the design and analysis of environmental investigations and geological studies.

1
  • I have seen a similar 3d study except it was with saltwater injection wells and subsurface geologic data showing possible saltwater contamination and movement.
    – Brad Nesom
    Nov 22, 2010 at 5:01
7

I TA for a 3D Visualization course, and in my opinion there is certainly a place for 3D visualization in GIS.

For example, bringing a DEM (Digital Elevation Model) in ArcScene (or an equivalent program) allows you to view the X, Y AND Z (elevation) values. You can then bring in additional layers that have no elevation reference and "steal" the DEM's Z-values. These layers would essentially be draped on top of the DEM. From here, you can do all sorts of neat things, like create a viewshed for a particular point (e.g., mountain top), create custom animations (e.g., fly-throughs), or simulate certain scenarios (e.g., flooding a given region).

Additionally, Google has an excellent 3D Warehouse which contains thousands of customized symbols that have been digitized by people across the world using Google Sketchup. Although many of these symbols may not be worth a second look, there are also some extremely detailed models that can be found and downloaded (a star rating helps identify these). Once downloaded, these images (common extension .skp) can be used as the symbology for a simple point feature found in ArcScene. Adding an offset to this point shapefile then allows your new image (e.g., a black hawk helicopter) to appear to be flying on top of your landscape.

Here are a few snapshots of these examples:

  • Black Hawk fly-through with custom layers draped over a DEM enter image description here

  • Flooding a scene

Initally, this is what the terrain looks like: enter image description here

This is what the terrain looks like after a 10m rise in water levels: enter image description here

And this is what the terrain would look like after a 30m rise in water levels: enter image description here

Again, this is just a visual representation of the data and should certainly be taken with a grain of salt. What is the resolution of the DEM that was used? How accurate is the data? Etc. These are questions that need to be addressed before being able to take anything away from this on an analytic point of view.

Still, when it comes down to it, it's important to be able to view your data in different ways, and 3D Vis provides a whole new way to view data that 2D platforms simply can't provide. After all, we live in a 3D world, and being able to capture this added dimension when viewing your map can really help to put things in perspective.

1
  • Can you point to tutorials or lessons on how to create such scenes as those you have posted here?
    – Catlike
    Jan 18, 2014 at 7:27
4

Presentations - to give the wow factor. A bit 1996.

People want 4D now X,Y,Z and T (Temporal data now).

As for 3D it can be helpful with Mobile (Cell) Phone mast placement (locating) even in urban areas with building structures.

Though Google Earth has expanded reach and become mainstream - likely because it is free for personal and commercial use.

4

For the purpose of visualization of time using 'space time aquarium' where in 3D space you can use X and Y to represent location in space and Y to represent time.

alt text

Two names that would be helpful exploring this approach would be

  • Mei-Po Kwan - see for example paper here and some more figures here
  • Menno-Jan Kraak - he calls it 'cube'; see for example here

Somewhere between atmospheric data geoviz (mentioned by mankoff) and wow factor (mentioned by Mapperz) is In the Air: "a visualization project which aims to make visible the microscopic and invisible agents of Madrid´s air (gases, particles, pollen, diseases, etc)"

intheair

Website provides link to interactive app, unfortunately it did not work for me.

2

For example, one can gain a better understanding of the effects of soil sealing and the behavior of water flow caused by the same sealing or to analyze the flow of land in case of a landslide

1

3D visualization is used along with animation in the transportation planning industry. Visualizing traffic movement and flow is extremely useful.

4
  • 1
    What's the third dimension used for in this case? Traffic movement can be displayed in 2D too. Is it used for additional information? If so: What kind of?
    – underdark
    Nov 19, 2010 at 8:25
  • my bad, it is 2d with animation.
    – Brad Nesom
    Nov 19, 2010 at 14:54
  • 1
    X,Y,T is as 3d as X,Y,Z.
    – jvangeld
    Nov 20, 2010 at 6:35
  • @underdark The third dimension could be driver behavior for example. Older people drive different than younger people, so as you have interaction between 100s of subjects, each of them have their own parameters. although they're in a x,y,(somewhat z), and time scale, the idea is that you see how the subjects, treat reality given all of their surroundings
    – dassouki
    Dec 19, 2011 at 2:59

Not the answer you're looking for? Browse other questions tagged or ask your own question.