Google Maps Examples

In addition to producing geovisualisations in print, and in static images online (such as those in the Colour Maps page), it is also possible to overlay layers created in ArcGIS onto Google Maps directly. This is done by creating and symbolising layers in ArcMap (I've used versions 9.0 to 9.2) and then converting them to KML format using a converter. I've used Kevin Martin's excellent Export to KML tool for ArcGIS, but there are other ways of doing it. Many of these can be found via ESRI's ArcScripts search page. Another useful, stand-alone, tool is Zonum's Shp2kml converter. It works independently from ArcGIS so all you need is suitable shapefiles in the right projection. As always with these kinds of technologies and techniques, there are many ways to do things.

How do you get kml files onto Google Maps? Simple. All you need to do is put your kml file on a web server so that it is publicly accessible and then enter the full URL for the file into the Google Maps search box. You can then embed the map in a web page or copy and paste the link to share with others. The only caveat here is that kml files do not have the same functionality in Google Earth as they do in Google Maps, so you can't do all the same things in your web browser on Google Maps that you could do with Google Earth. However, the restrictions are not too bad and things are developing all the time so perhaps in future Google Maps KML support will match that of Google Earth. For an overview of the differences and what can and can't be done in Google Maps, take a look at this list.

What I've done on this page is provide a number of links to different files I've created and then converted to KML. The links themselves are to the Google Maps but if you are looking for the original KML files (to view in Google Earth, for example, or just to look at the code) then see the Links to KML Files section on the Navigation menu. I've tried to keep it simple rather than map everything. The intention here is to illustrate what can be done, rather than attempt an exhuastive exhibition of my work. As you will see, the functionality of Google Maps integrates almost seamlessly with overlaid KML layers. You can spend a lot of time exploring and querying these maps in order to get a better understanding of the underlying spatial interaction processes associated with different kinds of places. It's still relatively early in the development of this web-based technology but from these few examples you can get the idea of what can be done (the first two examples are more simple).

1. The first example is a sub-set of inflows to the district of Carlisle in Northern England. In this example, I've shown all inflows to Carlisle of 10 or more persons. You can find more information about individual flow lines by zooming in and clicking on a line. At this simple level, we can see that larger volume flows to individual wards do not very often extend beyond the immediate region surrounding Carlisle, with a few notable exceptions of course. Note that you can turn on and off individual layers on the left hand side of the Google Map, available at this link: Carlisle Inflows - 10+

2. The next example is the same as the first, but I've structured the KML file differently so that you can view flows for different wards by using the plus and minus signs to the left of the map. Once again, you can zoom into individual lines and view more information about that flow. In this version, I've also set it to start at Satellite view rather than Terrain, but this can be changed by the viewer to suit individual preferences: Carlisle Inflows - 10+ by Ward

In the two examples above, the information that you can see when you click each line relates to a number of characteristics associated with each ward origin and destination, as follows:

FID is the feature ID from the original Shapefile
OCODE is the original origin ward code - not meaningful here, but used in creating flow line
DCODE is the original destination ward code - not meaningful here, but used in creating flow line
MIGRANTS is the number of people moving from the origin to the destination
OWARD_CODE is the official code for the origin ward
OWARD_NAME is the official name for the origin ward
OLA_CODE is the official code for the origin Local Authority
DLA_NAME is the official name for the destination Local Authority
OCOUNTRY is the origin nation within the UK where the ward is located
OPOP2001 is the population of the ward in 2001
DWARD_COD is the official code for the destination ward
DWARD_NAM is the official name for the destination ward
TYPE is the type of Local Authority of the destination ward (in this case, Non-Metropolitan District)
REGION is the official region of the destionation ward (in this case, North West)
DISTANCE is the flow line distance, in Metres
Start is the date one year before Census day 2001
End is Census day 2001

In the three examples below, I've done a bit of extra work and added in hmtl descriptions as I created the KML file so that the variable names make more sense.By default, Export to KML will use the field names in your table within ArcMap. I've also written a little description for each layer that appears in red text to the left of the screen.

3. The possibilities for displaying different features and attributes of features are almost endless with KML functionality in Google Maps. In order to demonstrate something different to the above, therefore, I've also extracted a layer for inflows to Manchester's Central ward between 2000 and 2001, then symbolised it by nation of origin within the UK. The feature attributes are viewable via clicking on a line in this case contain information about the ward of origin, since the destination is always Manchester Central ward. Once again, you can turn different parts of the layer on or off via the tick boxes to the left of the screen: Manchester Central Inflows

4. When dealing with migration flows for individual cities, it is often sensible to deal with different flows in different ways, since the many thousands of tiny flows can obscure the more important information. In order to demonstrate this, I've created a map showing only inter-ward flows of 50 or more persons in Glasgow between 2000 and 2001. In this instance, I've chosen to display gross flows (i.e. the total flow in both directions between a pair of wards). When you click on a line to find out more, therefore, you will see that I have not referred to origin and destination wards but to 'Ward 1' and 'Ward 2' since I'm interested in absolute connectivity here rather than direction. I've also added in the population of both wards for 2001: Glasgow Inter-Ward Flows, 50+

5. In addition to being able to display flow lines, it is also possible to display spatial interaction using points. In the next example, I've extracted a KML layer showing moves out of Glasgow of 100km (c.60 miles) or more. As you will see when you click on the link to the Google Map, they do not all appear at first, but once you zoom in and pan around you can see more. In this instance, I've coded the labels so that when you click on a point you see all the information about the origin ward and the destination ward. Using the layers plus and minus signs on the left you can filter by ward and/or by flow magnitude. In total, 8.8% of out-migration from Glasgow between 2000 and 2001 was to destinations more than 100km away. Particularly important here are moves out to London, which are clustered together and not all visible at first: Glasgow Outflows 100km+

Note: In all the examples above, you can identify features in the map by clicking the links in the map legend to the left of the Google Maps screen. When you click on a feature there (this would be the 'Table of Contents' in ArcMap), a callout bubble identifying the feature will appear on the map. It is also possible to integrate KML into Microsoft's Live Maps, but at this stage it would appear that Google Maps and Google Earth offer the best functionality. However, things change quickly so perhaps in future it may be possible to integrate these features with a wider variety of platforms.