Saturday, 16 March 2013

Exploring the Mt Dixon rock avalanche

Page Redirection

This blog has moved to a new home, if you are not redirected automatically, follow the link to

In my last post I discussed the growing volume of geodata available online - and in particular ways in which it can be accessed by freely available GIS software. This kind of accessibility has great potential for geoscientists, and among other uses (e.g. teaching, structural geology) perhaps is a step toward 'crowdsourcing' geotechnical expertise for geohazard identification and emergency response. We can take a look at the 2013 Mt Dixon rock avalanche to see an example of this (see Dave Petley's landslide blog for a very interesting discussion on this event).

A NASA EO-1 ALI image of the Mt Dixon rock avalanche deposit (credit: The Landslide Blog)

By draping the NASA image over topography in GE, and including the geological units, fault traces, and structural measurement layers from the GNS repository we can get some understanding of the lithological and structural setting of the failure.

Satellite image overlay of the rock avalanche including structural and lithological data
Click here to download the data for Google Earth 

The light blue shading indicates the failure occurred in interbedded greywacke and argillite (perhaps no surprise there), and we might infer that structure in the region of the landslide is possibly affected by the westerly-dipping faults present on the lower valley wall. The orange markers indicate the location of structural measurements, and support a regional dip azimuth of between 260 and 330 deg, with an inclination of between 60 and 80 deg. This is a similar orientation to the planar rock slope immediately south-east of Mt Dixon, and would indicate the initial movement direction was perpendicular to the local bedding orientation. Using the "Add path" tool in GE, we can draw a long profile to investigate the pre-failure topography of the runout zone. If we select the new path in the menu, the "Show elevation profile" tool will produce a plot of elevation vs. distance along the path.

Mt Dixon rock avalanche runout path, indicating an elevation loss of ~890 m, and reach angle of 29% (16 deg). The red arrow on the map indicates the transition to deposition, and is reflected by the vertical grey marker on the elevation profile. This appears to be the base of a crevasse field marking a steepened section of the glacier surface.

While these observations are relatively simple, I hope they provide some inspiration to realize the potential for what is a growing volume of readily-accessible geodata, particularly when it can be incorporated in a free 3D mapping platform such as Google Earth.

Wednesday, 13 March 2013

Visualizing geodata in 3D

Page Redirection

This blog has moved to a new home, if you are not redirected automatically, follow the link to

After spending some time working on the Bavaria maps pages (both the Earth and Maps pages work now!), I recently went in search of a way to project the same data in a vector format. At the moment I use the GDAL library (specifically gdalwarp and gdal2tiles) to convert georeferenced .tiff files to a .png format for projection in Google Earth (GE). These are raster files, just the same as the .jpg's that come from your digital camera, and each pixel is assigned a different colour. Using the data in this way has some issues, however, as the resolution of the images is limited, there's no way to attach metadata, and serving up all the image files puts a reasonable load on the server.... In contrast, vector formats allow us to define a shape by coordinates, then tell the computer a texture or colour to fill the shape, and  present the same data with essentially has no resolution limits (though it of course depends on the accuracy of the data collection). In addition vector files can contain metadata with links to external webpages, and are usually about 10% the size of the raster images. 

After passing through, I landed at the GNS GeoServer website. OneGeology is one of those 'logical' collaborative efforts much like Wikipedia or OpenStreetmap for geological maps... for some time now, Geological data has been provided online using the .wms format as a means of connecting remotely served geological data to whatever GIS software you're running on your PC (see here for an example). It's really a great step forward, I think you could say that this initiative now provides geological data of the whole world at some resolution for free. It is, however, still a bit slow and somewhat disorganized at the moment, at least on the OneGeology portal... But as it's a true cutting-edge combination of science and IT, I think the fact that it's there is really fantastic.

The OneGeology portal links .wms data from various servers around the world, one of which is located at GNS in New Zealand. While I'm new to this, the GNS site is one of the best geodata services I've come across. The portal (once you click on Data>layer preview) provides access to 1:1M and 1:250k geodata for the whole country in a wide range of formats... including .kml for GE. This can be accessed as network links streamed off the server (for GE, first download the .kmz file by clicking here), or (for example) as a .kml file to download to your computer (this is a bit faster, check the portal). 

Google Earth screenshot showing vector-based geology of the Mt Cook region 
(data credit: GNS & Google)

I'm told by the guys and girls at GNS that the data-serving side is a true 'work in progress' (a client upgrade will happen next week), and I think the future looks exciting for the users throughout the geoscience world.

Sunday, 3 March 2013

It's time to do this!

Page Redirection

This blog has moved to a new home, if you are not redirected automatically, follow the link to

Hi all,

Over the last few years I've spent a lot of time working in the Alps - learning about the valley geomorphology, slope instabilities, bedrock properties... enjoying the steep valleys, long trails, alternately hot and freezing weather, and supermarkets closed at lunchtime. 

Looking north from the Zugspitze - Feb 2013
A German version of steep valleys and freezing weather;
A view from the top of the inaccessible (for this scientist) north wall of the Zugspitze this Feb.

At the same time I've become reliant on web-based GIS as a means of maintaining field data, using tools such as Google Earth to keep it accessible even when I wasn't in the office - or when the office was not as ordered as it should have been. As I begun to generate modelling data, the simple format, and flexibility of .kml (the .xml?-based language of GE) lent itself to quickly producing results out of a number of softwares... 

The ability to compare field data to model results in a 3-D GIS environment, and share and discuss results with colleagues around the world without requiring any special software, provided some of the greatest insights during my time in Zurich.

I now have a reasonably large library of code, tools, and .kml-based geo data that grew with the project, but never really reached a 'mature' stage... Over the next months I hope to begin outlining some of the methods and tools I found most useful... and while I'm not an html-monkey I'll try to present some of the ideas, concepts, useful websites, and codes using the awesome google .api library.

To start, I've linked the 1:500 000 geological maps of Switzerland, (Liechtenstein), and Germany as Google Earth and Google Maps overlays on my "Online maps" page. In the next days I'll describe where to find the data, how to adapt it, and some of the benefits of having something like this available in a 3-D open source platform!

3D Geology of the Swiss alps looking out over Thun, Tichino, and south toward Italy