Residual Relief Separation: DEM Enhancement for Geomorphological Mapping

Hillier, J. and Smith, M.J.
Earth Surface Processes and Landforms, 33, 2266-2276.


Geomorphologically mapped data form a primary set of observations that can be used to infer former environmental conditions. Thus, objective and consistent mapping of landforms from remotely sensed data (e.g. satellite imagery, digital elevation models (DEMs) is paramount for reconstructing palaeo-environments.

This paper proposes a technique, “residual relief separation”, to enhance landforms in DEMs prior to visualization and digital mapping. This is applied to a ~600 km2 region surrounding Lough Gara, Ireland, where drumlins (~200 m wide) overlie a regional relief of hills (~10 km wide). Here, residual relief separation uses a difference in width-scale. Regional relief is approximated by a 1 km wide median filter, then subtracted to leave the drumlins in a “residual” topography. In a second step, the residual relief is normalized to allow for amplitude variations in the drumlins across the area (~5-40 m high). Finally, visualization uses a simple black-to-white colour scale for height.

Whilst not numerically out-performing other visualization techniques, this method performs equally well, and as the data are not “illuminated” there is no azimuthal bias. Additional benefits include the relatively simple calculation, intuitive visual comprehension, no emphasis of noise, and the possibility of using any desired visualization technique after the landscape has been topographically manipulated.

Introduction to the Journal of Maps Student Edition

Smith, M.J. and Lynch, K.
Journal of Maps Student Edition, v2007, 1-10

This editorial introduces material for the launch edition of the Journal of Maps Student Edition. The Journal of Maps seeks to create greater exposure, and provide publishing opportunities, for researchers that have produced map content that would otherwise remain unpublished. To this end the journal is open access and, due to the all-encompassing inter-disciplinary nature of “maps”, publishes on a wide-range of topics. The Student Edition is targeted at supporting students in furthering their research careers. The quantity and quality of journal citations is one of the primary measures of academic success, yet few research programmes actively teach their students about academic publishing. With maps often one of the first outputs of research, the Student Edition can leverage its’ unique position across disciplines to offer a “first publishing opportunity”, thereby providing students with an environment through which they can learn from their experiences. All peer-reviews are completed internally with a focus upon positive outcomes and actionable points that can significantly improve the quality of the submission. The material presented in this issue all resulted from the British Geomorphological Research Group’s Postgraduate Spring Symposium in March 2006 hosted at the University of Ulster.

Editorial: summary of activities 2004-2007

Smith, M.J.
Journal of Maps, v2007, 368-376


It has been an exciting period of growth at the Journal of Maps (JoM) overthe last three years and, as we come to the end of 2007, it seems particularlyappropriate to provide a summary of our activities, an outline of our futureplans and acknowledge the support from those involved with the journal.

GDAL/OGR 1.5 Released

THe latest version of GDAL/OGR has now been released and is available for download. This is a fairly big upgrade with 29 new and updated raster drivers and 11 new and updated vector drivers. This includes support for BigTIFF, TerraSAR-X, Generic Bbinary, ISIS3, WMS, WFS, GPX, GMT and KML. So well worth a look at. You can also see who uses GDAL in their software. Much to my surpise, ESRI’s ArcGIS 9.2 was listed and you can see that they are using GDAL to extend the capabilities of import/export alongside their standard RDO drivers. Have a look at this tech support article.

Deeply functional!

I was at the at the GRSG’s Annual Conference at the Geological Society in London today. This is one of my favourite conferences as it is low key, has commercial and academic members and brings together a wide variety of inter-disciplinary subjects. As ever, it was sponsored by the usual array of vendors, including InfoTerra. They were rather proudly showing a variety of Leica Geosystems Imagine brochures as from 2008 they will be the UK vendors. I’ve scanned the front page of the Leica Photogrammetry Suite (right) brochure which really did amuse me. Can anyone explain what “deep functionality” is?? And why is there a jet in the background with all its engines on fire, presumably plunging to Earth? Is it because the rather smug looking PC operator has just jammed all its navigation systems? Or perhaps he knows that the onboard maps were georeferenced with Imagine? Who knows… If anyone from Leica at Peachtree Corners (great address!) can enlighten me I’d be grateful.

Initial TopSat Data

TopSatAs part of the availability of TopSat tasking to UK higher education, we had our initial request for three sites in Armenia approved and have just taken delivery of the first of these. Whilst the conditions were not ideal due to the presence of a large amount of snow at higher elevations, the 2.9m panchromatic and 5.7m RGB should be ideal for our use; namely acquiring ground control for classifying DMC data of Armenia. So other issues aside, when you zoom in on the data, as the screenshot shows, there are horrible block artefacts in the imagery. Whilst we have asked why this is, my guess is that this is lossy image compression, possibly applied at the satellite prior to download. At present it remains unfit for purpose (because I can’t actually see any trees even at 2.9m resolution!). Watch this space.

ArcMap Geometry

One of the “exciting new features” in ArcMap 9.2 is vector geomtry. What hifalutin new feature is this I hear you cry? What spatial treats has ESRI unleashed upon us? Well, have a look at the graphic (right) to see a new menu item (Calculate Geometry) that is available in the attribute table. If you select that you will be presented (dependent upon whether you have a point, line or poly layer) with the following dialogue box (below). And, excitement upon excitement, those of us not blessed with extensive VBA skills can now find out the location of our feature (or its length or perimeter and area).

<

p>OK, so I had a rant about this a little while ago and it’s nice to see it in the final product….

Peyton and Byrne @ St “Meet me at” Pancras

I met a colleague at St Pancras recently and we decided to grab a quick coffee so headed to the nearest coffee bar which happened to be Peyton and Byrne, “a British Bakery”. All I can say is that if this is the state of British bakeries, please o please bring on the French. A luke warm cup of coffee, followed by a rock hard sandwich for my colleague was enough. And to cap it all off, someone walked off with my cuppa leaving a half-finished hot chocolate. Cheers!

Open Peer Review

At the Journal of Maps we operate an open peer-review system when reviewing papers that have been submitted. Peer-review has been around for over 300 years and, as far as the academic commumnity is concerned, is the best method to review the quality of material for publication. That is not to say it is ideal because it is far from it, but a viable alternative remains to be found. Of course peer-review is simply a “quality check” on an individual’s work by their peers and with such a simple definition there is scope for a myriad of operational systems. In practice the primary concern comes down to whether the authors and/or reviewers remain anonymous. These are term single-blind and double-blind systems and the reasoning is that anonymity increases the likelihood of an unbiased review. This of course is a mute point and there have been a number of studies that looked at how these systems can be abused. This occurs in the form of “gift authorship” (where authorship is given to a minor party), redundant publication (multiple publication of principally the same work), but also covers plagiarism and deliberate delay (by referees). Whilst it is usually hard to retain author anonymity because of relatively small research communities, it is much easier to remain anonymous as a reviewer. This promotes the “pot shot” culture where a manuscript is an easy target for critcism. This is not good and most authors are familiar with the crushing feeling that an abusive review manifests. JoM therefore operates an open peer-review system where neither author or reviewer remains anonymous. We believe that reviewers are responsible for their comments and should be able to stand by them. Whilst the general principle is sound, this is open to subtle abuse. For example, a reviewer of a manuscript from a “well known” author might be reluctant to be overtly critical regardless of the quality of the manuscript. I know of colleagues who, whilst being the most appropriate reviewers, have refused on the grounds that it might not be the best career choice!It is notable that there are general “norms” for peer-review across disciplines, with spatial subjects (geography, geology) being quite conservative. Single and double blind systems are common and there is a certain amount of the “pot shot” culture (although any good editor will mitigate against this). Other subjects are much more liberal in their reviewing methods. This can include the public posting of a manuscript for comment and revision by any interested parties, prior to publication. This opens up the opportunity for significant contribution by other researchers and, in certain circumstances, it is appropriate for co-authorship to be offered. Indeed, peer-review offers the opportunity for significant “value added” through the improvement of a manuscript. I believe at JoM that nearly all maps have seen improvement as a result of review, with a small proportion significantly improved. It would be nice to think that in these instances joint-authorship might be a consideration.

Standard Normal Transform

Following on from the previous post on animating a stem and leaf plot for teaching, here is another one on a standard normal transforms.

RFID Student Monitoring

This is an interesting article from Doncaster about monitoring student movement using RFID chips embedded in school badges (I assume washable!). Clearly there are issues relating to surveillance that some are unhappy about and this has to be balanced against the increase in security of children.

Passive systems (i.e. no batteries) have a range of less than 1 m, which means you would need to install readers where ever monitoring was to occur (i.e. at an entrance to a classroom). Which of course means such a system is no different from taking a register in each classroom, except its digital and its automated. At this level it is not surveillance, but monitoring and is a good thing. It means automated processing can immediately flag when students have entered the school (registration) and then when they are missing from a scheduled class.

I think there’ll be plenty more where this is coming from!

ASTER G-DEM

ASTER is an experimental sensor that sits onboard NASA’s Terra platform and is jointly run by NASA and Japan’s Ministry of Economy, Trade and Industry. Whilst it is an experimental instrument, it has provided remarkably good data (14 spectral bands ranging from 15-90m) and has been very reliable (since launch in 1999). In addition to these 14 bands it also has an aft looking sensor for the collection of stereo imagery and subsequent generation of 30m resolution DEMs. This was originally performed manually “as requested” but has now moved to an automated “on demand” service.

Because of the success of the mission, and near-global coverage of stereo data, funding has been found to compile a DEM of (nearly!) the entire globe at a spatial resolution of 30m and 7m vertical accuracy. The project, called ASTER G-DEM is currently in the preliminary preparation stages, with processing due to start in January 2008 and to take a year. The website is quite informative, showing the scope of G-DEM and how it compares to its rivals, in particular SRTM. In fact this page shows ASTER stereo coverage for the world and compares this to SRTM coverage (which is limited to 56S-60N). Also note that ASTER is photogrammetric, whilst SRTM is interferometric. One of the “problems” with the latter is that the sensor is side-looking and therefore you get data “holes” in steep terrain. This is not a particular problem for the vertical ASTER imagery. Some sample data is already available which people can have a look at. This will be distributed as 16-bit (which means integer values and therefore rounded to the nearest metre) TIFFs in 1 degree tiles.

Definitely a project to watch.

ArcGIS SP4

Well Service Pack 4 is now available, which includes a list of bugs fixed as well as “new” features. And at 176Mb for Desktop thats neh too bad; although the updated help file weighs in at a heftier 379Mb.

TopSat Opportunities for UK HE Staff

MIMAS maintain an archive of satellite imagery, primarily of the UK, for most UK HE institutions through the Landmap website. They have just announced an agreement with Infoterra Ltd for the acquisition of TopSat imagery.

TopSat is a Qinetiq satellite (as the prime contractor) that was designed by Surrey Satellite Technology Ltd and was jointly funded by the BNSC and MoD. The four key partners are QinetiQ, SSTL, Rutherford Appleton Laboratories and InfoTerra.

Up until 28th February 2008 academics can request specific “tasking” of the satellite, with up to 3 acquisitions per week being performed. This is a great opportunity to acquire both panchromatic (2.9m resolution, 17x17km) and multi-spectral data (RGB, 5.7m resolution, 18x12km) forresearch or teaching purposes.

Animated GIFs and Powerpoint

I wanted to put an animated GIF in to Powerpoint recently and, when imported, it simply showed a blank square. A little bit of Googling later and it would appear that, from Powerpoint2000 onwards, animated GIFs can be viewed (fully animated) in presentation mode only. When in editing mode, they simply show the first frame. Useful to know…

Stem and Leaf Plots

My wife was teaching stem and leaf plots to first year business students recently and wanted to dynamically show how they are built from a raw data set. We use Powerpoint 97 which doesn’t support motion paths, so makes any kind of “real” animation difficult. Open Office’s Impress has a half-hearted approach to motion paths, but they are pre-defined (although please correct me if I’m wrong!). As a result I ended up using the only animation software I have, which is the Flash-based Swish. For a quick effort, the animation works reasonably well.

Interesting factoid of the week: vinegar….

Vinegar dissolves the glue that holds the eggs of head lice to hair strands, thus making them easier to comb out. You can tell I have children … .

Quantum GIS 0.9 Released

Quantum GIS 0.9 was released last week. Things are getting close to a full “first version” and this update adds support for scripting Python plugins (and, indeed, Python programs that use QGIS libraries). If you haven’t had a go with QGIS it is well worthwhile and makes an admirable companion to GRASS.

Adpative Release (and the spiral of death)

I have blogged about using Blackboard before and, by large, things have been pain free. I hit a problem recently which should have been (and was!) easy to solve, but turned out to be quite obtuse. I use Respondus to upload multiple choice tests which my remote sensing class have to take on a weekly basis. This works well and, over the two day period the test is available, nearly all the students complete the questions. Occasionally there are good reasons why a student might not take a test which means making it available to them at a later date (there is no credit for the test so it doesn’t matter whether it has already been seen). The “Adaptive Release” facility on Blackboard covers this (and more complex) eventualities so I set up a new rule and … it didn’t work. After trying several combinations I gave up and chatted to our Blackboard people.

Adaptive Release has four sections that look like they all need completion, given the following text:

This content item is visible to all users until a [section] item criteria is created.

It suggests that each section requires completion which, as it turns out, is not the case. Indeed, for my simple scenario I only needed a date and the student IDs to make it available. This then nearly worked except that the date range set up by Respondus for the test overrides the Adaptive Release rule. Once that is removed it then does work correctly. And the spiral of death? Well, I had incorrectly set up the rule so that it entered the mark into, what I thought, was the correct gradebook item. It wasn’t and was actually waiting for a mark to be entered in to that item before it would allow the student to take the test to enter a mark into the item…

As ever, it’s easy when you know how.

Mapping amusing places

Thanks to my brother for this link to the Amusing Places website. Having spent many hours looking at OS maps (for example), this is an appropriate use of a Gazetteer. One wonders if the OS have already “flagged” up certain places in the UK…..