The changing face of Aotearoa

This talk was delivered at the National Library on 5 May 2016 as part of the series on Cartography being jointly presented by the National Library and Victoria University of Wellington. This is an edited transcription of the talk rather than a prepared written speech.

Kevin Norton is a Senior Lecturer of Physical Geography at Victoria University of Wellington. He studied chemistry and geology at The Ohio State University and geology at the University of Minnesota-Duluth in the USA. He later received a doctorate from the Leibniz University of Hannover, Germany and arrived in Wellington after a post-doctoral position at the University of Bern, Switzerland.

He specializes in the application of geochemical and numerical methods to surface processes. In 2015, he was awarded a Rutherford Discovery Fellowship from the Royal Society of New Zealand to study soil production and erosion and their role in stabilizing global climates.

Listen to the discussion

Stream the talk (52 min):

Download the recording (mp3, 60MB)

A brief introduction to geomorphology

Cartography and geomorphology

Now is as good of a time as any to define geomorphology. Geomorphology comes from geo which is the Greek word for earth. Morpho is the word for form or shape and logos is to write or to study. So geomorphology is really the study of the shape of the earth and what we do then, is look at changes in the shape of the earth.

Mackinder in 1895 had a beautiful quote that really sums up the early evolution of the field of geo saying that it was the “half artistic, half genetic consideration of the form of the lithosphere” and that’s really the link back to cartography. Geomorphology’s always been about art in some way or another, we see the landscape, we’re inspired by the landscape and it makes us want to understand how it got that way.

Shows New Zealand with islands named 'New Ulster or North Island', 'New Munster or Middle Island', 'New Leinster or Stewart Island or South Island'.John Rapkin, map of New Zealand, 1852. Ref: MapColl-830a/[ca.1852]/Acc.296.

If you look at early maps, quite often they’re annotated with these beautiful drawings of geomorphology, they may not have known they were doing geomorphic drawings but they were. They were drawing these features that were important for understanding how the landscape changes, this artistic tradition went in two different ways, it went into the slightly overboard cartographic illustration that we see here.

The role of resistance and mass movement

Drawn bird's eye view of the Black Hills, in the Western USA.

So this is a drawing by Newton from 1875 as part of John Wesley Powell’s Western U.S expedition and this is of the Black Hills in South Dakota and what I want to show you here is the utility of these maps, of these illustrations and that they show us the fundamental interplay between geomorphic systems.

In this case it’s between uplift, so the force that causes this mass to dome upwards and erosion, the force that’s eroding it back down. So we can see a big dome structure in the middle that’s been pushed up through tectonic forces and we see these rivers cutting back into it and a big eroded section in the middle and that’s really what geomorphology is all about. It’s all about the interplay between forces that lift the land up, tectonic forces and forces that erode the land back down.

The opposite end of that spectrum is my personal favourite end of this spectrum and that’s the simplistic landscape drawing and these remind me always of Chinese calligraphy, they’re such beautiful simple images and they capture everything that you need to understand about these landscapes, so maybe not a map per se but kind of in the cartographic realm at least.

Sketch of a hill slope, showing some fundamental concepts in geomorphology.

This is a drawing by GK Gilbert, one of the pioneers of geomorphology, an American geomorphologist at the end of the 19th century. It’s highlighting one of the other fundamental concepts in geomorphology and that’s the concept of resistance. What we see down here at the bottom of this hill slope are all these little forms and these are little rock promontories that did not get eroded away. So we have a nice smooth hill slope and then rocks poking out at the bottom. So we’re looking at the role of rock in lithology and in that erosional geomorphic system.

Gilbert really is one of the fathers of modern geomorphology, together with a person named William Morris Davis. William Morris Davis is actually shown in this photograph right here together with Charles Cotton, who really brought geomorphology to NZ. He was the first geomorphologist in the country it ends up; hired as a lecturer at Victoria University at the early part of the 20th century and really brought the ideas that Davis was building in the US around the formation and degradation of landscapes, brought those to NZ and here’s just a few of Cotton’s illustrations.

Photograph of William Morris Davis and other men, outdoors. Geographic cross-section sketch of the Hanmer Plains by Charles Cotton.

Here are the Hanmer plains. This actually looks shockingly like say the case around Wellington; we have a big fault here, we have an uplifted side, a side that’s downthrown and what that’s doing is controlling the landscape. We have ranges and we have valleys and you see that all around the NZ landscape where we have these strong tectonic forces that shape the landscape.

Cotton did a lot around education as well. He wrote a book that eventually became called ‘Geomorphology of New Zealand’; add a few names along the way, and that was used as a textbook around the world for about 30 years; it was a very influential book. So this is kind of the geomorphic tradition in NZ, and it’s the tradition that we kind of grew out of.

So why do we do geomorphology at all? The reason that we do geo is that it allows us to say something about the past and this is a concept that’s not completely unfamiliar to any earth scientist.

There’s a famous quote by Charles Lyle that says the present is the key to the past and what that means is that we can look at modern landscapes.

Photo of the Fusino River in Italy, showing the speaker. Photo of the Swiss Middle Land basin, illustrating the size and kind of rock formations.

In this case this is the Fusino river in Italy (me for scale) and you can see that there’s kind of boulders here, there’s cobbles, there’s all kinds of sediment in this river and if you look at the Swiss Middle Land Basin, you see the sequences of rocks that have cobbles, some of them quite big (with a geomorphologist for scale). So what we can say is that this landscape probably once looked like this so we can start understanding how these landscapes formed.

What becomes really exciting to me is that the other side of this is that if we understand how they formed, then we understand what the driving mechanisms are, and then we can actually project into the future as well. So we can not only understand the past then; now we can start making predictions about the future and that’s really where geomorphology has its sort of fundamental role in society; in understanding how changes in climate and tectonics and human behaviour are going to are going to affect the landscape.

That brings us to the point of what are these technological advances? How does all this change? And really the first major advance is digital cartography.

Making major advances

Google Earth image of the Black Hills, in the Western USA, compared to the earlier illustration.

So what you’re looking at here is just a basic Google Earth image. Google Earth is one of the most stunning tools for geomorphology that’s out there. So I just simply zoomed in on the Black Hills because we have a lovely map of the Black Hills as well, that we just looked at previously.

You can see actually Newton did a pretty good job of drawing it from above, even though there was no way for him to get above it so these are impressive illustrations. But we can see that same dome, we can see the rivers and size them to the outside but what this now allows us to do is go beyond theory and it allows us to start measuring what’s happening on the landscape. We can measure how high this mountain is, we can measure how deep the rivers are in size, we can measure hill slope angles and with all of that information we can start trying to understand how that landscape’s changing.

The really key fundamental backbone to digital cartography is digital elevation models. These are digital representations of earth’s surface and the first really big step forward was something called the Shuttle Raider Topography Mission that was flown in 1999.

Diagram of radar signals being transmitted and received by the space shuttle. Coverage map of the world. Photo of the space shuttle with the boom this transmitter was attached to.

NASA put up this big boom on one of the space shuttles, I think it's the Challenger shuttle (actually the Endeavor, in 2000 – space ed), and they shot a laser transmission - actually it's a radio transmission - down and then measured the reflection back up to the space shuttle. And what that allowed them to do was, based on travel time down to the earth’s surface and back, they got the distance between the space shuttle and all points on earth in between, so that allowed them to slowly build up a map of earth’s surface.

That's what the SRTM (Shuttle Raider Topography Mission) is. It’s really the first of what we call a medium resolution global data set. These things were released originally at 90 metre resolution. What that means is that each individual lock on the ground was and is represented by 90 by 90 metre square. So that's probably the footprint of the National Library more or less, (it) gets one elevation. So it was a huge advance forward because before this we were dealing with squares on the ground that were a couple kilometres big. It dramatically changed the way that we were able to view the earth and they did it for the entire earth between 80 North and 80 South.

What can we do with this?

Map of the North Island showing channel steepness, indicating the prevalence of steep waterways (including waterfalls) in the central island. Also, a nice photo of a waterfall.

Well we can do this sort of thing; so we can actually start to quantify what’s happening on the landscape. Here I’ve just taken the SRTM for the North Island and I’ve extracted river channel steepness from it. In this case what you’re seeing is the outline of the North Island; the yellows are very flat channels, the reds, and the purples are really steep channels and what you notice is that there’s steep channels focused around the Taupo volcanic zone up here and there’s steep channels focused around the axial mountains, sort of the Tararuas, Ruahines, things like that.

So things that look like this, these beautiful waterfalls that you see all along New Zealand; they’re related to either uplifting forces; these are like where the bits of lithosphere colliding or bits of rock are colliding and lifting up one section, like in the axial ranges. Or they’re related to down drop, to subsidence; so the Taupo volcanic zone, the big central volcanic area is actually dropping relative to everything else and it’s leaving behind these big beautiful waterfalls.

It gives us a chance to come back to those fundamental drivers of geo, things like tectonics and look at them in a quantitative way. So the SRTM has been a major revolution for us and that’s just getting more and more detailed now.

One of the other key advances that we’ve had started in kind of the late 60s (late 70s – Kevin-corrected-this-later ed), when the first satellites were launched, is global positioning systems. Pretty much everybody knows these things now, you’ve got them on your cell phone, they’re in your car, they can help you walk down to McDonalds, they are really everywhere right?

What astonishes me is that these things were initially put up in the 60s before we really understood that any of this was going to happen. They built such flexibility into these satellites that we can use them for navigation today even though they weren’t originally designed to be used for navigation.

The way that they work is through triangulation. You’re down here on the ground; you’re receiving signals from the satellites and based on the distance from the satellite they can triangulate your position to very high precision. If we use a lot of satellites and we use a few tricks then we can get extremely high precision.

In this case, this is work from Huw Horgan, who’s a senior lecturer at Victoria in the Antarctic research centre. He’s interested in glaciers and this is the Tasman glacier here and he put a bunch of GPS receivers out onto the ice and he just let them sit there. So if the ice isn’t going anywhere then they think they stay in the same spot forever right, they don’t move?

It ends up that if you measure the position of these things with time they’re moving towards the front, they’re moving towards the lake and what Huw did was looked at the rate at which they were moving, that’s this top line, and the rate would spike and then it would drop back down, it would spike and drop back down, so it’d go fast and slow, fast and slow, fast and slow. So he looked at this for quite a while and what he realised was that when the ice was moving faster we had a whole lot of rain. So the ice is responding to precipitation. This is information we never would’ve actually known if it weren’t for the GPS installation on these ice masses. This is the Tasman front right here, where they had their GPS. The height of that is maybe about 30 or 40 metres. This is a big wall of ice. So GPS has really helped us a lot with this type of high precision monitoring of locations.

One of the next large advances that we have is LiDAR.

Light Detection and Ranging

Illustration of a plan using LiDAR, sending out LiDAR pulses and spotting individual points on the landscape.

So LiDAR stands for light detection and ranging. It’s related to radar, which was radio detection and ranging, for those who remember the days of radar, and what light detection and ranging does, it takes a laser pulse, shoots that laser pulse out at something, you get a bounce off of your material and it comes back and is recorded again at the receiver. So you send out light, you receive light.

The time that it takes for that light to travel there and to travel back tells you how far away you are from that, so what LiDAR can do, is if you put one of these on an airplane and you fly that airplane over the surface, you get these returns from the ground and it allows you to make a map of the elevation of the surface.

So it works exactly like that SRTM did, which was a radar topography instead of a laser topography; works the exact same except now you’re closer to earth, which means you have a much higher resolution, so we’re not talking about 90m for one elevation, now were talking about metres, less than a metre, even down to centimetres for a single elevation pixel. This is really quite advanced.

One of the neat things about this is that were sending out millions of these laser pulses per second, usually about 4 million to 10 million is the pulse rate. What that means is that if you have vegetation a couple of those lasers pulses are going to make it down through the canopy and they’re going to bounce of the actual found surface and then there going to make it back to the receiver. They don't all do it but every once in a while one does it. Most get reflected back by the canopy, by the leaves, by branches in between, things like that, but every once in a while, one makes it through and because of that we can separate out the tops of the trees from the bottom of the ground and that's a huge advantage, something we cannot do with other techniques.

One of the other nice advances that's coming out right now, there’s not very much going on with it quite yet because it’s brand new, is that some people have changed the wavelength of these lasers that they’re shooting at the ground surface and if you use the proper wave length you can actually penetrate through water and what that means is that now we can actually map the riverbeds as well and that's something that has never been possible before.

So if we just look at LiDAR, this is a representation of what I was just talking about. Here’s the canopy and there’s the base of the canopy and this is the real ground surface. By shooting tens of millions of lasers at the ground surface, we’re able then to separate out the trees from the ground. There’s some interesting advantages for doing this; one thing that it does is allow us to actually see the true ground surface.

Maps generated from LiDAR, including a layer-separated map showing the tree layer and the ground layer.

This is a paper that was done by some people down south and what you see is that they've pulled the vegetation off and when they’ve pulled the vegetation off you can start to see traces of the Alpine fault. This is that exact same area in the field. Don't know about you but I don't really fancy finding a fault in there, it’s not really easy to do. So what LiDAR has done is actually revealed these hidden objects, they’re hidden by NZ dense bush. In other parts of the world you don't necessarily need this because you don't have thick vegetation, but here there’s a distinct advantage especially on the west coast, so were actually able then to kind of get an idea of what’s lifting up the Southern Alps.

One other side advantage to this that I’ll just talk about a little bit, is that you might have noticed that we have a mass of trees up here, so the other thing LiDAR can do is give us biomass. We can actually say, well how much vegetation do I have on this landscape? You can just subtract the trees off and measure how much trees you have. So if you are interested in say carbon sequestration by forest or understanding how much material’s available in a forestry system, these types of sensors are really massive advantages over traditional methods.

So that's airborne LiDAR.

There’s another type of LiDAR and we call that terrestrial LiDAR. It's the exact same sensor just a little tiny sensor on a tripod. This is a terrestrial LiDAR instrument, it’s kind of on a six foot high tripod. These things are shockingly little.

A hill by a river, with sensors set up on tripods. A raw point cloud showing the LiDAR output. The same cloud, now colour coded by computer analysis that identifies vegetation, rock surface, gravel, and water.

This is a group from Renne in France led by Dimitri Lague who has done a lot of work in NZ over the years and we’re here on the Rangitikei River setting up the LiDAR to scan the river. You set your LiDAR up on the tripod, you hit go and you walk away because if you stand in the middle of it then you show up in the image so you don't want that, right? We just let it turn in 360 degrees and scan the whole area and what we get is something that looks like this.

We get a point cloud of surface returns so in this case we’re focused in on one of these banks and you can see the boulders coming out, you can see the cobbles on the riverbed, you can see vegetation over hanging the river bed. One nice thing about it, is because we're not just getting return times, were also getting intensities. That allows us to segregate these points into what type of material they bounced off of.

So we can separate out the trees, we can separate out the rock, we can separate out fine rock, we can separate out water. We start to be able to actually pull apart the landscape and if interested in, say, only what size material is in a river; well we can just simply pull out all the stuff that's rock and see what size is in the river.

This is really a massive advantage for us because it’s for the first time allowing us to get at particle transport in rivers. This is something that is very hard to do, you might imagine that rivers tend to move a lot of material when they’re in flood right, they don't tend to move a lot of material when they’re not in flood, so trying to understand how much is being moved in each flood event is hard for us to quantify but repeat surveys with these terrestrial LiDAR’s systems are starting to make that possible.

Now to a terribly entertaining one that's really just come about in the last few years and that's structure from motion. Structure from motion is a concept developed out of traditional photogrammetry. Photogrammetry is a field that's been around for a very long time that involves taking photographs from two slightly different points of the same location and building up a 3d image of that location.

That's been around for a long time, but it’s always required that we know exactly where our cameras are. If we know exactly where our cameras are and where our cameras are pointed then we can construct a 3D image from our cameras. It's the same way your eyes work, they’re slightly separated, you look in the distance and that gives you depth of field. What’s happened with structure from motion, is that some very smart people have figured out ways of estimating camera positions by the image that's in the camera.

Structure from motion

Image indicating the process of taking public tourist photos of the coliseum and synthesising them into a 3D model.

From Cornell’s website, I’ve just pulled out this one; this is the coliseum. What they've done is gone online and taken a lot of tourist photos of the internet of the coliseum, and dumped it into their structure from motion programme. It looks and says which part of the coliseum I’m now looking at, places the camera – so each one of these little squares here is a camera position – places the camera based on its relative position to the rest of the scene and then slowly builds up correlations between those.

It’s an iterative process where you take two images, find out where they are relative to each other, put a third one in, find out where it is relative to the other two and keep doing that and with time, in this case just based on tourist photos, they were able to build a 3D model of the coliseum. You see this kind of thing popping up now on Google Earth pretty regularly, if you zoom in on Google Earth quite a lot, quite often now you’ll see 3D buildings coming up and they’re built through these same basic processes.

So what does that allow us to do? To give you an idea about the work load of this, a few people here and I were just in the South Island last week.

Taking a series of photographs of a hillside, and using motion DEM to develop a pretty accurate point cloud.

We took the ferry across, and while taking the ferry across, just when entering the Marlborough Sounds, I thought well I’ll take a few photos and see if we can get a structure from motion DEM off of those photos.

Here’s the ferry crossing, that’s all of the photo positions, so I took 12 photos of this hill slope as you just enter Marlborough Sounds; it’s on the right if you want to do it.

I took a series of photographs, put them into the programme and it created a 3 dimensional point cloud. Takes a couple hours to do this in total, so one of the things that this allows us to do, which is really spectacular, is go to anywhere in the world; remote locations where we can never possibly hope to get a big heavy GPS receiver in or a big LiDAR system in or anything like that. You’d take your camera that you’re going to have with you anyway because everybody has a camera and you just take a bunch of photographs and then you can build these elevation models off of them later.

So why is that useful? Well in this case I can take this 3D point cloud and I can draw profile lines down these hill slopes and these profile lines can be modelled and if I model those, what this hill slope tells me is that hill slopes in NZ, at least some of them, are continuously slipping. These are hill slopes that do not undergo the standard slip like we see in a lot of NZ. So if you go up to Gisborne, if you think about Cyclone Bola - all the material that fell off the landscape up there; these are hill slopes that have never had that happen because they have a perfect theoretical form.

So this is allowing us to kind of get to locations that you could not normally get to and start asking questions about those landscapes as well.

Related to structure from motion is really quite a popular topic and that's the use of drones or unmanned aerial vehicles (UAV), they have a lot of different names, they're banned in a lot of places now because they keep running into expensive things essentially, but this is one down here.

Photo of a landscape taken by quadcopter, and two charts showing how you can take two sets of data derived from quadcopter photos to clearly show the change in that landscape.

It’s just a little quad copter, you can buy these things for a couple hundred dollars, sort of duct tape your camera to the bottom of it, which is what we had done here and send it up and just put your camera on repeat and have it just keep taking photographs.

This is again work that we had done when Dimitri Lague was here a couple years back, taken again on the Rangitikei and what we got out of this, because Dimitri comes back every year to every two years and has been doing so for 10 years, is change. You can look at this point cloud that we created first, the elevations that we created the first time, the elevations that we created the second time and we can ask the computer what’s the elevation difference between these two points, and what it’s done is pointed out that some of these big meander bends like you see here on the Rangitikei, some of these things are undergoing significant mass loss. The purples here are mass loss, the reds are mass gain, so you can imagine that we just have material falling off the hill slopes and piling up at the bottom of the hill slopes and now the river has to deal with these and the rivers may or may not be capable of moving this material.

These things are really changing the way that we do geomorphology. What maybe used to require years of effort. Each campaign would actually take us multiples days, usually weeks, because we’d have to go out with a total station, which is a distance measuring device that has angles on it. We’d measure the distance and the angle, then we’d calculate the trigonometry, how high it was and then we’d measure the distance and an angle, use trigonometry to measure that elevation and we do that again and again and again and you can imagine how long it would take to do a 1km long section of a river that way.

So this idea of structure from motion is really taking off in geomorphology and it seems to be kind of where we’re going right now. One of the other things we can do is not take repeated photos at one time of a location, but we can take repeated photos through time of a location, known as time lapse photography.

This is up at Tongariro National Park and we had an idea that something interesting was happening up there but we weren’t real sure what it was and so I set my camera up one day and just let it take a photograph every minute. Every minute it took a photo, took another photo, took another photo. After a few hours I came back and grabbed my camera and dumped all the images into a processing programme and what we see is that overnight we have these beautiful 1,2,3cm-long ice needles that are growing on the surface. During the day they’re melting and if you look at what’s happening they’re going straight up and then they’re melting off to the side.

What’s happening is that we’re actually transporting material down the hill slope just by this ice growing and thawing and this is the reason that all of those slopes around the Tongariro National Park are denuded of sediment, why they don't have anything sitting on top because this type of a process is happening on a regular basis.

That one was pretty exciting but this one I find shocking, this is Bryan Anderson’s work.

Bryan Anderson is a researcher at Victoria in the Antarctic Research Centre, works a lot with Huw whose work I showed before. Bryan has a camera set up above the Fox Glacier and he set this up to look at the retreat of the Fox Glacier; nothing else, he just wanted to see how the Fox Glacier was flowing and how it was retreating through time and what came out was something that none of us knew was there. The whole hill side is just collapsing as the ice is melting back. As the ice is melting down it’s undercutting that hill slope and destabilising the whole thing.

This is over the course of almost a year so you kind of see the sun come in; this is the same, its 11am every day for a year. If you look at this, there’s people at the beginning that are wandering up, this is the tourist trade up on Fox Glacier, they used to kind of walk up there and Bryan had already told them at this point that this was happening and they kept taking people up and if you notice they don't actually stop until it starts collapsing catastrophically in the middle. Even after this bridge here breaks you see a few people wander up, so it’s pretty shocking.

What this has allowed us to do is look at say ice retreat and the destabilisation of hill slopes in locations that we never thought possible and we can now also quantify that. We know how long it is between each photograph, we can measure how far each rock has moved, so we can look at the rate and distribution of deformation on that hill slope, so it’s really powerful stuff. If you’re interested in seeing that again it’s on Vimeo, if you just type in Fox Glacier slip and VUW (time lapse photography) you’ll get there.

All of this is great for coming up with theories but what about the rates, the processes around that? There is a wonderful quote by Dick Charlie in the late 70s that I truly love because I’m this type of geomorphologist.

What about long-term changes?

A geomorphologist with a soil auger.

He said “whenever anybody mentions a theory to a geomorphologist he instinctively reaches for a soil auger.” And this is what we do, so we come up with these great ideas like, well maybe ice retreat is causing hill slope failure. It’s a sound theory but we can’t necessarily test it just with photographs, we have to go out to the landscape and we have to find a way to test it.

That's where my personal area of expertise comes in. In geomorphology, I’m kind of a geochemist as well, so I look at the chemistry of rocks and I can use something called a cosmogenic nuclide as dating technique on rocks; so it's a stopwatch. What these cosmogenic nuclides are, are just simply atoms that are formed from cosmic rays. Don't worry about it, but right now there are thousands of cosmic rays travelling through your body, these are high energy particles that are emitted from black holes, things like supernova and they’re flying through space, through the atmosphere, through us. If they hit an atom on earth, in a rock, in us, in the atmosphere, they split that atom apart.

Photograph of an atom being split. Photograph of a rock dropped by a glacier, illustrated with 'rays' showing how the accumulation of subatomic particles can indicate how long a it has been on the earth's surface.

This is one of the most exciting photographs in all of earth science I think. This is one of the very first photographs of an atom being split. It’s really just humbling to look at this thing. They shot a proton at a heavy nucleus and that nucleus flew apart, these big lines, the short lines, are cosmogenic nuclides, they’re stable atoms that came off of it and then all these tiny little lines are other little energetic particles that can go on to make more of these things. These cosmic rays come in from above, they create this huge cascade of particles that rain down on us, and through us and these all create these new atoms at earth’s surface.

This is a photograph of Northern Sweden and here’s a rock that was dropped by a glacier and the longer that rock sits there at earth’s surface, the more of these atoms accumulate. So we can use them as a stopwatch to find out how long something’s been at earth’s surface.

One place that we’ve done this rather extensively is in Antarctica.

The Spenser Mountains. The Mackay Glacier. Charting the subatomic particles in surface rocks over time and place, allowing the gathering of climate change data.

This is work that a PhD student of ours, Richard Jones, did. This is the Mackay Glacier; you can see where it is right here, so the Mackay Glacier is draining the East Antarctic ice sheet, it’s an outlet glacier for the largest ice mass on the planet. And what Richard did was collect these boulders, and then counted up the number of cosmogenic nuclides in these boulders and it told them how long ago the ice deposited these boulders, how long ago the ice was up here and then how long ago it was down here, and then how long ago it was down there and then further and further.

What he saw, if you look at the elevation that we collected these boulders at versus the age – the ice was relatively stable; and then about 6000 years ago it catastrophically thins. This is 200m of ice thinning and it happens in about 400 years. The ice simply just lowers almost instantly as far as geology is concerned, 400 years is no time at all. We lose a couple hundred metres of ice, I think that's the height of the Sky Tower, so imagine you are losing a Sky Tower worth of ice in just a couple hundred years, and this is draining the largest ice mass on the planet. This was pretty shocking and what we’re doing now is applying this technique to glaciers in New Zealand to try to understand how ice has grown and shrunk through time in NZ as well.

One of the other things we can do is go back and look at some of the classic geomorphic systems in NZ. This is work that Cam Watson, a master student at Vic is doing. One of the classic geomorphic problems in NZ is something called the K-surface. Charles Cotton proposed the K-surface over a hundred years ago; it’s right here in Wellington.

Chart and sketch indicating how historic data and models can be combined with newly-gathered data to better understand geological events.

If you’ve ever been up on top of Mt Vic and looked west, you probably have seen that we have kind of flat topped hills off towards the west. What Cotton proposed is that those flat top hills were once all connected and then they gotten cut by faults and that this big flat surface was eroded somewhere around 4ish million years ago plus or minus a bit. If that's the case, then all of these faults have moved over the last 4 million years, so these off-sets are 4 million years’ worth of accumulated movement.

What Cam did was collect rock on these K-Surfaces and then modelled how old it must be, based on how much of these atoms were in it, how many of these atoms there were and what he found is that these things are about 250,000 years old not 4 million years old. What that implies is that all of that movement, a couple hundred metres of movement has taken place over a quarter of a million years and not 4 million years. The Wellington region is moving a whole lot faster than Cotton thought it was, and probably a lot faster than a lot of us thought it was as well.

We can do other stuff with these things, so we can look at weathering and erosion. Erosion is breaking rock down into smaller particles, weathering is chemically changing that rock, so turning it from rock into soil and so we can use these cosmogenic nuclides to analyse this.

Collecting river sand.

Here is an old master student of mine Abby and I collecting river sand. We can then count the number of these atoms in this river sand and that tells us how fast the entire catchment is eroding. So say you’re interested in, how fast the ulterior gorge is eroding, we can do that by taking sand down at the outlet and finding out the average erosion rate of the upstream area. This has been a technique that has allowed us to understand large scale formation and degradation of landscapes.

Chart showing denudation rates of Swiss regions and recent rock uplift rates. Shows that the rate is high where the land is the highest, and where we have the most uplift, so tectonics flux.

I’ve done a lot of work in the European alps, in this case, in the Swiss Alps and in landscapes that look like this. This is the Val Lumenzia where we’ve collected river gravels at all these locations marked by the sample names here. If we plot those up, here’s the erosion rate, or the denudation rate. You see that the rate is high where the land is the highest, where we have the highest elevations and it’s also highest where we have the most uplift, so tectonics flux.

What that shows us is that erosion is driven by rock uplift, driven by tectonics. In NZ we sit at the boundary of the Australian and the Pacific plates; they’re coming together, they’re causing rocks to be uplifted, that means in NZ erosion should be driven by tectonics as well. If we understand how that's happening, how this uplift is happening, then we’ll understand how the erosion is happening.

The land surface has to keep up with the uplift; so the erosion, the downward erosion, has to keep up with what’s pushing it up and some colleagues from the US have looked at this recently on the West Coast.

The NZ West Coast. Map showing locations where soil samples were gathered.

What they did was scaled these mountains, which was a formidable effort and took soil samples all up and down mountains on the West Coast. Here’s a map of where they were; so they went up and down these things, collected a lot of samples and what they get is that – you don’t have to worry about that Y-axis or X-axis; just note that all these black dots, which are the NZ case, plot well above everything else. This is the global data set of chemical weathering and this is NZ. NZ weathers something like ten times faster than anywhere else in the world.

I’ve taken a stab at modelling this, trying to understand why NZ can weather so fast, and we do a pretty good job of modelling these really fast weathering rates on the west coast. So what are the controlling factors of that erosion and weathering? Weather, their uplift, we’ve seem that already.

Maps of New Zealand, focusing on different geological aspects.

This is an uplift map from Te Ara, the online encyclopaedia of NZ, just showing high uplift rates in the Southern Alps and then in Kaikoura and sort of up around the axial ranges as well. These are the places we’d expect to see really fast erosion, creating things like Mount Cook with uplift as well. Where we get high rainfall, we also get high erosion, you imagine big rivers means a lot erosionability; and also where we get very low temperatures, we tend to have a lot of erosion. The reason for that is shown up here; you have rocks shattering because of ice, crossed action just simply breaks the rocks apart. So all of these things together inspire to create very fast erosion rates and very fast weathering rates.

To finish up this part of it, which is almost the whole thing; if you plot these erosion rates that I’ve measured with cosmogenic nuclides in some of the big rivers against modelled soil production rates, our understanding of soil production and how it works chemically; what you see is that this black line here is a 1-1 line, that’s the line where the amount of soil produced equals the amount of erosion of the soil, anything below that line has more soil produced than were eroding. That’s a location on the landscape where soil is stable.

Charting the soil life of New Zealand. The North Island's doing okay, but the South is potentially losing far too much.

Anything above that line is where we have more erosion of that landscape than we have ability to produce soil. That scares me, so almost every single landscape I’ve looked at and if you plot these up, the ones in reds and yellows have excess erosion, the ones in blue have excess soil production, so were kind of okay. The North Island’s not going too bad but the South Island is seriously in trouble. What this suggests is that, if our modelling is correct, a lot of NZ landscapes have a limited soil life.

So the summary of all that is that NZ, first of all is a stunning natural laboratory for geomorphologists, it’s really the one place in the world that we can come and put our finger on something that’s changing today. The other thing is that we have some of the fastest uplift, fastest erosion, fastest weathering and fastest soil production here.

So that question of is soil sustainable, that’s something that I’ll be moving forward with in the next five years with a Rutherford Discovery Fellowship from the Royal Society. Hopefully I can come back and tell you guys that we’re all fine, we’ll see how that goes.

I’m going to end this talk just with some photographs and just to kind of highlight what geomorphology means. What we’ve learned from all of these techniques is that there’s a lot of challenges, a lot of hazards related to geomorphology, because we live on the landscape, and because the landscape is changing, humans and landscape are going to bump up against each other on occasion. So the better we understand geomorphic systems, like say rock strength, rock failure, river flooding, things like that, the better we would be able to predict how it’s going to change through time.


Retreating glaciers, showing bare rock and lakes.

One thing we have to deal with is retreating glaciers, so you have the Hooker Glacier, these things have proglacial lakes, because in the 1980’s they were all the way out here and up here and today they are all the way back there, and these things have retreated kilometres in the last couple of decades. That’s creating lakes, which are a bit of a hazard but also a lot of sediment and as you saw earlier, it’s probably destabilising all these hill slopes.

Increased sediment load, getting dumped into rivers and making them flood more.

This is leading to massive increase of sediment load. So these things, if something gets dumped into the rivers, so what does the river do with them? Well the river can’t quite transport it, so the river ends up flooding a little bit more often and we end up with these downstream effects of increase flood hazard.

Coastal hazard, for example in Wellington where storms can take chunks out of Island Bay.

One other thing that in Wellington is quite important is coastal hazard. I’m sure anybody who lived in Wellington a few years ago remembers this. The big southern storm took out a lot of Island Bay and only when we kind of get a better handle on wave frequency and wave magnitude and things like that are we going to really understand our coastal hazard pretty well.

Collapsing slopes, including into rivers.

The other thing that’s happening is, like we saw with the Rangitikei, these slopes are collapsing, they’re being undercut, they’re collapsing into the rivers, and so again we’re having to deal with increase sediment load and downstream transport of this material which is then creating elevation, it’s filling up the downstream part of these rivers with soot and clay.

Increasing slope instability making land difficult to build on or cultivate.

Then lastly slope instability like up here in Tongariro National Park, all of this stuff is just slopping of the hill slopes and you might say it’s not a problem around here because it’s in a national park but if you build a house on this stuff you’re going to have to contend with this movement.

A generally nice landscape.

So I’d just like to finish off with a quick thanks to everybody who’s kind of helped out with images and movies and what not. Thank you to Richard Jones, Cam Watson, Dimitri Lague, and Brian Anderson for maps, diagrams, and videos.

By Kevin Norton

Kevin Norton is a Senior Lecturer of Physical Geography at Victoria University of Wellington.

Post a Comment

(will not be published) * indicates required field

Be the first to comment.