Landlab teaching tools

I’m trying to develop teaching materials using Jupyter notebook (mainly inspired by my colleague Fabien Maussian) and its super interesting but quite hard work for me as a python beginner. I was looking for other things people have been doing with this format and I came across the teaching tools that accompany Landlab, which aims to create an environment in which scientists can build a numerical landscape model without having to code all of the individual components (Hobley et al., 2017). Landscape models have a number of commonalities, such as operating on a grid of points and routing material across the grid. Scientists who want to use a landscape model often build their own unique model from the ground up, re-coding the basic building blocks of their landscape model rather than taking advantage of codes that have already been written. Landlab offers python coded building blocks for developing your own model. Cool huh? Landlab is described in an open source paper.

The Landlab team currently shares teaching resources appropriate for geomorphology and surface water hydrology classes. The exercises use numerical models to illustrate physical processes. These exercises were designed as homework or laboratory assignments, but they could also be used to illustrate concepts in the classroom related to:

  • Hillslopes evolving according to the linear diffusion equation.
  • Drainage density sensitivity to the strength of hillslope and fluvial processes.
  • Fluvial channel morphology (steepness and chi-elevation relationships) sensitivity to rock uplift and rock erodibility.
  • Hydrograph sensitivity to watershed shape and storm characteristics.

These exercises do not require coding knowledge. They are written in Jupyter notebooks, which combine text and code, and are easy for students to use. The exercises include directed exploration exercises (i.e. students are told exactly how to change the code and run it) and thought and interpretation questions based on the resulting plots. The exercises can be tailored for your class.

Everything is open source so FREE! You can even run them online without any software installation.

For more information: https://github.com/landlab/landlab_teaching_tools (scroll down that page to see text), or just email the developer: Nicole Gasparini ngaspari@tulane.edu

References:

Hobley, D. E. J., Adams, J. M., Nudurupati, S. S., Hutton, E. W. H., Gasparini, N. M., Istanbulluoglu, E., and Tucker, G. E.: Creative computing with Landlab: an open-source toolkit for building, coupling, and exploring two-dimensional numerical models of Earth-surface dynamics, Earth Surf. Dynam., 5, 21-46, https://doi.org/10.5194/esurf-5-21-2017, 2017.
Posted in training | Comments Off on Landlab teaching tools

Der Standard “Eis und Klima” blog

Our research group has started blogging on Der Standard, an Austrian broadsheet. The aim is to let Austrian readers know a bit more about both our research, why we do it, and what is involved in being a working scientist.

Part of the idea is to offset the (for me a bit upsetting) trend of doubting scientists and their motives for doing their work, by showing a bit of ourselves in this blogging process, while also letting people know what is being done with our portion of the Austrian research funds we are allocated.

Hopefully it will be a positive experience, and I’ll update this post with the new articles as they come out every 4-6 weeks.

1. Kristin Richter has posted on Die Vermessung des Meeresspiegels (Measuring sea level), starting with the question of how come an oceanographer studying sea level rise ends up in the Alps, and going on to explain what data on sea level rise is available for investigations.
2. Johannes Horak has posted on Die Vermessung der Gletscher (Measuring glaciers), covering a broad sweep of what glaciers are, some reasons why they matter, how they are changing and how we can measure these changes.
3. Lindsey Nicholson has posted on Wie unberührt ist Grönlands Schnee? (Are the snows of Greenland pristine or polluted?) about an expedition to Greenland to see if pesticide pollution from lower latitudes is contaminating the snow on Greenland
4. Elisabeth Schlosser has posted on Antarktis: Leben und Forschen im ewigen Eis (Antarctica: Life and Research amidst the ice), about the revisiting the Neumayer III research station for her research after overwintering in the first station almost 30 years earlier.
Posted in Uncategorized | Comments Off on Der Standard “Eis und Klima” blog

3D numerical model of englacial transport

Anna Wirbels first PhD paper is out, and downloadable from my publications page (Wirbel et al., 2018). What she has done is take an existing freely available 3D model of ice flow (icetools; Jarosch et al., 2008) that employs the full Stokes equations to describe the flow of ice. This type of numerical treatment of ice flow offers the most realistic representation of the flow of complex mountain glaciers, assuming adequate inputs, such as terrain properties and ice temperatures are well known.

Anna worked, with guidance from Alex, to introduce a numerically robust, mass conserving, treatment of advection of deformable inclusions within this ice flow field. This allows her to represent englacial debris transport – illustrated in the video below by artificially imposing layers of initially circular inclusions into a cross-section of a glacier with fixed geometry and flow field (Time is in years):

Isn’t that cool?! (The full resolution video is downloadable as supplementary material to the publication)

Imagine the initial circular feature as a lump of rock material from a rockfall that has been buried in the accumulation zone, or sediment trapped in a hollow in an englacial channel, imagine the initial surface layer as an ashfall, and the initially vertical bands part way down the glacier as rocks and detritus that has fallen into crevasses, then look how over time they all get deformed and elongated into bands, which is what we see in real glaciers!

Also, notice how over time, even in this fixed flow field case, the location of debris emergence and the angle of incidence of the debris band with the surface both change over time. In these cases the model does not keep track of the debris once it emerges to the glacier surface, but its clear that both of these aspects will affect the debris flux to the glacier surface, and, combined with the surface ablation will determine the manner in which a supraglacial debris cover forms.

The model code is available from Anna on github: https://github.com/awirbel/debadvect/releases/tag/v1.0.0

References:

Jarosch, A. H. (2008) Icetools: A full Stokes finite element model for glaciers, Computers & Geosciences, 34(8), 1005–1014, doi:10.1016/j.cageo.2007.06.012.

Wirbel, A., Jarosch, A. H. and Nicholson, L. (2018) Modelling debris transport within glaciers by advection in a full-Stokes ice flow model, The Cryosphere, 12, 189-204, https://doi.org/10.5194/tc-12-189-2018.

Posted in Uncategorized | Comments Off on 3D numerical model of englacial transport

Featured on the cover of Geosciences Vol 7, No. 3

Well, a nice surprise at the start of the year: Hannah Prantls paper on mapping glacier snow cover extent and snow line elevation using terrestrial laser scanner signal returns is featured on the cover of Geosciences Vol 7, No. 3. Pretty cool as I think this is Hannahs first research article 🙂

The summary text for the paper is:

We demonstrate that Terrestrial Laser Scanner (TLS) return signals can be used to accurately map the snow cover extent over a glacier. A rule-based classification employing intensity, surface roughness and an associated optical image, achieves classification accuracy of 68–100%. Snow cover extent is valuable information for glacier surface energy balance models, which are sensitive to the glacier surface condition, however as the TLS intensity signal shows no meaningful relationship with surface or bulk snow density, the snow mass remains elusive.

Here is the featured figure:

Evolution of the TLS and AMUNDSEN model snowlines during summer 2014 and summer 2015. The order of the raster layer is: (A) 26 June 2014, (B) 18 July 2014, (C) 1 August 2014, (D) 25 August 2014, (E) 4 September 2014, (F) 23 September 2014, (G) 4 October 2014, (H) 21 April 2015 and (I) 1 October 2015.

And by all means consider reading the whole paper:

Prantl, H., Nicholson, L., Sailer, R., Hanzer, F., Rastner, P. and Juen, I. (2017) Glacier snowline determination from terrestrial laser scanning intensity data, Geosciences, 7, 60, doi:10.3390/geosciences7030060. [pdf]

Posted in Uncategorized | Comments Off on Featured on the cover of Geosciences Vol 7, No. 3

Randkluft and bergschrund

I only just found out what a randkluft gap is. For years I’ve been wrongly calling it a bergschrund. I love how these words are readily known by German-speaking mountaineers and I am so late to the game despite it being in my field of study.

Wikipedia sets me straight: “A randkluft is similar to, but not identical with, a bergschrund, which is the place on a high-altitude glacier where the moving ice stream breaks away from the static ice frozen to the rock creating a large crevasse. Unlike a randkluft, a bergschrund has two ice walls.” and provides this helpful graphic.

And now I am already know something more in 2018 than I did in 2017. Keep learning, keep growing, so it is said!

Posted in Uncategorized | Comments Off on Randkluft and bergschrund

Khumbu Himal drone footage

The Khumbu Himal is an impressive mountain environment – so much geomorphology and so many amazing glacier features to see!

Here is a fairly recent drone footage video of the 3 passes circular hike through the upper Solu Khumbu, which crosses the Kongma La, the Cho La and the Renjo La (‘La’ being a pass). I’ve no advanced scientific commentary on this – just wanted to share it, so I hope you enjoy the virtual flight around one of my favourite places.

Posted in Uncategorized | Comments Off on Khumbu Himal drone footage

Famous on Instagram

I don’t have an Instagram account, but thanks to having awesome, and talented friends, our work was featured on the National Geographic Instagram feed in October (I know – I’m slow to post this!) when photographer Robbie Shone joined us on a gorgeous day of fieldwork on Hintereisferner and the neighbouring glaciers. This high mountain valley hosts a scientifically valuable monitoring network, in which institutes including the University of Innsbruck (ACINN and Geography), the Bavarian Academy of Sciences (Commission for Glaciology and Geodesy), and the Tirolean Government (Hydrological Office) measure glacier change, permafrost change, meteorological conditions, precipitation, river runoff and more. These data are useful for understanding how the mountain environment is changing and can be used to develop numerical models of the environmental processes so that we can make useful predictions of future change.

The site is part of several international research networks: UNESCO IHP, GEWEX INARCH, ERB Euro-Mediterranean Network of Experimental and Representative Basins,  The international Long Term Ecological Research network (LTER-Austria, LTER Europe and ILTER), and the basecamp Station  is part of the EU Horizon 2020 INTERACT framework of Arctic (and a few Alpine) research stations.

We were doing two main things on this field work:

Firstly, we were measuring the glacier change. To do this we dig snowpits to record how much snow survived the summer and how much mass it adds to the glacier and also measure the length of stakes drilled into the glacier to record how much the ice surface has lowered, so they know how much ice the glacier has lost at that point over the year. They sum these mass gains and losses like a bank balance to see how the glacier has changed over a year, and extrapolate the changes at each measurement point across the glacier.

 Here I am sampling snow density through the fresh early autumn snow and the snow that survived the previous summer. We need the density of the snow so we can convert the snowdepth into a mass of water – the same depth of heavy wet snow is worth much more water than diamond powder. Photo credit: Robbie Shone.

The data are reported to the World Glacier Monitoring Service. Changes to the Hintereisferner have been measured this way by the University of Innsbruck for over 60 years, making it one of the longest detailed records of glacier change in the world!

Secondly, we were collecting data, and performing maintenance at our meteorological stations. There are six automatic weather stations operating at high elevation in this watershed, and a number of historical rain guages that have been measured for decades.

This automatic weather station is one of the more recent ones, which sits on the surface of Hinterieiferner. Data from this station tells us about the microclimate of the glacier and allows us to relate the pace of surface ice melt to the hour by hour weather conditions. The sensors are supplied by Campbell Scientific and the mast is a self leveling model designed in-house at the University of Innsbruck. Photo credit: Robbie Shone.

My personal fave photo of me from the day (was not featured on Instagram!) – a happy scientist just been put down by a helicopter on a glacier and a day of sunshiney data collection with good company ahead. Photo credit: Robbie Shone.

Posted in Uncategorized | Comments Off on Famous on Instagram

Visualizing the retreat of Hintereisferner

Hinteresiferner is the best-studied glacier in Austria with one of the longest records of glacier mass balance in the world. There are relatively abundant old maps and the oldest photographs of Hintereisferner date back to 1884, only 30 years after the Little Ice Age maximum in the Alps. The glacier extent has never been as great since then.

Philipp Rastner published a paper in 2016 (Rastner et al., 2016) describing how one can make appealing vizualizations from a timeseries of digital terrain models. In this publication an animation of the retreat of Findelengletscher in Switzerland was generated, and  using 11 topographic maps and making the assumption that the glacier geometry change between each surface model is linear – meaning the change between two digital terrain models is distributed evenly over the intervening years. This might not be realistic; for example instead of gradual retreat the glacier might have retreated rapidly and then stayed relatively constant, but for vizualizing the cumulative change over time the approach is still useful.

The method involves careful filling of any gaps in the digital terrain models, coregistering them and rendering the animations. From his paper this was done as follows: “All data pre- processing and main processing were performed in the ENVI 4.7 remote sensing and IDL  programming software (Exelis Visual Information Solution, USA). Esri ArcGIS 10.2 and Microsoft Excel were used for the co-registration and Adobe Photoshop for the DEM void filling and artefact removal. The DEM morphing was done in Abrosoft Fanta Morph and the rendering of the computer animation in Visual Nature Studio 3 (VNS; 3D Nature LLC). Finally, all movie  components were matched together in the video editing program Adobe Premiere.”, but you’ll have to read the paper for more details.

Studies based on Austrian glacier inventories (1969, 1979 and 1998) and reconstructions of glaciers at the Little ice Age maximum have revealed the scale of the glacier retreat over this period (Fischer et al., 2015), but Philipp Rastner has also made an animation of Hintereisferner change in the same way as he did for Findelengletscher using a number of alternative digital terrain models of the glacier surface, and here it is:

References:

Rastner, Philipp; Joerg, Philip C; Huss, Matthias; Zemp, Michael (2016). Historical analysis and visualization of the retreat of Findelengletscher, Switzerland, 1859–2010 Global and Planetary Change, 145:67-77.

Fischer, A., Seiser, B., Stocker Waldhuber, M., Mitterer, C., and Abermann, J. (2015) Tracing glacier changes in Austria from the Little Ice Age to the present using a lidar-based high-resolution glacier inventory in Austria, The Cryosphere, 9, 753-766, https://doi.org/10.5194/tc-9-753-2015.

Posted in Uncategorized | Comments Off on Visualizing the retreat of Hintereisferner

What can we say for sure about anthropogenic climate change?

Oh sometimes its so hard to find the right words to talk about climate science clearly. At least I struggle with this quite often, but I like the answer given by Dr Kate Marvel within this interview:

Q. What can we say for sure about anthropogenic climate change?

A. First, we know that carbon dioxide is a greenhouse gas. We know what its molecular structure looks like, and we know that this structure means that it absorbs infrared radiation. If we’re wrong about this, we’re wrong about the very basics of physics and chemistry.

Second, we know that burning fossil fuels increases carbon dioxide in the atmosphere. The chemical reactions that produce energy when we burn oil, gas, or coal inevitably produce CO2 as a byproduct. And that CO2 goes into the atmosphere. We have excellent measurements of atmospheric CO2, and they clearly show a dramatic increase since the industrial revolution.

Third, we know the climate has been changing. Multiple independent datasets show the global temperature rising. But that’s not all that’s been happening. There is more water vapour in the atmosphere. Spring is coming earlier. Rainfall patterns are shifting. Glaciers and sea ice are melting. There are more and deadlier heat waves.

Fourth, we know that these changes are very, very likely to be due to human activities. We know that the climate changes due to natural factors, but we also have a fairly good understanding of what the climate would look like without us. We can model this natural variability using powerful supercomputers, and we can also study the climate of the past using things like tree rings and ice cores. The changes we’ve observed are too large and too rapid to be attributable to any known natural factors. And they’re very consistent with what we expect increased carbon dioxide to do to the planet. An alternate explanation would have to come up with a plausible natural mechanism for these changes and explain why CO2 doesn’t act the way we think it should – and that’s a very tall order.

Here is Kate talking about climate modelling and clouds in a TED talk:

Also, from her website, a lesson in how to write succinct research summaries:

“My postdoctoral work identified a “fingerprint” of human influence on global precipitation patterns and showed that we are already changing rain and snowfall. This is both reassuring, because it suggests climate model projections are credible, and terrifying, because it suggests climate model projections are credible.”

 Ha ha! thank you Dr Marvel.

Posted in Uncategorized | Comments Off on What can we say for sure about anthropogenic climate change?

Determining supraglacial debris thickness from terrestrial photographs

Big shout out to my colleague Jordan Mertes for an excellent and fun collaboration – our new paper in Journal of Glaciology is online now.

Nicholson, L. and Mertes, J. (2017) Thickness estimation of supraglacial debris above ice cliff exposures using a high resolution digital surface model derived from terrestrial photography. Journal of Glaciology. doi:10.1017/jog.2017.68

Years and years ago when I was doing my PhD we wanted to get an idea of how thick the layer of rubble and dust was over the Ngozumpa glacier (in the photo below – this is the biggest glacier in Nepal and the lower 12-15 kms of it are covered in rock debris/rubble). To find out more about debris-covered glaciers like the Ngozumpa, have a look here.

Its pretty unrewarding work to dig through this kind of rock debris overlying the ice, but the debris covered part of the glacier is studded with steep bare ice cliffs, above which the rock debris layer can be seen in a vertical section. So, to avoid digging holes, we hit on a good idea that involved Doug Benn running round the crest lines of these exposed cliffs with a surveying reflector while I surveyed his position from the glacier side moraine, using a theodolite to measure the dip and distance to the reflector and also the dip to the debris-ice interface beneath him. Using simple trigonometry, and imagining the ice cliff geometry to be simpler than it really is, we could use these measurements to get an estimate of the debris thickness visible above the cliffs.

The world moves on. Now simple photographs can be used with some very cool software (Structure from Motion – Multi View Stereo processing software to be precise) to relatively easily make digital 3D surface models. It occurred to me that the kind of measurements I did in my PhD could be done readily, and probably more accurately, from a high-resolution terrain model. In spring 2016, Anna Wirbel and I took a bunch of photos of ice cliffs to see how it would go.

The short answer is: it turned out rather nicely. Below is an oblique view of the surface model we made from our photos and the coloured dots forming lines along the ice cliff crests show the implied debris thickness (hd) in centimetres along the three ice cliffs we took pictures of. For an idea of scale the cliff peak furthest away is almost 45m high. The numbered red and yellow dots are ground control points that were accurately surveyed and used to scale the model correctly.

For the long answer you can read the paper – its freely available online 🙂

I’d really like to see if the lower resolution, but closer-to-target, imagery from UAVs, that are now almost regularly flown over debris-covered glaciers can also be used to perform similar debris thickness assessments at a glacier scale. I don’t have a UAV though …. so I hope someone else will do that with their UAV images!

Posted in Uncategorized | Comments Off on Determining supraglacial debris thickness from terrestrial photographs