Antarctic Chasm One
Client: Matt Polaine, British Antarctic Survey <maine@bas.ac.uk>
Second attempt:
Antarctic Chasm One
The Brunt Ice Shelf in the Antarctic is growing a massive chasm at the rate of 1.7km a year, that may lead to an area of nearly 1,000 sq km breaking off and taking the Halley VI base with it. Your task is to make a 3D immersive visualisation of this growing chasm, allowing viewers to descend into it in the way that BBC’s Peter Gibbs did on camera (goo.gl/48cVds). You will have access to aerial scans, photographs, video, multispectral satellite imagery, ground penetrating radar of the crack and other data from technical specialists at British Antarctic Survey (BAS). You will use these to extrapolate places where data is missing, and also to model the growth of the chasm for a speeded-up view of the future. For a full immersive effect, spatialized audio would add further realism.
Original proposal
The Brunt Ice Shelf in Antarctic is growing a massive chasm, that may lead to an area the size of Wales breaking off. Your task is to make a 3D immersive visualisation of this growing chasm, allowing viewers to descend into it in the way that BBC’s Peter Gibbs did on camera, but using machine learning methods to extrapolate the growth of the chasm over time and into the future. You will have access to many sources of aerial scans, ground-based and water data collected by BAS. http://www.bbc.co.uk/news/science-environment-36197657
Extended version - needs to be edited down!
Antarctic Chasm One
The Brunt Ice Shelf in the Antarctic is growing a massive chasm at the rate of 1.7km a year, that may lead to an area of nearly 1,000sqkm breaking off and taking the Halley VI base with it. To avoid this BAS has decided to move Halley VI – a considerable undertaking. http://www.bbc.co.uk/news/science-environment-36197657
Your task is to work with British Antarctic Survey (BAS) to make a 3D immersive visualisation of this growing chasm, allowing viewers to descend into it in the way that BBC’s Peter Gibbs did on camera, but using machine learning methods to extrapolate the growth of the chasm over time and into the future to the point of eventual separation. This animation of the crack propagation to be compressed in time, with ~5 minutes representing ~50 years, from detection of chasm change to Brunt calving. You will have access to many sources of aerial scans/photographs/video, multispectral satellite imagery, (entire length) ground penetrating radar cross section of the crack and photogrammetric modelling of a ~300m section of the chasm.
Key challenges: The crack propagation trajectory is unknown, although Hilmar can advise on likely scenarios. Some image footage is missing; some of it is due to flat light (no shadow) problems, and also because of the scale (4k video, 25fps, 40km, at multiple drone passes limited to 300m at a time due to battery performance at low temperatures – not currently possible). Extrapolation of known data will be required to model unavailable/unknown aspects of Chasm One.
The single point of contact at BAS will be Matt Polaine, Innovation Manager. He will coordinate your data/technical requests to the key science, imaging and audio staff such as Hilmar Gudmundsson, Andrea Cziferszky and Pete Bucktrout.
The immersive effect will require spatialized audio. Some of this audio may be available from BAS, some may need ‘creation’.
Depending on the course requirements, BAS may be able to assist with the supply of some hardware, but funds are limited. We might consider for example the use of a console such as the PS4pro and VR headset to provide the VR experience. However given the scale of the landscape, considerable LOD modelling/horizon may be required to reduce CPU loading to meet the require fps of dual screens for VR. If the required fps can’t be met for the VR experience in the time available, then this is understood.
A final presentation of the project would be expected at BAS, describing how the project achieved its end result, and what might be possible in the future/with more time.
Summarising the resource side:
1. Student resource – team of 6 working for 7 weeks, starting mid January, about 600 person-hours 2. BAS resource – one person (the “client”) to attend 3 x 1-hour meetings during that period
Regarding future data, it is possible that BAS may have more drone footage taken in November, however this won’t be available until February 2017. BAS would also wish to use the UCam model for presentations at BAS, and there might be an opportunity to use the modelling in a future BBC Horizon documentary about the Halley VI move, but this is currently speculative.
