Skip to 0 minutes and 6 secondsI'm here today with Professor Stuart Marsh. He's the Deputy Director of the Nottingham Geospatial Institute. Stuart, great to have you here. You know, as soon as human beings were able to climb up trees and climb on the tops of hills, they were using that vantage point to look for things and find things. Is that what this technology is about? Yes. It's in the great tradition of searching for the high ground to see the bigger picture. So generals stood on hills and eventually using balloons, aircraft, and now we have satellites in space, which is pretty much the highest view you can get of the planet. So we're using it to see that bigger picture.
Skip to 0 minutes and 45 secondsBut we have the advantage now that we can combine that with very high resolution cameras. So as well as seeing the overview, we can zoom right in down to, say, an individual logging camp or something like that. That's great. Are there ways that satellites or drones or super high aircraft are already being used in human rights or emergency relief? So, yes. There's certainly been some use. And you've seen that, for example, in the conflict in the Balkans in terms of looking for some of the mass graves. And there are examples where these technologies are being used. But what I would say is that there's a lot of room for growth.
Skip to 1 minute and 22 secondsBecause I've been working in the use of satellite technologies for probably 30 years now in the environmental sciences and using them in the built environment and engineering and the interaction between those two worlds. And I would say that their use in this new field is probably a couple of decades behind what we've done in the environmental sciences. So there's a lot of room for growth and improvement in what we can do. Excellent. Now satellites, of course, have advantages and disadvantages. There are things that airplanes can do and drones can do that satellites can't and so forth.
Skip to 1 minute and 58 secondsCan you walk us through a little bit of how those different things fit together if we're applying this to a human rights violation, such as contemporary slavery? Right. So that's definitely the case. If you've got an aircraft or a drone, you can take it to where you want it to be when you want it to be there. And so if you're after a specific event, you've got a lot more control over the timing. A satellite passes over on a regular pattern. So we might get an image of a particular region every three days at ten o'clock in the morning.
Skip to 2 minutes and 29 secondsSo if the abuse is happening every second day at three o'clock in the afternoon and unless it's leaving some trace or marker in the environment, then we're going to struggle to see it. On the other hand, no one can stop us flying over. We have an open skies policy. No one can switch the satellites off. And unlike positioning a navigation type satellite that run your sat nav, you can't jam this very easily either. So the data are being acquired, whereas a country can refuse entry to an aircraft. They can confiscate a drone at customs. So getting the imagery can be harder using these more ground-based technologies, if you like. So the planet, though, is enormous.
Skip to 3 minutes and 17 secondsAnd the amount of area that needs to be covered is vast. And if you do that with human beings looking at images that are taken from space, they get eyestrain and it must be very difficult to look for needles in haystacks. Are there ways that we can teach machines to help us to locate particular images or signatures that would suggest to us that slavery might be existing in a particular place? So this is very good point. And some of the early work we've done here at Nottingham has been using human interpreters and looking for particular things. But what we observe is that you're quite right. There are signatures in the landscape. There are patterns in the imagery.
Skip to 3 minutes and 59 secondsAnd once we know what that pattern is, we can then look for it elsewhere in the imagery. Yeah, sure, using a human interpretive. But some of these patterns are so distinctive that we can programme computers to go look for them in much bigger datasets. So we have a whole area of science now, you could call it data mining. Some people call it big data. Data-driven discovery is what we call it in Nottingham, where we use these computer techniques to go find signatures in vast datasets. And so there is a degree to which you could automate this, at least as a first screening stage.
Skip to 4 minutes and 34 secondsYou probably need a human to come in afterwards and verify, yeah, that site's interesting, no, that one isn't. So I can understand that there are logistical problems, but are there other ways that you have to balance these technologies? Yes, sure. There are also technological issues. If you imagine putting a satellite up, that has a very big price tag. Once it's up there, we can't change the technology that's on it very easily and it doesn't happen. But if we invent some new sensor tomorrow, measure something new that we're really interested in, we could put that on an aircraft or a drone much more easily.
Skip to 5 minutes and 7 secondsSo there is an advantage in being able to update the technology and to tailor the technology for the thing you're trying to measure. There are also issues around the detail of what we can see, the resolution we call it. So we can see more from an aircraft or a drone purely because it's flying closer to the Earth's surface. And so from a satellite we might be getting down to about half a metre. We can go beyond that to centimetres with drones and aircraft. To put that in some context, take the example of the fish processing plant. So we could see the general layout of the plant and the drying racks and that distinctive pattern.
Skip to 5 minutes and 46 secondsWe can see that in the satellite data. But we wouldn't see individuals. If we had a drone fly over those facilities, then we could see the individuals who were working in them and get a much more detailed picture of what was going on on the ground. And we know, of course, from anti-slavery workers on the ground, that those are actually enslaved children. Which is a very powerful thing to be able to identify when very often that crime is being hidden away.
Skip to 6 minutes and 12 secondsSo our challenge becomes, how do we take the imagery that comes down from the technology, down from the satellites, possibly from the airplanes or drones, and use those images as a conduit to lead us to the point that we can achieve liberation for those children who are enslaved processing fish or whatever it is they're doing? Thank you so much for being with us. And we'll be exploring those notions of liberation later in the course.
Eyes in the Sky
Professor Stuart Marsh is a leading professor of Geospatial Engineering at the University of Nottingham, whose Nottingham Geospatial Institute has begun using satellite technology to track sites of slavery around the world.
In this film, Stuart and Kevin Bales discuss the geospatial imaging of slavery hotspots in particular countries and regions as a solution to ending slavery. Earth observation data offers spatial resolutions as close as one metre and so can evidence gatherings and movements of people, destruction of sites, and the appearance of mass graves or illegal surface mines. Radar data can see through cloud cover and penetrate several metres through sand. Thermal data can detect features in the subsurface, such as hidden excavations.
After watching the film, please read about the work of the Associated Press in using satellite data to track slave boats and ultimately help to free thousands of enslaved fisherman. Three articles from their investigative series Seafood from Slaves, which won the 2016 Pulitzer Prize for Public Service, are: “AP tracks slave boats to Papua New Guinea”, “AP investigation prompts new round of slave rescues” and “More than 2,000 enslaved fishermen rescued in 6 months”.
Tell us in the comments about your reactions to this use of technology, and if you can think of other technologies that we should be harnessing to combat slavery. For example, several antislavery groups are exploring the use of drone technology to monitor suspected sites of slavery. Others are working to develop software that will open up areas of the ‘dark web’ - the intentionally anonymous Internet hidden from search engines - where criminal networks conduct slavery and trafficking business. Still others are creating a version of ‘Yelp’ for migrant workers to review recruiters and employers, in order to shine a light on bad practices so that workers can avoid being abused or trafficked. What other technologies and platforms might help?